Condensed Matter Nuclear Science: Proceedings of the 11th International Conference on Cold Fusion: Marseilles, France, 31 October-5 November 2004

condensed mlatter nluclear slcience proceedings of the I I th international conference on cold fusion editor jean-paul ...

Author: Jean-paul Biberian

22 downloads 749 Views 59MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

condensed mlatter nluclear slcience proceedings of the I I th international conference on cold fusion

editor jean-paul biberian

clondensed mlatter n uclear cience proceedings of the I Ith international conference on cold fusion

This page is intentionally left blank

c ondensed m after n uclear

cience proceedings of the I Ith international conference on cold fusion

Marseilles, France 31 October-5 November 2004

Editor

Jean-Paul Biberian Universitede la Mediterranee, France

YJ? World Scientific N E W J E R S E Y • L O N D O N • S I N G A P O R E • B E I J I N G • S H A N G H A I • H O N G KONG • T A I P E I •

CHENNAI

Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

CONDENSED MATTER NUCLEAR SCIENCE Proceedings of the Eleventh International Conference on Cold Fusion Copyright © 2006 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

ISBN 981-256-640-6

Printed in Singapore by World Scientific Printers (S) Pte Ltd

FOREWORD Background It has been a great honor for me to organize I C C F l l in Marseille, France, my hometown. During the ICCF10 in Cambridge, it occurred to me t h a t Marseille was an ideal choice for the I C C F l l . T h e field had matured, and it was obvious t h a t the scientific demonstration of Cold Fusion had been made, and so I proposed to organize the I C C F l l . Certainly, a lot more is needed to be accomplished in the field of research and technology, but we had answers to many of the questions of the scientific community. We knew then for sure t h a t the phenomenon announced in 1989 by Professors Martin Fleischmann and Stan Pons was real. Moreover, they had not foreseen all the discoveries t h a t have been made since their announcement, in particular, the discovery t h a t hydrogen, not only deuterium may be nuclear active under certain conditions. Although much work needs t o be done, it had been shown t h a t transmutations of elements were occurring, indicating t h a t the simple D + D producing helium reaction was not the only reaction channel. It occurred to me at t h a t time t h a t more t h a n science and technology; we had to focus on public relations. My goal in proposing to organize I C C F l l was to bring Cold Fusion to the attention of b o t h the scientists and to the ordinary citizens. There are many things t h a t we do not know a b o u t our research, but we are sure of one thing; it is not only about fusion. We have observed fission and t r a n s m u t a t i o n beyond doubts but there are probably more reactions t h a n we currently know. T h e name of the conference is now International Conference on Condensed Matter Nuclear Science. For the sake of simplicity and continuity, we have decided to keep the old acronym I C C F . In order t o meet our goal of creating awareness a b o u t this, we had the great honor to have Brian Josephson come to the conference. He is the Nobel Prize winner in 1973 in physics for the discovery of the effect t h a t bears his name. I would like to use this opportunity to thank him for his support, as he took the risk by associating his name to a controversial topic. He came to the conference and lectured on "Good and Bad ways of doing Science." It was a great pleasure for me to meet him. He added a lot of weight to the conference. The city of Marseilles under the patronage of the Societe Frangaise de Physique (French Physical Society) granted him and Martin Fleischmann the medal of the city during a ceremony at the City Hall. One full day of the conference was held at the Faculte des Sciences de Luminy, the University where I teach and do research, so t h a t my peers at the university could attend. This has been an opportunity for students and faculty members to come and listen to other scientists t h a n myself on the subject. A demonstration unit was even displayed in the hall of the cafeteria, where many students could see for themselves the reality of the phenomenon.

vi

As we now known, the international thermonuclear experimental fusion reactor (ITER) will be built in Cadarache, which is less t h a n 1 h by car from the conference location. Support for I T E R was very popular in France at the time of the conference and it was very difficult to get the local press to talk about alternative energies. T h e television stations refused to cover the event, although 170 people from 20 different countries had come to the conference, and the keynote speaker was Brian Josephson, the Nobel Prize Laureate. This goes to show how science and politics are mixed up. Nevertheless, a couple of short papers were published in the newspapers, and a few months later a national business newspaper "Les Echos" published a half page article on the conference and CMNS. T h e conference was held at the hotel Mercure. This was a posteriori a good choice because it means mercury in French, and is one of the key ingredients used by the alchemists in the past to make gold! T h e Conference Several important new results were presented during the conference. The joint US/Israeli team, headed by MD Irving Dardik, confirmed t h a t the superwaves they use in their electrolytic experiments help in producing more heat. Also Iwamura et al. showed new t r a n s m u t a t i o n effects in their experiments of diffusion of deuterium gas through a complex structure of palladium and calcium oxide. In addition to the traditional Cold Fusion community, a t e a m of Russian scientists claimed t h a t their experiments show the existence of light monopoles. T h e theory was developed by Lochak from France. They t r y to explain the Chernobyl nuclear accident by the interaction of the monopoles with uranium nuclei, changing the half-life of the nucleus. A German team, comprising of Czerski and Huke, who were working in high-energy physics, discovered CMNS when they lowered the energy of the deuterium beam. They demonstrated t h a t the cross section of the deuterium with deuterated metals was much higher t h a n expected. To explain their experimental d a t a they needed to add a large screening potential. T h e y came to the conclusion t h a t they were doing Cold Fusion, and for the first time attended to the conference. Another important contribution was the one from the Vysotskii t e a m from Ukraine, who confirmed their biological t r a n s m u t a t i o n experiments. Certainly there is a lot more to be discovered. This is very exciting news for science, mankind, and us as scientists. O n t h e theory front, I must confess t h a t there appears t o be far too many. T h e initial idea of the necessity of high deuterium loading in metals to obtain the effect seems to be relevant only for cathodically loaded P d wires. The fact t h a t hydrogen is also active, and t h a t in some cases the loading is obviously low indicates t h a t something else is happening. Storms mentions "active sites," but what are they? Can we use classical q u a n t u m mechanics or q u a n t u m electro dynamics? Do we need poly-neutrons, neutron band structures, or magnetic monopoles? Nobody knows for sure, but every theory developer is convinced t h a t he/she is on the right p a t h to obtain the solution.

vii

The Proceedings When accepting papers for the conference, we decided to be open and to avoid filtering. This is in reaction to the attitude of the scientific community, in its large majority, regarding CMNS. If we publish everything, all kinds of foolish and false ideas can be put forward, but if we are too narrow in our choices, great ideas can be lost. By opening up, we took the position that everyone is capable of deciding for oneself what is good and bad science. We did not want to have a committee to decide and thereby take the risk of missing a great opportunity. These proceedings follow the same philosophy, and therefore the reader must use his/her own understanding to judge the quality of the works presented here. As the editor of this book, I take full responsibility for this choice, and I hope that the future will prove me more right than wrong. Peter Hagelstein from MIT, who organized ICCF10, was a great support in helping me with the quality of the proceedings. He found a company that reformatted all the papers that I had received in numerous formats. This was a precious help and I believe the readers will appreciate it. The Future The ICCF12, as has been decided, will be held in Japan, and the following one in Russia. This is good news because both the countries are very active in the field. After ICCF10, the International Society for Condensed Matter Nuclear Science was created, and helped us to organize the conference. In the future, this society is likely to play a major role in organizing international events. My feeling is that we have now entered a new era, and that the various effects of CMNS will revolutionize science and technology in the near future. Acknowledgements First of all, I would like to thank Jed Rothwell who has put in tremendous work by improving the quality of the papers. English has become the international language for science. However, more than two-thirds of the contributions were made by people coming from non-English-speaking countries. Jed read them all and did his best to understand their contents. Especially, difficult papers were theoretical papers. All my thanks to Vittorio Violante as well from ENEA Frascatti, who co-chaired the conference. I am also grateful to the scientific committee who trusted me and helped me in making decisions. This is also a good place to recognize the role of the sponsors: Infinite Energy, the City of Marseilles, the Departement des Bouches du Rhone, the University who financially helped and made this conference a success. Finally, I am very proud of the help of my two elder children, Melanie and Gabriel without whom this conference would have been a disaster. They worked day and night to the satisfaction of all the attendees. Dr. Jean-Paul Siberian, Chairman, ICCF11, Marseilles, October 2005

•

11 «

IBS

C1 "XJfl -ML

£ ^hnfl w

« * .

!

k

CONTENTS Foreword

v 1. G E N E R A L

A tribute to gene Mallove - the "Genie" reactor K. Wallace and R. Stringham

1

An update of LENR for ICCF-11 (short course, 10/31/04) E. Storms

11

New physical effects in metal deuterides P. L. Hagelstein, M. C. H. McKubre, D. J. Nagel, T. A. Chubb, and R. J. Hekman

23

Reproducibility, controllability, and optimization of LENR experiments D. J. Nagel

60

2. E X P E R I M E N T S Electrochemistry Evidence of electromagnetic radiation from Ni-H systems S. Focardi, V. Gabbani, V. Montalbano, F. Piantelli, and S. Veronesi

70

Superwave reality 7". Dardik

81

Excess heat in electrolysis experiments at energetics technologies I. Dardik, T. Zilov, H. Branover, A. El-Boher, E. Greenspan, B. Khachaturov, V. Krakov, S. Lesin, and M. Tsirlin

84

"Excess heat" during electrolysis in platinum/I^COa/Nickel light water system J. Tian, L. H. Jin, Z. K. Weng, B. Song, X. L. Zhao, Z. J. Xiao, G. Chen, and B. Q. Du

102

Innovative procedure for the, in situ, measurement of the resistive thermal coefficient of H(D)/Pd during electrolysis; cross-comparison of new elements detected in the Th-Hg-Pd-D(H) electrolytic cells F. Celani, A. Spallone, E. Right, G. Trenta, C. Catena, G. DAgostaro, P. Quercia, V. Andreassi, P. Marini, V. Di Stefano, M. Nakamura, A. Mancini, P. G. Sona, F. Fontana, L. Gamberale, D. Garbelli, E. Celia, F. Falcioni, M. Marchesini, E. Novaro, and U. Mastromatteo

108

Emergence of a high-temperature superconductivity in hydrogen cycled Pd compounds as an evidence for superstoihiometric H/D sites A. Lipson, C. Castano, G. Miley, B. Lyakhov, and A. Mitin Plasma electrolysis Calorimetry of energy-efficient glow discharge - apparatus design and calibration T. B. Benson and T. 0. Passell Generation of heat and products during plasma electrolysis T. Mizuno, Y. Aoki, D. Y. Chung, and F. Sesftel Glow discharge Excess heat production in P d / D during periodic pulse discharge current in various conditions A. B. Karabut Beam experiments Accelerator experiments and theoretical models for the electron screening effect in metallic environments A. Huke, K. Czerski, and P. Heide Evidence for a target-material dependence of the neutron-proton branching ratio in d + d reactions for deuteron energies below 20keV A. Huke, K. Czerski, T. Dorsch, and P. Heide

128

147

161

178

194

210

Experiments on condensed matter nuclear events in Kobe University T. Minari, R. Nishio, A. Taniike, Y. Furuyama, and A. Kitamura

218

Electron screening constraints for the cold fusion K. Czerski, P. Heide, and A. Huke

228

Cavitation Low mass 1.6 MHz sonofusion reactor R. Stringham Particle detection Research into characteristics of X-ray emission laser beams from solidstate cathode medium of high-current glow discharge A. B. Karabut Charged particles from Ti and Pd foils L. Kowalski, S. E. Jones, D. Letts, and D. Cravens

238

253

269

Cr-39 track detectors in cold fusion experiments: Review and perspectives A. S. Roussetski

274

Energetic particle shower in the vapor from electrolysis R. A. Oriani and J. C. Fisher

281

Nuclear reactions produced in an operating electrolysis cell R. A. Oriani and J. C. Fisher

295

Evidence of microscopic ball lightning in cold fusion experiments E. H. Lewis

304

Neutron emission from D2 gas in magnetic fields under low temperature T. Mizuno, T. Akimoto, A. Takahashi, and F. Celani

312

Energetic charged particle emission from hydrogen-loaded Pd and Ti cathodes and its enhancement by He-4 implantation A. G. Lipson, G. H. Miley, B. F. Lyakhov, and A. S. Roussetski

324

H-D Permeation Observation of nuclear transmutation reactions induced by D2 gas permeation through Pd complexes Y. Iwamura, T. Itoh, M. Sakano, N. Yamazaki, S. Kuribayashi, Y. Terada, T. Ishikawa, and J. Kasagi

339

Deuterium (hydrogen) flux permeating through palladium and condensed matter nuclear science Q. M. Wei, B. Liu, Y. X. Mo, X. Z. Li, S. X. Zheng, D. X. Cao, X. M. Wang, and J. Tian Triggering Precursors and the fusion reactions in polarized P d / D - D 2 0 system: effect of an external electric field S. Szpak, P. A. Mosier-Boss, and F. E. Gordon Calorimetric and neutron diagnostics of liquids during laser irradiation Yu. N. Bazhutov, S. Yu. Bazhutova, V. V. Nekrasov, A. P. Dyad'kin, and V. F. Sharkov

351

359

374

Anomalous neutron capture and plastic deformation of Cu and Pd cathodes during electrolysis in a weak thermalized neutron field: Evidence of nuclei-lattice exchange A. G. Lipson and G. H. Miley H-D Loading An overview of experimental studies on H/Pd over-loading with thin Pd wires and different electrolytic solutions A. Spallone, F. Celani, P. Marini, and V. Di Stefano

379

392

3. T R A N S M U T A T I O N S Photon and particle emission, heat production, and surface transformation in Ni-H system E. Campari, G. Fasano, S. Focardi, G. Lorusso, V. Gabbani, V. Montalbano, F. Piantelli, C. Stanghini, and S. Veronesi

405

Surface analysis of hydrogen-loaded nickel alloys E. Campari, S. Focardi, V. Gabbani, V. Montalbano, F. Piantelli, and S. Veronesi

414

Low-energy nuclear reactions and the leptonic monopole G. Lochak and L. Urutskoev

421

Results of analysis of Ti foil after glow discharge with deuterium I. B. Savvatimova and D. V. Gavritenkov

438

Enhancement mechanisms of low-energy nuclear reactions F. A. Gareev, I. E. Zhidkova, and Y. L. Ratis

459

Co-deposition of palladium with hydrogen isotopes J. Dash and A. Ambadkar

477

Variation of the concentration of isotopes copper and zinc in human plasmas of patients affected by cancer A. Triassi Transmutation of metal at low energy in a confined plasma in water D. Cirillo and V. Iorio The conditions and realization of self-similar Coulomb collapse of condensed target and low-energy laboratory nucleosynthesis S. V. Adamenko and V. I. Vysotskii

485

492

505

The spatial structure of water and the problem of controlled low-energy nuclear reactions in water matrix V. I. Vysotskii and A, A. Kornilova Experiments on controlled decontamination of water mixture of longlived active isotopes in biological cells V. I. Vysotskii, A. Odintsov, V. N. Pavlovich, A. B. Tashirev, and A. A. Kornilova

521

530

Assessment of the biological effects of "Strange" radiation E. A. Pryakhin, G. A. Tryapitsina, L. I. Urutskoyev, and A. V. Akleyev

537

Possible nuclear transmutation of nitrogen in the earth's atmosphere M. Fukuhara

546

Evidences on the occurrence of LENR-type processes in alchemical transmutations J. Perez-Pariente History of the discovery of transmutation at Texas A&M University J. O.-M. Bockris

554

562

4. THEORY Quantum electrodynamics Concerning the modeling of systems in terms of quantum electro dynamics: The special case of "Cold Fusion" M. Abyaneh, M. Fleischman, E. Del Giudice, and G. Vitiello Screening Theoretical model of the probability of fusion between deuterons within deformed lattices with microcracks at room temperature F. Fulvio Resonant tunnelling Effective interaction potential in the deuterium plasma and multiple resonance scattering T. Toimela Multiple scattering theory and condensed matter nuclear science"super-absorption" in a crystal latice X. Z. Li, B. Liu, Q. M. Wei, N. N. Cai, S. Chen, S. X. Zheng, and D. X. Cao

587

612

622

635

Ion band states Framework for understanding LENR processes, using conventional condensed matter physics S. R. Chubb

646

I. Bloch ions T. A. Chubb

665

II. Inhibited diffusion driven surface transmutations T. A. Chubb

678

III. Bloch nuclides, Iwamura transmutations, and Oriani showers T. A. Chubb

685

Bose-Einstein condensate Theoretical study of nuclear reactions induced by Bose-Einstein condensation in Pd K.-I. Tsuchiya and H. Okumura Proposal for new experimental tests of the Bose-Einstein condensation mechanism for low-energy nuclear reaction and transmutation processes in deuterium loaded micro- and nano-scale cavities Y. E. Kim, D. S. Koltick, R. G. Reifenberger, and A. I. Zubarev Mixtures of charged bosons confined in harmonic traps and BoseEinstein condensation mechanism for low-energy nuclear reactions and transmutation processes in condensed matters Y. E. Kim and A. L. Zubarev

694

703

711

Alternative interpretation of low-energy nuclear reaction processes with deuterated metals based on the Bose-Einstein condensation mechanism Y. E. Kim and T. O. Passell

718

Multi-body fusion 3 He/ 4 He Production ratios by tetrahedral symmetric condensation A. Takahashi

730

Phonon coupling Phonon-exchange models: Some new results P. L. Hagelstein

743

Neutron clusters Cold fusion phenomenon and solid state nuclear physics H. Kozima

769

Neutrinos, magnetic monopoles Neutrino-driven nuclear reactions of cold fusion and transmutation V. Filimonov

776

Light monopoles theory: An overview of their effects in physics, chemistry, biology, and nuclear science (weak interactions) G. Lochak

787

Electrons clusters and magnetic monopoles M. Rambaut

798

Others Effects of atomic electrons on nuclear stability and radioactive decay D. V. Filippov, L. I. Urutskoev, and A. A. Rukhadze

806

Search for erzion nuclear catalysis chains from cosmic ray erzions stopping in organic scintillator Yu. N. Bazhutov and E. V. Pletnikov

818

Low-energy nuclear reactions resulting as picometer interactions with similarity to K-shell electron capture H. Hora, G. H. Miley, X. Z. Li, J. C. Kelly, and F. Osman

822

5. OTHER TOPICS On the possible magnetic mechanism of shortening the runaway of RBMK-1000 reactor at Chernobyl Nuclear Power Plant D. V. Filippov, L. I. Urutskoev, G. Lochak, and A. A. Rukhadze

838

Cold fusion in the context of a scientific revolution in physics: History and economic ramifications E. Lewis

854

The nucleovoltaic cell D. D. Moon

868

Introducing the book "Cold Fusion and the Future" J. Rothwell

871

xvi

Recent cold fusion claims: Are they valid? L. Kowalski

879

History of attempts to publish a paper L. Kowalski

888

Author Index

895

A T R I B U T E TO G E N E MALLOVE - THE "GENIE" R E A C T O R

K. W A L L A C E A N D R. S T R I N G H A M PO Box 1230, Kilauea,

HI 96754,

USA

"Genie", a 40kHz sonofusion reactor consists of two opposing 40kHz piezos separated by 4 mm of D2O, with a centered Ti target foil, with one piezo transmitting, the other receiving and taking that signal, amplifying it, then feeding it back to the transmitter as the resonating frequency of the reactor. This process makes for efficient watt input, Q{, where 80% of these watts will be used as the acoustic input, Q a , to the "Genie" sonofusion reactor. In the reactor the transient cavitation bubbles, TCBs, produce billions of low-energy high-density jets per second that accelerate deuterons into foil targets producing excess heat, Qx. The Qx is determined by calorimetric measurements of experiments that use coolant water circulated to the surface of the well insulated reactor and data collected in the form of T-ln and Tout at steady-state temperatures and coolant flow rate. The total watts out, Q0, minus Q a ideally should equal zero, and we know that this calorimetry method has several losses that are not measured. This makes the method very conservative when looking for Qx. The Qx must make up those heat losses before making its presence known. The result from experiments of system I using flow x DT x 4.184 for Q0 — Q\ = Qx shows that Qx values over unity are the norm. System II used a more realistic calculation for Qx, where flow X DT X 4.184 for Qo — ô6T-(t-Hr-H

© o o

1 1 1

O

O

i

O

O

O

O

o LO o o o LO

CO

i

1 O

O

O

o o CO 1 1 1

LO

O

O

O

i

O

O

O

T"H

CN

o o p o o o o o o o p i o o o o o o o p o o o o o o o o o o o •a M o O X h ôo^Qdô6d^c«dH«««6oicowdHioTJ CO

^

C O C O C O I O ^ I O C M C O C O C O C O C O ^ C M C O C O C O C O C D C O C ^

.

^

^

.

^

^

^

^

^

.

^

O

Q™Q Q P Q Q Q P O

P

oCMoCN O

O

O

oCMoCM O Q

O

oCMoCM O

C N O O O

affi° ° ° Kffi ? Effi° W ° °

Q

oIM O

CN O

K ? EQ

cflcnmtncninracocn s o - T i t B o i t o h i - i i n n b ' i f f l n t o i / n B h j a n r j ]

H C N C O ^ i O ! O N 0 0 O l

7

q q

q q

06 *o

10 »o

q q

o

^f

CO

CO

o

q o O C C o q o q C ^ C < ] C N C N C N l ( N C N C N C N C N C N l C N ( N O O O O O O O O O O O O O O O

q q CO

CD

CO

CO

i—(

t—

i—(

i-H

CO

CO

^

10

K i-i

00 »-H

Ol <M

CO

CO

CO

qioioocsiNoio; doiiocodr-lNoi

l O H I N N ^ C S Q O Q O ^ l N C O I N

CO Ol lO H

CO CO CO Oi i-H O O O C O C O O O T T c O O O O i O C O i O O O i O i O IV 00 1^ O i-H r-H r H r ^ W O O O O O O O O i - H i - H i - 5

CO

^J •^ CO Ol rH I> CO 00 T-H

N N O I O O O C O C O C C O O C S C v i ^ - H O i - H i - H T - H i - H

i O t 0 i O 0 1 N ^ © q N C 0 W H ^ 1 0 1 0 1 f f i 0 ) 0 ) O i N N O O H H H ( N O C 0 0 0

ooococNcocsco'cocococococsoicqc^c^cocococo

O O O O O i O i O i O O O i O i O i C O C O ^ C O O T C O ^ O O t - H r H O O O O

I

I

I

I

I

I

I

I

I

I

I

I

lO O O iO O lO O O H (N H I I I I I

I

o o o o io io o O

>-H

rH

lO

CO

CN

i-H

I

I

I

I

I

I

O O O i C L O i O O O i O i O m o I ^ O i O i O O O O i O i O O O O O O i O O i O i O O r H ^ c H o O O O C ^ ^ O O C O C ^ O O i O C O O (M 4 He. • Helium is partially retained, and dissolved helium is released only slowly to the gas phase for analysis. Recent work at ENEA Frascati also supports these conclusions. 72 ' 73 Excess heat and helium observations consistent with a D + D —> 4 He + 23.8 MeV (heat) reaction mechanism, in a metal deuteride near room temperature, stands in stark contrast to the d(d,7) 4 He reaction known from nuclear physics.

0

1

2

3 4 5 6 Helium increase (ppmV/V)

7

8, I

Figure 6. Excess energy determined by gradient (boxes) and differential (diamonds) calorimetric methods plotted against the increase in 4 He concentration in a metal-sealed helium leak-tight vessel. The experiment was performed by heating palladium on carbon hydrogenation catalyst materials to ~190°C in ~3 atmospheres of D2 gas pressure (see Appendix B).

3.2. Reaction

Q Value

As the loss of deuterium in association with excess heat is not presently observable, and since there are no commensurate energetic reaction products, the argument in support of reaction mechanisms consistent with D + D —> 4 He is indirect. One can measure energy production, and assay for 4 He in the gas stream or the solid, with uncertainties introduced in the reaction energy Q because all of the helium produced may not be accounted for in the measurement. Experiments are prefered in which a total inventory of the helium is made in order to improve the accuracy

33

of the reaction Q value measurement. To this end, we discuss briefly an experiment in which helium was measured in the gas stream, and an additional effort was made to drive the helium out of the metal. The experiment under consideration was performed at SRI, and the excess heat measured is illustrated in Fig. 5. The experiment was performed in a helium leaktight, all-metal and metal gasketed calorimeter. Samples were transferred in metal gas sample flasks to be analyzed for 4 He by the U.S. Bureau of Mines at Amarillo, Texas. 70 The initial value of 4 He was 0.34 ± 0.007 ppmV/V in the D 2 gas used to charge the cell. Figure 7 traces the history of the cell, M4, from four helium samples taken after excess power was observed. The upper solid line is the expectation for helium concentration presuming: (i) an initial value of 0.34ppmV/V; (ii) that 4 He is produced in a reaction which delivers 23.8 MeV of thermal energy to the calorimeter. The first gas sample taken shortly following the second heat burst of Fig. 5 yielded a value of 1.556 ± 0.007ppmV/V 4 He, which is about 62% of its expected value, and consistent with the earlier observations by Miles, Bush and collaborators, 55 and also Gozzi and coworkers.71 A second sample taken about 6 days after the first showed a measurable increase in 4 He content instead of the decrease that would be expected since, to maintain positive cell pressure, the gas taken for the first sample had been replaced with cylinder D 2 containing a lower level of 4 He (0.34 ppm V/V). These findings support earlier observations that helium is released slowly from the palladium after an initial delay.

• [He] ppm' V/V End

He release

Value expected for 23.8 MeV/He

i

2.5

Q Sampled values

Sample 4

1,

or " CD to 1 /

52.0 E

»

CL CL

d.5

V

_„2.07 077 1H±0. 01

Sample 2 and D 2 make-up

Cathode heating and composition cycling

1 1.661 ±009 1.556 ± 007 * T62%~

M• 69%

Sample 1 and D„ make-up

" 1.0

D„ purge and Sample 3

/

• V

a>0.5

I

Background value of [4He] gf 0.34 ± 0.077

Initial value in D„ 0.0> 450

Jw°/
4 He + ~ 23.8 MeV (heat). This value remains the most accurately determined in this field (in the sense that contributions from both the gas stream and the metal are included), but it suffers from the criticisms that the numbers of samples were few, and the largest value of 4 He measured was less than 50% of that in air. We note that 4 He has been produced numerous times in excess heat experiments at levels above that of the concentration in air. One example is shown in Fig. 6. This plot illustrates the real-time correlation between excess heat and the growth of 4 He concentration in a metal-sealed, helium leak-tight vessel. The Q value of 31 ± 13 and 32 ± 13 MeV per 4 He atom measured is also consistent with the reaction D + D —> 4 H e + ~ 23.8 MeV (heat). Because of the importance of this result, it is discussed further in Appendix B. 4

4. Excess Heat Beyond the Basic Fleischmann—Pons Experiment The importance of the basic Fleischmann-Pons experiment, as discussed in the previous section, is primarily scientific in the sense that the research provides strong evidence of a new excess heat effect of nuclear origin. In addition, this basic experiment has provided a focus for a significant research effort, and many important aspects of the experiment have been studied, as discussed above. During the past 15 years, numerous variants of this basic experiment have been proposed and executed. In what follows, we examine a subset of such studies. 4.1. Self-Sustaining

Excess Heat

Effect

Excess heat production in the absence of electrical input was first discussed by Fleischmann and Pons at ICCF4. 74,75 It was found that cathodes continued to

35

produce excess heat in some cases after the termination of the electrochemistry in which heat had been produced earlier during "normal" cell operation. This effect has been reported in other experiments, 22 ' 76 ' 78 ' 79 and is often referred to as "heat after death." These observations suggest a possible difference in the conditions required for the initiation of an excess heat effect as compared to the conditions required for sustaining the effect.

4.2. Excess Heat in Other Metal

Deuterides

Excess heat in electrochemical experiments involving metal deuterides other than PdD has been reported in several modified Fleischmann-Pons experiments. Helium production in association with excess heat was studied in palladium alloys by Miles.80 Excess heat has been reported in TiD with a D2O-H2SO4 electrolyte using thermal imaging. 81 Excess heat and other effects were reported from an electrochemical cell with a Ti cathode by Bernardini and coworkers.76 Excess heat observations with Pt cathodes were reported by Dash 52 and by Storms. 82 Platinum does not form a deuteride, and the excess heat in this case is attributed to a surface layer.82

4.3. Indirect

Gas

Loading

Arata and Zhang introduced an interesting experiment in which palladium black is sealed within the interior of a palladium cathode, and the resulting composite (hollow) cathodes structure is electrolyzed.83 Deuterium gas within the interior of the cathode is in equilibrium with the deuterium in the highly loaded cathode, and the palladium black is loaded indirectly. Evidence for excess heat and helium production has been obtained from this kind of experiment. We discuss this further in Appendix C.

5. Nuclear Emissions The first report of emissions consistent with deuteron-deuteron fusion was made by Jones and coworkers in 1989,2 who claimed that low-level neutrons near 2.45 MeV were emitted from electrochemically loaded titanium deuteride. j Since that time, there have been numerous reports of neutron and charged particle emission generally at low emission rates from metal deuterides, which are consistent with a deuterondeuteron fusion reaction mechanism. In what follows, we will discuss some of this work. We note that there is in addition evidence in support of emissions that are clearly not associated with a deuteron-deuteron fusion reaction mechanism. A subset of these observations will be discussed below.

J

Additionally, present on the titanium surface were Pd, Cu, and Li due to deposition from the electrolyte.

36

5.1. The Jones

Experiment

Much effort has been devoted since 1989 to the Jones effect by Jones and coworkers, 84,85 and also by other groups. 3 2 ' 3 3 ' 8 6 - 9 0 Most of the effort has focused on neutron measurements, but there have been observations of charged particles as well.85 Low-level nuclear emissions typically occur in bursts lasting a few seconds to several days, 84 and at lower current density than that discussed above in connection with the excess heat effect (Jones and coworkers reported neutron emission at 20mA/cm 2 ). A distinct burst effect (see Fig. 8) was reported by Wolf in neutron emission from PdD electrochemical experiments. 91 This is reminiscent of a similar burst effect noted for the excess heat in the Fleischmann-Pons experiment. It was noted in the early 1990s that there were considerable similarities between the electrochemical protocols and palladium cathodes used by Wolf in neutron emission experiments, and those used at SRI in excess heat experiments. In discussions between the two groups, it became apparent that the most significant difference between the two experiments was the current density involved. At SRI, excess heat was observed at elevated current densities, with a current threshold in the range of 200-300 mA/cm 2 . At Texas A&M, Wolf increased the current to about 30mA/cm 2 to initiate neutron emission, but found no neutron emission at significantly higher current densities. 92 The low-high current protocol of Takahashi discussed above scans through both regimes, and is associated with both excess heat and neutron 28

Counts our

emission

.c

130 120 110 100 90 80 70 60 50 40 30 20 10 0

-

•^ •

6 mm cells 3-30-90

H

Neutrons open = 1 - 2 . 6 MeV closed = 1 - 5 . 5 MeV

• -

•

f'i

o

Curent density -

:

„ _ J _ _ _ _ _ _ 4 _ _ _ _

10

1

30

20

,

U _

40

50

Time (h)

Figure 8. Neutron burst from PdD loaded electrolytically from the experiments of Wolf and coworkers.

37

5 . 2 . Stimulation

of Nuclear

Emissions

with

Electrical

Current

Cecil and coworkers reported charged particle emission in experiments where current was run through TiD^ [using Ti(662) alloy]. 9 3 For example, a proton signal centered at 2.45 MeV was presented. 15 Jones and coworkers have recently used this same general approach. 9 4 , 9 0 T h e group evacuated a heated chamber containing 25-250 /mi thick Ti foils, introduced 1 atmosphere of D2 gas and partially loaded the foils, which were then placed in front of a particle detector, or in a neutron counter. A low background nuclear detector system was used. Electrical current was run through the foils during nuclear counting. Background subtracted counts were observed for neutrons at typical rates of 40 counts per hour. Background subtracted counts were observed for charged particles identified as < 3 MeV protons at typical rates of 100 counts per hour (and up to 2000 counts per hour), with emission rates varying strongly over t h e course of a few hour burst event. Results from a particularly strong burst are shown in Fig. 9. T h e signal is centered at 2.4 MeV and the particle identity is determined from a comparison with a similar spectrum obtained with a thin Al absorber foil. 94 Background subtracted coincidences between two particles, consistent with protons and tritium from deuteron-deuteron fusion, occur at a rate of one count per hour in foils with no electrical current running. In this case, there is an additional experimental procedure of LiOD and D2SO4 surface t r e a t m e n t and rinsing prior to nuclear counting. In all cases in the recent Jones experiments, excess nuclear counts diminish with time, positive signals are observed greater t h a n 50% of the time, and no excess nuclear counts are observed when H2 is substituted for D2.

5 . 3 . Neutrons

from

Ti Shards

in Deuterium

Gas

3

Neutron emission consistent with d(d,n) He reactions were reported in 1989 from gas-loaded metal deuteride shards with thermal cycling. 9 5 In these experiments the metal is loaded at high pressure and cycled between liquid nitrogen t e m p e r a t u r e and room t e m p e r a t u r e . Neutron emissions appear occasionally in burst e p s i s o d e s . 9 6 - 1 0 0 High efficiency neutron detectors were employed, 9 8 and some of the measurements were performed under conditions where the background was quite low. 9 6 ' 1 0 1 Research along these lines has not continued in recent years. T h e evidence in support of neutron emission from this kind of experiment is strong, but in such experiments the conditions are not well controlled. This, combined with a low associated emission frequency, made other approaches seem more attractive. 9 6

5.4. Relation

between

Neutron

Emission

and Excess

Heat

Effect

We note t h a t the amount of energy associated with these low-level emissions is not observable calorimetrically. W h e t h e r there is a correlation between nuclear emission effects and excess heat or not has been under discussion since 1989. 1 0 3 k

T h e proton from the d(d,p)t reaction initially has 3 MeV, and perhaps some energy degradation occurs if the proton is born inside the metal deuteride.

38

70'

• Plastic 0 Glass

60 50.

4a o

" 3a 2010-

-jlxi

-*t 11

16

21

26

31

36

41

46

JL 51

56

61

UL •N1* 66 71 J

Light output Figure 9. Proton spectrum near 2.4 MeV from T i D x taken over 21 min using a double scintillation detector, as presented by Jones at ICCF10. The two different signals are from two different scintillators (plastic and glass). The light output scale is related to the proton energy.

There have been several observations of excess heat with simultaneous diagnostics for neutron emission, and for x-ray and gamma emission. 5 ' 71 ' 104 An anticorrelation between excess heat and neutrons was reported by Okamoto and coworkers.105 We have noted that excess heat production in the Fleischmann-Pons experiment has an associated current threshold around 200-300 mA/cm 2 , while neutron emission is associated with lower current densities (around 30 mA/cm 2 in the experiments of Wolf, Jones, and coworkers cited above), suggesting that the different effects have different operating regimes. In experiments that operate with both high and low-current densities, there are reports of both neutron emission and excess heat production. 106 ' 107 There is a study with the low-high current protocol in which a correlation is reported where cathodes producing the largest excess heat effect also show the largest neutron emission.108 5.5. Nuclear Fusion

Emissions

not Attributable

to

Deuteron-Deuteron

There are several reports for weak nuclear emissions, which clearly are not associated with a deuteron-deuteron fusion mechanism. In 1990, Cecil and coworkers noted the presence of very energetic charged particles from thin TiD foils up to and beyond 10MeV 93 (see also the PhD thesis of Liu 109 ). Energetic alphas around 15 MeV have been observed by Lipson and coworkers from PdD. 1 1 0 Both alphas and protons near 14 MeV have been seen in experiments with TiD. 111 The appearance of such signals under conditions that are similar to those associated with excess heat

39

production and low-level deuteron-deuteron fusion is significant, and provides additional information that may be helpful in understanding the underlying physical processes responsible for the new effects. 5.6. Broad Proton

and Alpha Spectrum

from Deuterons

on

TiDx

Kasagi and coworkers have reported anomalous results from beam experiments using TiDj, targets in which product nuclei are observed with a very large energy spread. 112 This group bombarded TiD^ (a; > 1.2) foils with 90-150 keV deuterons and observed protons with energies from 6 to 17MeV and alphas with energies from 4.5 to 6.5 MeV at scattering angles between 135° and 155° at a rate of 10~ 6 of the deuteron-deuteron fusion rate. An aluminum absorber foil in front of the particle detector stopped the elastically scattered beam. The energy spectra for the protons and alphas are consistent with a three-body reaction mechanism d + d + d ->• 4 He + p + n,

Q = 21.62 MeV (three body)

Hubler and coworkers at the Naval Research Laboratory have partially confirmed the existence of this anomaly in that the protons have been observed in some, but not all, of the samples bombarded. A coincidence experiment between the proton and alpha would be required to confirm this reaction. The experiment is important, and thought by some to be related to other experiments that show anomalies in metal deuterides. It could be interpreted as a probe of the probability that two deuterons are close together in the solid when the bombarding deuteron encounters them. There is no viable conventional explanation for the existence of this nuclear signal. 6. Conclusions The research discussed in this paper provides evidence for effects in three categories: (1) The existence of a physical effect that produces heat in metal deuterides. The heat is measured in quantities greatly exceeding all known chemical processes and the results are many times in excess of determined errors using several kinds of apparatus. In addition, the observations have been reproduced, can be reproduced at will when the proper conditions are obtained, and show the same patterns of behavior. Furthermore, many of the reasons for failure to reproduce the heat effect have been discovered. (2) The production of 4 He as an ash associated with this excess heat, in amounts commensurate with a reaction mechanism consistent with D + D -> 4 He + 23.8MeV (heat). (3) A physical effect that results in the emission of: (a) energetic particles consistent with d(d,n) 3 He and d(d,p)t fusion reactions, and (b) energetic alphas and protons with energies in excess of 10 MeV, and other emissions not consistent with deuteron-deuteron reactions.

40

Experimental results for tritium production were noted, and anomalous results from deuteron beam experiments on TiD^ were discussed briefly. In each case, the effects cannot be accounted for by known nuclear or solid-state physics. The underlying processes that produce these results are not manifestly evident from experiment. The scientific questions posed by these experiments are, in the opinion of the authors, both worthy and capable of resolution by a dedicated program of scientific research.

Appendix A: Calorimetric Issues As enumerated by Storms, 30 four questions must be addressed when evaluating the state of heat observations: (1) Was the calorimetric technique used by Fleischmann and Pons sufficiently stable and accurate to see the claimed extra energy? (2) Have others independently replicated the claims using stable and accurate calorimeters? (3) Can prosaic sources of chemical energy or energy storage effects be ruled out? (4) Have reasons for success or failure been discovered? A clear, positive answer to the first question was available in massive detail 114 but only retrospectively. Fleischmann and Pons used an open cell from which energy was lost in a variety of ways, including by infrared radiation, and as chemical energy carried with the evolving gases. The resulting differential equation was awkward and subject to misunderstanding. Fleischmann addressed the mathematical issues to the point where the isoperibolic method used can be seen to have accuracy better than 1% (Fleischmann's estimate is 0.1%). With daily automatic calibration, the Fleischmann-Pons calorimetric method was able to establish and assure stability for the length of time (months) needed to perform their experiments. Because the cell was open to escaping gases, attention was also given to the possibility that an unknown and variable fraction of the evolving D 2 and O2 gas may recombine to D 2 0 within the calorimetric envelope. A reduction of this subtractive term would give rise to the appearance of "excess heat" when in reality it simply had a chemical origin. Quantification of the upper bound of this effect was made by Fleischmann and Pons, 115 by Jones et al.n6 and in final detail by Will. 117 With 15 years of hindsight, it is now clear that the initial heat claims of Fleischmann-Pons must be taken at face value as quantitatively sound and capable of standing alone. Fortunately, this last is not necessary. Hundreds, possibly thousands, of attempts were made to replicate the Fleischmann-Pons effect. Apart from those conducted in 1989, when little was understood about necessary conditions, most researchers who attempted to reproduce the effect claimed success. We do not attempt to evaluate the bulk of these experiments or claims, although readers are invited to do so and could benefit in this regard by starting with the reviews by Storms. 113

41

To address the second question above, we provide an example. Flow calorimetry with closed cells was selected at SRI at the outset in part to avoid ongoing criticisms of the Fleischmann-Pons (Fick's law) method. Electrochemical cells were operated thermodynamically closed, sealed, and immersed in the fluid flow. Rather than measuring heat flow as a temperature difference across a defined (and presumed stable) barrier, the emerging heat in the SRI experiments was measured as the temperature rise in a moving liquid mass that surrounded the cell. In this way, the governing equations become trivial: all considerations of heat source dependence and so-called "recombination" effects can be avoided in simple first principles operation. Where potential sources of error were anticipated or recognized, the calorimetric system was designed to yield conservative estimates. Over 50,000 h of calorimetry to investigate the Fleischmann-Pons effect have been performed to date at SRI, most of it in calorimeters identical or very similar to that shown in Fig. 10. Water

Inlet RTDs

- Hermetic 16-pin connector

Water in

Acrylic top piece Gasket

Gas tube exit to gas-handling manifold

Water outlet containing "veptun migine tube and outlet RTDs

Gas tube containing catheter"

Acrylic flow separator Catalyst RTD

Hermetic 10-pin connector Stainless steel dewar Gasket PTFE plate Quartz cell body - *' PTFE liner -

, Screws Recombination -• catalyst in Pt wire basket - PTFE spray seperator cone PTFE ring

-

— Quartz anode cage

Pd cathode Brass heater support and pins

_,,

Acrylic flow restrictor

-"

Stainless steel outer casing

'

Heater "

Pt wire anode PTFE ring

' * Locating pin

:

Figure 10.

Stand

SRI mass flow calorimeter.

The object in Fig. 10 is immersed in a constant temperature bath (typically held constant and uniform to within 3-mK long term and short term). The bath is situated in a constant temperature room (±1K). Calorimetric fluid is drawn from the environing bath, past two inlet temperature sensors (100 f2 platinum resistance

42

temperature sensor RTDs) through the flow labyrinth and past the cell under interrogation, which is encased in a brass fin structure for better and more uniform communication of heat. The fluid flow is then drawn upwards and constrained by the hemispherical top to flow through a small hole and labyrinth, past and closely proximate to two outlet RTDs. Additional sensors of other types have been used to further assure accurate values of the measured temperature rise. The fluid is then drawn through a heat exchanger to the inlet of a constant volume displacement pump, and then delivered directly to an electronic balance. The balance is polled by computer to determine the fluid mass flow rate as AM/At. In high power experiments requiring higher fluid flow rates, rotameter flow sensors were used in series with the mass flow measurement. Calorimetry at SRI normally employed an electrical Joule heater either inside the cell or wound tightly on the cell circumference, constrained by the radial brass heat fins. This heater is used to calibrate the thermal efficiency of the convective heat flow process, and to allow for constant power operation in the presence of a variable electrochemical heat input. The equation of the calorimeter is thus: Pin = (-fV)Electrochem + (-f^)joule = Constant, /

AM

Pont = [CP^f -* xs = -'out

\ + k)

(Tout - Tin),

*in-

In most experiments and all discussed here, the fluid was air-saturated H 2 0 for which the heat capacity, Cp, is well known. Current, voltage, and temperature sensors, and measurement instrumentation including the mass balance, were calibrated independently of the calorimeter. The only term requiring in situ calibration is the conductive loss term k, which is the coefficient sum of all heat that leaves the calorimeter by means (primarily conductive) other than with the fluid flow. This term is small. By careful insulation, controlled geometry and selection of the fluid flow rate, k was typically less than 1% of C p (AM/At). Since k is defined by geometry, it is also very unlikely to change and was observed to be stable. Using these precautions the calorimetric method developed at SRI was shown by calculation and experiment 27 to have an absolute accuracy between ±0.35% and 0.5% depending on input conditions. It is worth noting that this accuracy is actually worse than that calculated by Fleischmann 114 for the Fleischmann-Pons isoperibolic method when fully implemented. It is important to further note that the calorimetric methods are very different, the absolute accuracies comparable, and the results obtained consistent. Several results have already been presented (Figs. 1-5) to give a sense of the calorimeter signal-to-error ratio in heat-producing experiments. Below we summarize the findings and results from the major SRI effort to confirm (or refute) the existence of the Fleischmann-Pons effect. (1) Sustained excess heat effects were observed from Pd electrodes undergoing

43

(2) (3)

(4) (5) (6)

(7)

(8)

(9)

electrolysis on more than 50 occasions with confidence more than three times the measurement uncertainty (3a). Bursts or episodes of excess heat generation lasted from periods of several hours to more than a week. During a burst, the excess power was typically between 3 and 30% of the total electrical input power, with the largest sustained observation being in excess of 340%. The intensity of this effect was in the range 3-300 W/cm 3 . Sustained heat bursts exhibit an integrated energy at least ten times greater than the sum of all conceivable chemical reactions within the closed cell. When normalized to the number of cathode Pd (or D) atoms, the energy yield is of order 100-1000 eV/atom, with the largest observed yield 2076 eV/Pd atom. Endothermic effects were not observed. Calorimeters maintained a tight thermal balance at times when excess power was not being observed. During these intervals, calibration checks could be performed conveniently and reliably by adjusting the level of the complementary Joule heater to increase and decrease the system total input and output power. Except for the expected transient responses, the calorimeters were never observed to exhibit output heat power lower than input. Excess heat effects were observed to occur with D2O but not H2O electrolytes, under similar or more extreme conditions of loading, input power and current density. After an appreciable initiation or incubation time, heat generation appeared to be correlated to three variables: D/Pd loading above a threshold; electrochemical current stimulus above a threshold; and deuterium flux or other dynamic stimulus.

With these results of a clear enthalpic excess unaccountable by known chemical or physicochemical means, it was determined at SRI and elsewhere to undertake a thorough and systematic evaluation of possible nuclear processes, by careful screening of potential products. 1 At SRI alone, serious efforts were made to interrogate active electrochemical and gas loading cells118 for neutrons, X-rays, 7-rays, charged particles, beta and other charge emission, tritium, 3 He and 4 He. In over a decade of effort, evidence for all of these potential products was observed, except neutrons. Of the products observed, only 3 H, 3 He, and 4 He could in any way be correlated, quantitatively or temporally, with the enthalpy production rates.

This is a matter of considerable importance and perhaps some confusion. Many claims for nuclear reaction products exist in the field now more broadly termed "low-energy nuclear reactions" or LENR.

44

Appendix B: Results for the Case Experiment at SRI In 1998, Case reported results from an experimental technique that offered potential advantages over electrochemical Fleischmann-Pons experiments. 54 He exposed commercial supported platinum group metal (PGM) hydrogenation catalyst materials to hydrogen or deuterium gas at slightly elevated pressure (1-3 atm) and temperature (150-300°C) in sealed metal vessels. In some cases, temperature sensed in a thermowell situated in the catalyst bed was higher by an amount sometimes exceeding 10°C under two different conditions: during heat bursts in D 2 ; and in D 2 relative to H 2 at the same pressure. The difference in thermal conductivity between the gaseous isotopes does not appear to contribute appreciably to the observed temperature difference. The primary heat loss pathway does not significantly involve gas phase conduction. Furthermore, a temperature difference for the same heater power is observed only in unusual circumstances. Having surveyed a very extensive range of supported PGM catalysts, Case concluded that the effect: (1) Could be observed with carbon supported catalysts but not with nonconductive supports. (2) Was exhibited with all PGMs (Pd, Pt, Ir, and Rh) except Ru, which had not been well studied at the time of publication. (3) Occurred optimally in a narrow range of catalyst loading (0.5-1%); at higher and lower loading the magnitude of the effect diminished and disappeared. (4) Occurs in a narrow range of temperature, about 130-300°C, with activity peaking at ~50°C. (5) Was destroyed irreversibly when the catalyst was heated to temperatures much greater than 300° C. Case concluded that his results represented evidence of a nuclear process because the energy production estimated from his calorimetry was much greater than could be accounted for from chemical processes. In addition, one post-test gas sample was reported to contain approximately lOOppm V/V of 4 He in a 1 atm sample of D 2 . The elements of simplicity, small materials inventory, rapid initiation, clear thermal signature and possible quantitative nuclear signature inspired many to attempt replication of what became known as the Case effect. Some succeeded in repeating the isotopic temperature disparity, although it quickly became clear that the experiment was far more demanding than it first seemed.™ An effort was made at SRI to reproduce the Case experiment. Initial attempts to reproduce the effect with catalyst materials supplied by Case and in metal sealed m

O n e study by Clarke 1 1 9 did not measure any significant increase in helium levels in a mass spectrometer where levels much smaller than lOOppmV/V would have been easily recognized. Clarke, however, did not observe the procedures described by Case, 5 4 which were in any case incomplete. Neither was Clarke able to measure any temperature effects and his geometry, which consisted of milligram single samples of "Case-type" catalyst confined with D 2 or H2 in very small sealed P b pipe sections, differed greatly from that used and recommended by Case.

45

vessels somewhat smaller than those employed by Case failed to observe systematic temperature differences between H 2 and D 2 . Samples of gas withdrawn from H 2 and D 2 cells did not show an increase in 4 He when submitted to an on-line, highresolution and high-sensitivity Extrel C-50 Quadrupole Mass Spectrometer. Two clear results of this first phase of activity were: (1) The Case effect, whatever its cause, is not normally or always present in catalyst samples, even those certified by Case as being "active." (2) He is not a natural or normal component of these catalyst materials. Significantly different results were obtained in a second campaign after a visit by Case to SRI to demonstrate his experimental procedures. Experiments were performed in pairs in nominally identical 50 cm 3 Nupro stainless steel sample flasks modified by the e-beam welded addition of a 1/8" stainless steel thermowell and a 1 in. Cajon VCR fitting to permit admission of catalyst. An amount of 10 g samples of catalyst were loaded into the vessels, which were then mounted and supported as shown in Fig. 11.

Figure 11.

Configuration of the Case experiment at SRI.

46

A heating element 1-mm diameter and 117-cm long was wound helically on the circumference of the bottom third of the cells and temperature was maintained by supplying power from a computer controlled DC supply. The two cells were mounted axially in 11, 10 cm inside diameter stainless steel dewars, surrounded by granular insulation to maintain their geometry and impose a consistent thermal environment. A significant change in this second phase of activity was to expose just the top of the sample cells to ensure an axial heat flow; this was found to be important both for calorimetry and to recreate the Case effect. During the course of the experiment, the currents to and voltages across each heating element were monitored together with two thermocouples measuring temperatures in the thermowell of each cell, and an ambient temperature sensor. Typically, experiments were operated in pairs, with one blank and one test cell, with the heater power set to keep the catalyst bed temperatures at nearly equal values. Test cells contained Case-certified 0.4-0.5% Pd on C catalyst in D2. Blank cells contained samples of identical catalyst in H 2 or inactivated or non-catalyst carbon in D 2 . After hydrogen treatments and evacuation to clean the catalyst surfaces, cells were then subjected to an extended period "soak" in H 2 to ensure by measurement that they were 4 He leak-tight and that any labile 4 He sources adsorbed or absorbed on the cell inner surface or catalyst volume had been exhausted. This period was also used to establish a reference temperature. Cells were then charged with D 2 or recharged with H 2 to the reference pressure, and their heat flow and helium levels monitored. Experiments exhibited a range of behaviors. Figure 12 summarizes 6 of 16 results of helium measurements in paired cells; these fall into three classes of behaviors: (1) Cells that show no increase of 4 He over long periods of time (including all cells operated with H 2 ). (2) Cells that exhibit a slow, approximately exponential increase in [4He] with time, following a trajectory very different from that expected for convective or diffusional leakage in from the ambient. (3) Cells that display no measurable increase in [4He] for a period of several days, followed by a rapid, approximately linear rise in [4He] to levels sometimes exceeding that of the ambient background. Using data from temperature sensors situated in the catalyst and gas phases it was possible to make heat flow estimates in one of two ways: (1) Gradient method, based on the relationship between the temperature difference between the catalyst bed and confined gas, and the heater input power. (2) Differential method, based on the temperature differences between active and the reference catalyst bed sensors and room temperature, as a function of the relative input heater powers. The energy estimated in excess of that provided by the heater for these two calori-

47

metric methods is plotted in Fig. 13, together with the measured helium concentration during the time of greatest derivative, d[4He]/dt in experiment SC2. Excess heat and the apparent increase in [4He] seem to be temporally correlated.

Figure 12.

Results of 4 He measurements from the Case experiment at SRI.

In an attempt to establish a quantitative correlation, Fig. 6 plots the two calorimetric estimates of excess heat production interpolated from Fig. 13, vs. the measured increase in [4He] (the value plotted in Fig. 13 minus the 4 He initially present in the D2 gas). Regression lines through these data incorporating the origin have slopes: Q = 31 ± 13 and 32 ± 13 MeV per 4 He atom, respectively, for the gradient and differential calorimetric methods. The Q value of ~23.8 MeV expected for reaction mechanisms consistent with D + D —• 4 He falls with the assigned uncertainties. The apparent shortfall of 4 He may occur for the same reasons observed in electrolytic Fleischmann-Pons studies discussed in Sections 3.1 and 3.2. In that case the residual helium putatively formed in a new nuclear process is released from the metal only slowly to the gas phase for analysis. Another factor must also be considered in accounting for the 4 He mass balance. In Fig. 12 for experiments SC2 and SC4.2 (D 2 on United Catalysts G75D and E, corresponding to 0.4 and 0.5% Pd on C, respectively), the final trajectories of 4 He with time show essentially the same linear decreasing trend. It is apparent the 4 He is either leaving the cell, in which case it may enter if the partial pressure of 4 He is lower inside than the ambient 5.22ppmV/V, or else 4 He is being adsorbed or absorbed onto or into the stainless steel or catalyst solid components, in which case these must be considered as potential sources of the observed increase. The first is unlikely since the cells were demonstrated repeatedly to be helium

48

—«— ppm V SC2 — — 3-Line fit for 4Jde. —o— Differential ---a... Gradient

> Q. CM

5

o E 3

"ai 3 X

40 20 10 Time (days)

Figure 13. at SRI.

0 20

Excess energy and helium production as a function of time from the Case experiment

leak-tight. An experiment performed to test the second hypothesis demonstrated clearly that 4 He adsorbs or absorbs into this class of catalyst material and also into or onto activated charcoal of similar geometry, with a rate consistent with the declining trend in Fig. 12, at similarly elevated temperatures. There is reasonable confidence that the 4 He source of the rising trends in Figs. 12 and 13 is not a release of stored 4 He from the catalyst for the following reasons: (1) A helium storage mechanism is expected to be reversible. A rate such as that shown in Fig. 12 should manifest itself clearly in the 5-day hydrogen pre-soak intended to address this concern (among others) before experiment initiation. (2) The rate of helium sequestration in carbon and carbon catalyst materials, at room temperatures, appears to be negligible. (3) Multiple random samples of catalyst were subjected to direct assay for 4 He by heating to temperatures in excess of 2250 K in the mass spectrometers of Clarke 120 and Arata and Zhang. In all cases these experiments demonstrated that the volume of 4 He contained with Pd on C catalyst materials was less than that in an equivalent volume of air. We conclude, tentatively, that helium is produced in a process that involves deuterium, but not hydrogen, which evolves heat commensurate with a nuclear mechanism consistent with a D + D - t 4 He reaction. Even taking account of the rate of helium re-ad/absorption, the mean value of the 4 He falls below that expected (assuming 23.8 MeV reaction energy) from the measured heat evolution. These observations regarding released 4 He are consistent with those made in studying

49

electrolytic Fleischmann-Pons cells. Calculating the excess power density normalized to the volume of Pd for Cell SC2, we obtain a maximum value of -~50 W/cm 3 , also consistent with the values observed in electrolytic loading of Pd. Further attempts to replicate the Case effect and the above-described results are underway at ENEA Frascati. Appendix C: The Arata and Zhang Effort Arata and Zhang have been major contributors to the research area under discussion. Dr. Arata came from the plasma fusion community. He and his colleague Dr. Yue-Chang Zhang began working as a team on cold fusion in about 1989. Their researches have been carried out in cooperation with other material scientists, including H. Fujita, who gave the Honda Memorial Lecture on atom clusters in 1994. In the early 1990s their researches led them to develop the DS (Double Structure) cathode, which is a Pd bottle whose "outer structure" is the wall of a Pd bottle and whose "inner structure" is filled with Pd-black. In 1994, they published observations of excess heat using open-cell DS-cathodes 83 and water-flow Dewar calorimetry similar to that described in Appendix A. In a continuous 3-month run Arata and Zhang observed excess heat power averaging about 15 W. Data from another run shows a heat balance for about 11 days, followed by 12 days with excess heat power averaging ~80W, with the output heat ~1.8 times input energy. An initiation time with no excess heat effect was seen in these, and in all other experiments with DS cathodes. Two additional 100-day runs exhibiting continuous excess heat of about 10W were published in 1995. 121 Three additional runs showing ~10-20W excess heat were published in 1996. 122 , 123 One of these runs continuously produced excess heat for 6 months. In another run, a mechanical pressure gauge was fitted to the DS cathode and measured the gas pressure inside the Pd bottle. During much of the run, the pressure gauge pinned at 800 bar. Excess heat power was much reduced when the pressure was less than 200 bar. Excess heat studies continued through 2002. An especially important result is presented in a more recent publication. 124 In this run, the inner cathode consisted of metal powder derived from an oxidized ZrgsPdas alloy, showing that powders other than Pd-black can produce excess heat. In 1998, a pair of DS cathodes were run in series, with one DS cathode in D2O electrolyte and the other in H 2 0 electrolyte. The DS cathode in D 2 0 electrolyte showed excess heat rising to 20 W, while the one in H2O electrolyte showed no excess heat. 125

Replication

Effort at SRI

A pair of DS cathodes prepared by Arata and Zhang were operated at SRI under conditions closely similar to those employed in Japan. Two nominally identical cells were exercised cathodically within intentionally similar cells, one in 0.3 M LiOD and the other in 0.3 M LiOH. Calorimetric accuracy was less than that normally present in the SRI Mass Flow Calorimeters because of the need to operate at very high input

50

power levels to exceed the threshold current density on large area cathodes. It was also not possible to submerge the Arata-Zhang cell completely in the constant temperature bath. Nevertheless, the results obtained at SRI 118 confirm the excess heat results published earlier by Arata and Zhang. In the same range of input powers, the heavy water cell clearly yielded more output heat than the light water cell when operated simultaneously and monitored with the same instruments. The maximum excess power observed in D 2 0 was 9.9 ± 1.3% of the measured power input, with the average value being approximately half the maximum. The measured excess power exceeded the experimental uncertainty (1-2% depending on conditions) for a period of ~86 days to produce an integrated energy excess of 64 ± 6 MJ for the D 2 0 cell. For the H 2 0 cell in the same period of time, the measured energy excess was - 1 ± 6 M J . At the conclusion of the experiment, both cathodes were removed and placed successively in a sealed chamber where they were punctured mechanically, and the gas contents of the cathode void volumes extracted for analysis. The Pd black powders also were removed and the Pd metal walls of the hollow cathodes were sectioned for 3 He and 4 He analysis. Significant amounts of tritium and 3 He (from the decay of tritium) were found inside the interior of the DS cathode electrolyzed in heavy water. Small amounts of 4 He were attributed to atmospheric contamination. Detailed results and conclusions from this work 126 and the results of later analyses are summarized here:

(1) Production of tritium was between 2 and 5 xlO 1 5 atoms. (2) Assuming that the 3 He is from tritium decay, independent determinations of both quantities allows for a determination of when the tritium was generated. Modeled as a single event, the generation of tritium computed from the measured yields occurred during the period of cathodic electrolysis. (3) There is definite evidence of excess 3 He from tritium decay in all samples of gas, Pd bulk metal and Pd black from the D 2 0 experiment. (4) Samples of Pd taken from a similar and contemporaneous electrode run with H 2 0 show low 3 He levels consistent with blank Pd. (5) Measurements of the 3 He gradient through the 3.5 mm wall of the D 2 0 cathode show that the 3 He is the decay product of tritium, which diffused from a source inside the electrode void volume (see Fig. 14). (6) A ~30% increase in tritium levels in the D 2 0 electrolyte measured by liquid scintillation methods is quantitatively consistent with the integral flux of tritium departing the cathode void and registered in the cathode walls. (7) The total inventory of tritium in the initial electrolyte and its increase, while substantial, represents only 0.05% of the total tritium mass balance following the experiment. (8) No evidence was found for 4 He in the interior gas (in the hollow within the cathode) or in the metal that was quantitatively consistent with the measured excess heat.

51

0.4 0.5 Radial position (cm) Figure 14. The 3 H e profile as a function of radius observed in an Arata-type, double structured cathode experiment at SRI.

The results of gas, metal and electrolyte phase 3 He and tritium analyses compel the conclusion that tritium was sourced in, on or adjacent to the Pd black in the void of the D2O double-structure cathode. While neither sought nor expected, this result provides sufficient evidence of the formation of a wholly nuclear product in an essentially electrochemical experiment to merit further study. Of concern, however, is the apparent absence of 4 He. Excess Heat and

Helium

Having convinced themselves that non-chemical heat production within DS cathodes is a real phenomenon, Arata and Zhang sought to identify helium nuclear products using quadrupole mass spectroscopy. They constructed a welded stainless steel manifold vacuum system evacuated with clean turbo pumps. They also built a small furnace system for outgassing metal powder test samples, and later developed a protocol using a Ti getter pump for removing chemically reactive gases from test volumes of desorbed gas. The manifold system included two dedicated mass spectrometers: one normally used to repeatedly scan the mass-4 peak structure, and the other to repeatedly scan the mass-3 peak structure. The initial analysis program was restricted to a study of desorbed gases from furnace-heated, as received and post-run Pd-black. The quadrupole instruments were very clean and had moderate mass resolution. Arata and Zhang routinely observed the 4 H e + peak resolved from the D^" peak. 123 The gas desorption studies 127 provided compositional analysis of both strongly bound gases (released at T > 1000° C) and less strongly bound gases (released at T < 800° C). They repeatedly demonstrated that post-run Pd powder al-

52

ways showed easily measurable 4 He+ peaks, whereas as-received Pd powder showed no detectable 4 H e + peaks. In 1998, they detected 3 He+ within a resolved mass3 peak structure. They found that they could often measure 3 He even when the mass-3 peak was not resolved. The 3 He + fraction could be distinguished from the HD + fraction by measuring the peak height as a function of ion source voltage. 128 The helium signal always made its appearance at a higher voltage due to its higher ionization potential. In 1999, following the experiment at SRI 118 and post-test sampling 126 described above, Arata and Zhang expanded their studies to include analysis of intergranular and ullage gas from the interior volume of one of their DS cathodes. They added an aliquot sampling volume to their manifold and built an assembly for piercing the DS cathode and for collecting gas samples. They used their system to collect both prompt release gas samples from the room temperature DS-cathode and slow release gas samples from the moderately heated DS-cathode. The helium signals were strong and clean. Measurements corresponded to a total 4 He production ~0.05% of that required to explain the run-integrated excess heat measured for the test cathode. 129 Whether 0.05% as observed in Osaka or 0.00% observed in similar experiments performed at SRI, the discrepancy relative to the integrated excess heat is large, and this difference has not been resolved. Possible explanations are: (1) The measurement of heat in both laboratories was substantially in error and no 4 He should be expected. The heat associated with the generation of tritium (measured as 3 He by Arata and Zhang) would be unmeasurable in either calorimeter. (2) Heat was produced in the cathode void by nuclear reaction of D, but the bulk of product helium was lost before sample collection due to micro-fractures, which occur in surface-stressed Pd, as described by Farkas. 130 (3) Heat was produced as in Fleischmann and Pons electrolytic cells at the cathode outer surface where loading, deuterium chemical potential, and stimulation are the greatest. The helium produced was only tenuously attached to the Pd surface and vented to the atmosphere in the Arata Zhang electrolytic cells, which are helium permeable." In this case, one must ascribe a secondary, not primary purpose for the cathode void and its enclosed Pd black, which conflicts with the hypothesis of Arata and Zhang. This is an important question that will be resolved as a result of continued experimentation underway at Osaka University, SRI, and ENEA Frascati.

References 1. M. Fleischmann, S. P o n s , a n d M. Hawkins, J. Electroanal. Chem. 2 0 1 , 301 (1989); Errata 2 6 3 , 187 (1990). See also M. Fleischmann, S. P o n s , M . W . Anderson, L.J. Li, n

I n other words, helium from the outer surface is not collected and measured in these experiments.

53 and M. Hawkins, J. Electroanal. Chem. 287, 293 (1990). 2. S.E. Jones, E.P. Palmer, J.B. Czirr, D.L. Decker, G.L. Jensen, J.M. Thorne, S.F. Taylor, and J. Rafelski, Observation of cold nuclear fusion in condensed matter, Nature 338, 737 (1989). 3. E. Storms, Calorimetry 101 for cold fusion: methods, problems and errors, http://www.lenr-canr.org/acrobat/StormsEcalorimetr.pdf 4. S. Pons and M. Fleischmann, in Proceedings of the ICCF1, 1990, p. 1. 5. D. Gozzi, P.L. Cignini, M. Tomellini, S. Frullani, F. Garibaldi, F. Ghio, M. Jodice, and G.M. Urciuoli, in Proceedings of the ICCF2, 1991, p. 21. 6. M.C.H. McKubre, R. Rocha-Filho, S.I. Smedley, F.L. Tanzella, S. Crouch-Baker, T.O. Passell, and J. Santucci, in Proceedings of the ICCF2, 1991, p. 419. 7. D. Macdonald, M.C.H. McKubre, A.C. Scott, and P.R. Wentrcek, I EC Fundam. 20, 290 (1981). 8. M.C.H. McKubre, R.C. Rocha-Filho, S.I. Smedley, and F.L. Tanzella, Calorimetry and electrochemistry in the D / P d System, in Proceedings of the ICCF1, 1990, p. 20. 9. K. Kunimatsu, N. Hasegawa, A. Kabota, N. Imai, N. Ishikawa, H. Akita, and Y. Tsuchida, Deuterium loading ratio and excess heat generation during electrolysis of heavy water by a palladium cathode in a closed cell using a partially immersed fuel cell anode, in Proceedings of the ICCF3, p. 31 (1992). 10. N. Hasegawa, N. Hayakawa, Y. Tsuchida, Y. Yamamoto, and K. Kunimatsu, Proc. ICCF4 1, 3-1 (1993). 11. M.C.H. McKubre and F. Tanzella, inProceedings of the ICCF7, p. 230 (1998). 12. L. Bertalot, L. Bettinali, F. DeMarco, V. Violante, P. De Logu, T. Dikonimos Makris, and A. La Barbera, inProceedings of the ICCF2, p. 3 (1991). 13. S.E. Koonin and M. Nauenberg, Cold fusion in isotopic hydrogen molecules, Nature 339, 690 (1989). 14. J.O'.M. Bockris and A.K.N. Reddy, Modern electrochemistry; an introduction to an interdisciplinary area, New York, Plenum Press (1970). 15. S. Pons, M. Fleischmann, C.T. Walling, J.P. Simons, international PCT Patent Application PCT/US90/01328, International Patent number WO 90/10935 (1990). 16. L. Bertalot, A. DeNinno, F. DeMarco, A. La Barber, F. Scaramuzzi, V. Violante, Excess power production in electrolysis experiments and ENEA Frascati, inProceedings of the ICCF5, p. 34 (1995). 17. E. Storms, inProceedings of the ICCF6, p. 105 (1996). 18. E. Storms, Relationship between open-circuit-voltage and heat production in a PonsFleischmann cell, inProceedings of the ICCF7, p. 356 (1998). 19. M. Swartz and G. Verner, Excess heat from low-electrical conductivity heavy water spiral-wound P d / D 2 0 / P t and Pd/D20-PdCl2/Pt Devices, inProceedings of the ICCF10, (2004). 20. S. Pons and M. Fleischmann, inProceedings of the ICGF2, p. 349 (1991). 21. M. Fleischmann, inProceedings of the ICCF5, p. 152 (1995). 22. M.H. Miles, S. Szpak, P.A. Mosier-Boss, and M. Fleischmann, inProceedings of the ICCF9, p. 250 (2002). 23. inProceedings of the ICCF4, 2, 18-1 (1993). 24. E. Storms, Some characteristics of heat production using the cold fusion effect, Proc. ICCF4 2, 4-1 (1993). 25. L. Case, in his oral presentation at ICCF10. 26. M.C.H. McKubre, S. Crouch-Baker, A.M. Riley, S.I. Smedley, and F.L. Tanzella, Excess power observations in electrochemical studies of the D / P d system; the influence of loading inProceedings of the ICCF3, p. 5 (1992). 27. M.C.H. McKubre, S. Crouch-Baker, F.L. Tanzella, S.I. Smedley, M. Williams, S.

54

28. 29. 30.

31. 32. 33.

34.

35. 36. 37. 38.

39. 40. 41. 42.

43.

44. 45. 46. 47.

48. 49. 50. 51.

Wing, M. Maly-Schreiber, R.C. Rocha-Filho, P.C. Searson, J.G. Pronko, and D.A. Koehler, Development of Advanced Concepts for Nuclear Processes in Deuterated Metals, EPRI Report TR-104195, August 1994. A. Takahashi, A. Mega, T. Takeuchi, H. Miyamura, and T. Iida, inProceedings of the ICCF3, p. 79 (1992). T. Aoki, Y. Kurata, E. Ebihara, N. Yoshikawa, Proc. ICCF4 2, 23-1 (1993). E.K. Storms, A study of those properties of palladium that influence excess energy production by the Pons-Fleischmann effect, (1996) http://www.lenrcanr.org/acrobat/StormsEastudyofth.pdf F. De Marco, A. De Ninno, A. Frattolillo, A. La Barbera, F. Scaramuzzi, and V. Violante, inProceedings of the ICCF6, p. 145 (1996). A. Takahashi, T. Takeuchi, T. Iida, and M. Watanbe, Neutron spectra from D 2 0 - P d cells with pulse electrolysis, inProceedings of the Provo meeting, p. 323 (1990). A. Takahashi, T. Iida, T. Takeuchi, A. Mega, S. Yoshida, and M. Watanabe, Neutron spectra anc control lability by PdD/Electrolysis cell with low-high current pulse operation, inProceedings of the ICCF2, p. 93 (1991). J.O'.M Bockris, D. Hodko, and Z. Minevski, The mechanism of deuterium evolution on palladium: Relation to heat bursts provoked by fluxing deuterium across the interface, inProceedings of the ICCF2, p. 337 (1991). E. Yamaguchi and T. Nishioka, Direct evidence for nuclear fusion reactions in deuterated palladium, inProceedings of the ICCF3, p. 179 (1992). H. Sugiura and E. Yamaguchi, inProceedings of the ICCF7, p. 366 (1998). E. Yamaguchi and H. Sugiura, inProceedings of the ICCF7, p. 420 (1998). C. Bartolomeo, M. Fleischmann, G. Larramona, S. Pons, J. Roulette, H. Sugiura, and G. Preparata, Alfred Coehn and after: The a, /3, 7 of the palladium-hydrogen system, Proc. ICCF4 1, 19-1 (1993). F. Celani, A. Spallone, P. Tripodi, A. Petrocchi, D. DiGioacchino, P. Marini, V. DiStefano, S. Pace, and A. Mancini, inProceedings of the ICCF5, p. 57 (1995). F. Celani, A. Spallone, P. Tripodi, D. DiGioacchino, P. Maini, V. DiStefan, A. Mancini, and S. Pace, inProceedings of the ICCF6, p. 93 (1996). G. Preparata, inProceedings of the ICCF6, p. 136 (1996). M.C.H. McKubre, S. Crouch-Baker, A.K. Hauser, S.I. Smedley, F.L. Tanzella, M.S. Williams, and S.S. Wing, Concerning reproducibility of excess power production, inProceedings of the ICCF5, p. 17 (1995). J. Tian, X.Z. Li, W.Z. Yu, D.X. Cao, R. Zhou, Z.W. Yu, Z.F. Jiang, Y. Liu, J.T. He, and R.X. Zhou, Anomalous heat flow and its correlation with deuterium flux in a gas-loading deuterium-palladium system, inProceedings of the ICCF9, p. 353 (2002). J.O'.M. Bockris, et al., Proc. ICCF4 2, 1-1 (1993). X.Z. Li, B. Liu, N.N. Cai, Q.M. Wei, J. Tian, and D.X. Cao, inProceedings of the ICCF10, (2004). Y. Iwamura, M. Sakano, and T. Itoh, Jpn. J. Appl. Phys. 41 4642 (2002). Y. Iwamura, T. Itoh, M. Sakano, S. Sakai, and S. Kuribayashi, Low energy nuclear transmutation in condensed matter induced by D2 gas permeation through Pd complexes: Correlation between deuterium flux and nuclear products, inProceedings of the ICCF10, (2004). D. Letts and D. Cravens, Laser stimulation of deuterated palladium: Past and present, inProceedings of the ICCF10, (2004). D. Cravens and D. Letts, Practical techniques in CF research: Triggering methods, inProceedings of the ICCF10, (2004). T.O. Passell, inProceedings of the ICCF5, p. 603 (1995). T.O. Passell, inProceedings of the ICCF9, p. 299 (2002).

55 52. J. Dash, inProceedings of the ICCF6, p. 477 (1996). 53. T. Bressani, Nuclear products in cold fusion experiments, comments and remarks after ICCF-6, inProceedings of the ICCF6, p. 703 (1996). 54. L.C. Case, Catalytic fusion of deuterium into helium-4inProceedm[, of each of the super-imposed waves are adjustable.

3. Electrolytic Cells Figure 2 shows the overall layout of the electrolytic cells designed and constructed by E T . T h e cathode is 8-cm long ( 6 c m in effective length), 0.7-cm wide and 50 or 100-/xm thick P d foil. Two platinum foils of similar dimensions - 2 0 m m x 80 m m x 0.1mm, are used for the anode. They are located 5 m m from each side of the cathode. T h e c a t h o d e - a n o d e assembly is immersed in an electrolyte made of 0 . 1 1 M LiOD in D2O. T h e electrolyte and the c a t h o d e - a n o d e are inside a cell made from two concentric aluminum cylinders with alumina powder thermal insulation in between the cylinders (Fig. 2). T h e cell is immersed in a constant t e m p e r a t u r e water b a t h usually set at 2.5 ± 0.25°C. Three sensors monitor the t e m p e r a t u r e in different locations in the cell. The cell has an external recombiner. Presently, there are six electrolytic cells working in parallel; three cells per water b a t h . Each dedicated computer drives three cells. T h e same computer also collects, stores and analyzes the experimental data. Temperatures readings are done at a rate of 25scans/s, whereas voltage and current readings are taken at a rate of 50,000 scans/s. T h e cathode resistance is measured to infer the level of deuterium loading. A Labview program was developed to perform all these functions. A typical Superwave generated by this computer control system is given in Fig. 1. More t h a n one hundred of P d foils have been used in our experiments so far. Dr. Vittorio Violante of the Frascatti research center in Italy provided a significant number of the foils. T h e foils received from Dr. Violante were degreased and annealed at 870° C for 1 h. Then they were etched first with nitric acid of 65-67% for l m i n and then for 1 min in Aqua Regia 1:1 water solution. Following t h a t the foils were rinsed four times in heavy water, twice in 95% ethanol, and once in ethanol absolute. Finally the foils were dried in vacuum at ambient t e m p e r a t u r e for 24 h.

86 F0(f) = A0smz Hi

FoW 1. 0.8 0.6 0.4 0.2

n 1.5

2 I

Ff(f) = A0 sin 2 (co0f) (1+A, sin 2 (o^f))

F2{i) = A0 sin 2 (w o 0 [1+-A1 sin 2 (o^Q (1+,42 sin 2 (w 2 /

F(ti

F2f) ( l + A , sin 2 (co3()))]

F3»

,AA~.

.•r*r^*S—»^_

Figure 1.

Principles of Superwave formation.

87

Figure 2.

Schematic layout (left) and a photo (right) of the ET electrolytic cells.

Excess heat is measured using an isoperibolic calorimeter. Each electrolytic cell was calibrated using an electrical heater. A typical calibration curve is given in Fig. 3. The "AT" of Fig. 3 is T4 — T5, where the temperatures are measured in the locations shown in Fig. 2. The accuracy of this calorimeter is estimated to be 1-2%.

75 70 86 M 55 50 45 40 35 30 25 20 15 10 5 0 0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

ir

Figure 3.

A typical calibration plot of the calorimeter of the electrolytic cells.

4 . S u p e r w a v e Effect o n D e u t e r i u m L o a d i n g in P a l l a d i u m C a t h o d e s

A set of experiments was performed to investigate the effect of current modulation on the rate of deuterium loading into the Pd. Figure 4 shows the three levels of

88

wave modulation considered, while Figs. 5-7 show loading rate dependence of foil No. 56 on the modulation level. A clear and reproducible correlation is observed; the higher the level of modulation, the faster is the deuterium-loading rate.

" "

8! I *..;,

/

Ik."

V

eters

Fiisi laodulaiso

Constant parameter;

•z ir:c _•: ww^-v,-?,^.

S e c o n d modulation

• ' " . S -,V'.-j(-

Third modulation

Figure 4.

Three levels of modulation of Superwaves used in the experiments.

IMWWtfSKM*;. *««.«

s^«.

8.**

^

';,".' "

j,*i

Third level

—, J- '» • -mm it a a Figure 5.

R/Ro

resistance ratio in loading foil No. 56 as a function of level of modulation.

»> — »

•HIF^

Increasing of current

m

Increasing of current

I'

First level of modulation

First level of modulation'

Second level of modula

Third level of modulatipn

Second level of modulation

i *.}.. ' 8

Figure 6.

«

R/Ro resistance ratio dependence on the level of wave modulation in foil No. 56.

5. Excess Heat Generation Significant amount of excess heat was generated by two of the foils: foil Nos. 56 and 64. Both of these foils were prepared by Dr. Violante. The rest of this paper will focus on describing these two successful experiments. Figures 8 and 9 show the Superwaves used in these experiments. The excess heat in foil No. 56 was generated in the fourth loading. It started after 80 h of loading and lasted for approximately 300 h. Figure 10 shows the evolution of input and output power during this period and the excess power generated. The average input, output and excess power generated during the period are estimated to be, respectively, 3.9, 6.6, and 2.7 W. The total excess energy generated is estimated to be 3.1 MJ.

90

1 2 3 3 2 1 1 2 3 3 2 1 2 3 2 1 2 3 3 Level of modulation

Figure 7.

Resistance ratio dependence on modulation level and current density (foil No. 56).

-

i\

, i

1!

i! J

i! U! !i !* •$

!(i il

^1 im'i'-d

* * » a i i n «fce o h * ofat whs nnb» W B » « • • ————.— ——aaagk— -—r ^ : " 2 ; ' '.'.: '-'! * r ' ^ C a + 10.3MeV.

3. Conclusions "Excess heat" can be repeatedly obtained through the electrolysis in Pt/K^COs/Ni light water system. Comparing the heating coefficients, ke is always smaller than kT in Na2C03 electrolysis; On the contrary the ke is always higher than fcr in K2CO3 electrolysis. Taking the whole experiment into consideration, the ke is always higher in K2CO3 electrolysis than ke in Na2C03 electrolysis. The thermal output energy is 280% more than electrical input energy when the current was 0.078 A. The "excess heat" is about 2.5 x 104 J over 24 h. After electrolysis, the increases of pH value and absorbency in K2CO3 solution (which are, respectively, by ApH = 0.15 and AA = 0.108) implied that some H+ combined with K + to form Ca 2 + .

107

References 1. V. Noninski, Excess heat during the electrolysis of a light water solution of K2CO3 with a nickel cathode, Fusion Technol. 2 1 , 163-167 (1992). 2. R. Mills and W.R. Good, Fractional quantum energy levels of hydrogen. Fusion Technol. 28, 1697-1706 (1995).

INNOVATIVE P R O C E D U R E FOR THE, IN SITU, M E A S U R E M E N T OF T H E RESISTIVE T H E R M A L COEFFICIENT OF H ( D ) / P d D U R I N G ELECTROLYSIS; C R O S S - C O M P A R I S O N OF N E W ELEMENTS D E T E C T E D IN THE T h - H g - P d - D ( H ) ELECTROLYTIC CELLS

F R A N C E S C O C E L A N I , A. S P A L L O N E , E . R I G H I , G. T R E N T A , C. C A T E N A , G. D ' A G O S T A R O , P. Q U E R C I A , A N D V. A N D R E A S S I INFN-LNF

Via E. Fermi 40, 00044 Frascati, Rome, E-mail: [email protected]

Italy

P. M A R I N I , V. DI S T E F A N O , A N D M. N A K A M U R A EURESYS,

Via hero 30, 00129 Rome,

Italy

A. M A N C I N I ORIM

Sri, Via Concordia

65, 62100 Piediripa,

Macerata,

Italy

P.G. S O N A Via S. Carlo 12, 20090 Segrate,

Milan,

Italy

F . F O N T A N A , L. G A M B E R A L E , A N D D . G A R B E L L I Pirelli Labs SpA,

Viale Sarca 222, 20126 Milan,

Italy

E . C E L I A , F . F A L C I O N I , M. M A R C H E S I N I , A N D E. N O V A R O CSM SpA,

Via di Castel Romano

100, 00129 Rome,

Italy

U. M A S T R O M A T T E O ST Microelectronics

SpA,

Via Tolomeo

1, 20010 Cornaredo,

Milan,

Italy

In the framework of cold fusion studies one of the most important parameters is the deuterium (D) to palladium (Pd) ratio, D / P d . It is well known that the value of this parameter is related to the normalised resistivity (R/RQ)

of the D - P d system. When at high D / P d ratios (i.e. at low R/RQ

values) some excess heat occurs, the Pd wire temperature increase and, as a consequence, the apparent R/Ro value also increases. This effect might give raise to ambiguous data interpretation: similar results are in fact expected in case of a Pd wire degassing (i.e. decreasing of D / P d ratio). To solve this problem, we developed an innovative procedure and a suitable experimental set-up for the in situ measurement of the Resistive Temperature Coefficient (which is affected only by the real D / P d ratio) during electrolysis. We will report the results on the hydrogen and deuterium loading of thin (50 urn), and long (60cm) Pd wires, immersed in a solution of C 2 H 5 O D (or C2H5OH) and 108

109 D2O (or H2O), with addition of thorium (Th) and mercury (Hg) salts at micromolar concentrations. Evidence of "transmutations" of some elements occasionally present on the Pd surface, and sometimes also in the electrolytic solution, have often been claimed in cold fusion experiments. In the present work, unexpected elements have been detected by high-resolution ICP-MS analysis. Some of these elements have also an isotopic composition different from the natural one.

1. Introduction: Reasons to Measure, In Situ, the Resistive Thermal Coefficient of the Pd Cathode To detect excess heat in cold fusion experiments one of the most important parameters is the deuterium (D) over palladium (Pd) ratio, D/Pd. It is well known that the value of the D/Pd ratio is related to the normalised resistivity (R/Ro) of the D-Pd system: e.g. its maximum value, 2.0, is equivalent (at room temperature) to a D/Pd ratio of about 0.75. By further increasing the D/Pd ratio, the R/RQ starts to decrease. If at high D/Pd ratios (^> about 0.90) there is some excess heat, the Pd temperature increases; as a consequence, the apparent R/RQ value also increases, which is the same result that would be obtained if degassing occurred (a decrease of D/Pd ratio). This effect might give raise to ambiguous data interpretation and fruitless discussions. To try to solve such a problem, we developed an innovative experimental set-up and a specific procedure for the in situ measurement of the Resistive Temperature Coefficient. As such a parameter is affected only by the real D/Pd ratio, any ambiguous interpretation in case of sudden variations of the D/Pd ratio will be avoided. We will show, and discuss in some detail, a test that lasted about 2 days, made during a typical loading-unloading experiment. The results were obtained on deuterium loading of thin (50 ^m), and long (~60cm) Pd wire, immersed in a solution of ultrapure, vacuum distilled and ultrafiltered C2H5OD and D2O with addition of thorium (Th) and mercury (Hg) salts at micromolar concentrations. 2. The Unconventional Frascati-INFN Procedures for Electrolytes, Electrolysis and Cathode Shape and Preparation Here are some of key points regarding the Frascati-INFN electrolysis procedure. The aims of experiments were essentially: (a) To develop innovative and reproducible electrolytic techniques capable to maximise the values of hydrogen (H) and deuterium (D) concentrations in palladium (Pd) (i.e. the so-called "overloading": H(D)/Pd 3> 0.90 as mean value). (b) To shorten the time from beginning of experiment to reach overloading (4h).

110

Table 1. Summary of the composition of the nine experiments performed, date and some comments about results Comments on results OVL = overloading (D/Pd > 0.9) S/N — signal/noise Pr = 1 count = 6E10 atoms

Experiments

Date: begin —> end (dd/mm/yy; 1

Electrolyte composition (mol)

1

20/12/02 -> 16/01/03

CsNOa (5E-5) SrCl2 (1E-5) LiOD (1.5E-5) H 2 S 0 4 (5E-6) NH3OH (1E-4)

Almost no OVL Anode = Pd wire 250 fim Pr = 80 =* S/N = 2

2

17/01/03 -> 14/02/03

CaCl 2 (21E-5) SrCl 2 (1E-4) HgCl2 (2E-4) H2SO4 (1E-5)

Two times OVL Residual Cs? Pr = 170 =>• S/N = 4

3

18/02/03 - • 05/03/03

CaCl 2 (1E-5) SrCl2 (1E-4) HgCl2 (2E-4) H2SO4 (1E-5) Th(N0 3 )4 (8E-6)

Three times OVL R/Ro = 1-706 Pr = 370 =s- S/N = 9

4

04/04/03 -» 14/04/03

HgCl2 (1E-5) Hg 2 S0 4 (E-4) Th(N0 3 )4 (21E-6)

No OVL

5

15/04/03 - • 19/05/03

T h ( N 0 3 ) 4 (34E-6)

Several times OVL; excess heat Pr = 300 =>• S/N = 6

Hg 2 S0 4 (2E-6) 6

29/05/03 - • 31/07/03

T h ( N 0 3 ) 4 (65E-6) Hg 2 S0 4 (5E-6)

7

31/10/03 -> 08/12/03

T h ( N 0 3 ) 4 (21E-6) Hg 2 S0 4 (3E-6)

8

29/01/04 - • 15/03/04

C 2 H 5 Od = 740 cm 3 3

D 2 0 = 61cm Th(N0 3 ) 4 (30E-6) Hg 2 S0 4 (8E-6) 9

15/03/04 -» 21/05/04

C 2 H 5 OH = 710 cm 3 H2O = 37cm

3

D2O = 12cm 3 T h ( N 0 3 ) 4 (21E-6) Hg 2 S0 4 (7E-6)

Several times OVL; excess heat Pr = 1.4E3 => S/N = 21 208 Pb anomaly; ™3^Tl = 4.6E3 and 209 Bi = 9E3 Few times OVL: Na contamination Pt deposit on Pd Q = 40E+3C Few times D overloading (R/Ro < 1.8) Na contamination Pt deposit on Pd Almost "light" experiment - • BLANK Large current adopted for long time (Q = 125E+3C) Few times H overloading (R/Ro < 1.45) Pt deposit on Pd

Ill

2.1. Electrolytes

and Electrolysis

Procedures

We decided first to employ hydro-alcoholic electrolytic solutions. The reasons for this unconventional choice were described in detail in our previous papers of ICCF series (Refs. 7 and 8). In short, we used, and are still now using, a solution of 90-95% heavy ethyl alcohol (C2H5OD) and 10-5% heavy water (D2O). The main dissolved cations were strontium (as SrCLj) and mercury (as H g C y ions, at some micromolar concentrations (10-100/xM and 1-10 fjM, respectively). The pH was kept at about 4 (acidic) by addition (if needed) of few drops of concentrated HNO3. For the sake of comparison, most of the electrolytic "cold fusion" experiments carried out in other laboratories use 0.1-1 M LiOD/D20 solutions, i.e. strong basic solutions (pH around 13/14), and a cathode current density between 60 and 600mA/cm 2 . Such procedure follows the pioneer teachings and long experimental work of Fleischmann and Pons 9 (Univ. Utah, USA), since 1989. We emphasise that in our experiments, because of the mildly acidic electrolyte (pH ~4) and the very low current density (only 5-20mA/cm 2 at the cathode surface and five times less at the anode) it is possible to reduce, to a large extent, the corrosion of components inside the electrolytic cell, borosilicate glass beaker included. Corrosion effects are typical of "conventional" electrolytic experiments: they are usually operated at high current for long time at strong basic pH. We anticipate that despite our efforts, problems coming from Pt dissolution on the anode are still not solved in a reliable way. 2.2. Elemental

Analysis

As a consequence of our procedure, at the end of the experiment it is possible to make accurate elemental analysis by inductively coupled plasma-mass spectroscopy (ICP-MS); inductively coupled plasma-optical emission spectroscopy (ICP-OES); scanning electron microscopy (SEM) with micro-analysis. The following were analysed: the residual powder filtered off from the electrolyte (by vacuum distillation), and the components of the Pd wire itself. The deleterious "matrix effect" in ICP-MS analysis is strongly reduced. After proper (long) background subtractions and inter-calibrations just after each ICPMS measurements, the results were quite accurate and reliable. The results were "safe" because they did not call for special, sophisticated (and always dangerous) mathematical elaboration. Refs. 5-7 give a detailed overview and a general discussion on the interpretation of ICP-MS results. As a historical note, because of obvious economic considerations, we developed our experimental procedure with hydrogen overloading using low cost light hydroalcoholic electrolytes at different concentrations. In the next step, we tried to adapt, and transfer, the most successful methods to the heavy water and heavy alcoholic solutions, which are about 400 times more expensive. In our deuterium-based experiments, we were looking for anomalous production of excess heat, tritium and particularly "transmutations". Very recently in fact,

112

"transmutation" phenomena received a significant acceptance in the scientific community mainly because of the reproducible experiments carried out at Mitsubishi Heavy Industries Laboratories (Yokohama, Japan). 1 - 4 We would like to remark that some of the Mitsubishi Group results (headed by Yasuhiro Iwamura) were independently confirmed also by our experimental group at INFN Frascati National Laboratories, by using with respect to that of Iwamura a complementary methodology: wet electrolytic environment instead of dry gaseous one. A detailed description and comments on the results can be found in Refs. 5-7. 2.3. Cathode Shape and

Preparation

Based on suggestions made by Giuliano Preparata and Emilio Del Giudice (University and INFN Milan, 1994), in our loading tests we use Pd cathodes consisting of wires 50-100 cm long with diameter as thin as 0.05 mm. Generally, researchers use rods (following the example of Fleischmann and Pons) or plates (25 x 25 x 1 mm, following Akito Takahashi, Osaka University, Japan, 1992). Before use, the Pd wires were carefully cleaned by dipping them in sequence in organic solvent, water, nitric acid, and water. Afterwards, by means of a specific protocol of joule heating and subsequent slow cooling down to room temperature, the wires are stress relieve annealed and, at the same time, properly oxidised as to form a thin film of Palladium oxide on the wire surface. Such a complex protocol for treating wires was developed and continuously improved by our group since 1996. By the way, we note that since 1993, we have exploited the peculiar characteristic of PdO at surface of palladium plates. The plates (Takahashi type geometry) were air oxidized at about 700° C, with a butane gas flame in a proper alumina crucible. Results from fast deuterium loading were presented at the ICCF4 and ICCF5 conferences. At that time the understanding of phenomena, although quite intriguing and reproducible, was not deep enough to justify further research. 2.4. Effects of Electrolysis

Procedure

on

Results

As reported in our papers presented at previous ICCF conference, JCF Meetings and Asti Workshops, we performed a series of experiments with hydro-alcoholic solutions containing small amounts of Sr and Hg salts. We found excess heat (see Ref. 8) and tritium (see Ref. 10) well above background. We also found that in the hydro-alcoholic ambient, during the anodic phases (that is for some hours every 13 days) of our loading cycles, the Pd electrode was partially eroded, producing small solid debris. Significant amounts of very small Pd particles are found at the bottom of the cell at the end of the experiments. In such black coloured powder ICP-MS analysis showed the presence of Pd together with some unexpected elements (see Ref. 5). Moreover, after several electrolytic loading-unloading cycles, we could observe a very unusual phenomenon: the eroded (and therefore very active?) surface of the

113

Pd wire was able to rapidly absorb the small amount of the deuterium gas dissolved into the solution without applying any electrolytic current. In our cell the maximum overpressure is only 50 mbar. Such spontaneous absorption was remarkable: a D/Pd ratio up to 0.75 was often reached. The observed D self-loading of the Pd wire up to a value equivalent to a gas pressure of over 10 bar, is obviously connected to our new and specific electrolysis procedure. Because the observation of such experimental effect, we designed a new electronic circuit able to manage, at will, the cathodic —> anodic —> cathodic cycles, in a continuous way. Such circuitry worked well and is under consideration for a patent application. Coming back to the Pd cathode, we observed by SEM analysis that the surface after electrolysis, was deeply modified; many small bubbles and holes were sprayed over the surface, suggesting the formation of a nanostructural and/or fractal geometry. See Ref. 11 for further comments about the importance of nano-structure in cold fusion experiments as pointed out by Yoshiaki Arata (Osaka University, Japan). Based on Arata's views, we are also convinced that the formation of nanostructural and/or fractal geometry at Pd surface plays a key role in the production of all the anomalous effects detected in cold fusion experiments. 3. Analysis of N e w E l e m e n t s Another important result in cold fusion studies is the experimental evidence of "transmutations" of some elements present on the cathode surface (Th and/or Hg in the present work), along with the Pd itself. In our experiments, we used both light and heavy water electrolytic solutions. In some experiments, characterised by high deuterium loading over long periods (days) and repeated loading-unloading cycles, unexpected elements (sometimes in quite large amounts) have been detected by high resolution ICP-MS analysis. Some of these elements have also an isotopic composition different from natural one. We would like to recall that in 2001, Iwamura et al. 1 was the first to show, in a very elegant experiment, that strontium (Sr) is apparently transmuted into molybdenum (Mo), or cesium (Cs) into the rare earth praseodymium (Pr) when: (a) D2 gas is forced to flow for enough long time (several hundreds of hours), (b) the D 2 gas flow is at a high enough rate (>2sccm), (c) the D 2 gas flow through a proper multilayer of Pd/Pd-CaO/Sr or Cs. We recall that proper multilayer was fully developed by Iwamura's team at Mitsubishi Heavy Industries (Yokohama Laboratories, Japan), starting in 2000, and later patented at International level. We tried to check whether such a "transmutation" can also occur after repeated D-Pd loading/unloading/loading cycles in our experimental set-up. In July 2002, we were ready to perform an independent variant of the Iwamura experiment.

114

Before starting we performed ICP-MS analysis of all the components present in the cell (C 2 H 5 OD, D z O, SrCl 2 , DC1, HgCl 2 , and Pd), and pieces of the two Pt wires (anode and reference electrodes, taken from the same batch). At the end of the D-Pd loading/unloading experiment, the electrolytic solution was vacuum dried, the residue was collected and again analysed by ICP-MS together with the Pd cathode, all dissolved in hot-concentrated aqua regia. Excess Mo was found in amounts far above any conceivable contamination. The isotopic composition of the Mo was different from the natural one (see Ref. 5). It appears that the phenomena previously discovered by Iwamura in a flowing deuterium gas system also occur in our electrolytic cell, operating for a time length of 500-1000 h, according to our loading-unloading-loading procedure.

3.1. Thorium

Salts as

Electrolyte

In January 2003, we decided to substitute the strontium salts, previously used as electrolytes, with thorium salts. The two main reasons for such the change were as follows: (a) Some results published also by our group in 1997-1998, seemed to show Th "transmutations" during high-electric power, at high temperatures and pressure, with AC (50 Hz) electrolysis with massive zirconium electrodes (both anode and cathode, see Ref. 12). Accordingly, we decided to test whether something similar could happen in our new experimental apparatus based on thin Pd wires and a very strict control of impurities. (b) Th ions (like Sr 2 + ), because of the local alkalization produced by the passage of the electrolytic current, can precipitate on the cathode surface as Th(OH) 4 (solubility product Ks = lO" 5 0 ). Th ions (like Sr 2 + ) should not be galvanically deposited on the cathode because of the high negative value of their standard potential (EQ = —1.899 V). In any case, as we want Th to be present on the Pd surface, even though this element could be co-deposited as Th deuteride (instead of as Th hydroxide) through some unknown process, no incompatibility with the aim of our specific experiment should occur. Taking into account that very low values of current density are required to deposit the proper Th(OD)4 and/or ThD s layer(s) on the cathode surface, the occurrence of some anomalous excess heat, should be easily detected. Operating with Th containing electrolytes, we intended to assay the following: (a) the occurrence of anomalous excess heat; (b) the presence of foreign elements: at the Pd cathode surface and/or into the bulk, into the liquid solutions, in the insoluble agglomerates generally present at the end of the experiment in the electrolytic cell.

115

4. Short Description of Experimental Set-up: Electrolytic Cell and Flow Calorimeter In order to understand the resistive thermal coefficient (RTC) measurement procedure (see Chapter 5) and experimental results obtained, we report some of key points of our electrolytic cell. The configuration of the cell was shown in Fig. 1 of Ref. 6. The sample holder, a PTFE tube, is placed in a 1000 ml borosilicate glass (type 3.3, brand FORTUNA) cylinder (diameter 67 mm and height 460 mm). The cathode and anode are both "U"-shaped and are located on the opposite walls of the holder, facing each other. The cathode is a thin (diameter 0.050 mm) long (60 cm) Pd wire (total surface about 1 cm 2 ). In the lower part of the Pd "U"-shaped cathode, at its centre, a small weight (6g PTFE cylinder) keeps the wire tense during the loading so as to compensate its 4-6% elongation. The anode is a Pt wire: diameter 0.250 mm, length 60 cm, purity >99.99% (Aldrich Chemicals, Germany; provided with an analysis certificate made with the ICP-MS method). A second Pt wire (length 30cm, same type of previous one) is put exactly in the middle of the "U" -shaped cathode for purposes of electrolytic tests. To measure cathode resistance, an AC current (16 mA, 10 kHz, square wave, equivalent to a current density along the wire as high as 800 A/cm 2 ) is superimposed by an array of metallic polyester capacitors to the low intensity electrolysis DC current (2-20 mA). The AC resistance value, resulting from AC voltage drop measurements (about every 10 s), is computed and acquired by a computer. The square-wave AC current, with a low current of about 15 mA, is kept constant by a proper array of fast (ns response) constant current diodes, kept at constant temperature to minimise their strong thermal dependence. A high quality LM135H thermometer (sensitivity 0.05°C), inserted in a PTFE tube, is placed in the middle of the cell, perpendicular to the cathode and anode. A Joule heater (maximum 20 W), used to calibrate the calorimeter, is located between the electrodes in a peripheral position. It is inserted in a PTFE tube (diameter 8 mm and length 30 cm). The cell is slightly pressurised (50mbar at maximum) and thermally insulated. The electrolysis gases and vapours are allowed to flow through both twin cold-traps and silicon oil bubblier before reaching the atmosphere. Corrections for these losses of energy are not yet applied, consequently all the data for excess heat are slightly under-estimated. The heat exchanger within the cell consists in a 500 cm long PTFE pipe, outer/inner diameter 4/2 mm, wound around the PTFE holder through which the cooling water flows. The temperature of the distilled water flowing in the pipe is continuously measured at the inlet and outlet of the heat exchanger with two LM135H thermometers. A computerised peristaltic pump (Masterflex 7550-62) provides a constant flow of distilled water (0.200 ml/s, with day-to-day stability of ± 1 % , routinely measured every 12 h). Since November 2003, the water is pumped from a 4-1 temperature-stabilised bath (brand Thermo NELSAB, model RTE 201), through thermally isolated tubes, kept usually at 24±0.1°C, to which it returns from the cell.

116 18 February 2004.row

0.0045

0.001 1.4x105 1.6x105 1.8x105 2x105 2.2x105 2.4x105 2.6x105 2.8x105 3x105 Time (s)

Figure 1.

Row data. R/Ro and Resistive Temperature Coefficient, versus time (s).

Cell and pump are placed in a small volume temperature stabilised chamber (kept at 24±0.15°C) to further improve the accuracy and reliability of the calorimetric measurements.

4.1. Composition

of the Electrolyte

and Cleaning

Procedure

A mixture, typical 93% volume of heavy ethyl alcohol (C2H5OD) and 7% water (D 2 0), with a total volume of 750ml, was used as electrolyte. The ethyl alcohol (Aldrich) was previously vacuum distilled at 30-35° C with a vacuum distillation system (Buchi, Model 134, Switzerland) to eliminate mainly sodium and iron. It was also ultrafiltered on line using a 100 nm, Millipore PTFE filter. The distillation system was significantly modified in our Frascati Laboratory in order to keep vacuum conditions in a static state. The density was routinely measured by densimeter (Mettler Toledo, Model DA-110M, Japan) before and after distillation, to confirm that no significant H 2 0 contamination occurred during the distillation operations. Heavy water (99.97%) isotopic purity, reactor grade (Ontario-Hydro, Canada)

117

was distilled at 40-45° C in vacuum and ultrafiltered before use, in a procedure similar to t h a t used with the alcohol. Density was measured before and after distillation. T h ( N 0 3 ) 4 (5/15 mg) was added to the electrolyte and the pH of the resulting hydro-alcoholic solution was adjusted to a value of about 3 by adding few drops of (highly concentrated, 14.5 M) HNO3 (diluted in pure D 2 0 ) , in order to avoid uncontrolled precipitation of Th(OD)4. T h e cell was cleaned after each experiment using repeated cycles of water/organic solvents/water/nitric acid/water in an ultrasonic b a t h . After experiment # 2 (February 14, 2003; see Table 1 in Ref. 6), we increased the duration of the immersion of the cell in concentrated (65%) warm (60°C) HNO3 from 2 m i n to 14 h (all night) because we suspected t h a t residual traces of Cs might remain t r a p p e d around the strictly connected spires of the cooling serpentine. 5. R T C M e a s u r e m e n t : P r i n c i p l e s of O p e r a t i o n T h e innovative procedure developed for the, in situ, measurement of the R T C , is based on a regular (every 60-200 s) changing of the intensity of AC current (at 10 kHz) along the wire, from the "low intensity" (15 mA) normally used to measure the R/RQ values, to the "high intensity" (about 120 m A ) . T h e change of current intensity affects the power dissipated in the wire (i.e. from about 7-15 m W at low intensity regime, to about 800 m W at high-intensity regime). Obviously, the dissipated power depends on the wire resistance variation, due to H / D absorption. W h e n the "high intensity" is on, the wire increases its t e m p e r a t u r e . T h e variation is large enough (although the wire is immersed in a good thermally conductive solution, like alcohol-water) to be detected, with accuracy of the order of some percent by our R/RQ measuring system. It is reasonable to assume t h a t during the "high intensity" cycle the actual loading ratio D / P d remains substantially unaffected. It is also reasonable to assume t h a t the relationship: W = hS(Tw-Ta),

(1)

where W is the dissipated power, h the thermal exchange coefficient, S the wire surface, and Tw — Ts is the t e m p e r a t u r e difference between wire and solution, remains substantially unaffected when the wire resistance changes because of the H / D loading. T h e value of the R T C is known for pure P d ( a = 3.8 x 10~ 3 ). It is, therefore, possible to calculate the (1) values at H ( D ) / P d = 0 by applying t h e low-high current intensity: Riow

= Ro(l + aTs),

(2)

^high = .Ro(l + &TW), Rio^/Rhigh

= (1 + aTs)/(l

+

(3) aTw).

(4)

118

In (4) the only unknown is Tw: such a quantity can be calculated as a function of W (Tw = f(W)) and can be determined at any moment of the experiment (in fact, the actual dissipated power at high level of current intensity is always known; consequently, Tw is always known; the only unknown is therefore the value, function of the loading ratio, of the resistive coefficient a). Taking into account that the change of a as a function of loading ratio is very large (>200%, i.e. from 0.38%/°C at D/Pd = 0 to 0.18%/°C at D/Pd = 0.75) the assumed approximations can be largely accepted. 6. Experimental Results on RTC Measurements from experiments # 8 (deuterium experiment) and # 9 (hydrogen experiment) are compared here because the experimental set-up was exactly the same, making cross-comparisons easy. 6.1. Exp. 8 (File 18 February

2004)

Deuterium

In Fig. 1, the R/RQ ratio (vertical left axis, from 0.95 to 2.05), shows over time (from 140,000 to 310,000s), and the Resistive Temperature Coefficient (vertical right axis, from 0.001 to 0.004). The data clearly show the twin values of R/RQ due to the cycles of low power (about 7-15 mW) and high power (700-800 mW) because AC (electromigration) currents injected along the Pd wire. In Fig. 2, the same data of Fig. 1 are shown after off line filtering. This procedure, up to now, has been made selecting the data one-to-one, and is very long time consuming and tedious. Figure 3 is just a magnification of Fig. 2 around the unloading-loading area. In short, the cell was kept in cathodic condition up to time 196,000 {R/RQ = 1.82). At that time the electrolytic current was disconnected and the R/RQ value, after a small decrease, started to increase again because unloading. About RTC, it was at 0.0020-0.0021 during loading condition and decreased to 0.0015 at the point where it reached the maximum of R/RQ. At time 194,000 unloading was forced by anodic current (2 mA). The maximum of R/RQ was 1.94 and not 2.0 as expected because the anodic stripping was quite strong and the unloading was not homogeneous along the wire. At time 204,000 the anodic current was ended and the R/RQ reached the minimum value of about 1.015. This value is little bit larger then 1.00 because some stress, due to previous loading, was induced into the wire. The RTC reached a value of about 0.0037, close to the value reported in the literature (0.0038 at 293 K). From about time 204,000 to time 206,000 the wire was in floating condition and absorbed some deuterium dissolved in the alcohol-water solution. At time 206,000 was given again cathodic current, and it started usual loading. At time 213,500 the R/RQ reached the maximum value, about 2.015, larger than 2.00 because previous residual stress. The RTC reached it minimum value,

119 18 February 2004_corr 0.004

—,—

ll1

V V — —

0.0035

-L

»Ro_d |

5

'

—

7 * ytf

mi

»

™

-TcN_L->H

|

0.002

'xypv" 1

]

-

[JJPPH

0.001

1.4x105 1.6x105 1.8x105 2x105 2.2x105 2.4x105 2.6x105 2.8x105 3x105 Time (s) Figure 2.

Same as Fig. 1, except the data are filtered for clarity.

about 0.0014. The value of 0.0014 is clearly different from previous 0.0015 (at time 198,000) found in our experiment because of forced, rapid inhomogeneous unloading. Such values are lower then that reported in the literature (0.0018). We think that our measurement conditions (electrolysis) are quite different from that reported in the literature (clean gas environment) and can explain the differences. Moreover, we think that the value of heat exchange constant, measured at the beginning of the experiment, increased because of better thermal conductivity. This effect reduced the amount of signal detected. Obviously, more sophisticated analysis and experimental procedures will be needed in the future. At time 290,000 the R/RQ value reached the same value of time 145,000, i.e. 1.90. We can observe that the value of RTC was lower at time 290,000 compared to time 145,000. A possible explanation is that at time 145,000 some excess heat occurred that increased the value of R/RQ of the wire. The effect of excess heat can be seen in Fig. 4. The value was over 200 mW.

120 18 February 2004.cor 0045

0.001 1.9x105

1.95x105

2x10 5

2.05x105

2.1x105

2.15x105

2.2x105

Time (s) Figure 3.

Magnification around maximum-minimum of

R/RQ.

In conclusion, the combined observation of R/RQ and RTC can help strongly to judge if there is some excess heat, especially at low level of power.

6.2. Exp. 9 (File 25 March 2004)-

Hydrogen

Experiment

In Figs. 5 and 6, similar to Figs. 2 and 3, are shown the behaviour over time (in s) of R/RQ (vertical left axis, from 0.95 to 1.85) and RTC (vertical right axis, from 0.001 to 0.0045). The electrolytic current was as low as 5 mA. The wire was previously almost overloaded because of the combined effects of Th and Hg, apart from cathodic current. At time 930,000 it was almost steady at R/R0 = 1.43 from several hours and RTC had a value of about 0.0024. At time 933,000 electrolysis was discontinued, and the wire immediately started to deload. At time 93,340, because the loading was not very fast, we applied anodic current (2 mA).

121 18 February 2004_row 0035

0.004

0.0035

0.003

H

v

0.0025

X

0.002

0.0015

0.001

1.4x105 1.6x105 1.8x105 2x105 2.2x105 2.4x105 2.6x105 2.8x105 3x105 Time (s) Figure 4. Row data. Evidence of excess heat, at time 150,000, in respect to time 290,000. In fact, the R/Ro values were the same but RTC were different.

At time 934,000 the R/RQ value reached it maximum (1.80) and RTC its minimum value (0.0013). The value of RTC was 0.0013 lower than the expected value, 0.0018. As discussed in the previous deuterium experiment 6.1, the differences of values about RTC can be understood if we accept that the heat exchange coefficient of wire with the solution increased a large amount in respect to "time zero" of the experiment, when we calibrate (keeping the value for data elaboration). At time 940,500 the R/RQ reached its minimum value (1.05) and RTC its maximum (about 0.0033): we disconnected the anodic current. We did not wait long in anodic conditions because we were afraid the wire might break from the anodic corrosion effect. The minimum of R/RQ was larger than 1.00. At time 941,500, after observing weak self-loading, we restarted the cathodic current at 5 mA. At time 951,000 the R/RQ reached it maximum value of 1.62 and RTC its minimum of about 0.0017. Both values are different from expected. We think that large differences are due to inhomogeneous loading of the wire.

122 25MAR04 corr 0.0045 1.8 0.004 -

RIR0_d 1.6

if 5

0.0035

1I

1.4

0.003

£& Ifali r

0 0025

f ffl

0.002

1.2 I

I

jpPfrpi w™# IF*"* "H" 0 0015 J„L->H |

1

1 9.3x105

9.4x106

9.5x105

9.6x105

9.7x105

0.001

9.8x105

Time (s) Figure 5.

Filtered data. Hydrogen loading. Overview of 1 day of experiment.

At time 985,000 the loading was still improving (R/Ro about 1.53) and after about other 30,000s reached the value of 1.42, similar to the value that we observed at time 935,000 (the beginning of unloading —> loading cycle). In other words, the mild unloading —> loading cycle did not damage the surface of Palladium wire. 7. Experimental Results with ICP-MS Experimental conditions and ICP-MS analysis results about the nine experiments, up to now performed that included ICP-MS analysis, are reported in Tables 1 and 2. 7.1. Summary of Experiment about 5 X 1 0 1 0 atoms)

#7 (Analysis

by ICP-MS,

1 count

Date: Oct 31, 2003-Dec. 08, 2003. Current: Usually 10mA, up to 50mA along 5 days (Q = 40 x 10 3 C). Overall results: Only few times overloading.

123 25MAR04_corr

0.001 9.3x105

9.35x105

9.4x105

9.45x105

9.5x105 9.55x105

9.6x105

Time (s) Figure 6. Filtered data. Hydrogen loading. Magnification around: minimum (relative) —» maximum (absolute) —> minimum (absolute) —> maximum (relative) of R/RQ.

Reasons: Large 23 Na in C2H5OD (poor vacuum distillation); excessive Pt deposition from anode. Electrolytes (750cm 3 solution: C 2 H 5 OD 93%, D 2 0 7%): • Th(N0 3 ) 4 = 21.5 x 10~ 6 mol (-> 2.6 x 108 counts) recovered 70%. • Hg 2 S0 4 = 3 x 10" 6 mol (-• 3.6 x 107 counts) recovered 3%. Main elements detected by ICP-MS: • • • • • •

Cu = 3.9 xlO 6 ; 6 3 Cu/ 6 5 Cu = 2.0; Cu from Hg? Zn = 3.6 xlO 6 ; Cu/Zn = 1.08. Rb = 2 xlO 4 . Cs = 2 xlO 4 . Pb = 1.25 xlO 6 ; Pb from Hg and/or Th? Bi = 8 xlO 3 .

124 Table 2.

Summary of some of most frequent elements found by IGP-MS in the nine experiments 64

K (93.3)

C u (69.2) 65 C u (30.8) 63/65 = 2.25

Zn Zn 67 Zn 68 Zn

63

39

66

(48.6) (27.9) (4.1) (18.8)

206

P b (24.1) P b (22.1) 208 P b (52.4) 206/208 = 0.46 207

133

Cs (100)

Experiments

31

1

7E3

2.2E6

2.4E6

2.7E6

>3E8

1E4

2

0

0

2.1E6 63/65 = 1.94

1.8E6

6E5

4E5

3

4E5

2.7E6

6.1E6 63/65 = 1.93

1.7E6

1E6

9.5E5

4

2.5E4

1.4E6

2.9E6 63/65 =: 2.05

7.5E3

1.94E5 6.9E5

5

1.8E6

2.3E6

2.4E7 63/65 = 2.07

1.2E7

9E4

1.5E6

6

2.72E7

6.4E6

9.3E8 63/65 = 2.08

5.1E8

2.2E5

2E8 206/208 == 0.39

7

0

0

3.9E6 63/65 =: 2.0

3.6E6

2E4

1.25E6

8

2.5E5

1.6E8

2.7E6 63/65 =: 2.14

1.77E7

3E5

7.1E5 206/208 == 0.49

9

0

0 (found 6.6E6 from C2HsOH)

2.3E6

6.2E6

0

4.5E5

P (100)

63/65 =: 2.12 Background subtracted.

7.2. Summary

of Experiment

#9

Date: March 15, 2004-May 21, 2004. Current: Usually 20mA, up to 50 mA along 15 days (Q = 125 x 103 C). Overall results: Only few times H overloading (R/RQ < 1.45). Reason: Excessive Pt deposition from anode. Electrolytes: (1) 760 cm 3 solution: C 2 H 5 OH = 712 cm 3 , H 2 0 = 36 cm 3 i.e. commercial 95% concentration, ethyl alcohol; D2O = 12cm 3 ; (2) Th(N0 3 ) 4 = 21.5 x 10- 6 mol (-> 2.6 x 108 counts) recovered 60%; (3) Hg 2 S0 4 = 7 x 10- 6 mol (-> 8.4 x 107 counts) recovered 58%. Main elements detected by ICP-MS: • Cu = 2.3 xlO 6 ; 6 3 Cu/ 6 5 Cu = 2.12; Cu from Hg? • Zn = 6.2 XlO6; Cu/Zn = 0.37; Zn from Pd? • Rb = 0.

125

• Cs = 0. • Pb = 44.5 xlO 5 ; Pb from Th and/or Hg? • Bi = 0. 8. Key Parameters and Comments After a lot of tests in different experimental conditions we have identified the following key parameters and experimental conditions to be fulfilled: (a) Deuterium, must be carefully purified, especially if it comes from a liquid compound. (b) Large overloading, for long time (days) and flux of deuterium through the Pd surface. At present the stability of the D overloading in our experiment is not yet completely satisfactory. (c) Presence of "nanostructures" on the Pd surface (Pd + Th + Hg + "impurities" ). (d) Pt dissolution should be avoided. Pt ions are galvanically collected on the Pd surface and inhibit the loading process (a drop in the cathodic overvoltage): at present this important item is not fully controlled in our experiments. (e) Cleaning the Pd surface during the loading process by means of short periods of anodic stripping. 9. Comments on Transmutations • Pb seems to be related to Th: enhanced decay of Th by deuterium? (also according to Cincinnati Group preliminary experiments, by Zr-H, in 1997); • Cu seems to be related to Th, Hg and Pd: fission, according to Takahashi model and, very recent, specific results using Y. Iwamura multilayer device (test on D-Hg-Pd, Japan-Italy joint Collaboration, this Conference); • Zn seems to be related to Pd: fission, according to Takahashi model; • It is necessary to increase, in a large amount, the equivalent cross section. At present, it is about 1 barn on Iwamura Cs to Pr (ion beam deposition) gas experiments, reproducible; up to 600 barn in our electrolytic (Th-Pb, experiments # 6 reported in Tables 1 and 2) but not fully reproducible. 9.1. Comments

about Th-Hg

Addition

• The Th-Hg solutions, at micromolar concentration, give transmutations results "better" than Sr-Hg solutions. • The main drawback is the criticality of Th addition because, quite frequently, it form a thick layer at Pd surface that decrease largely the H (or D) absorption into Pd, at least at room temperature. • The main advantage is the increased mechanical properties of Pd wire: our historic problems of thin wire ruptures disappeared.

126

• T h e combined effect t h e previous two points can be very useful in t h e case of technological application of cold fusion effect: high t e m p e r a t u r e and over-all reliability are needed.

10.

Conclusions

By comparing the results of nine experiments, it can be argued t h a t are real and not an instrumental artefact. In fact, t h e appearance of new elements depend on:

transmutations

(a) deuterium presence (not hydrogen) in t h e P d lattice; (b) deuterium overloading for a long time together with inward-outward deuterium flux trough t h e P d surface; (c) elements added to the electrolyte t h a t are capable of being fixed a n d / o r absorbed at the Pd-electrolyte interface. Work is in progress at I N F N - L N F to build a special calorimeter and electrolytic cell transparent to I R radiation. Such radiation will be detected from a I R thermocamera, model TH7102MX (from N E C , J a p a n ) , provided also with a special optical instrument (Nikon) able t o detect our thin wire (50 /xm) u p to 2 cm of distance. Our goal is to get a "full movie" of t h e "excess heat" coming out from the P d surface, along the experiment, starting from t h e beginning when no excess heat exists, u p to its eventual occurrence. Further work is necessary in order t o reduce t h e dissolution of P t at t h e anode in our cell.

Acknowledgments We are deeply indebted with Prof. Akito Takahashi (Osaka University, J a p a n ) and Dr. Yasuhiro Iwamura (Mitsubishi Heavy Industries, Yokohama, J a p a n ) , for very useful suggestions a n d critical comments about b o t h ICP-MS measurements and discussion about models t h a t can help to explain t h e experimental results. We cannot forget t h e patience, a n d encouragements given t o us from Prof. Jean Paul Biberian, chairman of ICCF11 at Marseille. He waited our paper for really long time, without complaining us too much. At I N F N Frascati National Laboratory we cannot forget the enthusiastic support given to our experimental activity, in all aspects, by Dr. Sergio Bertolucci (Main Director of the Laboratory) and Dr. Lucia Votano (Director of Research D e p a r t m e n t ) . At I N F N t h e F R E E T H A I experiment was performed with a grant from G V National Group.

References 1. 2. 3. 4.

Y. Y. Y. Y.

Iwamura, Iwamura, Iwamura, Iwamura,

et et et et

al, al, al, al,

Jpn. J. Appl. Phys. 4 1 , 4642-4648 (2002). Proceedings of the ICCF10, 24-29 August 2003. The 5th Meeting of Japan CF-Research Society (JCF5). Proceedings of the ICCF11, 2004.

127 5. F. Celani, et al, The Ith Meeting of Japan CF-Research Society (JCF4), October 17-18, 2002, Iwate University, Japan, pp. 17-21. 6. F. Celani, et al, Proceedings of the ICCF10, 24-29 August 2003. 7. F. Celani, et al, The 5th Meeting of Japan CF-Research Society (JCF5). 8. F. Celani, et al, Condensed matter nuclear science, in Proceedings of the ICCF9, 19-25 May 2002, Beijing, China, pp. 29-35; ISBN 7-302-06489-X. 9. M. Fleischmann, S. Pons, and M. Hawkins, J. Electroanal Chem. 261, 301-308 (1989). 10. F. Celani, et al, Condensed matter nuclear science, in Proceedings of the ICCF9, 19-25 May 2002, Beijing, China, pp. 36-41; ISBN 7-302-06489-X. 11. Y. Arata and Yue Chang Zhang, Proceedings of the ICCF10, 24-29 August 2003. 12. F. Celani, P.G. Sona, and A. Mancini, Proceedings of the ICCF7, April 19-24, 1998, Vancouver, Canada, pp. 56-61.

E M E R G E N C E OF A H I G H - T E M P E R A T U R E S U P E R C O N D U C T I V I T Y IN H Y D R O G E N CYCLED P d C O M P O U N D S AS A N E V I D E N C E FOR S U P E R S T O I H I O M E T R I C H / D SITES

ANDREI LIPSON, CARLOS CASTANO, AND G E O R G E MILEY University

of Illinois

at Urbana-Champaign,

IL,

USA

ANDREI LIPSON AND BORIS LYAKHOV Insititute

of Physical

Chemistry,

RAS,

Moscow,

Russia

ALEXANDER MITIN P. Kapitza

Institute

for Physical

Problems,

RAS,

Moscow,

Russia

Transport and magnetic properties of hydrogen cycled PdHz and P d / P d O H ^ (x ~ (4/6) X 10 —4 ) nano-composite consisting of a Pd matrix with hydrogen trapped inside dislocation cores have been studied. The results suggest emergence of a high-temperature superconductivity state of a condensed hydrogen phase confined inside deep dislocation cores in the Pd matrix. The possible role of hydrogen/deuterium filled dislocation nano-tubes is discussed. These dislocation cores could be considered as active centers of LENR triggering due to (i) short D-D separation distance (~Bohr radius); (ii) high-local D-loading in the Pd and the corresponding effective lattice compression; (iii) a large optic phonon energy resulting in a most effective lattice-nuclei energy transfer.

1. Introduction The study of the trapping and interaction of hydrogen atoms inside metallic nano-structures (dislocation cores or "nanotubes") is important in order to shed light on the possible origin of LENR in highly loaded palladium hydrides/deuterides. Recently, Ashcroft presented a strong argument that hydrogen dominant metallic alloys, might demonstrate high-temperature superconductivity (HTS) even in a modest external pressure range. 1 According to Ashcroft's reasoning, HTS of highly loaded metal hydrides would be due to the overlapping of metal-hydrogen bands. The high-electron concentration and optical phonon energy result in a strong electron-phonon coupling. The advantage of hydrogen dominant hydrides for achieving HTS is in a chemical sense they have already undergone a sort of "precompression," and once impelled by external pressure enter into metallic phase, the electrons from both the hydrogen and the metal may participate in common overlapping bands. There is, however, another approach for achieving a compressed metal hydride 128

129

with a high-coupling constant, that may serve as a model alloy to search for HTS in hydrogen-dominant metallic systems. The deep dislocation core filled with hydrogen atoms could be considered as a sort of "nanotube" with an effective diameter of about two Pd Burgers vectors. Moreover, it is implied that within the deep dislocation core (R^ < b) the local hydrogen concentration might be large enough to provide the ability of overloading the Pd beyond the x = l. 2 Notice that the pressure inside such deep cores of the edge dislocation would be comparable to the local palladium bulk modulus, i.e., up to 100 GPa. Therefore, both conditions for hydrogen "precompression" and external pressure impelling would be satisfied. If hydrogen is trapped inside deep dislocation core sites it is anticipated that these sites could be characterized by - superstoichiometric hydrogen loading (loading ratio x = H/Pd > 1.0 or precipitation of condensed metallic hydrogen in the form of H„, (H2) n ; - high-local pressure/compression P ~ 120 GPa; - high-optic phonon frequency (hu> ~ 120 meV); - strong electron-phonon coupling (Ae-ph ~ 1.0); - strong Pd-H(D) band overlapping. In summary, such enhanced parameters of condensed hydrogen in Pd lattice suggest an excellent conditions to achieve both HTS and triggering LENR processes (to be active sites of cold fusion in Pd lattice). The observation of a diamagnetic/superconducting behavior from a hydrogen phase precipitated inside the dislocations confined to the Pd matrix could be feasible when two referenced conditions to dislocation network are satisfied:3 (1) there is a sufficiently large number of dislocations within the Pd crystal that contain tightly bound compressed hydrogen and (2) the network of dislocations is organized in the form of closed loops, which can carry a persistent current. In this work, we studied the structural and magnetic properties of PdHa, single crystal and Pd/PdO:!!^ heterostructure samples with a very low-average loading ratio of (x) ~ (4.5 - 6.0) x 10" 4 that were produced by cycling the pristine Pd single crystal or Pd/PdO in a H2 atmosphere or during electrolysis, respectively. Thermal desorption analysis (TDA) showed that the residual hydrogen, precipitates as a condensed phase within the deep dislocation cores. Magnetic measurements recorded the appearance of a strong diamagnetic contribution of the condensed hydrogen phase [PdH^-Pd] in the Pd matrix at T < 30 K (in a weak magnetic field, H < 5.0 Oe), and an antiferromagnetic behavior in the higher magnetic field.3'4 Transport and magnetic measurements performed with Pd/PdO:H x samples provide direct evidence of superconducting transition at T < 70 K. 2. E x p e r i m e n t a l To minimize the effect of impurities on the magnetic properties of the PdH^ system, a 99.999% pure Pd-single crystal produced by Metal Crystals and Oxides of

130

Cambridge, UK was employed in this work. The cylindrical ingot was grown using the Czochralski method with a [110] axis, a length of 10 cm, and a diameter of ~ 1.0 cm. The sample with dimensions of 2.7mm 3 x 2.7 mm 3 x 0.6 mm 3 and a weight of 52 mg. was cut from the as-grown ingot by using a low-speed diamond saw and mechanically polished to remove surface irregularities. Then, this single crystal was annealed in a high vacuum (p = 10" 8 Torr) at a temperature of 800°C for 5h. The annealed pristine single Pd crystal served as a reference and is labeled the "background" sample or (bgr), and the appropriate TDA and magnetic measurements were conducted on this sample prior to the hydrogen cycling. In order to create a condensed hydrogen phase in the Pd sample, a H2 gas cycling procedure was applied. The sample was loaded and degassed twice. The pristine Pd(bgr) sample (after TDA and magnetic measurements) was loaded at a pressure of 930 Torr at a temperature of 390 K and degassed in a vacuum of 1 0 - 6 Torr at a temperature of about 400-430 K. After the cycling, the PdH^ sample was subjected to a final annealing at a temperature of 570 K and a pressure of 10~ 8 Torr for 2h. This post-cycled PdH^ crystal is labeled "foreground" sample or (fg). The cold worked 12.5//m Pd foil (99.95%) of Nilaco, Japan has served as a basis for Pd/PdO heterostructure production. In accordance with ICP analysis performed by producer, the concentration of ferromagnetic impurities (mainly Fe) was found to be 700°C, the hydrogen desorption (curve b) increased and the post-cycled (fg) sample produced a pronounced peak with maximum at a temperature of 870° C. (Notice that the TDA peak in this experiment was not completed because the heater in our apparatus could not heat the sample beyond 900°C.) The analysis showed that H 2 pressure at 870° C is about six times higher than that observed at the same temperature range in the background measurements. Integration of the PdH x (fg) H2 peak (with the background subtracting), and comparison of its area with calibration data obtained for TiH 2 powder allowed the estimation of the effective loading ratio x = H/Pd, averaged over the PdHa:(fg) sample volume as x — (4.5±0.5) x 10~ 4 . The averaged value of x is smaller than expected for any known stable phase of Pd hydride. Considering that after the H 2 cycling, the sample underwent additional annealing at a

133

temperature of 570 K, which caused the decomposition of the residual a-phase in the lattice, we assume that all remaining hydrogen detected in the cycled Pd single crystal is located in the dislocations, but not in the regular lattice. Extrapolating the results of SANS measurements performed in Ref. 2, for various residual hydrogen concentrations (x > 1.7 x 10~3) in Pd polycrystal it is possible to estimate a radius Ru of residual hydrogen distribution with respect to dislocation cores. At (x) = H/Pd = 4.5 x 10~ 4 in our sample, the P H ~ 2.75 A is close to the Burgers vector or minimal radius of H capture in Pd.

1.20E-0071.00E-007S

8.00E-008-

=

6.00E-008-

CO 0

°-

4.00E-008-

I 2.00E-008 0.00E+00O 0

200

400

600

800

Temperature (°C)

Figure 1.

In order to estimate the binding energy of hydrogen within the Pd, the hydrogen activation energy was calculated. The formal kinetics Garlick-Gibson model may be used to estimate the kinetic parameters of second-order thermal activation processes by accounting for the rising part of the hydrogen TDA peak: 2 ' 5 £R~ = ks^Ti/(T2 —Ty)} ln(P2/P\), where fee is the Boltzmann constant; Pi, P2 and T 1; T2 are the hydrogen pressures and temperatures, respectively, corresponding to those pressures, at the rising portion of the desorption curve (so that T2 > 7\ and Pi > Pi)- In accordance with the formula, the activation energy of the desorption which reflects the effective binding energy of hydrogen atoms in the Pd lattice was found to be e H = 1.6 ± 0.2 eV. The magnitude of activation energy of the hydrogen desorption that was found for the PdH x (fg) crystal is well above the H-trapping activation energy derived for H-cycled polycrystalline cold-worked Pd that has an extremely low-bulk hydrogen concentration captured inside the deep dislocation cores (en ~ 0.7eV). 2 ' 3 This provides us with a strong argument that the hydrogen is bound solely inside the deepest core sites. Thus, the TDA results show that the atoms of hydrogen are tightly bound inside the deep dislocation cores and can be fully removed only at

134

very high temperatures (T > 1000°C), which would suggest a full recrystallization of the Pd sample. The results of TDA allow us to conclude that PdHj; samples that are produced by H 2 gas cycling of the Pd single crystals followed by annealing at T = 570 K contain no hydrogen atoms in their crystalline lattice except those that are tightly bound inside the deepest dislocation cores (or "dislocation nanotubes") with a minimal radius RH = 2.75 A. Thermal desorption analysis performed with Pd/PdO(bgr) samples in the temperature range of 20-700° C showed an absence of any hydrogen desorption above the background level. The hydrogen pressure in a vacuum chamber in the whole studied temperature interval did not exceed 2 x 10~ 8 Torr (Fig. 2, curve a). This H2 pressure value is in good agreement with a residual hydrogen background in the chamber. In similar measurements performed with Pd/PdOirL^fg) sample, a broad H 2 thermal desorption peak with T m a x = 430° C is appeared and pressure still not returns to the background level up to T = 700° C (Fig. 2, curve b). Comparative TDA of a the Pd/PdO:H x (fg) main peak with T r a a x = 430° C (with subtraction of H2 background) and calibration peak obtained for TiH2 powder allowed to estimate an effective average loading ratio x = H/Pd in Pd/PdO:H :r (fg) sample, which was found to be (x) = (5.5 ± 0.5) x 10~ 4 . This effective value of x = H/Pd is rather smaller than that which could be expected for any known stable phase of Pd hydride. 2 Taking into account that after the H2 cycling the sample was underwent to additional annealing at T = 573 K caused a decomposition of remaining PdHj; aphase, it is reasonable to assume that all residual hydrogen remaining in the cycled sample is tightly bound with dislocations, but not in the regular lattice.

1.60E-0071.40E-007•£ 1.20E-007==- 1 .OOE-0073 8.00E-008CO

i . 6.00E-008C\i

1

4.00E-0082.00E-008O.OOE+0000

100

200

300 400 500 Temperature (°C)

600

700

800

Figure 2.

Thermal activation analysis of the rising part of the TDA curve with subtracting of a hydrogen background in the chamber gives an activation energy of the Hatom desorption e H = 0.63 ± 0.10 eV. The magnitude of this activation energy is well above the H-trapping activation energy in the H2-cycled single crystal Pd,

135

with residual a-phase and is in good agreement with the activation energy of Htrapping in polycrystalline Pd after electrochemical cycling. Notice that dislocations in Pd/PdOiHz are mainly stored at the interface between Pd and PdO because the maximum temperature of the hydrogen desorption peak (T m = 430° C) is still lower than that in a bulk Pd cycled with hydrogen (T ~ 800°C). 5 The traces of a bulk hydrogen presence in the Pd/PdOiHz sample could be seen as broad noncompletely resolved maxima in TDA spectrum (Fig. lb) at T = 540 and 640°C. Thus, we can conclude that H-atoms in Pd/PdOrH^ tend to trapping mainly inside deep dislocation cores at the boundary between Pd and PdO (where dislocation concentration must reach maximum value compared to Pd bulk). A large hydrogen binding energy £H = 0.63eV/atH is well above the H-binding energy or enthalpy of stoichiometric Pd hydride formation (| AH\ ~ 0.18 eV/at H), suggesting a significant Pd-H common band overlapping at the Pd-PdO boundary inside deep dislocation cores. TDA data allow to propose a simple model for the deep edge dislocation core filled with trapped hydrogen (Fig. 3). In accordance with Figs. 1 and 2 after the H-cycling, data the regular lattice of Pd contains no hydride and thus, all residual hydrogen is localized inside the deep core of the edge dislocation (in the direction of Pd [121]) determined by Burgers vector b[l 0 1 ] = 2.75 A. Depending on average residual concentration (x) and dislocation density Nd, the effective loading ratio inside the deep core in Pd f.c.c. lattice is determined by simple formula: xeS = \/2(x} /Ndxb2. Accordingly, to this formula at Nd ~ (1.0 2.0) x 10 11 cm~ 2 4 eS and (x) ~ (4 —6) x 10~ the x would be in the range of 1.0 < a;eff < 3.0, suggesting superstoichiometric hydride formation in the deep dislocation cores. 3.2. Magnetic

Measurement

Results for

PdHx

The ZFC experiment in the DC-mode showed that in a weak magnetic field (H = 0.5Oe), the magnetic moment of H2-cycled PdH^(fg) samples in the temperature range of 2 < T < 50 K is significantly lower than M(T) for the original Pd(bgr) single crystals (Fig. 4). The difference between the moments of the cycled sample and the original single crystal [PdH^-Pd] reflects the contribution of the condensed hydrogen phase inside the deep dislocation cores. This difference tends to increase at T < 30 K. There was no temperature dependence observed for the [PdH^-Pd] difference of moments in FC measurements at high-magnetic field (H = 1000 Oe) measurements. The moment of the PdH x is about 4-5% lower than that of the Pd(bgr). Thus, a nearly constant negative AM difference is detected within all temperature ranges 2-298 K. In order to ascertain the origins of the dramatic decrease in the paramagnetic properties of PdH^ compared to the Pd single crystal at T < 30 K, we studied the magnetization of the PdHa:(fg) and Pd(bg) samples as a function of the applied magnetic field. These measurements were performed at constant temperatures T = 2.0, 10.0, 50.0, 100.0 and 298K in magnetic fields ranging from 0 to ±200 Oe. The M(H) dependences at T = 2.0 K for the Pd(bgr) and PdH^fg) samples that

136

Pd (121) 1D(T01) = 2 . 7 5 A Figure 3.

are presented in (Fig. 6) have an essentially different character, especially below H ~ 10.0 Oe: (1) The total width of the hysteresis loop at M = 0 for the PdHE(fg) sample (AH = 7.45 Oe) is much higher than that of the Pd(bgr) single crystal {AH = 3.7 0e).

0.00006- , 0.00005-

A. , . . ...

«

.

t

.

.

*1,

0.00002-

•

—©~Pd/PdHx (fgi * Pd(bgr) . . -* PdHx-Pd

0.00004; 0.00003-

'

.

•" V• *

.

i

.

.

. 1.0) sites suggesting high-temperature superconducting HTS properties. In view of our findings, the deep dislocation core sites produced during deuterium loading in Pd could be referenced to the active centers of LENR triggering, which assume significant enhancement of DD-fusion probability in Pd compared to the regular lattice sites. Triggering of LENR at dislocation core sites would be strongly appreciated due to: - shortest DD-distance (close to Bohr radius as it predicted for metallic deuterium) ; - highest D-loading accompanied by a lattice compression comparable with Pd share modulus; - high-electron concentration resulting in a most effective electron screening of deuterons in Pd; - large optic phonon energy (huiv ~ 120 meV) resulting in a most effective lattice-nuclei energy transfer. Acknowledgements This work was partially supported by the NSF under Grant No. DMR-9982520. The magnetic measurements were performed at the Center for Microanalysis of

146

Materials at t h e Frederick Seitz Material Research Laboratory at UIUC. This facility is supported by the U.S. Department of Energy under Grant No. DEFG02-91ER45439.

References 1. N. W. Ashcroft, Phys. Rev. Lett. 92, 187002 (2004). 2. B. J. Heuser and J. S. King, Metal. Mat. Trans. A29, 1594-1598 (1998). 3. G. Lipson, A. Bezryadin, C. H. Castano and G. H. Miley, Bullet. APS 48 (1), 983 (2003). 4. G. Lipson, B. J. Heuser, C. H. Castano, A. Celic and G. H. Miley, J. Phys.: Cond. Matt, (submitted) 5. D. A. Van Leeuwen, J. M. Van Reitebeek, G. Schmidt, and L. J. De Jongh, Phys. Lett. A170, 325 (1992).

CALORIMETRY OF E N E R G Y - E F F I C I E N T GLOW D I S C H A R G E A P P A R A T U S D E S I G N A N D CALIBRATION

THOMAS B. BENSON The Greenview Group, Pleasanton, CA, USA E-mail: [email protected] THOMAS O. PASSELL TOP Consulting, Palo Alto, CA, USA E-mail: [email protected]

1. Introduction This work aims to develop a "family" of low-powered calorimetrically accurate glow discharge units, similar to that reported by Dardik et al. at ICCF-10, and to use these to test a wide range of cathode materials, electrode coatings, gas types, gas pressures, and power input levels. We will describe the design and calibration of these units. The strategy is to use a large number of very similar units so that the calorimetric response does not vary significantly for a given power level. The design is metal or sealed glass cylindrical tubes, charged with 0.4-50 Torr mixtures of deuterium, hydrogen, argon, or helium gases. Units operate from 0.2 to > 2 W power input. The units have low mass (2W. Configuration 1. Brass or other metal pipe approximately 40 mm in length and 30 mm in diameter serves as a cylindrical outer cathode, with an axial inner rod of

149

1-2-mm diameter, running the length of the cylinder, as the anode. The device is made vacuum-tight and includes a viewing window to monitor the progress of the glow. The anode and cathode are composed of a variety of materials. Configuration 2. Similar to configuration 1, except the entire system is enclosed in a sealed glass tube enclosure. Rolled metal foil or metal mesh is inserted into the glass sleeve, forming a cylindrical outer cathode of approximately 12-mm diameter and 100-mm length. The anode is an axial inner rod of 1-2-mm diameter, running the length of the cylinder. The anode and cathode are composed of a variety of materials. Configuration 3. A glass tube enclosure is built, encasing two opposing "hollow cup" electrodes of approximately 30-mm length and 10-mm inner diameter. Electrodes are made from tungsten with a variety of coatings. The opposing electrodes are linked by a glass tube surrounded by a wire coil; this coil is energized with RF high-frequency signal which ionizes the gas in a highly efficient manner. Highvoltage current of various types (DC, AC, and Modulated DC) is applied between the electrodes to further ionize the gas.

- RF ionizing coil

til)

Electrode connecting wire

Left fi'j.ti'.-J.in glass sleeve

Figure 2.

2.3. Gas

Opposing "cup" electrodes in sealed glass.

Handling

In configuration 1 (brass tube), standard commercial parts are assembled and made sufficiently vacuum leak tight to allow a bleed-through system to maintain 1-50 Toripressures of chosen gases or gas mixtures in a number of tubes simultaneously from a single manifold without significant levels of atmospheric or internal off-gas impurities to interfere. Configurations 2 and 3 use an initial fixed gas pressure in a sealed glass envelope which is then independent of a vacuum system or a gas

150

0 Figure 3.

Cutaway drawing of cup electrode in glass.

source and thus can be more fully enclosed for more accurate calorimetry than is practicable in the first method. 2.4. Driver

Power

Supply

The driving power for the glow discharge comes from DC high voltage (700-1200 V) or variable waveform, multi-frequency modulated DC (700-1200 V, with modulation of ±300 V in frequencies from 5 to 5000 Hz). Also an AC high-voltage unit (40 kHz, 1000 V) is used to help ionize the gas with RF energy. In all cases these driver power supplies are miniaturized and encapsulated in a small metal enclosure (on the order of a 10-cm long, 5-cm wide) which is used in online, continuous calorimetry in order to measure the true power delivered into the tube. Input power to these driver power supplied comes from regulated DC power supplies with voltage and current measured to 3-figure accuracy. The total system is capable of 3 W output but can be run stably down to less than 0.2 W by reducing the input DC. 2.5. Accurate

Measurement

of Input

Power

The calorimetry of the a low-power glow discharge system can be done with acceptable accuracy, as long as we understand some key issues related to input power. • Issue 1. In order to do calorimetry on a glow discharge system, some method of measuring input electrical power must be employed. • Issue 2. Direct measurement of "driving" power is extremely difficult, given the complex unpredictable waveforms of a glow discharge system. Dardik et al.1 has reported using a 50kHz-l MHz sampling data collection system to measure the current and voltage of a complex high-voltage AC (or modulated DC) circuit. This approach is quite expensive and requires an unusual degree

151

of effort, which many researchers may not be able to afford. Therefore, direct measurement of the AC/modulated DC power may be impractical. • Issue 3. Measure simple DC power into the power supply. The solution therefore is to measure the DC power flowing into the driver power supply that creates the high-voltage AC (or modulated DC) power. For example, in our system, we might input approximately 6 V DC, 150 mA (about 0.9 W) of steady electric current into our oscillator/amplifier circuit, which converts this DC input into high-voltage modulated DC or AC. This approach, however, requires that we understand how much of that input power is being delivered to the glow tube, and how much is lost as waste heat in the electronic circuit. • Issue 4- Conversion efficiency of the power supply varies constantly. For any given glow discharge, the amount of power lost as waste heat in the driver power supply can fluctuate from 30 to 90%. This conversion efficiency is based on gas pressure, temperature, power level, gas composition (D2, H2, or Argon), surface conditions of the electrodes (metals in glow discharge will sputter, dynamically creating and destroying nanopores in the electrode surface and changing the ionization potential of the system) and many other factors. Essentially there is no way to predict or rely on any predicable power supply conversion factor. • Solution. Real-time calorimetry of power supplies. In our system, we miniaturize all power supplies and perform calorimetry on them in parallel (usually with the same apparatus) as the glow tube itself. Therefore, we can very easily measure the low-voltage DC power and current into the system, giving us an unambiguous energy input {Qi), then measure the proportion of this DC input power that is lost as waste heat in the power supply (Q p ). The total energy into the glow tube (Q a ) can easily be seen as Q\ — Q p . In Fig. 4, we see data from our "zero" control tubes; input power Qi can be compared to the sum of the two calorimetrically measured powers Q a and Q p , demonstrating an accuracy of better than 5% accuracy. If there were excess heat being produced in the tube, we would see it easily as excess in Q a .

Power in

Power out (calculated from calorimeter)

Qi DC (W)

Q p (power supply loss) Q a (tube)

Total DC power (%)

0.1378 0.988

0.1246 90.39% 0.5854 59.25%

0.138 1.00

Figure 4.

0.0138 10.00% 0.4142 41.92%

100.39 101.17

Typical breakdown of power to supply or glow tube (control tube).

152

2.6. Calorimetry

of Tube

Calorimetry is based on measurement of equilibrium temperature of the device in air. We use two methods, based on screening speed vs. accuracy of measurement. Calorimetry Method 1. Both glow tube and power supply are enclosed in separate copper or aluminum enclosures, with the only connection being wires to conduct the low- or high-voltage currents. Thermocouples and/or thermistors (5kfi, 2% precision) are attached to the outside of the metal enclosures, which measure the equilibrium temperature reached in the ambient air environment. Calibrating heaters are installed inside the metal enclosure (adjacent to the glow tube and/or power supply) and the system is calibrated for several given levels of power. In some cases a thermal fluid is poured into the enclosure to ensure even distribution of heat inside the unit. This method is very fast, for large-scale screening of systems, but provides at best 10% uncertainty (and at worst 30%) because of low-temperature differences between the tube and ambient, variable air currents. Calorimetry Method 2. The above metal enclosures are inserted into a thick insulating body (either Dewar or Styrofoam) with only a small portion of the metal exposed to the air. This insulated body is placed in a controlled laminar airflow box with temperature-controlled ±0.1°C. The temperature of the exposed metal surface is measured by thermocouple or thermistor at equilibrium, compared to ambient. System is again calibrated for a number of input powers to an internal heater. This is similar in principle to the "double wall isoperibolic calorimetry" as described by Storms. 2 Because the heat flux is concentrated into a smaller, more controllable point of exposed metal, and the airflow and temperature are more precisely controlled, this method provides much better accuracy, with uncertainty less than 5%. Note that any number of other calorimetric approaches such as liquid isoperibolic, flow, or Seebeck, could be used with equal or better success to the above. Refer again to Storms 2 for discussion of relative merits. Data are collected using a Keithley 2700 40-channel data collection device, which feeds data into MS Excel spreadsheet on any PC. This system is simple to use and robust.

2.7. RF

Noise

Another common problem of the glow-discharge system is RF noise from the highvoltage signal or RF ionization signal. In our system this problem is avoided by enclosing both glow tube and power supply in copper or aluminum enclosures. These shield the external thermocouples or thermistors from RF noise, and also, as described above, serve as integral parts of the calorimeter. Finally, the Keithley data collection system are quite effective at filtering stray RF noise.

153

2.8. Time

Constants

Time constants vary from 30 to 45 min when exposed with no insulation to ambient air to 3h, encased in insulation, in thermally controlled enclosure. Time constants are defined as the time to reach thermal equilibrium for steady input power. Our approach assumes the excess heat, when and if detected, will be of sufficiently long and steady nature to be observable with systems with relatively long time constants. Highly fluctuating excess power would be detectable but not easily quantified by these methods. 3. Results Figures 5 and 6 give the time history of temperature of the calorimeters surrounding the power supplies feeding 40 khz AC to two "Configuration 1" tubes. These temperature differences, while generally precise to ~ 0.2°C are accurate to ~ 0.5°C. The differences in the various thermistors are not significant when looking for > 50% values of excess heat and are the primary cause of the uncertainty of up to 30% in these measurements. Of the four tubes operated simultaneously, thermal balances varied from 87 to 111% where these percentages are the observed energies in the discharges divided by the power delivered to the tubes. Similar runs were performed with Configurations 2 and 3 tubes, with similar results. No heat was seen in excess of known uncertainties. Figures 7-10 are photos of a selected set of power supplies and tubes used or developed in this research effort. 4. Discussion The above results are the first set of measurements attempted by this approach and the absence of excess heat is not unexpected. We purposely tried a much simplified experiment from that described by Dardik et al.,1 but with a great number of changes to process variables to allow us to see a wide range of behaviors in the apparatus. In the first place, we tried placing the active metals (Pd) leaf wrapped around the center electrode, with none on the cylindrical electrode surface. Then we tested palladium powders and wire. Furthermore, we used AC power, DC power, and 3-frequency modulated DC power to drive the glow. Finally, we used mixtures of argon and deuterium, in addition to pure deuterium, to test a variety of discharge modes. Different configurations were shown to have different advantages: (1) Configuration 1 (gas bleed through system) allows the measurement of any product gases generated within the discharges to be captured and subjected to various measurements of composition on the fly. It comes to equilibrium quickly and is quicker to build, so that very large numbers of materials to be screened very quickly, although it is less accurate. (2) Configurations 2 and 3 (sealed glass tubes) have important advantages. They do not need a continuously operating vacuum system or gas supply.

154

They can employ much cleaner electrodes, more free of surface contamination. Their calorimetry is slightly more complex with a longer time to reach thermal equilibrium, but they are small enough to fit inside a small temperature-controlled chamber, which gives them greater accuracy. Additionally the sealed glass tube can be run for extended periods of time, allowing product gases to accumulate to higher concentrations. 5. Conclusions Methods have been developed for efficient screening of candidate materials on electrode surfaces of glow discharge tubes that will provide an environment conducive to nuclear reactions between deuterium and itself or other light elements. This method can detect excess heats ratios >1.2 with more than 95% certainty. It provides a valuable new platform for large-scale exploration of excess heat effects in the gas phase, using low-power inputs in the 0-3 W range. This method proves to be inexpensive, quick, accurate, and easy to perform once the basics are mastered. Note: The authors are interested in testing electrode materials from other sources, especially those that have already been successful in a liquid (electrolytic) environment. Also we would be pleased to provide advice or equipment to other researchers. Interested parties please contact the authors. Acknowledgments We are indebted to many of our colleagues in the research community for helpful discussions. In particular Arik El-Boher, Mike McKubre, and Russell George were generous in answering our questions. Bill McCarthy with whom we share laboratory space and Ed Wills have been helpful in building our apparatus. The skills of persons employed by Microscientific Glass Blowing Co. of Milpitas, CA have contributed to apparatus design. References 1. I. Dardik, H. Branover, A. El-Boher, D. Gazit, E. Golbreich, E. Greenspan, A. Kapusta, B. Khachatorov, V. Krakov, S. Lesin, B. Michailovich, G. Shani, and T. Zilov, Intensification of low energy nuclear reactions using superwave excitation, in Proceedings of the Tenth International Conference on Cold Fusion (Camridge, MA, USA, August, 2003); Text available at LENR-CANR.org. 2. E. Storms, Calorimetry 101 for Cold Fusion; Methods, Problems and Errors, 2004; LENR-CANR.org.

155

Additional Lists of Figures

O CO CD Q.

- P w r 4#1 -Ambient

E

T /

•

•

!

•

•

j •

fcj3***.f*T«+*

M "

v v i. •

.

i

18-, 50

100

150

200

250

Time (min) Figure 5. Temperature vs. time for the calorimeter surrounding the power supply for tube # 1 . Along with ambient temperature.

156

O = 23 •

»

-•• » • *

f

—

•

*

*

,/

18, 0

50

100

150

200

250

Time (min) Figure 6. Temperature vs. time for the calorimeter surrounding the power supply for tube # 2 . Along with ambient temperature.

%

m

-•**?s*r

Figure 7.

Photo of D C / A C high-voltage driver power supplies.

158

Figure 8.

Photo of brass pipe, flow-through type glow tube.

159

Figure 9.

Photo of sealed-glass type glow tube.

160

Figure 10.

Photo of sealed-glass type glow tube (mounted on vacuum station).

G E N E R A T I O N OF HEAT A N D P R O D U C T S D U R I N G P L A S M A ELECTROLYSIS

TADAHIKO MIZUNO AND YOSHIAKI AOKI Hokkaido

University,

Kita-ku

Kita-13

Nishi-8,

Sapporo

060-8628,

Japan

DAVID Y. C H U N G A N D F . S E S F T E L Department

of Physics,

Howard

University,

Washington

DC 20059,

USA

Direct decomposition of water is very difficult in normal conditions. Hydrogen gas can be usually obtained by electrolysis or by a pyrolysis reaction at high temperatures, starting at approximately 3700°C. However, as we have already reported, anomalous heat generation can occur during plasma electrolysis, and this process makes it rather easy to achieve both electrolysis and pyrolysis simultaneously. In this paper we describe anomalous amounts of hydrogen and oxygen gas generated during plasma electrolysis. The generation of hydrogen far in excess of amounts predicted by Faraday's law is continuously observed when conditions such as temperature, current density, input voltage, and electrode surface are suitable. NonFaraday generation of hydrogen gas sometimes produces more than 80 times as much hydrogen as normal electrolysis does. Unfortunately there have been few claimed replications of excess hydrogen, even in rare cases in which excess heat is claimed. In most cases, no excess heat or hydrogen is observed. The reaction products found after electrolysis were different after excess heat generation.

Keywords: plasma electrolysis, pyrolysis, hydrogen generation, transmutation 1. Introduction Hydrogen gas can be easily obtained by electrolysis. However, direct decomposition of water is very difficult in normal conditions. The pyrolysis reaction occurs at high temperatures, starting at approximately 3700°C. 1,2 We have already reported anomalous heat generation during plasma electrolysis.3'4 Some researchers have attempted to replicate the phenomenon, however, they report difficulty generating a high level of excess heat. They tend to increase input voltage to a very high value, hundreds of volts, a technique we do not recommend. It is very important to replicate the excess heat and other products during plasma electrolysis. The generation of hydrogen in excess of Faraday's law is continuously observed then conditions such as temperature, current density, input voltage and electrode surface are suitable. Non-Faraday generation of hydrogen gas sometimes produces more than 80 times as much hydrogen as normal electrolysis does. Usually, the plasma state can be easily started if input voltage is increased up to 140 V at a rather high temperature. 5 ~ 7 when the plasma forms, a great deal of vapor and 161

162

hydrogen gas are released from the cell. At the same time, this effluent gas removes heat (enthalpy), which then cannot be detected with calorimetry based only on temperature. It is difficult to calibrate the exact enthalpy balance. A mixture of gas and heat is especially complicated and difficult to measure. In this paper we describe a heat measurement system used to during plasma electrolysis that accounts for all enthalpy. 2. Experiment 2.1. Electrolysis

Cell

Figure 1 shows the experimental set up. We measure many parameters including sample surface temperature, neutron and X-ray emission, mass spectrum of gas, input and output power, and so on. Figure 2 shows the schematic sketch of the cell and measurement system. 1 ' 2 The cell is made of the Pyrex glass 10 cm diameter and 17 cm in height and 1000 cm 3 in solution capacity. It is closed with a Teflon rubber cap, 7cm in diameter. The cap has several holes in it, three for platinum resistance temperature detectors (RTD) (Netsushin Co., Plamic Pt-lOOfi), two for the inlet and outlet of the flowing coolant water, and one to hold a funnel that captures the effluent gas from the cathode. The funnel is made of quartz glass, and is 5 cm in the diameter at the top of the cell, and 12 cm in length. Gas leaving the top of the funnel flows into a water-cooled condenser, which is connected to the funnel with another Teflon rubber cap. This is shown in Figs. 3 and 4. 3. Measurement of Hydrogen Gas A mixture of steam, hydrogen and oxygen (from pyrolysis) passes from the cell to the condenser. The steam condenses and falls back into the cell. An 8-mm diameter Tigon tube is coupled with the gas exit of the condenser, connecting it to a gas flow meter (Kofloc Co., Model 3100, controller: Kofloc Co., Model CR-700). The flow to voltage transformer element is a heated tube of thermal flow meter system, the minimum detection rate of hydrogen gas flow is 0.001 cm 3 /s, and the resolution is within 1%. The power output from the measurement system was led to the computer through a logger. After path through the flow meter, the gas goes to a mass spectra analysis system. A small amount of constant volume of the gas such as 0.001 cm 3 /s paths continuously through a needle valve and was analyzed by a quadruple mass analysis method. The main composition of gas released from the cathode was then continuously analyzed by above-mentioned method. 3.1.

Calorimetry

Temperature measurements were made with 1.5-mm diameter RTDs. Calorimetry was performed by combining the flow and isoperibolic method.

163

Figure 1.

Photo of the experimental setup.

Flow calorimetry is based on the temperature change of the cooling water. The cooling water is tap water flowing through Tigon tubing. It passes first through a constant temperature bath to keep the temperature constant. From there, it flows through the outer jacket of the condenser, and then through the coil of tubing wrapped around the funnel. (The outside of this cooling water coil is covered with the anode and a platinum mesh.) The flow rate is measured with a turbine meter (Japan Flow Control Ltd., Model T-1965B). The inlet temperature is measured before the cooling water enters the condenser, and outlet temperature is measured where it exits the cell. Heat from both condensation and glow discharge electrolysis are combined together. Isoperibolic calorimetry is performed by placing three other RTDs were in the cell electrolyte at different depths in the solution to measure the temperature. The solution is mixed with a magnetic stirrer. Figure 5 shows the notional sketch for heat measurement. Heat out can be divided into several factors. These are energy for water decomposition, heat of

Figure 2.

Sketch of experimental setup.

164

To gas (low meter and mas spectrometer Pooling water out Cooling water in

*M - ti'll

--L_JLJ

I H2 + 0 2 + vapor

*

J mixing gas

j

Teflon rubber stopper Elactrolyte level

Shnnkable teflon cover

Mixing gas bubble

Figure 3.

Detail of the gas measurement.

electrolyte, heat bring by the coolant, heat releasing from the call wall, and heat releasing with the vapor through the cell plug. The heat balance is estimated by input and output formulas, input and output power is given in the following equations: Input (J) = / (current) x V (Volt) x t,

Out =

Hg+Hw+HC+Hr+Hy,

here, Hg = Heat of decomposition — / 1.48 x dl x di,

Hw = Electrolyte heat = / Ww x Cw x ST,

-/-

Photo of cell R T D : Pt resistance thermometer, O.OOIdeg

X. * j

Glass dome Coolant coil Pt anode Rectangular Pt had an integral lattice constructed using a 15 cm length of 0,1 cm in diameter. 15 cm in height

Figure 4.

Photo of cell.

165

where Ww is the electrolyte weight, C w heat capacity, and ST is the temperature difference. Hc = Heat of coolant =

Wc x Cc x ST,

where Wc is the coolant weight, C c heat capacity, and ST is the temperature difference. HT = Heat release = / (Ww xCw + Wcx

CC)T,

where Tr is the temperature change. Hv = vapor = Wv x Cc.

Fxcess gas Hc: Heat of coolant Hg: Heat of decomposition

Figure 5.

Schematic representation of heat balance.

The heat balance calculation is straightforward. Input power is only from the electric power source. Output is divided into several parts. The first factor is heat of water decomposition (designated Hg). It is easily calculated from the total electric current. The second factor is electrolyte enthalpy (Hw). It is easily derived from the solution temperature difference. The third factor is heat removed by the coolant (Hc). This is measured from the temperature difference between the coolant inlet and outlet, and the coolant flow rate. The fourth factor is heat release from the cell (-ffr)- This is rather complicated and can be estimated with a semi-empirical equation. The fifth factor is heat release by vapor (Hv). This is difficult to measure precisely. However, I have measured most of the heat in the condenser directly by monitoring the inlet and outlet temperature of the cooling water that passes through the condenser outer jacket.

166

I Figure 6.

W sample. Left, before; Right, after glow discharge electrolysis.

If there is excess hydrogen and oxygen gas, we have to measure the gas volume precisely, because even a small volume of gas removes a large amount of enthalpy. This is done with the precision gas flow meter. The first factor, water decomposition (jffg) has a large effect on the rest of the equation. 3.2. Electrode

and

Solution

The electrode is tungsten wire, 1.5 mm in diameter and 15 cm in length. The upper 13 cm of the wire is covered with shrink-wrap Teflon and the bottom 2 cm is exposed to the electrolyte and acts as an electrode. The light water solution was made from high purity K 2 C 0 3 reagent at 0.2 M concentration. 3.3. Power

Supply

Data from the electric power supply (Takasago, Model EH1500H) were collected with a power meter (Yokogawa Co., Model PZ4000) and averaged every 5s. The sampling time was 40 us and the data length were 100 k.

Figure 7.

Photo of power supply.

167

Figure 8.

3.4. Data

Photo of EDX.

Collection

All data, including the mass of cooling water flow from the flow calorie measurement, the temperature of coolant entrance and exit, electrolyte temperature measured by three RTDs, input voltage, current, electric power and the amount of the hydrogen gas generated were collected by a data logger (Agilent Co., Model 34970A), and stored in a personal computer.

3.5. Element

Analysis

The sample electrodes and the electrolyte were subjected to element detection by means of energy dispersive X-ray spectroscopy (EDX), Auger electron spectroscopy (AES), secondary ion mass spectroscopy (SIMS) and electron probe microanalyzer (EPMA).

4 ' - « ICPS-7500

Figure 9.

Photo of ICP mass.

# >

168

4. Results We changed the input voltage stepwise as shown in Fig. 10. In this example, plasma formed at 120 V. Once plasma formed, input current suddenly dropped. Meanwhile, the solution temperature reached roughly 80° C. We increased voltage in stages to 350 V and then decreased it to 100 V. Plasma continued at 100 V and ceased at 80 V.

cfdaia»w209ioti

Figure 10.

Time/1000 s

Change for input voltage, current and Temperature.

The gas collected by the funnel over the cathode is collected by the Q-massspectrometer continuously as shown in Fig. 11. It is mainly composed of three gasses: hydrogen, oxygen, and nitrogen were always detected. Other gases were under the background levels and were not detected after electrolysis started. Oxygen gas increased after plasma electrolysis occurred. Hydrogen gas appeared when electrolysis started. When plasma electrolysis began, hydrogen production decreased while at the same time oxygen increased. The increased oxygen gas means that

1E-08r

¥cfdata¥w20910&1 wk4

Figure 11.

Time/1000 S

Changes of the ion current of various masses.

169

cfdala910&1 .wk4

Figure 12.

Time/1000 S

H2 and O2 ion current and the ratio over time in 1000 s.

direct water decomposition by pyrolysis had begun. (Although total hydrogen production decreases when glow discharge begins, "excess hydrogen" increases. In other words, the ratio of hydrogen to input power sharply increases, glow discharge consumes much less power than ordinary electrolysis, and power falls even more than hydrogen production does.) The behavior of the hydrogen isotope molecules is the same as hydrogen molecular ion. Figure 12 shows changes over time for hydrogen and oxygen molecules and their ratio. The ratio increases as input voltage increases, reaching 0.45 at 2500 s and 350 V. This means the gas from the cathode had begun direct decomposition by pyrolysis. The ratio of the hydrogen in gases released from the cathode is exactly same as the Faradaic value from the current. The equation is (i7!—0.1161) x 0.667 + 0.1161. Here, F\ is the rate of hydrogen gas estimated from the flow meter and, I is current. The factor of 0.116 is the rate of ordinal hydrogen generation, i.e., 0.5 x 22 414/F (F, Faraday constant of 96 500C/mol). The factor of 0.667 is atomic ratio of hydrogen for water.

Figure 13.

Hydrogen generation over time.

170

100 f

&MB^Mm3 Figure 14.

.

1

1

.

a10

Time/1000 s Current efficiency and ratio of oxygen.

Figure 13 shows the changes of hydrogen generation estimated from current, and measured with the gas flow meter. These rates were same with ordinary electrolysis. However, the value measured by the flow meter shows an upward deviation compared the value estimated from the current. The change of the ratio between these two values, i.e., current efficiency (e) and the ratio of the oxygen gas to the total generation with hydrogen from the cathode are shown in Fig. 4c. Here, e exceeds unity when the plasma electrolysis started, gas generation increased with the input voltage. It reached as 8000% at 350 V input voltage. The ratio of oxygen reached 30%. This means that almost all of the hydrogen came from water decomposition (pyrolysis) during high voltage plasma electrolysis. Figure 14 shows the change of the ratio between these two values, i.e., current efficiency (e) and the ratio of the oxygen gas to the total generation with hydrogen from the cathode. Here, the e exceeded unity when the plasma electrolysis started; gas generation increased a great deal with the input voltage. It reached as much as 8000% at 350 V. For the entire run, the theoretical value of hydrogen generation calculated from the input current was 1144 cm 3 , and the value measured during plasma electrolysis was 2190 cm 3 . That is, the generation of excess hydrogen during a whole electrolysis reached 1046 cm 3 . For measurements made only during times when the plasma was present, the measured value was 1470 cm 3 compared to the theoretical value were 460cm 3 , so the excess was 1010cm 3 . Figure 15 shows the e and V relationship. Here, it can be seen that e has a tendency of increase with input voltage. Some points of e value under 100 V in the figure show up to twice the theoretical value of unity; these points were obtained during plasma electrolysis. On the other hand, e remains at unity during normal electrolysis. It can be expected that if the input voltage is increased up to several hundred volts, then e would far exceed unity. Figure 16 also shows the ratio of excess hydrogen to input power. These values were subtracted from the faradic hydrogen from the previous graph and converted into units of energy. It shows steep increase with the input voltage. It exceeds 10% and sometimes reaches 30%.

171

100 :

e

F

10 r

-!--•

!

;

r

!

1

t

o • 0.1 10

100

cfdata¥W20910#1

Input voltage

Figure 15.

1000

Voltage dependency of current efficiency.

c

Figure 17 shows the heat balance for the run shown in Fig. 16. This indicates no excess heat. However, during this run, enthalpy release from excess hydrogen and oxygen evolution was not measured. So, there may be excess enthalpy, but in this case it can only be said that the heat balance was almost 100%. Figure 18 apparently indicates a tendency of the excess hydrogen generation with the negative excess heat. The total endothermic heat was calculated as 6540 J during plasma electrolysis. Figure 18 shows an example in which no excess heat was generated. We changed the input voltage stepwise as shown in Fig. 19. We used 0.1 M of K2CO3 electrolyte. Plasma formed at 120 V. Once plasma formed, input current dropped. The solution temperature reached 70°C. We increased voltage in stages to 320 V and then decreased it to 100 V. Plasma continued at 140 V and ceased at 100 V. Figure 20 shows the changes of hydrogen generation estimated from current and flow meter. The value estimated by flow meter exceeded the value of the current estimation.

(%)

. ---

f Excess Hydrog en/in

3

,

10

.

„

. •

,

. ~.. ~ » . ^ -~-~*

, „ . . . .

1

J.'£_!.. .-?-•

•": .10

cfdata*W20910#1

Figure 16.

„ „

100 Voltage

1000

Energy efficiency of hydrogen gas to input power.

172

1.5

%^¥^>^m "spkx»?

£. 0.5 ' '

0 l" 0 tcfdaa¥W20910#lI

Figure 17.

j

"

"

' ' '

2

1

•

!

' ' '

-'

''

10 7 . Close to the reaction threshold there are two 1~ resonances in the compound nucleus 4 He. They can be excited by deuterons with an orbital angular momentum of 1. This is the reason for the unusually strong anisotropy of the angular distribution of the ejectiles even at the lowest energies. In our previous works 1 ' 2 we have found a strongly enhanced electron screening effect for the d + d reactions in metallic environments. Angular distributions and relative intensities of the proton and neutron channels investigated for d + d reactions taking place in Al, Zr, Pd, and Ta targets were, however, in agreement with the results of gas-target experiments. Here, we present new results obtained for Sr, Li, and Na targets giving a first evidence for an alteration of the neutronproton branching ratio and the angular distributions. Initial results were presented in Ref. 3 and hitherto completely published in Ref. 4 but now a first theoretical explanation for this surprising observation can be presented. 5 210

211

2. Experimental Results The experiment has been carried out at a cascade accelerator optimized for low-energy beams. The targets were pure metal disks becoming self-implanted deuterium targets under the deuteron irradiation. Four Si-detectors at the laboratory angles of 90°, 110°, 130°, and 150° were used for the detection of all charged particles, p, t, 3 He, of the reactions 2 H(d,p)t and 2 H(d,n) 3 He. 6 ' 2 The detectors needed to be shielded from the backscattered deuterons in order to prevent a congestion of them and the data acquisition system. Therefore, grounded Al-foils of thicknesses from 120 to 150 /xg/cm2 were placed in front of the detectors insulated from them. The thickness is sufficient in order to stop backscattered deuterons up to 60 keV. The low-energy part of some representative spectra from the 90°-detector is depicted in Fig. 1 magnifying the two lines of the recoil nuclei 3 He and t. The spectra are normalized to an integral value of one in order to make them commensurable. The energies above the peaks are the kinetic energies of the ejectiles in the laboratory system. They drop for increasing projectile energies, which is especially significant for the back angle positions. The gray filled spectra are from Ta targets while the black and gray step lines are for Sr and Li and Na, respectively. The two plots compare the form of the spectral lines at a low-projectile energy of 8 keV to a high energy of 30 keV. At 8 keV, the t-line of Ta is well separated while the 3He-line sits on an exponential background. The background is subtracted by fitting an exponential function to the lowest energy part and then by extrapolating it to the high energies. The spectral lines for Sr are already broader with an enhanced lowenergy tail leading to an overlap of both lines. This effect becomes even stronger for Li and Na. At 30 keV the Ta lines are broader but the tails of the Sr, Li, and Na lines are much more distinctive. The overlap of the two lines is even higher. For Li and Na the 3He-line is hardly more than an edge. The p-line at 3 MeV has also a long low-energy tail but it vanishes before the t-line. The appearance and the properties of these tails can be explained by a phenomenon known from the physical chemistry of the metal hydrides, called embrittlement7 which means that the crystal structure of the metal is bursted by the recrystallization process that accompanies the formation of the metal hydride crystal. Reactive metals change their crystal structure while forming the metal hydride. If the hydration proceeds not in a thermal equilibrium and relatively slow, the material cannot compensate the tension of the recrystallization process and bursts. Since deuteron implantation is far off the thermal equilibrium, embrittlement is a hardly avoidable concomitant phenomenon for reactive metals. How embrittlement effectuates the tails is elucidated with the sketch in Fig. 1. Assuming, the projectile travels through the target along a path covering many empty regions, the energy loss becomes smaller and consequently the nuclear reactions occur deeper below the target surface than in the case of compact materials. Therefore, the ejectiles that in turn can travel through more compact target regions loose more energy additionally contributing to the low-energy tail of the particle spectrum. The increase of the tail with the projectile energy arises from the simultaneous increase in the overall range of the projectiles. The material

212

dependence can be explained, too. Ta is almost a noble metal with low reactivity but nonetheless able to chemically bind hydrogen to high amounts. It just stretches its lattice dimensions but does not recrystallize like the highly reactive metals of the groups I and II of the periodic system. So there is no embrittlement and hardly a tail visible. On the other hand the effects of embrittlement and the tail increase from Sr over Li to Na with decreasing electron negativity. The symptoms were even visible, e.g., dust particles crumbled from a strontium target, the thickness of a natrium target grew considerably. The low-energy tail formation complicates the integration of the spectral lines till unfeasibility in the case of Na. The 3He-line sits on the tail of the t-line. All efforts to describe and extrapolate the tail of the t-line to the lower energies analytically failed, since the form of the lines is dependent on the nucleus species, ejectile and projectile energy. Uncertainties and imponderabilities in the integral of the spectral lines are taken into account in the errors additionally to the counting statistic. Consequently they are the dominating error source. If in doubt, events were attributed to the 3He-line only, gaining a conservative estimate at least. Fortunately, the tails are small at the low-projectile energies were the asymmetry in the branching ratios becomes observable. Neglecting I > 2 contributions the angular distribution can be described as follows: ^($)

= A0 + A2cos2$.

(1)

U.LU

Because of the identical bosons in the entrance channel the angular distribution is symmetric around 90°. $ and u> are the polar angle and the solid angle in the CM system, respectively. Since the experimentally determined thick target yield is dominated by the high-energy contributions below the Coulomb barrier a similar expression is valid for the differential counting number: — (tf) =a0 + a2 cos2 •&. (2) dw The expansion coefficients a$ and a2 now include a constant factor containing a product of detector and target properties and the number of incident projectiles. A measurement at 20 keV for Sr exemplary shows the results for the differential counting number and the corresponding fitting function in Fig. 2. The fit is computed with a non-iterative generalized linear fitting algorithm employing singular value decomposition thus allowing for more accurate values and better error handling. The data points obtained for the protons are included. As can be seen protons and tritons follow the same angular distribution. One observes a significantly stronger angular anisotropy for the neutron channel. The angular anisotropy is quantitatively given by the ratio (a2/ao = A2/AQ). Again with the previous argumentation the two fitting coefficients can be used to calculate the branching ratio of the two mirror reactions with ^d(d,n)3He = y( 3 He) = JV(3He) = a 0 ( 3 He) + |a 2 ( 3 He) crd(d,p)t Y(p) N(p) a 0 (p) + |a2(p) where in the last step N is simply the integral over the unit sphere of the differential counting number (2). When calculating with the fitting coefficients one

213

must consider that they are not independent variables. Then the Gaussian error propagation formula needs to be completed by a term containing the off-diagonal element of the covariance matrix from the fit. The results are plotted in Fig. 3. The branching ratios and angular distributions determined for Ta, Al, Zr, and Pd agree with the results of the gas target experiment. 8 Not so for Sr and Li. While for p there are no pecularities, for 3 He the anisotropy raises at lower energies (see also Fig. 4). Simultaneously, the n branch is suppressed. The results for protons and tritons are concordant. The low quality of the Li points results from the ambiguity of the integration of the spectral lines with large low-energy tails. For the same reason, the spectra obtained for Na could not be analyzed quantitatively, though the spectra indicate a strong suppression of the neutron-proton ratio at low energies, too. For details refer to Ref. 4.

!

0

I 1

L,,

,,l

50

S

1

I

I

1

>

. [ i i I i J i i H j i i n j i n

1.8

o(r.t) = a 0 + a.j cos 2 *

a Ta - P A Ta -6He • Sr - P J A Sr - H e • Li- P J * Li- He

1.6 1.4 1.2 CO

q-rrrrj

1.0 0.8

*

0.6 0.4

:

*

Î** *

0.2 0.0

.-H+UH11 n n |ni ill i ill int 1 mi hi i ih m l i n i l n '

1.1

1.0 0.9

0.8

"§& O Ta • Sr • Li

0.7

t. !.'...!. i . i l J

0.6

10

15

20

25

30

LLUDJ.I

JJJJJXI

35

45

40

50

55

lab (keV) a

E

Figure 3. The upper part displays the anisotropy from the detection of the p and 3 He ejectiles. The lower part shows the branching ratio for the two mirror reactions.

electricity. So this could be a conceivable reason for the experimental observations. Enantiomorphy is a necessary condition for such effects. However, the point groups belonging to LiD, SrD 2 , and NaD do not allow for this. From the theoretical point of view the cross section for the mirror reactions 2 H(d,p) 3 H and 2 H(d,n) 3 He at deuteron energies below 100 keV can be described with 16 collision matrix elements, corresponding to S-, P-, and D-waves in the entrance channel. The matrix elements for incoming D-waves cannot be omitted as frequently asserted since they are mandatory to describe the angular anisotropy down to the lowest energies. The values of the matrix elements are relatively well known and were obtained by fitting experimental cross sections, vector and tensor analyzing powers measured in gas target experiments. 10 ' 11 The differential cross section for both reactions can be presented by a coherent superposition of all 16 matrix elements 5 (dashed line in Fig. 4) and agrees with our results obtained for Al, Zr, Pd, and Ta. In the case of Sr (also for Li) a polarization of the deuterons

216 0.8 0.7 0.6 0.5 0.4 0.3 0.2

A Neutrons: filled Sr, open Ta • Protons: filled Sr, open Ta , . , Normal curve: a(S=0,1,2)=1 •»™ Polarization

0.1 0

LiLi

Lib_A±i_h

1.05 1.00 1.95 0.90

• Li • Filled Sr, open Ta . . . Normal curve: o(S=0,1,2)=1 • Polarization

0.85 0.80

i-4~tl~U~UU l I E U , . l ~ t i i . i I, S }

0.75 5

10

15

20

25

30

35

40

45

50

ix.LiÎj^~i~

55

60

EfVv) d Figure 4. The dashed line represent the normal curve. The solid lines result from a deuteron polarization corresponding to a suppression of the S = 0 channel at lower energies.

in the crystal lattice had to be assumed. A suppression of the channel spin 5 = 0 (spins of the deuterons are anti-parallel) and allowing the other channels with spins S = 1,2 to be undisturbed permits to describe simultaneously the enhancement of the angular anisotropie of the 2 H(d,n) 3 He reaction and the decrease of the n/p branching ratio at very low energies down to 0.83. The results of corresponding calculations are presented in Fig. 4 as full lines. Here we have assumed that the deuteron polarization takes place gradually below the Fermi energy (for Sr about 25keV), reaching its maximum value already below lOkeV. A strong quenching of the neutron channel might also be explained by different screening energies for relative angular momentum L = 0,1. Supposing that the screening energy for the L = 1 contribution is much smaller than for L = 0 one gets a decreasing n/p branching ratio at low energies reaching a minimum of only 0.93. In this case, however, the anisotropy of the angular distribution for both channels will be reduced till isotropy, in contradiction to the experimental results. For details refer to Ref. 5.

217 4.

Conclusion

We presented a first experimental evidence for a alteration of the branching ratios in the d + d fusion reactions obtained in an accelerator experiment which can be theoretically explained by polarization of the reacting deuterons in the crystal lattice while other rather trivial causes could be excluded. T h e reason for the deuteron polarization is, however, still unknown. A distinctiveness of the (earth) alkaline metals is the formation of an ionic bond to hydrogen, which might be a starting point for a possible explanation based on the spin-spin interaction. In view of the experimental indications for a strong n e u t r o n - p r o t o n asymmetrie in the d + d reactions at room temperature, our experiment supports an understanding of the cold fusion phenomena although further efforts are necessary. An experiment with more sophisticated particle detection techniques is in progress in order to refine the data.

References 1. K. Czerski, A. Huke, P. Heide, M. Hoeft, and G. Ruprecht, Nuclei in the Cosmos V, in N. Prantzos and S. Harissopulos (eds.), Proceedings of the International Symposium on Nuclear Astrophysics, Volos, Greece, July 6-11, 1998, p. 152 (Editions Frontieres). 2. K. Czerski, A. Huke, A. Biller, P. Heide, M. Hoeft, and G. Ruprecht, Europhys. Lett. 54 (4), 449-455 (2001). 3. A. Biller, K. Czerski, P. Heide, M. Hoeft, A. Huke, and G. Ruprecht, in Verhandlungen der DPG, Vol. 1, Gottingen, DPG-Fruhjahrstagung, 1997, p. 28. 4. A. Huke, Die Deuteronen-Fusionsreaktionen in Metallen, PhD Thesis, Technische Universitat Berlin, 2002. 5. Tatiana Dorsch, Theoretische Untersuchung der anomalen Asymmetrie im Verzweigungsverhaltnis der Reaktionen d(d,p) H und d(d,n) He im Verbund der Hydride der (Erd)Alkalimetalle. Diplomarbeit, Institut fur Atomare Physik und Fachdidaktik der Tech-nischen Universitat Berlin, 2004. 6. A. Huke, K. Czerski, and P. Heide, Accelerator experiments and theoretical models for the electron screening effect in metallic environments, in Proceedings of the International Conference on Condensed Matter Nuclear Science, ICCF-11, Marseille, France, November 2004. 7. W. M. Mueller, J. P. Blackledge, and G. G. Libowitz (Eds.), Metal Hydrides (Academic Press, New York, London, 1968). 8. R. E. Brown and N. Jarmie, Phys. Rev. C41 (4), 1391 (1990). 9. A. Biller, Einflufi der Vielfachstreuung auf Dicktarget-Yields in niederenergetischen Kernreaktionen. Diplomarbeit, Institut fur Atomare und Analytische Physik der Technischen Universitat Berlin, 1998. 10. H. Paetz gen. Schieck, and S. Lemaitre, Ann. Phys. 2, 503 (1993). 11. O. Geiger, S. Lemaitre, and H. Paetz gen. Schick, Nucl. Phys. A (586), 140-150 (1995).

E X P E R I M E N T S ON C O N D E N S E D M A T T E R N U C L E A R E V E N T S IN KOBE U N I V E R S I T Y

T . M I N A R I , R. N I S H I O , A. T A N I I K E , Y. F U R U Y A M A , A N D A. K I T A M U R A Division of Environmental Energy Science, Graduate School of Science and Technology, Kobe University, 5-1-1 Fukaeminami-machi, Higashinada-ku, Kobe 658-0022, Japan E-mail: [email protected]

We review three kinds of experimental works underway in our laboratory to investigate nuclear events in solid or liquid materials. The largest effort has been given to experiments to confirm the 7 Li(d, n2a) reaction rate enhancement reaching 10 1 5 in liquid lithium which was reported by H. Ikegami and R. Pettersson, Evidence of Enhanced Nonthermal Nuclear Fusion (Bulletin of Institute of Chemistry, BENF No. 3, Uppsala University, Sweden, September 2002). Li liquid droplets are formed as targets, and to keep them as pure as possible, we built a liquid Li loop. Thus far, in all cases of irradiation at the temperature from 520 to 570 K with 1024keV deuterons, we have not been able to reproduce the Ikegami enhancement for the 7 Li(d, n2a) reaction.

1. Introduction Experimental studies on condensed matter nuclear reaction using ion beams at Kobe University are reviewed. First, we are continuing experiments on keV-deuterium ion irradiation of deuterated Au/Pd samples, searching for possible anomalous nuclear reactions in solids. 1 ' 2 The Au-coated surface of the Pd sample is irradiated with the keV-deuterium ions and with MeV ion beams for simultaneous characterization of the surface, while the rear surface is exposed to D2 gas at atmospheric pressure to load the sample. In addition, a modified version of the sample system without the implanter has been installed at another beam line of the accelerator exclusively in order to reproduce the so-called Iwamura effect.3 Iwamura et al. observed transformation of 133 Cs into 1 4 1 Pr during forced permeation of deuterium through a multi-layered film of Pd and CaO. In the present system of ours the nuclear transformation is monitored by in situ PIXE analysis during forced permeation of deuterium. The third system has been installed to reproduce the experiments made by Ikegami and Pettersson, 4 in which they observed an enormous enhancement of 7 Li(d, n2a) reaction rate in liquid lithium. To maintain high purity, we have set up a keV-deuteron irradiation system with a liquid Li loop, which enables us to continuously bombard pure Li in a closed container. 218

219

2. D(d, p)t Reaction Rate Enhancement The D(d,p)t reaction rate under irradiation of deuterium ion beam has been measured in the keV energy range in various samples to determine whether the nuclear reaction rate can be enhanced in a condensed matter. Kasagi et al. reported that in PdO the enhancement factor relative to the reaction rate for the deuteron pair in vacuum increases up to 50 at E 1 keV) and at the low energy {Ecm < 10eV). The ratio between the low-energy and the high-energy C/eff value amounts to about 0.58, being nearly independent of the actual deuteron-deuteron potential. The total screening energy is the sum Ueg + Ucoh the value of which taken at the zero projectile energy UQ corresponds to 0.78 of the high-energy limit Ue. At deuteron energies below 10 eV, the effective screening energy remains almost constant, hence the expression for the cross section takes a very simple form. Starting from Eq. (1) we obtain ME of 2.2. If we gang 32 of these 4-unit systems together a 10 kW device results (see Fig. 6).

251

LM reactor Q| = 40W Q x = 40 W

80 W low-grade heat Reactor i 120 V & 60 Cycles Qj

4 LM reactors 320 W Power supply

32

4-

4 LM reactors 10,240 W

fjdk^bdS

Heat exchange liquid

q 120V —p1

4-unit LM reactor

in

I G D I f ] l C D I f ]

CM WDCMWD EMW3EMW CMMZlCm-MD

out

OUCDiCI OpfOOHD

Figure 6. A 4-unit piezo driven LM SF reactor. Ganging 32 units together produces 10 kW. Such a device delivers low-grade hot water at 80° C.

6. Summary We have today an infant technology that will grow at a very fast pace. Sonoluminescence is a fascinating subject and its study forms the basis of the SF technology. A paper 9 that gives many good references on SL of the single bubble can also be applied to the multibubble systems. Also see the references of.7 We can represent the utility of the sonofusion reactor as a device that can act as a power multiplier using the multiplier effect, ME, to measure the advantage over today's power costs (see Fig. 5b). There is one piece of important information missing; evidence of the other products associated with Q x production. So we go back to our old data from our 20 kHz system that reported the products of 4 He and 3 T (Ref. 2) and look for some of these products in the new analysis, by Francesco Celani, of our new target foils and we thank him for any results he may find. The Q x watts/(gram of reactor) is 2 and we expect that improvements will follow. If, for example, Q{ — 40 W and the Qa/Qi is 0.5 and Q x is 40 W, then the total watts output for one LM sonofusion unit is 80 W. This 20 g LM reactor can be used as a powermultiplier, ME, of value 2.0. If we have four LM reactors working together as a system the total output is 320 W and 25 of these 4-unit systems working together produces 8000 W with a reactor mass of 2 kg. When 1000 units are ganged together we can produce 80,000 W that would cost $300/day ($0.17 for a kWh). However, this system has a ME of 2 and the cost for Q; would be $150/day. With a 1000

252 unit reactor the Q x t h a t is produced would be worth $55,000/year and the initial cost for a sonofusion reactor might be $20,000 making the cost benefit for the first year $35,000. If the 40,000 W for Qi are solar generated, then we are looking at an added initial cost t h a t would be compatible with our sonofusion reactor with longt e r m dollar savings. T h e total heat production is Q\ + Qx with a cost reduction of 50% for low-grade heat used in building's utility heating. This is all possible with the sonofusion technology we have today and this technology is continually improving.

Acknowledgments I would like to t h a n k the individuals t h a t helped us get this 1.6 MHz system to the point it is in today: Dick Raymond, Richard America, Kip Wallace, Lynn Marsh, Julie Wallace, Ted Mill, Fran Tanzella, and Mike McKubre. References 1. R. Stringham, in Proceedings of the International Symposium on IEEE Ultras, 5-8 October 1998, Vol. 2 (Sendai, Japan), p. 1107. 2. R.S. Stringham, in Proceedings of the ICCF-8, 21-26 May 2000 (Lerici (La Spezia), Italy), pp. 299-304. 3. R.S. Stringham, in Proceedings of the ICCF-9, 19-24 May 2002 (Beijing, China), p. 323. 4. M.P. Brenner, S. Hilgenfeldt and D. Lohse, Rev. Mod. Phys. 74, 425-484 (2002). 5. K.R. Weninger, C.G. Camara and S.J. Putterman, Phys. E. Rev. 63, 016310-1 (2000). 6. G. Vazquez, C. Camara, S.J. Putterman and K. Weninger, Phys. Rev. LETT. 88 (19), 197402-1 (2002). 7. R.S. Stringham, in Proceedings of the ICCF-10, 24-29 August 2003 (Boston, USA, to be published). 8. R.A. Hiller and S.J. Putterman, Phys. Rev. Lett. 75 (19), 3549 (1995). 9. M.P. Brenner, S. Hilgenfeldt and D. Lohse, Single-bubble sonoluminescence, Rev. Mod. Phys. 74, 425-484 (2002).

R E S E A R C H INTO CHARACTERISTICS OF X-RAY EMISSION LASER B E A M S F R O M SOLID-STATE CATHODE M E D I U M OF H I G H - C U R R E N T GLOW D I S C H A R G E

A L E X A N D E R B. K A R A B U T FSUE

SIA "LUTCH", 24 Zheleznodorozhnaja Street Podolsk, Moscow Region 142100, Russia

X-ray emissions ranging 1.2—3.0keV with dose rate up to l.OGy/s have been registered in experiments with high-current Glow Discharge. The emissions energy and intensity depend on the cathode material, the kind of plasma-forming gas, and the discharge parameters. The experiments were carried out on the highcurrent glow discharge device using D2, H2, Kr, and Xe at pressure up to lOTorr, as well as cathode samples made from Al, Sc, Ti, Ni, Nb, Zr, Mo, Pd, Ta, W, Pt, at current up to 500 mA, and discharge voltage of 500-2500 V. Two emission modes were revealed under the experiments: (1) Diffusion X-rays was observed as separate X-ray bursts (up to 5 x 10 5 bursts a second and up to 10 6 X-ray quanta in a burst), (2) X-rays in the form of laser microbeams (up to 10 4 beams a second and up to 10 1 0 X-ray of quanta in a beam, angular divergence was up to 10 — 4 , the duration of the separate laser beams must be r = 3 X 1 0 ~ 1 3 - 3 X 1 0 ~ 1 4 s, the separate beam power must be 10 7 -10 8 W). The emission of the X-ray laser beams occurred when the discharge occurred and within 100 ms after turning off the current. The results of experimental research into the characteristics of secondary penetrating radiation occurring when interacting primary X-ray beams from a solid-state cathode medium with targets made of various materials are reported. It was shown that the secondary radiation consisted of fast electrons. Secondary radiation of two types was observed: (1) The emission with a continuous temporal spectrum in the form of separate bursts with intensity up to 106 fast electrons a burst. (2) The emission with a discrete temporal spectrum and emission rate up to 10 1 0 fast electrons a burst. A third type of the penetrating radiation was observed as well. This type was recorded directly by the photomultiplier placed behind of the target without the scintillator. The abnormal high penetrating ability of this radiation type requires additional research to explain. The obtained results show that creating optically active medium with long-living metastable levels with the energy of 1.0-3.0 keV and more is possible in the solid state.

1. Introduction Experiments were carried out previously to define a possible mechanism of high-energy phenomena in the solid-state cathode medium during high-current glowdischarge. The experimental results showed that the character of the detected X-ray radiation essentially differed from the known X-ray emission types. It indicated the importance of research into the performances of the detected X-ray radiation from the solid-state medium of the cathode sample material of the high-current glow discharge. 253

254

Penetrating radiation passing through the discharge chamber walls (5-mm thick steel) was recorded during high-current glow discharge (Fig. I). 2 The experiments showed that it was the secondary radiation occurring when interacting with the primary X-ray laser beams from the solid-state cathode medium with the material of the chamber walls and construction elements and lead protective shields.1 The created 100% reproducible technology for generating the X-ray laser beams, which would allow research into the performances of the secondary penetrating radiation.

Figure 1. Various variants of the experimental device: (a) system of a PM-scintillator placed 21cm from the cathode, (b) system with the PM-scintillator placed 70 cm from the cathode, (c) system with the PM-scintillator - PM placed 70 cm from the cathode with superimposition of the cross magnetic field.

2. Experiment Method and Results The experiments were carried out with the device that produces high-current glow discharge using deuterium, hydrogen, Kr, and Xe. The cathode samples made of Pd and other metals were disposed on the cathode-holder above which a window for output of penetrating radiation was placed. The window was closed by 15 /jm Be foil for protecting the detectors against visual and ultraviolet radiation (Fig. 1). A periodic-pulse power supply was used for the glow discharge, with a rectangular current pulse. The duration of the discharge current pulses were 0.27-10.0 ms, period between impulses was 1.0-100ms. The discharge was carried out in D2, Xe, and Kr. The X-rays recording was carried out with use of the thermo-luminescent detectors, X-ray films placed above the cathode at various distances, and scintillation detectors supplied with photomultipliers. 1

255

Thermoluminescence detectors are not sensitive to electrical noise and allow registering the absorbed radiation dose quantitatively in absolute units of dose measurement. Thermoluminescence detectors (TLD) based on A1 2 0 3 crystal register penetrating radiation beginning from the background values of radioactive radiation of the environment. These were to measure energy intensity and to evaluate the average energy of a soft X-ray emission in the discharge. To determine the average energy a special cassette (seven-channel spectrometer) was used. Seven channels (holes) with a diameter of 5 mm each were made in the cylindrical body of the cassette. The detectors in the form of disks with the diameter of 5 mm and thickness of 1 mm were arranged in the holes. A set of the beryllium foils having thickness 15, 30, 60, 105, 165, 225, 225, and 300 /im (in the each hole the foil having the special thickness was arranged) was arranged on the side of emission input in the body holes. Two TLD detectors were arranged outside the camera for the registration of background value of the emission dose. A pinhole camera gives a spatial resolution of X-ray emission and an opportunity to determine where the radiation emerges from. The time characteristics of the penetrating radiation were determined with the scintillation detectors supplied with the photomultipliers (PM). The signal from the PM was transferred to a fast preamplifier with an amplification constant of

/ x-ray, 4x10

Photons/burst

6

3x106

(a) 2x106 1 x106 0 ' X-ray, i Photons/burst 3x106

\k.

Figure 2. Typical oscillograms of the X-ray emission signal from the system PM-scintillator covered with the Be foil with the different thickness: (a) with covered the 15 ^ m Be shield, (b) with covered the 30 |im Be shield. The Pd-D2 system, the discharge current — 150 mA.

256

k = 7 and then to the two-channel computer digital oscillograph with the limit resolution frequency of 50 MHz per a channel. Organic scintillators on the base of polymethylmetacrelate (PMMA) with luminescence time of 3-5 ns were used. The time resolution of the entire path from the PM up to an oscillograph (experimentally) was 70-80 ns. Electrical noise was observed only when passing the front and back fronts of the current impulses feeding the glow discharge (Fig. 2). Three various variants of assembling the discharge chamber with the channel for the radiation extraction was used (Fig. 1). In the first variant (Fig. la) PMscintillator was placed 21cm from the cathode surface. The channel diameter for extracting the radiation was equal 1.7 cm. In the second variant the PM-scintillator was placed 70 cm from the cathode, and the diameter of the channel for extracting the radiation was 3.2 cm (Fig. lb). To define the type of penetrating radiation, the third variant of the experimental assembly included the magnetic system consisting of a constant magnet and an elliptic iron magnetic circuit (Fig. lc). The axis of the magnetic system poles was 35 cm from the cathode perpendicular to the axes of the radiation extraction channel. The magnetic field induction in the gap between poles was 0.2 T. For the determination of quantitative registration characteristics the thermoluminescence detectors were calibrated in the gamma-emission fields. The experiments were carried out using the following systems of cathode plasma-forming gas: Pd-deuterium. The obtained results show that the doses obtained by the corresponding detectors decrease exponentially with increasing the thickness of Be foil. The main component of the X-ray emission energy is in the range of 1.0-2.5 keV, but there is a component with a higher energy too. X-ray intensity was registered for the different values of current and voltage. The procedure of recording and measuring was developed as applied to two modes of the X-ray emission: a mode of the diffusion radiation bursts, generation of X-rays as laser microbeams. The intensity of the luminous flux from the scintillator when it was in the mode of generating X-ray laser beams was approximately 1000 times as much as the intensity in the mode of the diffusion bursts. In this case, the amplification constant of the radiation recording system changed by changing the supply voltage of the photomultiplier and by changing the amplification constant of the oscillograph. Under some experiments the luminous-absorbing filter attenuating the luminous flux coming to the PM was installed between the scintillator and PM. Two types of the filters were used, that attenuated the luminous flux by 50 times and by 2500 times, respectively. The intensity of the X-rays (number of photons a second) coming to the detector was determined by dividing the energy radiation power absorbed by the detector by the energy of an X-ray photon. Further the intensity falling to one detector was given to 2p solid angle. For the PM-scintillator the relative intensity of the X-rays was determined as the total of the amplitudes £ A ; of all the X-ray bursts within the time interval of

257

I s (Fig. 1). Then the relative intensity was given to a physical magnitude by the intensity value measured by the TLD detectors. The experiments using the PM-scintillator and shields made of the beryllium foil with thickness of 15 and 30 /*m gave the assessment of the X-rays energy value of Ex-ray ~ 1.0-2.5 keV (for different cathode materials, Table 1) that matched to the TLD detectors results well. 'x- ray, 4x106 • 3x106 2x106 1 x106 0 /(mA), 200 *

Photons/burst

(a) ! , , , . » *• •

;L •

!

_T^

0 A f = 10 n s . ' X-ray,

Photons/burst

t^s)

(b)

Af=10us

f(ns)

Figure 3. Typical oscillograms of bursts of the diffusive X-ray emission (PM-scintillator) during passing the discharge current. Ta-D2 at 175 mA. (a) PM-scintillator 21cm from cathode (as in Fig. la), (b) PM-scintillator 70cm from cathode (as in Fig. lb).

to Cn 00

Table 1. Material of cathode Glow discharge voltage (V) Glow discharge current (mA) X-ray energy during passing the discharge current, Ex-ray (keV) X-ray energy without current, _Ex-ray (keV) X-ray energy flow density, ip ( x l O - 4 W / c m 2 ) Number of X-ray pulses per second, Np ( x l O 5 pulses/s) Max energy of one X-ray pulse, i ? m a x ( x l O - 1 0 J) Number photons in one pulse, n (xlO 5 )

1650 130

Sc 1540 130

Ti 1730 170

Ni 1650 150

Mo 1420 210

Pd 1650 138

Ta 1600 138

Re 1520 125

Pt 1650 138

Pb 1610 138

1.54 1.68 1.2 3.8 1.2 0.50

1.26 1.5 1.7 3.7 1.5 0.74

1.45 1.46 3.18 6.0 1.9 0.83

1.91 1.96 1.2 3.4 1.5 0.49

1.48 1.33 1.36 2.7 1.5 0.63

1.98 1.71 1.4 4.0 1.3 0.41

1.62 1.62 2.13 5.1 1.4 0.55

1.36 1.38 0.74 2.2 1.1 0.87

1.47 1.75 1.9 4.4 1.6 0.68

1.36 1.45 1.7 4.4 1.3 0.94

Al

259

The dependence of changing the radiation intensity on the distance was determined using the experimental devices according to the diagrams in Fig. la, b. Magnification of the distance between the PM-scintillator detector and the cathode from 21 up to 70 cm resulted in reducing the radiation intensity more, than under the law l / r 2 (Fig. 3). Such result could be explained to the fact the radiation indicatrixes of the separate bursts had the elliptic shape with enough narrow angular orientation. The high intensity of X-ray emission allowed obtaining an optical image of the emission area. The pinhole camera with the hole with the diameter of 0.3 mm (as an optical lens) was used. The image shows that the cathode area with the diameter of 9 mm and especially its central part has the largest luminance (Fig. 4). The X-ray laser beam generation occurred under precisely fixed parameters and conditions of glow discharge.

(a)

(b)

Ncathode

Figure 4. The image of the X-ray cathode obtained using the camera obscura (pinhole camera). The objective with 0.3-mm diameter closes by the 15 /mi Be shield. With Pd—D2 and the discharge current of 150 mA, and the exposure time - 1000 s: (a) voltage is 1350 V, (b) voltage is 1850 V. The image is positive.

(1) The generation occurred only when periodic-pulse current was supplied. It did not occur with direct current, although X-rays as bursts of diffusion radiation did occur with direct current. (2) Some critical parameters of occurring the generation by the gas pressure of PGD in the discharge chamber and by the voltage of the discharge (7QD- The generation occurred at PQD(^GDcrit, UGD) ^GDcnt- A small change in the discharge pressure or voltage led to the occurrence of generation (the change in pressure was A P Q D = 0.2-0.3 Torr, and in voltage AUGD — 30-50 V). (3) These parameters were different for various cathode materials (Fig. 5). For example, when using Pd cathode, the X-ray laser generation occurred at pressure being twice as much as when using Ti cathode. (4) The parameters of occurring the X-ray laser generation depended also on the plasma-forming gas (Fig. 6). (5) When operating, the generation intensity gradually decreased (obviously

260

because of degradation of the cathode surface) and stopped in the course of time. This phenomenon was especially clear for cathode materials with a large coefficient of a material sputtered in the discharge plasma (e.g., Al, Pd, and Pb).

Photons/beam

X-ray. 1.2 x 1 0 8 0.8 x 10 8

(a)

0.4 x 10 s 0

*

Photons/beam

' X-ray.

I "11

8

1.2 x 10 (b)

0.8 x 10 8 f0.4 x 10 8 ' x - r a y | Photons/beam

1

1.2x1080.8 x 10 s 0.4 x 10 8

- I

,. 1 II 'I ' X-ray.

Photons/beam

!!

(c) ml''

ii:.,.

l

!

1.2 x 10 8 0.8 x 10 8

A t = 200 us

f(us)

Figure 5. The typical oscillograms of bursts from X-ray laser beams (PM-scintillator) in a D2 discharge for different kind of cathode samples: (a) Al, (b) Sc, (c) Pb, and (d) Ta. Assembly is by Fig. l a (the distance from the cathode to the detector is 21cm). *Pulse peaks are selected via a discriminator amplifier.

The X-rays as laser beams consisted of the separate beams, presumably, having a small diameter (up to 10 6 -10 10 photons in a beam). These magnitudes were obtained in assumption that the system of the PM-scintillator operated in the linear area, taking into account the magnitude of reducing the amplification constant of the path when recording the X-ray laser radiation. The X-ray laser beams emission occurred during the discharge burning and within up to 100 ms after turning off the current. At the specific parameters of the discharge the generation of the X-ray laser beams was observed only some ms later after turning off the discharge current (up to 20-30 beams after each current pulse). The time oscillograms type of the generated beams depended on the type of the plasma-forming gas (Fig. 4) and type

261

' X-ray> . Photons/beam " ] [ ' " '

1.2 x 1 0 8 -

; : m ^ < ^^_

/(mA) 200 0 A f = 100 ns t(ns)

Figure 8. The typical oscillograms of bursts from X-ray laser beams (PM-scintillator with optical filter) in the discharge for different kind of assemblyes. (a) The cathode sample is Ta, (b) the cathode sample is Mo, current, 100mA; D 2 . (a) Assembly is by Fig. 7a (the distance from the cathode to the detector is 21cm). (b) Assembly is by Fig. 7b (the distance from the cathode to the detector is 70 cm). *Pulse peaks are selected via a discriminator amplifier.

made of Al, Sc, Ti, Ni, Nb, Zr, Mo, Pd, Ta, W, and Pt at current up to 500 mA, and the discharge voltage of 500-2500 V. The pulse-periodic power supply of the glow discharge with the pulses duration of the discharge current of t = 0.3-1.0 ms and period of T = 1.0-100 ms was used. The targets as the shields made of various materials foil (Al, Ti, Ni, Zr, Yb, Ta, and W) with thickness of 10-30 mm and of 1.0-

Figure 9. The increased negative image of the flare spots of the roentgen laser beam tracks for the different distances from cathode, the roentgen film (Kodac and XBM) covered with the 15/im Al shield. The system P d - D 2 , the discharge current - 130mA; the exposure time - 1000s; (a) for 100 mm from the cathode surface, (b) for 210 mm from the cathode surface. The image is negative.

264

3.0 mm were arranged at a distance of 21 and 70 cm from the cathode (Fig. 10a, b). The scintillation detector supplied with the photomultiplier was used for recording the secondary radiation. The device with a channel for extracting the radiation with the length of 70 cm and the magnetic system creating a cross magnetic field relatively to the radiation axis at a distance of 35 cm from the cathode was used for defining the type of the secondary radiation (Fig. 10c).

(a, b)

(c)

Figure 10. Schematic representation of an experiment with X-ray targets (secondary penetration radiation research). PM-scintillator system. 1, cathode sample; 2, anode; 3, 15 /jm Be foil screens; 4, scintillator; 5, photomultiplier; 6, magnetic bar (magnetic induction between magnetic poles is about 0.2 T); 7, X-ray targets made a foil of various materials.

The procedure of the secondary penetrating radiation registration and calibration of the detector was similar to the procedure of registration the primary radiation beams. 1 Under the experiments the recording of the time radiation spectrums was carried out within the time between the back and forward fronts of the discharge current (being free of the discharge current). The radiation type was defined with the device with a channel for extracting the radiation with the length 70 cm when being free of the cross magnetic field and the imposed magnetic field was available (Fig. 10b, c). When free of the magnetic field, significant attenuation of the PM-scintillator signal was not observed when increasing the distance from the cathode to the detector from 21 to 70 cm (Fig. 11a). Superimposition of the magnetic field with an

265

'ei. beams, electrons/beam _ - * v

J 1' 1;

1.2 x10 8 0.8 x10

8:

|

1 1 1

0.4 x10 8 0 /(mA) 200 : 0 3lectrons/beam ' el. beams'

|—k.

i

i T1

._

if

ii

II i

j l

f

r

i

|

i

i

f

i

30707-08

„

IYI

1.2 x10 8 0.8 x10 8 0.4 x10 8 0 /(mA) 200

o

T1

•

^~_

k_

.1 i

30707-20 in*

Af=100us*4-4 -

— 0 *

Figure 11. The typical oscillograms of bursts from secondary penetration radiation beams (fast electrons) in the discharge for different kind of assemblies. The cathode sample is Ta; current, 100 mA; D2. (a) Assembly is by Fig. 7a (target arrange at a distance of 21cm from cathode without superimposition of the cross magnetic field), (b) Assembly is by Fig. 7c (target arrange at a distance of 21 cm from cathode with superimposition of the cross magnetic field), secondary penetration radiation (fast electrons) don't registered. *Pulse peaks are selected via a discriminator amplifier.

induction of 0.2 T led to complete disappearing of the signal in the PM-scintillator (Fig. l i b ) . Thus, the secondary radiation was the flux of the charged particles (presumably fast electrons) with a small angular divergence. W h e n increasing the distance from 21 t o 70 cm, t h e primary X-ray laser beams kept an ability to generate the secondary radiation when interacting with the targets made of various materials (Fig. 12). These results were the additional confirmation of the fact t h a t the X-ray laser beams had a small angular divergence. T h e type of the oscillograms of the primary radiation bursts was defined by the cathode material. T h e secondary X-ray radiation of two types was observed. (1) Radiation with a continuous time spectrum as separate bursts with intensity up to 106 photons per a burst. This emission began 0.5-1.0 ms later after turning off the discharge current. (2) Radiation with a discrete time spectrum and radiation intensity up to 10 9 photons per a burst. Distribution of the bursts by the time of this radiation was defined by the target material. T h e generation of the secondary penetration radiation is supposed to occur from solid medium of the lead shield. T h e results of recording the radiation bursts were used for the time spectrums construction. T h e dependence of the radiation bursts amount on the time interval between the back front of the discharge current impulse and forward front of the radiation bursts was under construction. T h e time spectrum of the primary X-ray laser radiation had a discrete character. T h e time spectrum of the primary X-ray

266 / x-ray. Photons/beam 1.2 x 10 8 0.8 x 1 0 8 0.4 x 1 0

8

IL

0 /(mA) 200 0 / (el.beamsi electrons/beam

f

(a)

r

1.2 x 1 0 8 0.8 x 10 8 0.4 x 10 8 0

!l

el.beams; electrons/beam

1.2x108 0.8 x 10 8 0.4 x 10 8 0 A t = 200 us

HP+-

?(us) Figure 12. The typical oscillograms of bursts from (a) primary (X-ray laser beams from cathode) and secondary penetration radiation beams (fast electrons) in the discharge for different kind of target materials. The cathode sample is Ta, D2; current, 180 mA. Assembly is by Fig. 2b (target arrange at a distance of 70cm from cathode), (a) primary penetration radiation from cathode, (b) Al target of 1.4-mm thickness, (c) Yb target of 1.8-mm thickness. Pulse peaks are selected via a discriminator amplifier.

was a function of the cathode material. Separate bursts were recorded within 85 ms after turning off the current. The time spectrum of the secondary radiation also had a discrete character, but the type of this spectrum was a function of the target material. Also this secondary radiation is registered using the X-ray film arranged behind the lead shield (Fig. 13). A third type of the penetrating radiation was observed as well. This was radiation recorded directly by the photomultiplier placed behind of the target without the scintillator (Fig. 14). In this scheme, the target was arranged between the shield with the thickness of 3 mm made of plastic and the PM detector. The type of the secondary radiation was defined by the detector material. This emission began 20min after turning on glow discharge current and emitted after turning off the discharge current of 20min and more (Fig. 14). An abnormal high penetrating ability of this radiation type requires additional research. 4. Discussion The features of the X-rays recorded in these experiments are as follows: • The X-rays leaves the solid-state medium of the cathode material.

267

Figure 13. The X-ray film exposed by the secondary penetration radiation from the lead screen with the thickness of 2 mm. The system M0-D2, current 220 mA, the exposure time - 720 s. 1, X-ray from cathode; 2, 15 /jm Al shield; 3, 2 mm lead target; 4, X-ray film; 5, area of X-ray film behind lead shield, it is the photoemulsion solarization presumably; 6, area of X-ray film behind 15 jjira Al shield only. The image is negative.

• The intensity of the X-rays increases 5-6 times when increasing the discharge voltage by 1.3-1.4 times. • The quantity of the X-rays energy is essentially not changed in this case. • The X-rays emission occurs within 100 ms after turning off the discharge current. The obtained results are the direct experimental proof of existing the excited metastable energy levels with the energy of 1.5-2.5 keV in the solid of the cathode sample. Presumably, these excited metastable levels are formed in the volume of separate crystallites. These excited metastable levels exist for the time of AT mst (up to 100ms and more). Then the relaxation depopulation of these levels takes D, mGy/s (dose rate)

0 «4* H + •

A

a ? 1

0.01

400

t (s)

20 |im Mo target 30 urn Ni 35 urn W 10nmTa 25 |im Zr 50 urn Nb 100 urn Al

800

(after GD switch off)

Figure 14. The third type secondary penetration radiation dose rate dependence upon the time after turning off the discharge current. The cathode sample is Ta, D2; current, 100mA; voltage, 2000 V. 1, 15/im Be foil shield; 2, scintillator; 3, X-ray targets made of foil of various materials; 4, photomultiplier.

268

place, being accompanied with the emission of the X-rays and fast electrons. These beams generation occurs from the solid-state cathode medium presumably for one passing in the mode of super luminance. In this case the duration of the beams should be of K r n - 1 0 - 1 3 s . Understanding the mechanism of these level of formation will require additional research. T h e existence of one of the two physical phenomena can be assumed: (1) Excitation of the interior L, M electronic shells without ionization of the outer electrons. (2) Vibrational deformation of the electron-nuclear system of the solid ions. T h e core of electronic shells is displaced relative to a nucleus with forming a dipole (an optical polar phonon). T h e frequency of the formed phonon is much greater t h a n the plasma frequency in a metal.

5.

Conclusion

Experimental research into this fundamental phenomenon has allowed us to create what is essentially a new type of device: the X-ray solid-state laser with a wave length of the radiation of 0.6-0.8nm, duration of separate pulses of 1 0 _ 1 1 - 1 0 _ 1 3 s and beam power in pulses up to 10 7 W. T h e results show t h a t creating optically active medium with long-living metastable levels with the energy of 1-3 keV and more is possible in the solid state.

References 1. A.B Karabut, Research into powerful solid X-ray laser (wave length is 0.8-1.2 nm) with excitation of high current glow discharge ions, in Proceedings of the 11th International Conference on Emerging Nuclear Energy Systems (Albuquerque, New Mexico, USA, 29 September-4 October 2002), pp. 374-381. 2. R.B. Firestone, Table of Isotopes, Eighth Edition, Vol. 2, Appendix G-1 (Wiley, New York, 1996). 3. R.C. Elton, X-ray Lasers (Academic Press, New York, 1990).

C H A R G E D PARTICLES FROM Ti A N D Pd FOILS

LUDWIK KOWALSKI Montclair State University, Upper Montclair, NJ, USA STEVEN E. JONES Brigham Young University, Provo, UT, USA DENNIS LETTS 12015 Ladrido Ln, Austin, TX, USA DENNIS CRAVENS Cloudcroft, NM 88317, USA

After familiarizing himself with the use of CR-39 detectors, about a year ago, the first author asked Steven Jones to send him a TiDj; foil, similar to that described at the Tenth International Conference on Cold Fusion.1 It was an attempt to detect 3MeV protons with the CR-39 chips. The idea was to develop an experiment suitable for student-oriented cold fusion projects. That is how the first author became a cold fusion researcher. After receiving the foil he sandwiched it between two CR-39 detectors for the period of 55 days. The area of each detector was one square inch. The exposure started 3 days after the sample was prepared (by keeping the titanium foil in deuterium gas at high temperature and pressure). The number of tracks counted on the face of the CR-39 detector that was applied to Jones' foil turned out to be 225. The opposite side of the same CR-39 detector (exposed to air) was used to count tracks due to our background. The number of background tracks turned out to be 132. Such results, if generated by a Geiger counter, for example, could be used as evidence of charged nuclear particles being emitted from the foil. The difference 225 — 132 = 93 is 4.9 times larger that the standard deviation of 18.9 (calculated as the square root of the sum of 225 and 132). A Montclair State University student, Marcee Martinez, was then asked to count the tracks again. First she "trained her eyes" by observing a CR-39 chip with tracks from alpha particles. Then she started counting "similar tracks." Her result was 165 for the signal and 124 for the background. This time the difference, 41, is 2.4 times larger than the standard deviation. Are conventional standard deviations, 19 and 17 (as above), appropriate indicators of uncertainty? They are probably not. A human being counting tracks must 269

270

frequently decide to either count or not to count a particular track-looking spot. The uncertainty associated with counting, the error of rejection, must be added to conventional standard deviation. The conventional standard deviation becomes negligible when the number of counts becomes very large but the error of rejection remains a constant fraction of the total number of counts. Suppose the counting situation is such that hesitation happens in 10% of cases and that the total number of counts is 900. In that case the error of rejection is probably close to 45 (5% of all counts). The conventional standard deviation (the square root of 900) is 30 and the sum becomes 75. The result should be reported as 900 ± 75, rather than as 900 ± 30. Richard Oriani 2 found a way to practically eliminate the error of rejection. But his method is more labor-intensive than the method we used. Instead of two detectors, one for the signal and one for the noise (background), as we did, he uses the same detector for both. This is accomplished by etching a single detector twice: before the experiment and after the experiment. After the first etching the surface is photographed through a microscope, field by field. In that way the preexisting background is recorded. The second etching takes place after the experiment and the same fields are photographed again. Then the photos are compared, again field-by-field. Only tracks that appeared after the experiment are counted. By his method the net signal of 132, e.g., would indeed be much more convincing. Fortunately, this was not a problem to worry about in another project involving track detectors. That project resulted from correspondence with Dennis Letts. He has a team of scientists investigating excess heat produced in electrolytic cells. Knowing about their apparent electrolyte boil-off,3 the first author asked for a chance to look at a possible "nuclear signature." Three palladium cathodes: Pd613, Pd-616, and Pd-615 were sent to the first author and he exposed them to the CR-39 detectors. The technique was the same as for the TiD x foils; the cathodes were sandwiched between pairs of detectors for several weeks, detectors were etched, and tracks were counted, under the microscope. The results were: (a) about 500,000 tracks on the two detectors sandwiching Pd-613, (b) about 11,000 tracks on two detectors sandwiching Pd-616, and (c) no tracks above the background for the Pd-615 cathode. These numbers are rough estimates, errors by the factor of two, or so, are not important in this particular context. Only then was the first author informed that the Pd-613 generated an unusually high amount of excess heat, the Pd-616 generated much less excess heat, and Pd-615 generated no excess heat at all. He was also informed that all three cathodes were cut from the same sheet of pure palladium and that the electrolyte used in the cells was prepared at the same time and kept in a container. The only difference was that several drops of an additive, labeled "sauce," were added to the electrolyte in which the Pd-613 cathode was used. That additive was known to contain a tiny amount of uranium (Fig. 1). After learning about this the first author asked for a sample of this sauce. Several drops of it were placed on a sheet of plastic and dried under a lamp to produce a

271

Figure 1. The photograph taken through a microscope, shows tracks over the area of the detector equal to 1.30 X 1.00 m m 2 .

layer of a dark residual. The CR-39 was at once deposited over that residual. At the same time another CR-39 was applied to the most active side of the Pd-613 cathode. Three weeks later the detectors were removed. They immediately revealed a large number of tracks. In fact the maximum track density at the cluster from the Pd-613 was essentially the same as three weeks earlier. These facts are consistent with the idea that excessive tracks were due to the contamination of the electrolyte to which the "sauce" was added. The very large difference between track densities from Pd616 and Pd-615, on the other hand, could not be blamed on contaminations because m these cases the electrolyte (and other materials) were exactly the same. The Pd616 produced excess heat and generated nuclear tracks; the Pd-615 did not produce excess heat and it did not generate nuclear tracks, above the natural background level (Fig. 2). Nuclear Signature Seems to be Real: It is important to emphasize that the "contaminating sauce" was not added to the electrolyte in which the other two cathodes (Pd-616 and Pd-615) were used. And yet the number of tracks due to the Pd-616, roughly 11,000, was found to be about 100 times higher than the number of tracks due to the Pd-615. This indicates that nuclear particles were detected at the surface of the Pd-616 cathode, long after the electrolysis. A skeptic might suspect that another alpha-radioactive contaminant (not the "sauce") might have been accidentally added to the electrolyte in which the Pd-616 cathode was used. If this were the case then both surfaces of the Pd-616 cathode would produce about the same number of tracks. In reality one side of the Pd-616 produced 8000 tracks while the other side produced 3000 tracks. How can this be explained? The electrolytic

272

Figure 2. The photograph through a microscope, shows tracks over the area of the detector equal to 0.25 X 0.19mm 2 .

cell was essentially mirror-like symmetric (a small cathode near the center and a spiral platinum anode, parallel to the walls of the beaker). Furthermore, clustering of tracks was discovered on the more active surface of the Pd-616 cathode. The tracks due to the Pd-613 cathode, by the way, were also distributed very unequally. One side produced nearly 500,000 tracks while another side produced only about 4000 tracks. Most of the 500,000 tracks were found in a cluster whose area was only a small fraction of the cathode area (see Fig. 2). It is difficult to explain strong clustering in terms of the contamination of the electrolyte. A more natural explanation is to assume that a very high concentration of tracks in a small area (about 2 or 3 mm) coincided with the spot at which heat was generated during the experiment. A tentative conclusion is that uranium contamination, in the case of Pd-613, was responsible for only a small fraction of what was actually observed. The first author agrees with the second author that excess heat demonstrations, designed to convince that something highly unusual (cold fusion) is taking place, should always be accompanied by attempts to display nuclear signatures. After all, there are many non-nuclear ways to generate excess heat, especially at the power level below one watt. A complete examination of all chemical processes taking place in a setup (to rule out chemical origin of excess heat) is much more demanding than using a nuclear detector of some kind. Cold fusion effects, if they are nuclear, must generate nuclear reaction products, either radioactive or stable. On May 2005 Ludwik Kowalski sent an e-mail message to Dennis Letts. He wrote: "I am going to galley proof our ICCFll presentation in Marseilles. This puts me in an awkward situation. If I were reporting on my own work I would add a short paragraphs, something like this: 'No additional experiments were conducted

273

to confirm observations reported 6 months ago. T h e unexpected delay is due to . . . ' B u t in this case I was only a messenger; you are real players. A reader is likely to be interested in the current status of our investigation. I think t h a t it is not right to report positive results only and keep negative results hidden. Do you agree? . . . " T h e following reply was received several hours later. "No additional experiments were conducted to confirm observations reported 6 months ago. T h e unexpected delay is due to the fact t h a t experiments seldom work on a schedule. The MOAC had to be modified slightly to re-store design stability and precision. Also, we have not observed laser-triggered excess power since August 2003. Of course I agree [with your last statement] - since changing metals at the end of 2004, my success rate has been zero. This is compared to a success rate of 87% during the years of 2000-2004. Other t h a n changing Palladium stock, I don't know what has caused the sudden loss of the laser effect. Experiments have been conducted in a high quality calorimeter (MOAC), in a moderate quality calorimeter (my Avanti) and on the open bench. T h e laser effect has not re-appeared under any of the above calorimetric conditions. Experiments are being conducted now to re-establish the laser effect or to explain why it stopped working. You may use this information in any way you wish, including an addendum. W i t h regard to reporting negative results, consider this: Cravens and Letts discovered the laser effect in September 2000 and reported the positive results publicly in August 2003. We spent 3 years testing the credibility of our result before reporting publicly. We anticipate behaving in a consistent manner now - we have negative results b u t we're not in a rush to report until we're sure t h a t we have negative results and t r y to provide some reasons why the results are negative. I believe t h a t reporting results formally by 2007 will be consistent with our previous work and should not be considered 'keeping negative results hidden'." References 1. S.E. Jones et al., Charged-particle emissions from metal deuterides, in Proceedings of the 10th International Conference on Cold Fusion, 2003 (Cambridge, MA, USA). This paper can be downloaded from the library at http://www.lenr-canr.org. 2. R.A. Oriani and J.C. Fisher, Energetic charged particles produced in the gas phase by electrolysis, in Proceedings of the 10th International Conference on Cold Fusion, 2003 (Cambridge, MA, USA). This paper can be downloaded from the library at http://www.lenr-canr.org. 3. D. Letts and D. Cravens, Laser stimulation of deuterated palladium: past and present, in Proceedings of the 10th International Conference on Cold Fusion, 2003 (Cambridge, MA, USA). This paper can be downloaded from the library at http://www.lenrcanr.org.

CR-39 TRACK D E T E C T O R S IN COLD F U S I O N E X P E R I M E N T S : REVIEW A N D PERSPECTIVES

A. S. ROUSSETSKI P. N. Lebedev Physical Institute, Russian Academy of Sciences E-mail: [email protected]

1. Introduction Earlier experiments 1,2 have showed emissions of DD-reaction products (3-MeV protons) and energetic charged particle emission (a-particles) during exothermic D(H) desorption from the Pd/PdO:D(H) heterostructures. The occurrence of these emissions was confirmed by independent experiments using both Si-surface barrier and CR-39 plastic track detectors. 3 ' 4 Over the last few years, the CR-39 track detector has become a popular method to measure charged particle emissions in cold fusion experiments. The use of CR-39 is quite simple and cheap, but this technique demands some special conditions. As shown below, the observance of these conditions allows the researcher to not only detect charged particles, but also to identify their types and estimate their energy spectrum. On the other hand, when these conditions are not met, the method can fail, mainly because of problems with radioactive nuclides contained in the surrounding materials (electrolyte, air, cathode, etc.), mechanical defects, and the development process after etching. The necessary conditions for correct CR-39 measurement include the following: - purification of CR-39 detector surface, - low density of background tracks before measurements, and protection of CR-39 from external radioactive contamination, - correct calibration procedure, - control of background during the measurement and the use of "clean" materials (without radioactive nuclides), - protection of detector surface from mechanical and thermal influences, high intensity UV radiation, - in some tests, shielding foils with known stopping ranges identify particle types, - correct etching conditions in each measurement. Access to purified track detectors along with knowledge of track characteristics produced by various types of particle allow the use of CR-39 chips in long-duration experimental exposure during and after electrolysis with Pd and Ti cathodes. We 274

275

used purified CR-39 detectors produced by Landauer Inc. (USA) and Fucuvi Chemical Industry Co. (Japan) with very low-background track density (N^ < 20cm - 2 ) to detect the emissions with very low intensity (10~ 4 —10 -3 c m - 2 s _ 1 ). 2. CR-39 Track Detector CR-39 is a polymer (Ci2H 18 0y) with a density of ~1.3g/cm3. When a charged particle crosses the detector surface it causes radiation damage along trajectory. This zone of structure damage may be increased to 10~ 4 — 10~2 cm by etching in a chemical reagent. We used optimal etching conditions for CR-39 detector: 6N solution NaOH at 70°C over 7h. V = V T / V B = f{dE/dx) is the etch rate ratio (where dE/dx is the stopping power of particle, Vrand VB is the track and bulk etch rates). A track is formed if VT > VB. If VB = VT, dE/dx = (dE/dx)s is the threshold stopping power. The critical angle of detection 6C is determined as the minimal angle between the particle trajectory and detector surface when track formation is still possible: 8C = arc sh^Ve/Vr). It is easy to shown that if detector crossed by particles in different directions the detection efficiency is determined as rj = 0.5 * (1 — sin# c ). 3. Experimental Technique The measurements were carried out using PAVICOM4 - a completely automated device for track detector processing. It consists of an optical microscope with a digital video camera; a sample holder stand with three-dimensional displacement provided by stepper motors with accuracy of 0.25 /mi; manual controls; and a personal computer for automatic measurements. We used a-sources and a cyclotron alpha beam (Ea = 2-30 MeV) to calibrate the CR-39. Figure 1 shows the photomicrograph of tracks from a-particle cyclotron beam (Ea = HMeV) normally incident on CR-39 detector. For proton calibration we used a Van de Graaf accelerator (Ep = 0.6-3.0 MeV). The results of the calibration - the dependence of track diameter from particle energy - are presented in Fig. 2 for a-particles, and Fig. 3 for protons. We used the following types of samples in our experiments. Electrochemically loaded Au/Pd/PdO:D x heterostructures were used for charged particle emission during deuterium desorption after electrolysis. Ti (30 /mi), Pd (30 /mi) cold worked foils and Pd/PdO heterostructures (50 and 100 /mi) were used for in situ measurements during electrolysis in a solution of 1M L12SO4/H2O (current density j = lOmA/cm 2 ). Pd (30 (an) and Pd/PdO (50/mi) samples with He implanted on

276

Figure 1. Tracks from o-partjcle cyclotron beam detector. Image size - 120 X 90 /im.

I 1 MoV) normally incident on CR-39

(/•;„

the surface 2 x 10 17 and 2 x 1016 at./cm 2 , respectively, were used to investigate the influence of He impurities on alpha-particle emission during electrolysis. The example of long duration background measurements with a 0.1 M Li2S04 solution is presented in Fig. 4. The main part of background tracks spaced in the range of diameters 8-12 fim. We can estimate the mean level of background a-emission as (fib) = 2.7 x 1 0 _ 4 s _ 1 c m - 2 .

Alpha calibration curves for Fukuvi and Landauer CR-39 detectors 13 12 11 10 — Fukuvi - Landauer

9 8

'U-

7 6 5

10

Figure 2.

15 20 E a (MeV)

25

30

Alpha-sources and cyclotron alpha.

277

Prolon calibration curves (or Land tier and Fukuvi CR-39 detectors 9

1

I

1

8.5 8 7.5 7 6.5 6 5.5 5 4.5 05

•V 1

1.5

2

- Landauer CR-39 •

2.5

3

Fukuvi CR-39

3 .5

Ep(MeV)

Figure 3. Proton calibration with Van de Graaf beam calibration (2-30 MeV) of CR-39. Accelerator (0.6-3.0 MeV).

4. R e s u l t s and Discussion The measurements with Pd/PdO (50/tm) sample after electrolysis showed that hifi.li energy charged particle emission lake place not only on the whole sample surface, but also it can concentrate in some small zones ("hot zones") with dimensions of a few hundred microns corresponding to places with maximum internal strains. The photomicrographs of "hot zone" (250 x 500//in 2 ) with ~10 3 tracks of a-particles and protons are showed in Fig. 5a,b. The distribution of track diameters for particles normally incident to the detector is showed in Fig. (i. We can see three peaks that corresponded to following particles and energies: d = G.O

Open CR-39 background, f = 335 h Open CR-39 background. (= 216 h Open CR-39 background. (= 118 h

10 Track diameter (um)

Figure 4.

I MIL 11

12

Example of background during electrolysis in 0.1 M L12SO4.

278

(n)

(b)

Figure 5. CR-39 measurement after electrolysis of P d / P d O i H ^ (50jitm). Shielding of CR-39 11 /im of Al. Photomicrographs of "hot zone" (250 X 500 /jm 2 ) with tracks of a-particles and protons. Image size — 120 X 90 fim.

6.5/xm (Ep = 1.4-1.7MeV), d = 7.0-7.6/xm (Ea = 9.2-14.0MeV), d = 7.8-8.6/xm (Ep = 5.6-7.8 MeV). The distribution of track diameters for the entire detector surface (not just in the "hot zone") is presented in Fig. 7. We observe the same peaks in this distribution. This means that the charged particle emission in the "hot zone" is the same as the rest of the sample, but the emissions are concentrated in space (and, possible, in time). We estimated the mean intensity of charged particle emission (protons and a-particles, without "hot zone") during all time of exposition (5h) as (n) = 6 x 10~ 2 s" 1 c m - 2 . a-Particles emitted from one point on the sample surface. We observed such events both after electrolysis and during long duration electrolysis measurements. Examples of these events with Au/Pd/PdOiD^, (40 /iin) and PdHe

I H Pd/PdO:Hx, S = 250 x 500 (im, CR-39/5.0 Jim Al Exothermic H-desorption Burst of emissions (photo)]

20T J

18-'

1614-

6.5

Figure 6.

7.0 7.5 8.0 Track diameter (fim)

9.0

Distribution of track diameters in.

279

|Pd(bgr)CR-39.open,l = 48h I Pd/PdO:Hx(lg). CR-30. open, r= 5.0 h I (Pd/PdO:Hx(lg) - Pd(bgr))

30

25

•f 20

I

& g

15

*

10-

6.0

6.5

7.0

J

7.5

3.0

8.5

9.0

Track diameter (urn) Figure 7. D i s t r i b u t i o n of track d i a m e t e r s on "hot zone" (250 X 5 0 0 / / i n 2 ) C R - 3 9 surface w i t h o u t "hot zone".

(30/nn. 2 x 1 0 , 7 a t . H e / c m 2 ) samples are presented in Figs. 8 a n d 9, respectively. In these cases we can t o estimate the energy of emitted a-particles, measuring their ranges from the point of emission to the end of tracks. Taking into account the thickness of detector shielding (50 and 6 0 / m i P E , respectively) and etched layer of CR-39 ( ~ 9 / m i ) . we can t o estimate the energies of emitted a-particles: EQi = En2 = 11.2 MeV; Ea3 = Eo4 = 9 M e V ; Ea5 = Ea6 = 8.3 MeV; Ea7 = lOMeV (see Fig. 8) a n d Eai = 9.2 MeV: Ea2 = 8.6 MeV; Ea3 = 9 MeV; Ea4 = Ea5 = Ea6 = 8.4 MeV (see Fig. 9). T h e energies of all these a-particles are more t h a n energies of n a t u r a l background and they can be identified as long range particles emitted from the sample surface under high internal strains. These events d e m o n s t r a t e the release of few tens MeV of energy concentrated in small volume of sample. If we suppose t h a t the energy concentrated in the region compared with dimension of the lattice, and released in a time period of 1 0 - 1 2 ~ 1 0 - 1 4 s , we get the power ~ 1 0 1 8 - 1 0 2 0 W / c m 2 .

5. C o n c l u s i o n (1) We formulated the necessary conditions for successful using of CR-39 detector in typical (•
cP c % 3

°

o

°.00"Q

Figure 1. Etch pits on the surface of a CR.-M9 plastic detector chip .suspended in the vapor over an active electrolysis cell. Each pit marks the location of a track of damaged material where a charged particle has penetrated the chip. A roughly conical pit has developed during etching because the etchant attacks the damaged material of the track more rapidly than it attacks the adjacent undamaged material. The area shown measures approximately 0.29 mm X 0.22 mm. The mean diameters of the darker circular pits are approximately 24 /an.

We counted 29,800 etch pits in the portion of the chip available for counting, and estimate by interpolation that another 3300 particles passed undetected into unavailable areas, giving a total of 33,100 charged particles impinging upon or passing through the 1044 images. By analyzing the shadow cast by the support hook we estimate that an additional 240 particles were stopped by the wire and did not reach the detector chip. Adding these the final total is about 33,300 charged particles impinging on the area of 1044 images.

284 A

\ )

\

I

/

I

//n>

/ ' / I

\

^

\

///C..

rtitti ;\.J l' .'

)<mw>^

am m m

Figure 2. Distribution of etch pits on the surfaces of CR-39 plastic detector chips suspended in the vapor over an active electrolysis cell. (A) The more heavily pitted chip. Density in pits per image is indicated by the color scale for a mosaic of 1044 images that span the approximately 8-mm 2 surface. Outlined images could not be counted and etch pit densities were interpolated for them; those in the upper portion of the chip correspond to the support hole and those in the lower portion correspond to laser-inscribed identification numbers. The solid contour lines are spaced at 10 pits/image and range from 10 pits/image along the right side to 110 pits/image near the peak. The dashed contour line denotes 3 pits/image. Arrows indicate mean orientations of tracks having elliptical etch pits. Arrow lengths are proportional to the mean cosines of the angles between individual track projections on the surface of the chip and the corresponding arrow orientation. (B) The facing chip at a distance of about 8 mm. A 2 mm supporting rod ran vertically between the detectors. The smoothed contour lines are spaced at 3 pits/image and range from 3 pits/image on the left to 30 pits/image at the peak. The area of low etch pit density over much of the detector lies in the shadow of the supporting rod. It indicates that the shower originated in a small volume close to and nearly behind the support rod. The edge of the shadow is not parallel to the edge of the detector chip, suggesting that the chip was canted with respect to the rod or that the active volume moved sideways as it rose.

A smoothed contour plot of the density of etch pits is shown in Fig. 2A. The units are etch pits per image. (Because the area of an image is about 6.3(10)~4 cm2 one must multiply by 1600 to obtain etch pits per cm 2 .) The solid contour lines are spaced at 10 pits/image and range from 10 pits/image along the right side to a maximum of 110 pits/image near the lower left corner of the chip. From its maximum the density of pits falls to below 10 pits/image near the right side of the chip, reaching a level of 3 pits/image at the dashed contour line. This is the etch pit density we customarily find in chips exposed to the vapor in experiments where we do not see a massive shower such as the one under discussion. The dashed contour line thus marks a boundary beyond which no shower particles left tracks capable of producing pits upon etching. In addition to its prominent position along the right side of the chip this boundary comes close to the lower left corner of the chip. (The image at the lower left corner of the chip contains 10 pits. The 10 pits/image contour line should pass through it, although this is not shown in the figure because the

285

contour algorithm requires more than a single data point to determine a contour segment. The gradient of pit density suggests that the 3 pits/image contour line then lies just beyond the corner of the chip.) Also shown in Fig. 2A is an indication of the directions of the tracks near the perimeter of the densely pitted area. Track orientations can be determined from the shapes of the etch pits as described more fully below. Eight areas were selected around the perimeter and one near the center of the chip, each consisting of four contiguous images. Within each such area the orientations were determined for those etch pits for which a clear measurement was possible. Where necessary the microscope was focused at several levels from the surface into the interior of the plastic to aid the determination. Each pit with a measured orientation was assigned a unit vector in the direction of the track as seen in its projection on the surface of the chip. These vectors were added to obtain the mean track direction, and the cosine of the angle between each constituent vector and the mean track orientation was determined. The mean cosine provides a measure of the extent to which the vectors are aligned. When the mean cosine is near unity the constituent vectors must be nearly parallel. When the mean is near zero the constituent vectors must tend to point equally in opposite directions. The arrows in Fig. 2 indicate the mean orientation of the tracks in each area, and their lengths are proportional to the mean cosines of the track projections along these directions. Clockwise beginning at the lower left corner the mean cosines and (in parentheses) the numbers of tracks from which they were determined, are 0.956(35), 0.387(43), 0.726(30), 0.453(49), 0.686(20), 0.925(30), 0.838(20), 0.798(72); and for the central arrow they are 0.331(58). We see that in the lower left corner of the chip, and along the lower contour lines on the right side, the tracks are strongly aligned pointing away from the central region of high track density. On other parts of the perimeter the tracks also tend to align pointing away from the central region but with more scatter as indicated by the smaller values of mean cosine. It is clear that the tracks originated somewhere in the vapor above the densely pitted surface of the chip. But they cannot have arisen from a stationary source because the extended region of high track density does not have rotational symmetry. Working back from the boundary determined by the dashed contour line, we can obtain a rough idea of the height of the particle source above the surface of the chip. During etching a roughly conical pit develops because the etchant attacks the damaged material of the track more rapidly than undamaged material. Etching causes the vertex of the cone to move into the plastic more rapidly than it causes the surface to recede. The vertex points in the direction that the energetic charged particle traveled as it entered the plastic. The axis of the cone coincides with the track of travel and the shape of the etch pit depends on the orientation of the axis and on the half-angle of the cone. When the axis is nearly perpendicular to the surface the intersection of the etch pit with the surface is nearly circular. It becomes increasingly elliptical as the orientation of the axis tilts away from the perpendicular.

286

Now consider the etching process when the damage trail makes only a small angle with the surface of the chip. Consider a spot on the damage trail inside the chip. This spot lies closer to the surface than to the beginning of the damage trail where the particle entered the plastic. As the etching process proceeds the surface of the chip etches toward the spot at a steady rate. Etching proceeds along the damage trail at a faster rate but it has farther to go. At a critical angle, equal to the half-angle of the cone, the surface and the apex of the cone reach the spot at the same time. For damage trails with angles smaller than the critical value relative to the surface of the chip the surface gets there first and no etch pits can form. All shower particles must have been generated near enough to the chip that none of them left an etchable track beyond the dashed boundary in Fig. 2A. This implies that beyond the boundary none of them made an angle as large as the cone half-angle with the chip surface. Measurement and analysis of cone angles indicates a half-angle of about 19° as described below, so we can deduce that all particles were generated below a sloping surface that rises from the boundary at a slant angle of 19° in the direction opposed to the arrows in the lower left corner and along the right side of the chip. The tent-like roof suggested by these rafters reaches a height of about 1 mm from the chip surface in the neighborhood of the peak, indicating that the source of energetic particle generation did not extend above this level. It reaches a height of about 3 mm near the upper edge of the chip suggesting that the particle source moved away from the chip or grew in diameter as it progressed upward toward and past the support hole. We envision an active volume having an initial diameter of a fraction of a millimeter that began emitting particles in the vapor about 1 mm from the chip near its lower left corner, just touching the tent-like roof, then moved upward along the chip in a wandering path occasionally touching and defining other portions of the roof along the way. Because the densely populated area extends beyond the dimensions of the chip in some directions, only a rough estimate can be made of the total number of charged particles generated in the shower. We estimate that about 50,000 etch pits would have been counted had the chip been sufficiently extended, and considering that tracks making angles less than 19° with the surface do not produce etch pits we estimate about 150,000 charged particles as the total number in the full 4ir steradians of the shower. The second chip is slightly larger than the first. It was photographed in a mosaic of 285 photographs, from which a montage of 1140 images was obtained by quartering each photograph. Following the same procedure as with the first chip we found that a total of about 10,700 charged particles left etch pits in the detector chip or passed undetected into areas unavailable for counting. A contour plot of the density of etch pits for the second chip is shown in Fig. 2B. A portion of the chip exhibits a maximum in track density roughly opposed to the peak density in Fig. 2A. In this area the variation of track density with position mirrors that in Fig. 2A with track density values that are about one-third as great. The rest of the chip shows a very low level of track density, comparable with the background level in Fig. 2A,

287

that we interpret as lying in the shadow of the 2-mm rod from which the chips were suspended. It appears from the orientation of the boundary of the shadowed region that the chip was hung in a canted orientation with respect to the support rod, or perhaps that the active volume moved sideways as it rose between the detector chips. Because the shower originated in the vapor above a relatively warm electrolyte we expect convection currents that would carry away the vapor near a chip in a matter of seconds. The fact that we observe a somewhat confined and well-defined volume in which the particles originated suggests that the duration of the major part of the shower did not exceed a few seconds. The arrows in Fig. 2A and B provide aggregate measures of the orientations of particle tracks in selected small target areas. Each aggregate includes tracks that produced elliptical etch pits for which track orientations could be determined. Tracks having nearly circular pits, for which no orientations could be established, were necessarily omitted. If we consider a target area directly under a distant source, such that a perpendicular from the source to the plane of the target lies within the target, tracks formed in the target can make only small angles with the perpendicular. The more distant the source the smaller the angles will be and the more nearly circular the etch pits will be. Depending on the size of the source, its distance from the target area, and the ability of microscopic examination to detect small differences from circularity, it can turn out that no orientations at all can be established in a target area directly under the source. In this event orientations can be determined only if the target area lies off to one side of the source. Figure 3 summarizes individual orientations for the six target areas indicated by arrows in Fig. 2B. Tracks in target areas A-C on the left-hand side of the heavily pitted portion of the chip point to the left, away from the heavily pitted area, indicating that the responsible particles came from the direction of the heavily pitted area to their right. For these target areas track orientations were obtainable for an average 62% of all etch pits. The patterns of tracks are quite different in the three targets on the right-hand side of the chip, for which orientations were obtainable for only 27% of etch pits. Tracks in target area F in the lower right are confined to a pair of narrow angular spreads pointing upward and downward. The upward-pointing tracks correspond to a source below the target and the downwardpointing tracks correspond to a source above the target. We interpret these tracks as originating in a compact source that moved upward in the vapor approximately 7 mm distant from the chip at the beginning of its trajectory. Before the source reached a position above the target area the tracks were directed ahead of it in the upward direction. After it had passed the target the tracks were oriented behind it in the downward direction. When the source was directly overhead all tracks were nearly perpendicular to the target, the etch pits were nearly circular and their orientations could not be determined. The track orientations in target area E indicate that the compact source passed nearly over this area as well, but slightly to the right beyond the edge of the chip. The orientations in target area D suggest that the source continued its motion along

288

Figure 3. Individual orientations for tracks that produced elliptical etch pits in target areas A - F in Fig. 2B. As described in the text the pattern of tracks in areas D - F indicates that a compact source of energetic particles drifted in the vapor in the general direction from F toward D, passing over and progressively to the right of these target areas.

more or less the same trajectory. In each of the images D-F where there is a clear division into two distinct spreads of orientations, the upward-pointing tracks were generated earlier than the downward-pointing ones. Because the activity of the source had a short lifetime, we expect the number of upward-pointing tracks in each target to exceed the number of downward-pointing ones, reflecting a decline of activity between the times at which the leading and following tracks were formed. The data are consistent with this expectation, recognizing that the compact source first became active near location F leading to a shortfall of upward-pointing tracks at that location from what would be expected had the source become active farther away. The patterns seen in Fig. 3 have counterparts in the data from Fig. 2A, but because the compact source was much closer to the first chip than to the second, the angle subtended by the source as seen from a target area in Fig. 2A was large enough that particles originating from the perimeter of the source produced oriented tracks at a location even when the center of the source was directly over that location. These tracks from a nearby overhead source add clutter to the record of a moving source that otherwise is so clearly evident in Fig. 3 for a more distant source.

289

4. Particle Track Analysis Etch pits arise from various causes. Some result from alpha particles from decay of radon in the air or from superficial damage to the plastic detector and others from the energetic particle tracks that are the subjects of this analysis. Track pits initially are conical in shape with the vertex located on the track and the axis of the cone aligned with the track. The conical shape arises because damaged material along a track etches away more rapidly than the surrounding undamaged material. At first the depth and diameter of a pit increase proportionately as etching proceeds, but when the end of the track is reached the etching rate in the track direction slows to that of undamaged material and the point of the cone rounds out. The pit has "bottomed out." The portion of the pit near the surface of the plastic continues for a time to retain its conical shape and its diameter grows at an unchanging rate. Then as the surface of the plastic etches away to the depth of the end of the track the pit completely loses its conical shape and it progressively approaches the shape of a nearly circular dish. Pits from superficial damage bottom out very soon because the damage extends only a short distance into the plastic. They become shallow circular dishes shortly after etching begins. Examination of tracks produced by radon from a pitchblende source indicates that the track lengths for radon alphas are only slightly longer than the depth of plastic that is removed by the etching process. Tracks that are nearly perpendicular to the surface of the detector produce sharp-pointed pits, but tracks that make a modest angle with the perpendicular bottom out because the surface etches down past their far ends. Bottoming out is particularly pronounced when tracks from radon alphas are etched a second time. After the second etch all pits are circular or nearly circular dishes with diameters nearly twice that of single-etched tracks. One such pit is visible at the top of Fig. 1. In order to eliminate background events and to distinguish between families of etch pits, we employ various combinations of the following data restrictions. Let R be the ratio of the major axis to the minor axis of a pit. By imposing the restriction R > 1.1 to remove circular and nearly circular pits we can eliminate most of the pits that arose from sites of superficial damage, and also the pits that were formed in the initial etch and then were etched again after electrolysis. By imposing the restriction i? < 1.5 we retain only those pits whose geometric mean diameter is a good approximation to the diameter of a circular pit of the same energy impinging normal to the chip surface. By imposing a restriction to include only those etch pits that are observed under the microscope to be sharp-pointed cones we retain only pits that have not bottomed out. And conversely by imposing a restriction to exclude pits with sharp-pointed conical shape we retain only pits that have bottomed out. In the following analysis we always apply the restrictions 1.1 < R < 1.5 and we sometimes additionally apply restrictions relating to the presence or absence of sharp-pointed conical pits. We expect that some of the tracks recorded in the shower chips were caused by alpha particles from decay of radon in the laboratory environment. The magnitude

290

of such contamination was explored in a number of control runs in which detector chips were suspended in the vapor over the electrolyte in the absence of electrolysis.6 The chips were etched to reveal pre-existing tracks, and then they were photographed in tagged areas that could be identified later, were mounted in the inactive cell, exposed for three days, removed, etched and photographed again to reveal the tracks associated with influences of the laboratory environment during the initial photography, mounting, exposure, and etching procedures. The densities of etch pits from tracks formed during these procedures amounted on average to 150 ±70pits/cm 2 . In searching for tracks from radon contamination of the shower chips we first photographed, counted, and determined the shapes and mean diameters of all etch pits on the back side of the primary shower chip in Fig. 2A. The back side was shielded by the chip itself from the shower on the front side. The mean etch pit diameters are shown in Fig. 4A. Before applying the restrictions 1.1 < R < 1.5 to remove nearly circular and strongly elliptical pits the peak near 17/on contained about 160pits/cm 2 . This density lies within the range of control densities previously noted. Hence we conclude that the 17/xm pits indicate tracks from radon in the laboratory environment during photography before electrolysis and that they provide a standard for pits on the shower sides of both chips. Measurements of etch pit dimensions on the shower sides of the chips give quite different results. First, we examine the etch pits on the second shower chip in the shadow of the support rod which covers half the area of the chip. Here we expect alphas from radon decay as observed on the back side of the primary shower chip, and possibly other particles from decay of a few long-lived products of the shower reaction that may have drifted to where their decay products could reach the shadow region. Figure 4B shows the size distribution of these etch pits in the shadow region. Two peaks are evident, one near 17/xm corresponding to the peak in Fig. 4A and a new peak with larger etch pits near 24 /im having no counterpart in Fig. 4A. Microscopic examination shows that the larger etch pits have bottomed out near the end of the etching process, but to a lesser extent than the pits that result from double etching of radon alphas. We can analyze the two peaks by considering the three-dimensional shapes of the pits. In Fig. 5, we retain only those pits that are observed not to have bottomed out. Specifically, they must have the shapes of sharp-pointed cones. This restriction removes the family of larger etch pits, and the remaining distributions (Fig. 5A and B) for sharp-pointed conical pits are statistically indistinguishable. Both clearly reflect contamination by alpha particles from radon decay. Next, we turn attention to the family of larger etch pits that have just begun to bottom out. In Fig. 6, we plot distributions of pit sizes where now we retain only those pits that are observed to have bottomed out (i.e. they must not have the shapes of sharp-pointed cones). Figure 6A shows the distribution for pits on the shaded area of the chip in Fig. 2B, and Fig. 6B shows the distribution for pits on the front surface of the primary shower chip in Fig. 2A. These distributions

291

1

20

CD Q.

'o.

fi 10 LU

5

J*~\\ 5 10 15

20

25

rfl r'V... n... 30

35

40

Mean diameter (Mm)

50 .E

40

t 30 CD

*i

"Q.

o

20

LU

10 _ E 3

5

10

15

20

25

30

35

*"»

•

40

Mean diameter Gum)

Figure 4. Etch pit sizes (geometric means of maximum and minimum pit diameters) on two chip surfaces. (A) Back side of the chip in Fig. 2A, shielded by the chip itself from the particle shower on the front side. (B) Shaded area of the front side of the chip in Fig. 2B, shielded from prompt shower particles by the support rod. These plots include all etch pits with ratios of maximum to minimum diameters in the range 1.1 < R < 1.5. The restriction 1.1 < R eliminates the circular and nearly circular pits from pre-existing tracks. The restriction it < 1.5 eliminates highly elliptical pits for which the geometric mean diameter ceases to be a good approximation to the diameter for normal track incidence.

are statistically indistinguishable, strongly suggesting that the family of larger etch pits identifies tracks of charged particles emitted from precursors whose lifetimes are sufficiently long that a few have drifted into the region between the support rod and the shadowed area of the second shower chip. We now can confirm the 19° value of the half-angle of the shower pits whose trajectories bounded the location of the shower source. From Fig. 5 the mean diameter of the radon alpha tracks in the peaks of the distributions is 17.4 ± 1.0 /xm. From Fig. 6 the mean diameter of the new tracks in the peaks of their distributions is 24.1 ± 1.2 fira. Comparing the diameter of the new tracks to that of radon alpha tracks the ratio is (24.1 ± 1.2)/(17.4 ± 1.0) = 1.4 ± 0.1. Analysis of the etching process leads to a relationship between the half-angle 9 of a conical etch pit, the pit diameter D at the surface of the detector chip, and the depth of etching S of the flat surface, D/S = 2(1— sin 6)/ cos 8. Sharp-pointed conical pits from radon alphas occasionally are found on surfaces perpendicular to the detector chip such as the edges of the chip and the edges of laser-inscribed numerals. The side views of these pits facilitate measurement of cone angles, and suggest a half-angle 9 = 36° for

292

B

10

rL_X L r ^ 5

10

15 20 25 Mean diameter (jUm)

10

15

1. 20

25

40

30

30

35

40

Mean diameter fritn)

Figure 5. Etch pit sizes on the two chip surfaces of Fig. 4, subject to the data cut 1.1 < R < 1.5 as in that figure and additionally restricted to include only sharp-pointed conical pits that have not bottomed out. (A) Back side of the primary shower chip in Fig. 2A. (B) Shadowed area on the front side of the secondary shower chip in Fig. 2B.

radon alphas. Substituting this value of 8 in the equation gives (D/S)iadon = 1.02. Because the new alphas have diameter 1.4 times as great we have (£>/5) new = 1-43 corresponding to 9 = 19°. Particle identities and energies can be determined using methods described by Fleischer et al.7 We first consider the relationship between etch pit half-angle and particle energy for protons and the corresponding relationship for alpha particles. Quantitative relationships for CR-39 plastic were provided by Fleischer.8 The shower particles cannot be protons because there is no energy for which protons produce tracks with 19° half-angles. For alpha particles the energy corresponding to a 19° half-angle is 2.0 MeV, suggesting that the shower particles could be alphas. Energies also can be deduced from measurements of etch pit diameters. Roussetski et al.,9 in support of their research on the emission of charged particles in various systems, have determined etch pit sizes for alphas having a wide range of accurately known energies. Their calibration curve for diameter D corresponding to energy E is well fit by the relationship D35E = constant over the energy range 1.85 < E < 7.19MeV. From this relationship we have E2/Ei = (L>i/D 2 ) 35 .With D2/D1 = 1.4 ± 0.1 this gives E2/Ei = 0.31 ± 0.04 for shower particles interpreted as alphas relative to radon alphas. Radon decays to stable 2 1 0 Pb in a cascade of reactions including three that generate alpha particles: 222 Rn —> 2 1 8 Po + a,

293

50

.£

40

10

^ P T 10

15

20

25

30

40

Mean diameter (JJ.VT\)

30 25 fe o.

20

CO

•5.

15

E

10 5 5

Jl

10

15

20

25

30

35

40

Mean diameter (fim)

Figure 6. restricted Shadowed surface of

Etch pit sizes on two chip surfaces subject to the cuts 1.1 < R < 1.5 and additionally to exclude sharp-pointed conical pits. The remaining pits have bottomed out. (A) area on the front side of the secondary shower chip in Fig. 2B. (B) Sample of the front the primary shower chip in Fig. 2A.

2 i s P o —„ 2i4p b + ^ a n d 2i4p 0 —> 2 i o p b + a T h e g e a l p n a s h a v e energies 5.5, 6.0, and 7.8 MeV, respectively, and are expected with equal frequency. Taking the mean value E\ = 6.4 MeV we find that the energy of the shower particles would be E2 = (0.31 ± 0.04) (6.4) = 2.0 ± 0.3 MeV if they were alphas. This confirms the determination from half-angle analysis. It is unlikely that nuclei with higher charge and mass would have sufficient range in the vapor to reach the detector chips and form etchable tracks, and hence in view of the available evidence we tentatively identfy the shower as 2 MeV alpha particles.

5. Conclusions Our observations furnish compelling evidence for a nuclear process that generated a shower of charged particles in an oxygen-hydrogen vapor. It appears to have consisted of a rapid reaction that generated a cloud of unstable intermediate particles whose decay products were the observed shower particles, tentatively identified as 2 MeV alpha particles. We can think of no explanation for this phenomenon in terms of conventional nuclear theory, and believe that an extension of the theory is required.

294

Acknowledgments We thank M. E. Fisher for assistance in data presentation. References 1. M. Fleischmann, S. Pons, M. Hawkins, J. Electroanal. Chem. 261, 301 (1989) and 263, 187 (1989). 2. E.K. Storms, J. Sci. Exploration 10 (2), 186 (1996) (full text in http://www.scientificexploration.Org/jse/articles/storms/l.html). 3. E.K. Storms, Infinite Energy 4 (21), 16 (1998) (full text in http://pwl.netcom.com/~storms2/review5.html). 4. R.A. Oriani and J.C. Fisher, Jpn. J. Appl. Phys. 4 1 , 6180 (2003) and 42, 1498 (2003). 5. R.A. Oriani and J.C. Fisher, Trans. Am. Nuc. Soc. 88, 640 (2003). 6. R.A. Oriani and J.C. Fisher, in Proceedings of the 10th International Conference Cold Fusion, 2003. 7. R.L. Fleischer, P.B. Price, and R.M. Walker, Nuclear Tracks in Solids, (University of California Press, Berkeley, CA, 1975). 8. R.L. Fleischer, Private communication. 9. A.S. Roussetski, A.G. Lipson, and V.P. Andreanov, in Proceedings of the 10th International Conference Cold Fusion, 2003.

NUCLEAR REACTIONS P R O D U C E D IN A N ELECTROLYSIS CELL

OPERATING

R. A. O R I A N I University

of Minnesota, Minneapolis, MN 55419, E-mail: orianOOl Qumn. edu

USA

J. C. F I S H E R 600 Arbol Verde, Carpinteria, CA 93013, E-mail: [email protected]

USA

We report the results of experiments in which CR-39 plastic particle-detection chips were exposed in various environments within and surrounding operating electrolysis cells. Because CR-39 detectors record only particles with energies in excess of about 0.2 MeV the detected particles must have arisen in nuclear reactions. Evidence for such reactions was found in deuterium gas behind a palladium cathode that served as part of the cell enclosure, in air behind a similarly disposed nickel cathode, in air beyond the glass wall of the electrolysis cell, and in oxygen gas above the anode when anode and cathode were placed in separate arms of a Utube cell. These results, augmented by earlier work indicating nuclear reactions within the electrolyte and in the hydrogen-oxygen gas over the electrolyte, cannot be understood in terms of conventional nuclear theory.

1. Introduction Energetic charged particles can be detected by the damage tracks they generate when penetrating various solid materials. 1 When the surface is attacked by a suitable etchant damaged material is removed more rapidly than undamaged material and etch pits are formed where tracks intersect the surface. In our studies we employ detector chips that are commercially available for recording alpha particle tracks from radon decay. After appropriate calibration, analysis of etch pit sizes, shapes, and cone angles can indicate the types and energies of the responsible particles, which can be protons, alpha particles, or more massive ions. In prior work tracks of particles having energies of a few MeV have been observed in CR-39 detector chips immersed in various electrolyte solutions in operating electrolysis cells, indicating that nuclear reactions have occurred in these electrolytes. The systems studied have included D20/Li 2 S04 as electrolyte with palladium as the cathode, 2 H 2 0/Li 2 S04 electrolyte with palladium as cathode, and H 2 0/Li 2 S04 electrolyte with nickel as cathode. 3 Nuclear tracks were also found in detector chips suspended in the oxygen-hydrogen gas above an H 2 0/Li 2 S04 electrolyte using palladium or nickel as cathode material. 4 In all these experiments controls were carried 295

296

out with detector chips immersed in the electrolyte solutions or suspended in the vapor above the solutions as appropriate, but in the absence of electrolysis, for the same length of time as the duration of the electrolysis experiments. The density of etch pits was observed to be on average greater for chips exposed during active electrolysis than for chips exposed in control experiments. The probabilities that the results from the electrolysis experiments and those from the controls could have arisen by chance from a single population ranged from 10~ 4 to 10~ 10 . Thus we demonstrated that a nuclear reaction of some sort in the electrolyte or in the vapor over the electrolyte can indeed accompany the electrolysis of either heavy or ordinary water using either palladium or nickel cathodes. Beyond this we have observed clusters of tens of thousands of nuclear tracks on detector chips in the H2 + O2 + H2O vapor above the electrolyte of an operating electrolysis cell.5 Analysis of one such event has shown that the tracks were caused by high energy charged particles that originated in the gas a few millimeters from the surface of the closest chip. The reaction produced a shower of about 150,000 alpha particles with energies of approximately 2 MeV of which we recorded about 40,000 on a pair of opposing detector chips. Figure 1 shows the pattern of etch pits on the more heavily pitted chip in a region having about half of the maximum track density. Figure 2 plots the variations in track density over the surfaces of the two opposing detector chips. It is evident that the particle source lay between the chips and closer to the more heavily pitted one. These experimental results cannot be explained by nuclear physics as currently understood, nor can the generation of excess energy during electrolysis first observed by Fleischmann et al.6 An extension of nuclear theory is required. The present work

Figure 1. Etch pits on a detector chip supported in the O2 + H2 + H2O vapor over an active electrolysis cell. 5

297

Figure 2. (a) Contour plot of the density of etch pits on the surface of the detector chip in Fig. 1. Density in pits/image is indicated by the color scale for a mosaic of 1044 images that span the approximately 8 mm 2 surface. Outlined images could not be counted and etch pit densities were interpolated for them. The solid contour lines are spaced at 10 pits/image and range from 10 pits/image along the right side to 110 pits/image near the peak. Arrows indicate mean orientations of the tracks in various locations, (b) Contour plot of the density of etch pits on the detector chip facing the chip in Fig. 2a at a distance of approximately 1 cm. Here the solid contour lines are spaced at 3 pits/image and range from 3 pits/image on the left to 30 pits/image at the peak.

was carried out to further extend the range of electrolysis phenomena that must be explained by any successful theoretical treatment. 2. Detectors Below the Cathode Electrolyses were carried out in a small tubular glass apparatus shown schematically in Fig. 3. The cathode, either palladium or nickel sheet, was clamped between O-rings fitted into a flanged joint whose lower half connected to a gas-handling and vacuum system. The anode was a platinum wire spiral. All electrolyte solutions had the approximate concentration of 0.02 g Li 2 S0 4 per ml of either D 2 0 or H 2 0 . By means of the gas-handling system the lower surface of the cathode could be maintained in an atmosphere of air or deuterium. CR-39 plastic detector chips obtained from the Fukuvi Chemical Industry Company, Japan, were etched in 6.5 N KOH for about 20 h at 65°C, the standard etching procedure for all controls and experiments. They then were examined at 100 x for pre-existing nuclear tracks including those produced by alpha particles from radon in the laboratory air. A pre-etched chip then was placed below and parallel to the cathode sheet, and in some instances pre-etched chips were also placed in the vapor above the electrolyte in the cell. The detector chip situated below the cathode was enveloped either by air or by deuterium gas at one atmosphere pressure. In experiments with palladium cathodes and light water for the electrolytic solution the air below the cathode was admixed with hydrogen gas by permeation of hydrogen

298

To power supply Thermocouple

Detector chips Nichrome ; wire heater Ni disc

Pt anode Detector chips

—3sr 4 in H2O.

3. Detectors Above the Anolyte An electrolysis cell in the form of a U-tube was devised, fitted with wire spirals within each leg to serve as electrodes, and with nichrome wire heaters surrounding the upper portions of the legs of the U-tube. The anode was made of platinum and the cathode was either platinum or palladium. The electrolyte was contained in the

V Figure 6. Shower of etch pits on a detector chip suspended under a palladium cathode in air lday, then in deuterium gas 2 days. The electrolyte was IJ2SO4 in D2O.

i. €

c

O

c

Figure 7. Shower of etch pits on a detector chip suspended under a palladium cathode in air 1 day, then in deuterium gas 2 days. The electrolyte was Li2S04 in D2O.

lower portion of the U-tube and extending upward to cover the electrodes in each leg. Detector chips were suspended in the legs above the level of the electrolyte where they were exposed either to the vapors generated in the anolyte at the anode or to the vapors generated in the catholyte at the cathode. T h e detectors were maintained at about 60°C by the nichrome heaters. Prior to carrying out electrolysis the detector chips were as usual etched and examined for pre-existing tracks. Electrolysis with 0.1-0.4 A / c m 2 current density was carried out for 2 or 3 days, after which the detector chips were again etched and examined. New nuclear tracks were observed in chips exposed above the anolyte as well as above the catholyte. Of fourteen experiments with chips surrounded by oxygen and water vapor above the anode, five produced new tracks in numbers much larger t h a n were produced in control chips held in the vapor above an electrolyte solution for 2 or 3 days without electrolysis. T h e successful experiments in O2 + H 2 0 over the anolyte all employed H 2 0 for the electrolyte. Figure 8 shows a small shower of etch pits on one of these chips.

4. D e t e c t o r s O u t s i d e t h e Cell We have in addition observed the production of nuclear tracks, in numbers significantly larger t h a n in controls, in detector chips placed in near contact with the outside surface of the glass cell wall at the level of the electrolytic solution. This phenomenon was seen in 6 of the 11 experiments of this type t h a t were carried out. Figure 9 shows some of the tracks produced in one of these experiments. T h e electrolyte employed H2O as solvent.

302

\

\

Figure 8. Shower of etch pits on a detector chip suspended in the O2 + H2O vapor over the anolyte in a U-tube experiment. The electrolyte was IJ2SO4 in H2O

5. Challenges to Nuclear Theory We note here the sporadic nature of charged particle generation in our experiments. As stated in previous publications we infer that the cause for the difficulty of replication is our ignorance of the full range of experimental parameters that should be

c

. \ \

Figure 9. Etch pits formed near the center of a large shower on a detector chip mounted just beyond the outside surface of the glass wall of the linear electrolysis cell. The electrolyte was L i 2 S 0 4 in H 2 0 .

303

controlled for nuclear reaction to occur. In addition, and probably equally important, is the strong probablity that detector chips may not be positioned sufficiently near to the location where reactions may have occurred. Our past work has amply demonstrated this possibility. Often one side of a detector chip bears a large number of new tracks while other side does not. We note also that the energy carried by the particles we detect is many orders of magnitude smaller than the total energy release as determined by calorimetric methods. Although we have presented strong evidence for nuclear reactions in the electrolyte and in various gases remote from the electrolyte, they cannot be the primary reactions that generate the reported excess energy. We presume that sustained primary reactions take place near the surfaces of the electrodes and that we observe transient secondary reactions associated with or triggered by products of the primary reactions. Yet in spite of the difficulty of replication and the modest energy represented by the observed nuclear tracks our experiments pose significant challenges to theory. We have shown that nuclear tracks can be generated during electrolysis in detector chips positioned within the electrolytic solution, in the H2 + O2 + H2O gas over the electrolyte in the straight-tube cell, in the O2 + H2O gas over the anolyte of the U-tube cell, in air below the cathode, in air just outside the glass wall of the electrolysis cell, and in deuterium gas below the cathode. These experiments indicate that nuclear reactions can be suported by oxygen and by deuterium, and that reactions can be triggered by unidentified agents that are able to pass through nickel and palladium cathodes and through the glass cell wall. Acknowledgments We thank M. E. Fisher for assistance in data presentation. References 1. R.L. Fleischer, P.B. Price, and R.M. Walker, Nuclear Tracks in Solids (University of California Press, Berkeley, CA, 1975). 2. R.A. Oriani and J.C. Fisher, Jpn. J. Appl. Phys. 41, 6180 (2003); Erratum 42, 1498 (2003). 3. R.A. Oriani and J.C. Fisher, Trans. Am. Nucl. Soc. 88, 640 (2003). 4. R.A. Oriani and J.C. Fisher, in Proceedings of the 10th International Conference on Cold Fusion (Boston, USA, 2003). 5. R.A. Oriani and J.C. Fisher, in Proceedings of the 11th International Conference on Cold Fusion (Marseille, France, 2004). 6. M. Fleischmann, S. Pons, and M. Hawkins, J. Electroanal. Chem. 261 (1989); Erratum 263, 187 (1989).

E V I D E N C E OF MICROSCOPIC BALL LIGHTNING IN COLD FUSION E X P E R I M E N T S

E. H. L E W I S E-mail:

P.O. Box 2013, Champaign, [email protected]; Web:

IL 61825, USA www.scientificrevolutions.com

There is evidence of microscopic ball lightning in several methods of cold fusion and transmutation. Thus far the experiments of Matsumoto, Miley, Shoulders, Savvatimova, and Urutskoev et al. have shown evidence of these objects that range in size from sub-atomic to about 1 mm in diameter. This article presents pictures and evidence collected by these groups, summarizes the evidence found by other groups, and discusses the significance of microscopic ball lightnings. The implications for atomic physics and physics in general are discussed.

1. Introduction During the last 13 years, the evidence for microscopic ball lightning in CF experiments has been building. Little objects that behave like natural ball lightning in many ways leave characteristic markings. Such markings have been found in Matsumoto's experiments since about 1991, Miley's experiments about 1996, Shoulders experiments about 1996, Savvatimova's experiments about 2000 and Urutskoev's experiments about 2001. I am not sure on the exact dates these markings were found. The markings are characteristically about 1 /im to 1 mm in diameter, and are usually either spots or long trail-like markings as though the ball lightning moved along the surface of something. Sometimes they are grooves or bore holes. Similar markings produced by plasmoids and EVs have been studied for decades, sometimes for fusion research; and through his experimental observations, Shoulders showed that the plasmoids studied by Bostick, Nardi, and other researchers are composed or EVs or are EVs. The significance of these objects for the field of cold fusion and transmutation in general is that these are a cause of transmutation and cold fusion reactions, but on a deeper level they point to the plasmodal identity of atoms. In this paper, the very similar markings found by these groups who did somewhat dissimilar experiments independently are discussed, compared and contrasted, and the general meaning of these objects to physics is discussed. Such markings are seen clearest in Matsumoto's Acrylite plastic sheets and in the clean targets used by Ken Shoulders. It seems that the X-ray films used by Savvatimova and the CR-39 sheets used by Urutskoev do not yield as clear markings. The trails may sometimes be millimeters long. In his early articles on cold fusion and transmutation, as part of the regular method of research on nuclear reactions, 304

305

Matsumoto set up Acrylite plastic sheets coated with an emulsion that allowed him to capture the tracks of reaction products. He found strange and large markings and began to explain them by a theory called the Nattoh model, which is a type of Japanese bean pudding. He focused his experimental research on these objects creating the strange traces, especially after he learned about the idea of microscopic ball lightning, and published about six articles in Fusion Technology from 1991 to 1995 about photographs of these markings and their meaning. However, he was generally ignored and ridiculed, even by cold fusion researchers, and felt that he couldn't even get a hearing at cold fusion conferences. In 1993, he accepted the idea of microscopic ball lightning and started research on this idea specifically, but did not get his ball lightning articles published. In 1996, after an earthquake in Hokkaido where he lives, he noticed that microscopic tracks were left on plastic sheets and perhaps discovered naturally produced microscopic ball lightning. Here are two pictures of his tracks from his experiments.

Figure 1. Ring and track markings left on a sheet of acrylite plastic. 1 It is evident that at least one toroidal object about 50 fj,m wide slid and jumped around. If there were more than one such object, it was remarkably similar in size. Acrylite sheets were outside of a discharge container filled with water. The sheets of plastic were about 100 jira wide by 50 mm X 50 mm, and this sheet was one of a set of such sheets set in parallel with about a 3 mm gap in between each sheet. These emulsions were set outside a cylindrical glass cell which had an Acrylite bottom 1 mm thick. They were set outside this plastic bottom. This suggests that the plasmoid phenomena traveled through the Acrylite or the glass. One of the anomalous abilities of ball lightning is the ability to travel through glass and other insulators. This evidence points to the identity of these objects as microscopic ball lightning. It is possible that the object responsible for these markings on one side of one sheet of plastic also left similar markings close in size on the backside of the preceeding sheet. Tornadoes hop in a similar manner. The object was 50 fim wide.

306

Figure 2. Ring and pit markings left on an electrode after discharge. 2 These markings are very similar to the markings shown previously by Nardi and Bostick of plasmoids. The markings are less than 50 fim wide. Matsumoto preformed electric discharging in ordinary and heavy water. The anode was a platinum pin and the cathode was a copper plate.

In 1996, a researcher working in Miley's lab investigated the microsphere experiments. These experiments involved little beads coated with thin layers of nickel or other elements about 650 A thick via a special, patented electrode sputtering technique. These pictures are from Nickel on plastic Run # 8 . This cell and the anomalous appearance of a wide range of elements has been described by Prof. George Miley in several articles. 3,4 In Miley's ICCF10 lecture, he said that this Run # 8 exhibited by far the most excess heat than the many other cells of this type. The electrolysis was performed in his laboratory at the University of Illinois. The pictures were taken by using a digital camera that was attached to a good optical microscope. The photographs in Fig. 4 were taken by Savvatimova5 of markings on X-ray film that were both inside and outside her discharge device. She used X-ray film to catch these BL markings. The objects in Shoulders' experiments that he calls EVs behave in ways similar to BL. They are a type of BL. Their special characteristics such as very fast travel may be due to the method of their production with the type of electrodes he uses. His method of getting the tracks shown here produces clear markings of the ball lightnings and their effects. He uses flat, clean materials, and sometimes coats his "witness plates" as they are called in this field with temperature sensitive material like wax in order to see how warm these objects are. Like many ball lightnings, the objects that left the markings in Fig. 5b might not have been warm at all, because in

307

4

«. i

-;• M >•

(e)

W

(g)

Figure 3. These seven photographs are at a magnification of 200 X or 400 X and are of markings on the two Lexan casings, two microspheres, and the Ti anode and cathode of Run # 8 . (a) and (b) are of the post-run microspheres, (a) Two ring marks in the metal. The marks look like two rings of ball lightnings, like beads in a necklace, left little pits in the metal. These are much like most of Shoulder's ring markings, (b) One faint ring mark in the plastic bead where the metal had flaked off. (c) and (d) The titanium plate cathode and the titanium plate anode of the cell. These plates enclosed the microspheres in a Lexan plastic casing, (c) A very faint mark pointed out by the black line. It is about 18 )im wide, and is clearer in the original photographs, (d) A round ring mark in a pit that formed in the anode. The pits were interesting and deep.

308 Figure 3. (Continued) The pit itself is about 200 fim wide at the surface of the anode, but narrowed down to the white area that is about 100 /im in diameter. The ring mark is about 20 or 30 /an wide. Grooves and strings of little dots are visible as well. Ball lightning-like plasmoids often form little strings and rings. Tornadoes and ball lightning also form such lines or circles. See for example this link: http://www.ernmphotography.com/Pages/BalLLightning/OtherBLPages/fbparks/BL_Triple-Waterspout.html. (e-g) Ring marks on the casings. Two different casings were used for this run. (e) Ring mark in the plastic casing, in a photograph that was computer processed to make the edges more apparent, (f) Lexan Plastic Casing # 2 of Ni/Plastic Run # 8 . Magnification 200 X. Shows a group of small rings. The markings are seen from the outside of an intact casing. The picture shows the convex impression left by a microsphere that was in contact with the inside of the casing. The bead developed both ridges and ditch markings. To the left of the bead impression are the faint marks of rings that are about 20 /jm wide, (g) Lexan Plastic Casing # 2 of Ni/Plastic Run # 8 . Magnification 200 x. Shows a ring mark to the right of the bead impression. The picture shows the convex impression left by a microsphere that was in contact with the inside of the casing. The bead developed both ridges and ditch markings. A BL may have left a trail mark in the plastic or bored through. The ring mark is about 25 /jm in diameter. Details about the experiment and the markings are discussed in other articles. The photographs shown here were taken by E. Lewis of various components of Ni-Plastic Run # 8 in the Laboratory of Professor G. H. Miley at the University of Illinois at Urbana-Champaign in 1996. His cooperation in allowing this work is gratefully acknowledged.

1umm

Figure 4. X-ray film outside (A and B) and inside the vacuum chamber after deuteron irradiation in glow discharge.

other similar tests with aluminum oxide coated with wax, there was apparently no melting of the wax at all, although it is obvious the underlying material, aluminum oxide,6 moved as if there was a splash. Similar effects are seen around ball lightning and in lightning strikes sometimes, and in other electrical discharges experiments, and also happen during transmutation. What all this points to is a state of matter that needs to be understood. In this plasmoid state, the atoms act like ball lightning.

309

(b)

(c)

Figure 5. (a) Ring Mark in Witness Plate. This is a typical type of ring marking (from Ref. 6). (b) Strike Marks on Lead Glass. These marks show the heatless motion of atoms (from Ref. 6). (a) and (b) are from the same article. Shoulders reports that there is no evidence of heat. This figure shows the effects of several ball lightnings hitting this lead glass surface, (c) Impact Site on palladium foil loaded with deuterium. Shoulders reports that chemical analysis of this spot showed many transmuted elements (from Ref. 6).

X-ray analysis of track

Energy (keV)

25

Figure 6. Impact Site X-ray analysis of the ball lightning strike shown in Fig. 5(c) (from Ref. 6). Shoulders reports that chemical analysis of this spot showed many transmuted elements.

In the next photograph, Fig. 5c, a chemical analysis was performed at the site of this type of atomic sloshing (motion), and a wide range of chemical elements are discerned, as is shown in Fig. 6. Recently, Urutskoev has published pictures similar to these as shown in Figs. 7 and 8. In Fig. 7 which is from his article published in 2002, "Observation of Transformation of Chemical Elements during Electric Discharge," 7 the marking looks much like the bottom two wispy lines in Savvatimova's picture in Fig. 4a. They look like some of the long trail tracks in Matsumoto's article as well.

310

1 lOmcm

Figure 7. Trails like those shown by Matsumoto and Savvatimova. Figures 7 and 8 are from Urutskoev et al. (from Ref. 7).

a 500 mem Figure 8. A comet like trail. They wrote that markings like these had a very high energy. These may be like the type of BL markings that Matsumoto called "white holes." The white marking is about 1000 mem long, and remind me of some of Shoulders pictures of fast moving EVs. The authors attribute a nuclear-reaction origin to the object.

These pictures show similar markings. Rings or trails or combinations of rings and trails or blotches or pits. This is because the microscopic ball lightning objects may move on surfaces, as does ball lightning in nature. Or they may impact surfaces, leaving the pits. Sometimes as in nature, they leave deposits. Sometimes tornadoes and ball lightning leave grooves in the ground or in materials. In their article from 2002, Urutskoev mentioned finding markings like scratches. Dash has also written about and shown a picture of a scratch-like mark in an electrode. And Shoulders has shown pictures of such grooves. In nature, ball lightning sometimes bore tunnels in walls, and Egely8 has published descriptions of these. The existence of these objects presents cold fusion researchers and physicists in general both a cause and an effect of atomic reactions. They are a newly discovered type of particle. At a deeper level, they call into question the assumptions of nuclear physics and the QM paradigm atomic model. These objects are anomalous to QM theory in that this type of matter is energy. It points to the idea that atoms are structured like them. Research is needed to determine the effect of these ball lightning objects on time, gravity, magnetism, electricity, and atomic reactions. We need accurate determinations of the physical relationships in order for theory to develop further.

311

In conclusion, thus far five groups working on three continents and one hemisphere have published the markings of tracks like t h a t of microscopic ball lightning t h a t were associated with successful t r a n s m u t a t i o n and cold fusion experiments t h a t exhibited unusually high t r a n s m u t a t i o n and excess energy. In fact, Savvatimova especially emphasized t h a t there was a correlation between the amount of t r a n s m u t a t i o n in a reaction run and the number of markings produced, and I think t h a t Miley's, Matsumoto's, Shoulders and Urutskoev's experiments also bear witness to such a correlation. I think t h a t people working in other continents and the other hemisphere will find such products and similar correlations. Some experiments may not produce these larger types of BL objects. B u t in all experiments anomalous to QM, the atoms themselves transform to a state like ball lightning.

Acknowledgments I would like to t h a n k all these researchers for their permission to reproduce these photographs.

References 1. T. Matsumoto, Observation of Tiny Ball Lightning During Electrical Discharge in Water, Manuscript Article, 1994. 2. T. Matsumoto, Experiments of One-Point Cold Fusion, Manuscript Article. 3. G.H. Miley and J.A. Patterson, Nuclear transmutations in thin-film nickel coatings undergoing electrolysis, in Proceedings of the Second International Conference on Low Energy Nuclear Reactions, 13-14 September 1996 (College Station, TX, USA). 4. G.H. Miley et al., Quantitative observation of transmutation products occurring in thin-film coated microspheres during electrolysis, in Proceedings of the ICCF-6, October 14-17 (Hokkaido, Japan). 5. I. Savvatimova, Reproducibility of experiments in glow discharge and processes accompanying deuterium ions bombardment, in Proceedings of the ICCF-8, 21-26 May 2000 (Lerici, Italy). 6. K. Shoulders, Charged Clusters in Action, Manuscript Article, 1999. 7. L.I. Urutskoev, V.I. Liksonov and V.G. Tsinoev, Observation of transformation of chemical elements during electric discharge, Annales Fondation Louis de Broglie 27 (4), 701 (2002). 8. G. Egely, Hungarian Ball Lighting Observations (Center Research Institute of Physics, Hungarian Academy of Sciences, 1987).

N E U T R O N EMISSION FROM D 2 GAS IN M A G N E T I C FIELDS U N D E R LOW T E M P E R A T U R E

TADAHIKO MIZUNO, TADASHI A K I M O T O , AKITO TAKAHASHI, AND FRANCESCO CELANI Division

of Quantum University,

energy engineering, Graduate School of Engineering, Kita 13 Nishi 8, Kita-ku, Sapporo 060-8628, Japan

Hokkaido

AKITO TAKAHASHI Emeritus

professor

of Osaka University,

Japan

FRANCESCO CELANI INFN-LNF,

via E. Fermi 40 00044, Frascati,

Rome,

Italy

We observed neutron emissions from pure deuterium gas after it was cooled in liquid nitrogen and placed in a magnetic field. Neutron emissions were observed in ten out of ten test cases. Neutron burst of 5.5counts/s were 1000 times higher than the background counts. These bursts occurred one or two times within a 300 s interval. The total neutron emission can be estimated from the counting efficiency, and it was 10 4 -10 5 counts/s. The reaction appears to be highly reproducible, reliably generating high neutron emissions. We conclude that the models proposed heretofore based upon d-d reactions are inadequate to explain the present results, which must involve magnetic field nuclear reactions.

1. Introduction There have been many reports of neutron generation during cold fusion experiments. 1_3 Although there have been a few negative reports, 4 most show some neutron emission. However, it seems hard to replicate, and reaction rates are very low. Shyam et al.5 reported on conventional light and heavy water electrolysis with a palladium electrode. They used 16 BF3 neutron detectors to increase the chance of detection. They observed a difference in neutron emission rates between light and heavy water electrolysis. The neutron count rate was slightly higher for heavy water. Shyam et al. conducted a series of experiments to detect production of neutrons from a commercial palladium-nickel electrolytic cell operated with 0.1 M LiOH or LiOD as the electrolyte, at a current density of ~80mA/cm 2 . A bank of 16 BF3 detectors embedded in a cylindrical moderator assembly detected neutron emission. A dead time filtering technique was used to detect the presence of neutron bursts, if any, and to characterize the multiplicity distribution of such neutron bursts. It was found that with an operating Pd-D20 cell located in the center of the 312

313

neutron detection setup the daily average neutron count rate increased by about 9% throughout 1-month period, over the background value of ~2386 counts/day. This indicated an average daily neutron production of ^2220 neutrons/day by the cell. In addition, analysis of the dead time filtered counts data indicated that about 6.5% of these neutrons were emitted in the form of bursts of 20-100 neutrons each. On an average, there were an additional six burst events per day during electrolysis with LiOD over the daily average background burst rate of 1.7 bursts/day. The frequency of burst events as well as their multiplicity was significantly higher with D 2 0 + LiOD in the cell when compared with background runs and the light water control runs. Oya et al.6 used a precise method to determine the relationship between neutron energy and excess heat. They use flow calorimetry measure excess heat generation. They showed a clear relation between heat and neutron generation. Neutron energy was in the MeV order when the excess power was generated. The key parameters for the occurrence of the anomalous phenomena, especially excess heat generation and the emission of excess neutrons, have been investigated through a series of electrolytic experiments in Pd-LiOD (H) systems. Seven key parameters have been identified: (1) (2) (3) (4) (5) (6) (7)

purity of Pd cathode, shape and size of Pd cathode, processes of pretreatment of Pd cathode, electrolysis mode, electrolyte, purity of the medium, initial open-circuit voltage.

In the present work, a series of systematic experiments have been carried out with some fixed parameters. By controlling key parameters completely, an appreciable correlation between the excess heat generation and the excess neutron emission can be replicated successfully. We have sometimes seen neuron emission with a phase transition method. This typically occurs in non-equilibrium conditions. Chicea and Lupu 7 showed the neutron emission from Ti metal loaded by deuterium gas absorption. Chicea used a simple measurement system. The sample holder includes Ti powder. The Ti metal absorbed deuterium gas and sporadic neutron generation occurred. In several experiments, Chicea and Lupu loaded titanium samples with deuterium in gas phase, and the temperature of the samples was changed over a wide range, while neutron emissions were monitored. Neutron emissions were recorded in very low intensity bursts, but still significantly above the background. This revealed that low energy nuclear reactions in condensed matter can be produced at a low rate, which is occasionally high enough to become detectable. They observed very strong neutron emission occurred more than 10 times during 20 h. At times, the emission exceeded four times background counts.

314

Jones et al.a used a similar method, and they reported neutron emission from Ti metal that absorbed deuterium gas. Jones' results are very clear, showing that neutron emission only occurs with deuterium gas, not hydrogen. They presented evidence for neutrons emanating from partially deuterided titanium foils (TiDx) subjected to non-equilibrium conditions. A previous paper presented data for complementary charged-particle emissions. Metal processing and establishing non-equilibrium conditions appear to be important keys to achieving significant nuclear-particle yields and repeatability. It is very important to confirm nuclear products to prove that cold fusion is, in fact, some kinds of nuclear reaction. Neutrons are especially suitable for this purpose. We have already published transmutations results from the electrolysis method. We have confirmed isotopic shifts in elements. We have also confirmed neutron emission during various methods of cold fusion. We have measured the neutron energy distribution during heavy water electrolysis with a Pd electrode with a closed-cell system. 9 The cell temperature and pressure can be raised to increase deuterium absorption. We observed a clear neutron energy peak at 2.5 MeV. This indicates a possible d-d nuclear fusion reaction. The reaction rate was estimated as 10 _ 2 3 /dd/s. We have used other methods to increase the probability of neutron generation. We used very high purity heavy water absorbed into a Pd wire. After the wire absorbed deuterium, hydrogen gas was admitted into the wire to stimulate the neutron generation reaction. 10 The neutron count, the duration of the release and the time of the release after electrolysis was initiated all fluctuated considerably. Neutron emissions were observed in five out of ten test cases. In all previous experiments reported, only heavy water was used, and light water was absorbed only as accidental contamination. Compared to these deuterium results, the neutron count when hydrogen is deliberately introduced is orders of magnitude higher, and reproducibility is much improved. Several analytical methods suggested some characteristic elements appearance in the electrolysis system after the neutron emission. After filling the Pd wire with deuterium in heavy water, we took the wire and immersed it in the heavy water system. Figure 1 shows the time change for input voltage, current and electrolyte temperature. At 3000 s, we changed the voltage from 32 to 85 V. Figure 2 shows the neutron emission during this voltage change. The neutron count was 100 times larger than the background count. The rate of neutron emission depended on the purity of heavy water. We can see neutron emission occurred at more than 90% of purity as shown in Fig. 3. We can say that we have to pay attention if you want to generate neutron emission. Because that the heavy water easily absorbs light water. The rate of neutron count was estimated as 1.5 x 10 _ 1 7 /dd/s. The rate was increased 106 by the conventional deuterium gas absorption method.

315

~ 250

10

5 200

- Curr ;nt

§ 150 O "P CD

E a)

100

: Electn ilyte ^ * f tempe •ature

—

*

•

I

| I

~ " " ~ V )ltage

50

r"

1

i

i i Li

1 1 »

1

>

1

i

^ ^

* i

2 3 Time/1000 s Figure 1.

Parameter changes for electrolysis.

Time (s) Figure 2.

Neutron burst.

2. Experimental The reaction cell was a Pyrex glass tube of 6 mm diameter, 3 mm inner diameter and 100 mm in length, filled with pure D2 gas. A coil wound around the tube supplied the magnetic field. This magnetic coil is made from 10,000 turns of 1.5mm diameter copper wire. Another Pyrex glass vessel of 50 mm diameter was put around the reactor tube, and filled with liquid nitrogen. The whole system was put in a stainless steel vessel 1.5-mm thick. The outer surface of the steel vessel is insulated with Styrofoam, and another layer of 1.5-mm thick stainless steel plates were placed on top of the Styrofoam insulation to prevent electromagnetic noise

316

2000 o o °

O

1500

1000

500

o CD

1 80

Figure 3.

„

• [ i fa t r i i i " i 85 90 95 Deuterium purity in electrolyte (%)

i

l

l

100

Dependence of neutron on purity of D2O.

from reaching the neutron measurement system. The vessel was filled with liquid N2 to cool the coil and the reactor tube. The magnetic field was 8 kg at the center of the reaction tube. Power for the magnetic coil was supplied by a stable direct current power supply through a resistive wire, to control the current. The magnetic field passes through the reaction tube along its length. The height of the coil is 100 mm; the same as tube length. The current passing through the coil was increased from 0 to 100 A, which gives the change of intensity of the magnetic field from 0 to 8 kg. Neutrons were measured with three external He 3 detectors placed around the cell, 20 cm from the vessel walls. The method seems rather simple. We filled the glass tube with pure D2 gas. The pressure was several atmospheres, typically 3 atm. The glass tube was then cooled by liquid nitrogen. After that, we supplied a magnetic field. The temperature was kept under -196°C. The magnetic field was periodically changed, and this produced a sporadic neutron burst. Figure 4 is a photo of the experimental system, power supply, and neutron measurement system. We used Aloka neutron survey meter TPS-451S and three He-3 detectors. The He-3 proportional detector has the energy sensitivity from 0.025 eV to 15 MeV. The sensitivity was calibrated using a standard Cf-252 neutron source. Figure 5 shows a schematic representation of the measurement system. The liquid N2 gas cooled the reactor tube. The maximum magnetic field was 10 kg in the center of the reaction tube. The current for the magnetic coil was supplied by a stable direct current power supply through a resistive wire. The magnetic field passes through the reaction tube along the length. The height of the magnetic coil is 100 mm, that is, the same as tube length. The current passing

317

Figure 4.

Photo of D2 gas experimental setup.

Reactor tube

rt.c. nowc

_ j

Mfirn." Ik '-.r.i ,...~*5 {•**•

I )c-?i dctcctc Hc-3dctoctor

]|-

Cd film

I le-J rinroctor

Vacuum pump

Figure 5.

D 2 gas J cylinder |

SI.P/'MI'l !

I-J

N

Schematic representation of the D 2 gas experimental setup.

through the coil was changed from 0 to 100 A; changing intensity of the magnetic field was changed from 0 to 10 kg. Neutrons were measured with three external He 3 detectors, each 2 cm in diameter and 10 cm in length. They were placed around the cell, separated 20 cm from the

318

cell. All the detectors were surrounded by a cylindrical plastic neutron moderators, 12-cm diameter and 15-cm high. The detectors were inside the moderator, with the open end of the cylinder facing the cell. To reduce noise, the detectors were covered by electromagnetic shielding. After calibration, neutrons and noise were distinguished by covering one of the detectors with 0.5-mm thick Cd film. A neutron entering through the plastic moderator will lose energy and be absorbed by the foil, while electromagnetic noise easily passes through the Cd material. The detectors were calibrated with a standard Cf-252 neutron source (2.58 x 10 4 decay/s). The background count was estimated as under 0.008 ± 0.003 counts/s. A typical count under these conditions was 5 ± 1 count/s from the standard neutron source. This means the total counting efficiency is estimated as 0.0002. Figure 6 shows the typical neutron counting rate over 10 min after 3 atm of D2 gas filled the tube, a magnetic field of 8 kg was imposed, and the cell has been cooled in liquid nitrogen. The magnetic field was changed to 10 kg at 1200 s by increasing the current. About 20 s, a low-level neutron emission began, and after 50 s, a sudden neutron burst was observed. In this experiment, the reactor tube was filled the pure deuterium gas up to 3 atm, and the liquid N2 was put into the vessel holding the reactor tube, and the magnetic field was imposed in the last step. In other experiments, these steps were taken in a different order.

DD cluster reaction distance between cell and detector: 30 cm

20 1

1

1

2

lBackup\dd30725#.wk4| Figure 6.

r

3

4

XSMO CS Time/1000 min

Typical rate of neutron count in 10 min, 3 atm D2 gas, 8 kg.

In this example cooling of the deuterium gas was continued for a considerable

319

time and neutron emission was sporadically observed when the electromagnetic field was changed. However, in other runs, neutron emissions were observed immediately after liquid N2 was added. Figure 7 shows the real time-representation for the previous graph. Neutron emission occurred very sporadically over a very short period. So, the rate of the neutron emission changed by the accumulation time. The real counts calculated by inverse time of each emission intervals is shown here. This demonstrates that the neutron emission is very strong and very high and it sometimes almost 1000 times higher than the background counts.

0 0001 ' ' I ' M

0

I, A, A I I I i, I I I I I I 1 I. I I I I I I ,)„,! I

100

Figure 7.

200 Time (min)

300

Real-time representation of Fig. 6.

Figure 8 shows the case of hydrogen gas at liquid N2 temperature under 8 kg of magnetic field. First, the tube was evacuated and the magnetic field was fixed at 8 kg. After that, at 220 s, hydrogen gas was introduced into the tube, and the hydrogen gas was removed at 3430 s. No neutron burst was observed during the time hydrogen gas was present in the tube. We can see there are no neutron emissions exceeding background counts during the test. Figure 9 shows another typical neutrons emission when the tube was first supplied the magnetic field and then cooled by liquid N 2 . Here, the neutron emission occurred immediately after liquid N2 was added. The count rate increased up to a peak within a few seconds and decreased a few seconds later. Total neutron emission for this brief period is estimated as 5 x 105. However, no more neutron emissions were observed after that, even when the input magnetic current was increased up to 100 A for 4000 s. In other examples, the total neutron count ranged from 104 to 105, and emissions lasted 1-4000 s. All cases were marked by a characteristic high

320


20 eCouit/10 min * 50 min mov. avg. »90 min mov. avg.

15 c

" H2 gas in, i atm

o

10

c 13 o O

,.

H2 gas out

r

i4ji i *

2

*1*J

1

ft

f

i 4f

3 XSMO CS

j Backup\dd30805#.wk4

Time/1000 min

Figure 8. A case of H2 gas at liquid N2 temperature under 8 kg.

level of neutron emissions at first, which gradually declined.

DD cluster reaction

10

'i

0.01

/3

z V

0.001 :

>

0

1

>

!

i

1.

;

1

20

,

i

40

,

i

!

60

i

80

1

100

120

Time (min) Figure 10.

Real time representation of Fig. 9.

Figure 11 shows an example when the temperature was kept at room temperature, 20°C. Deuterium gas was kept in an 8 kg magnetic field. However, there were no neutrons above background. The neutron emission measurements under various conditions are shown in Table 1. The necessary conditions to make a neutron burst were: deuterium gas, a magnetic field and a low temperature. Neutrons were not generated when one of these conditions was not met. The generation of neutrons when the intensity of magnetic field was changed has not been measured systematically. We usually kept the intensity of the magnetic field constant to avoid noise from the current change and magnetic influence on the measurement system. Table 1. Gas Air Air Vac. Vac. H2 H2 D2 D2

Neutron emission measurements under various conditions.

Mag. field (kg) 8 8 8 8 8 8 8 8

Temperature (°C) 20 -196 20 -196 20 -196 -196 20

Maximum neutron count 0.016/s 0.01 count/s 0.01 count/s 0.009 counts/s 0.009 counts/s 0.013 counts/s 5 counts/s 0.015 counts/min

322


20

c 'E °

•

15

-

10

. ._ a

o O 5

* Count/10 min a 50 min mov. avg. » 90 min mov. avg.

-I * If 9 n 9
8.0 MeV is supposed to be highly desirable, solely, because the background events would not be covered the picture of possible ECPE. Much shorter exposure of Au/Pd/PdOiD^ sample in front of individual SSB-1 (low geometrical efficiency detector) in vacuum showed few counts at E = 9.1 MeV (Fig. 5). Notice that the similar result of high energy counts was obtained with Au/Pd/PdO:Ha; sample. The annealing of deuterium/hydrogen from the sample (at T = 400° C) resulting in spectrum, which is similar to the Background (Fig. 4). Application of high geometrical efficiency, large area SSB-2 detector in similar experiment with Au/Pd/PdO:D,j; sample in vacuum gives higher number of counts at E > 8.0 MeV already for the shorter time of sample's exposure in front of detector (Fig. 6). In order to identify type of high-energy particles emitted from the surface of Au/Pd/PdO:D(H).,. samples the dE — E detector pair was employed in air at ambient condition. The background experiment with reference Au/Pd/PdO sample showed two-dimensional count points (corresponding simultaneously to particle energy loss in thin dE and its stopping in thick .E-detectors) with coordinates (dE, E) at channel numbers N&E > 1000 ch. and NE < 600 ch. These channel numbers are corresponded to detection of alpha particles with energy in the range of 5.0 < Ea < 7.5 MeV or protons of 1.3 < Ep < 1.5 MeV and are anticipated in terms of natural radionuclide and cosmic background (Fig. 7).

329

Au/PdO:Dx, spontaneous D-desorption: T=300K, P=10"6torr, SSB:S= 100 mm2, f= 133,000 s

3

o O 4 Radon ???

p(DD)

• •Ml

10

12

Energy (MeV) Figure 5. A u / P d / P d O : D x after t h e electrolytic D-loading. Sample is in front of detector (d = 15mm). Foreground run, SSB-1 detector efficiency 3.3% (arrow marks the expected radon alpha edge).

The bands of alphas and protons in Fig. 7 are calculated accordingly to stopping powers of these particles in silicon. This result only confirms the results of background measurement with individual SSB-1 detector (Fig. 4), which showed

20 18H

Detector: SSB, e = 12%, vacuum: 10~3 torr

16

Sample: PdO/Pd/Au:Dx40 urn; fexp = 2.4 x 104s

14 12-i §10 o

00

1

-j—r~r~ 3 4

5 6 7 8 Energy (MeV)

9

10 11 12 13

Figure 6. Large area high efficiency SSB-2 detector: S = 900mm 2 , e = 12.0%. Foreground run. The counts with E > 8.0 MeV are collected for a shorter time than that with SSB-1.

330

dE/E Au/Pd/PdO Background run (dE/ETDC 750920 ch, gate 20 ns) •

200ot

Au/Pd/PdO Background run Alpha energy (5.2-14.0 MeV)

1500

Proton energy (1.3-3.5 MeV)

',.—> <x> in)

•iP. r

•

1000

!UJ ;U

500^

• 500

1000

1500

2000

E (Channel

Figure 7. dE/E Background spectra of charged particles from pristine A u / P d / P d O sample: T,t = 500 h. The sample is in front of dE detector (d = 10mm).

absence of any real ECPE at E > 8.0 MeV. Notice that real particles produce only two-dimensional poi with coordinates [E, dE). All one-dimensional events on the dE and E axis could be referenced to electronic noise. The events with dE coordinate N dE > 0 and E axis NE > 2000 ch. may be considered as a passage of very high-energy particle through the detector pair. In contrast, experiment with both Au/Pd/PdOiD^ and Au/Pd/PdO:H x heterostructure samples showed significant count number at the strip corresponded to energetic alpha-particle emission with E > 8.0 MeV. In Fig. 8, the data obtained for exposure of 18 Au/Pd/PdOiD^ samples (T,t = 550 h) in front of dE/E detector is presented. Taking into account geometrical efficiency of dE/E detection, the number of counts which are corresponded to calculated alpha strip for stopping power of used dE/E detector pair and have energy E > 8.0 MeV was found to be (Na) = (6.4 ± 1.2) x 10" 4 (s _ 1 ) in 4TT ster. Thus, the SSB measurement, confirmed by dE/E noiseless detection showed that Au/Pd/PdO samples loaded with deuterium/hydrogen emit energetic alpha particles in the energy range of approximately 9.0-14.0 MeV. If CR-39 technique is sensitive enough to detect low-intensity energetic alphas, then some signatures of ECPE should also be anticipated in experiments with plastic detectors. In order to check feasibility of CR-39 to measure ECPE we used the same Au/Pd/PdOiD^ samples loaded with deuterium to be exposed in the exothermic deuterium desorption regime. To enhance D-desorption the additional mechanical loading was applied to the sample with attached CR-39 detectors (Fig. 9a, b). In order to determine feasibility of ECPE in other metals (with low hydrogen solubility) and compare such results with ECPE from Pd/PdO samples in this experiment we

331

2000\

d£/£two dimensional spectra of charged particles emitted from Au/Pd/PdO:Dx;Ii = 500h. dE-Etime gate = 20 ns

dOETDC 750-920 ch

\\5.5MeV 1500 1.4WeV- 7.0 MeV

c |

IP

1000

E(a) ,8.0 MeV x " i T A * l 0.0 MeV 1.?5JyleV ^ l - - . j f l L v

a

N

••"•-.

500-

12.0 MeV

14.0 MeV

3.0 MeV

3.5 MeV

^fii\'

2.0 MeV~ - „

_ "

2.5 MeV* ""

Bp) 500

1000

1500

2000

E (channel) Figure 8. AE/E (SSB: AE = 20/u,m, E = 100 ^m) 2-dimensional spectra of charged particles from A u / P d / P d O i D s (after electrolysis): S i = 550h; (N Q ) = (6.4 ± 1.2) x l O " 4 ^ " 1 ) in Air ster.

also detected charged particles using Stainless steel, Cu and Al cathodes subjected to electrolysis. Measurements with CR-39 detectors showed presence of tracks with diameter ranging of 6.0 < d < 8.0 fim for Au/Pd/PdOiD^ system and total absence of tracks in those diameter range for St. steel, Al and Cu cathode exposed on the same

Au/Pd/PdO:D x (40 urn) t = 170 h, P= 80 g St. steel (100 nm), t = 275 h, P = 200 g Cu (50 urn), f = 275 h, P = 200 g Al (50 urn), t= 275 h, P = j?00 g

40-

a

3530 25-

P/d?

20Radon

15-

z 1

Z<X<Us-

Sample

-» Clamp - • CR-39

10 5 0

us 6

p • 7

Track diameter (urn)

1. I, * M ^

a Load m = 80-250 g

Figure 9. a: CR-39 data on ECP emissions (8 < Ea < 16 MeV) in A u / P d / P d O i D j , (after electrolytic D-loading): (Na) = (5.6 ± 0.5) X l O - 4 ^ - 1 - cm _ 2 )/47r ster (compare with AE/E result). In Fig. 9b the diagram of experimental arrangement is presented.

332

manner. Accordingly t o our CR-39 calibrations 3 the tracks with diameter about 7.0 ± 0.2/tm are usually associated with energetic alpha particles in the range of 12.0 ± 3.0 MeV. T h e tracks with il = 6.2//in could be considered as a signature of 1.7 MeV protons or 2.8 MeV deuterons. Notice t h a t all tested samples showed presence of radon line at d = 8.0 / a n corresponding to alpha energy E = 7.8 MeV. Using Al and Cu samples results as the background counts one can calculate the yield of energetic (E = 12.0 ± 3.0 MeV) alpha particles, taking into account CR-39 efficiency with respect to critical angle of alphas. 3 This calculation gives the result (jY a ) = (5-6 ± 0.5) x 1 0 - 4 ( s - 1 c m - 2 ) in 4TT stcr. which is quite comparable with energetic alpha yield found from the dE/E detection data. Therefore, both SSB and CR-39 techniques allow to observe low intensity emission of energetic charged particles in Pd deuteride/hydride and show comparable results. T h e fast alpha-emission yield obtained in dE/E measurement has the same magnitude in CR-39 experiments confirmed by CR-39 detection. This observation gave rise t o our next step in E C P E study: application of CR-39 technique t o in situ electrolysis detection of E C P E using P d and Ti cathodes. Here we briefly remind some important results t h a t were obtained during in situ detection of E C P E in electrolysis experiment (the diagram is shown in Fig. 2) with the thin film P d cathode on top of alumina substrate. T h e background experiments showed proportional growth of track density vs. time for CR-39 immersed in electrolyte (Fig. 10). For large exposure time (/ > 10 days) two separate alpha peaks with track diameters located at 8.0 and 9.0/tm. respectively, were observed. The positions of these peaks are in good agreement with natural alpha background and are normally corresponding to ~ 7 . 0 MeV radon (8.0/(in tracks) and 5.0 MeV thoron (9.0/tin tracks) series of alpha.-nuclides contained in electrolyte. The foreground runs (f ~ 2.0 30 days) with electrolysis of Pd-thin film cathodes showed the appearance of unusual diameter tracks t h a t were not found in the back-

BB|Opon CR-39 background, t=335 h • • O p e n CR-39 background, / ." i

1 "'»""(

(

"

i

1

•

i

•

t

'

I

i

I y *

Data points Quadratic fit

y = -7.10914E - 6 + 0.00377X + 0.35837X2

M

-

0.00012"5" S> 0.00010-

•

D + , complex and precursor formation occur in an s-electron rich environment, i.e., where q+ < ^ . The IR imaging of the electrode surface shows that excess heat is generated at discrete locations, which, in turn, implies the formation of domains. The existence of hot spots indicates the presence of highly energetic fast reactions, which, in turn, produce pressure and temperature waves travelling through the electrode. Indeed, such waves were observed by the response of a pressure sensitive material onto which the Pd/D films were deposited. 5 1.2. System

Far from

Equilibrium

The characteristics of systems far from equilibrium are: (i) the formation of new structures is always the result of an instability which may be due to either internal or external fluctuations to the system, (ii) fluctuation is always followed by the response which may bring the system to its original conditions or may produce a new structure, (iii) the system's stability is determined by a complex interplay of kinetic and thermodynamic quantities (i.e., no statement can be made that is independent of kinetic considerations), (iv) chemical instabilities lead to spontaneous "self-organization" if the system is able to exchange part of the energy or matter with the outside world in order to establish a microscopic internal order (an open system must be maintained, if self-organization is to occur), and (v) as the overpotential is increased, the probability of cluster formation increases (increase in the rate of formation of hot spots). Parenthetically, in systems far from equilibrium the complexes can be viewed as "supermolecules" where the physical laws, as formulated for systems at or near equilibrium, may not apply. To quote: "... there exist new dynamic states of matter induced by a flow of free energy far from equilibrium. Such states are governed by a new physical chemistry on a supermolecular level, while all laws referring to the molecular level, remain essentially unchanged. In all cases considered, the coherent behavior on the supermolecular level corresponds in fact to an amplification of

361

specific molecular properties (such as kinetic constants) in far from thermodynamic equilibrium conditions" (Ref. 6, p. 290).

1.3. Field Interactions

with Cell

Components

1.3.1. Conductors Introduction of an uncharged conductor into the field reduces the total energy of the field. An uncharged conductor located outside the field is attracted towards the field. A conductor, charged or not, when placed in an electric field cannot remain in stable equilibrium. Consequently, if a conductor is constrained then it will suffer shape change, either reversible or permanent, depending upon conditions at the surface and the time involved.

1.3.2. Electrolyte The electrolyte phase contains mobile positive and negative ions distributed in a manner that assures charge neutrality (except at boundaries). It is known that an ion in contact with water is solvated, which means that the central ion is surrounded by an oppositely charged ionic cloud. When subjected to an electric field, each ion is acted upon with a net force representing the difference between the accelerating force arising from the applied field and the opposing forces, viz (i) the electrophoretic contribution associated with the structure of the moving entity and (ii) the force connected with the relaxation of the ionic cloud.7

1.3.3. Interphase Charging of the Pd lattice with hydrogen isotopes by electrochemical means occurs through a number of consecutive processes, i.e., charge transfer, adsorption, absorption, etc. These processes define the thermodynamic structure of the interphase (as opposed to its physical structure). The set of processes involved is as follows: D+ ( b ) -+ D+ (r) -^ D e - • D D « - D [(D+ • e") n - D+] denoting charge transfer, adsorption, absorption, placement in Pd lattice, ionization, complex formation.8

1.3.4. The Bulk Pd/D Any charge on a conductor must be located at its surface. Charged mobile species (D+ complexes) are also present in the bulk Pd/D material. In general, they will not be affected by an external field, since no field can exist there. However, in the present case, they might be affected by the field generated by the flow of the cell current, i.e., the electrodiffusion might occur.

362

1.3.5. Internal Stresses - Shape Change The relationship between the surface forces and the bulk response is given by / d iivvAA&dTr = p ,Anda,

(1)

where the div operator derives a vector from tensor. The term on the left-hand side term is the algebraic sum of all sources/sinks continuously distributed over the volume element. The right side defines the outflow, if positive and the inflow, if negative. Equation (1) indicates that forces acting on any finite volume in a body can be reduced to forces applied to the surface of that volume and vice versa. Consequently, it follows that the shape change at constant volume is associated with motion due to internal forces acting on the surface. Thus, the deformation will be determined by the distribution of surface forces, while the rate of deformation by their magnitude. Internal stresses can be present without the presence of external loads, e.g., due to inhomogeneities, imperfections, etc., a likely situation in the codeposited film and the continuous evolution of deuterium.

1000 - 3000

Vcm-1 —

•

Figure 1. An electrochemical cell. 1 - clear plastic (acrylic) wall. 2 - P t screen. 3 - Co-deposited PdD layer. 4 - Au foil. 5 - Cu foil. Cell connected to a galvanostat; electric field maintained by a regulated high voltage source (not shown).

1.3.6.

Location/Size

The presence of discrete, randomly distributed sites (hot spots, craters, boulders, etc.) implies the existence of volumes within the electrode material where conditions

363

promoting the highly energetic reactions exist. In estimating their magnitude, one must make a certain number of assumptions, e.g. (i) energy per single event is that of the reaction D + D —> He, (ii) the number of single events to produce a crater is on the order of 104 or higher, depending upon its radius, 9 and (iii) the number of single events needed to generate the "hot spot" displayed by IR imaging is on the order of 104 or higher, depending upon its size and brightness. Under these conditions and assuming the loading ratio greater than unity, one can calculate the radius of this volume to be on the order of 100 A or higher. The events take place within the bulk material in the close vicinity to the contact surface. 2. Experimental/Results An operating P d D / D 2 0 , 0.3 M LiCl/Pt cell was placed in an electrostatic field generated by a parallel plate capacitor where the field strength was maintained and controlled by setting the potential difference at a specified level. The cell geometry is shown in Fig. 1. The Pd/D electrode was prepared by the Pd deposition onto an Au foil from a solution of 0.03 M PdCl 2 + 0.3 M LiCl dissolved in D 2 0 . The electrodeposition was under galvanostatic control with the current profile as follows: 1.0 mA c m - 2 for 8 h, 3 mA c m - 2 for 8 h, and at 5.0 mA c m - 2 until all P d 2 + ions were reduced. Upon completion of the Pd deposition, the cell current was increased to a value needed to maintain a visible gas evolution (usually 30-50 mA c m - 2 ) for the next 2-3h followed by placement in an external electric field (1000-3000 V c m - 1 ) with the cell current increased to about 100mAcm~ 2 for the next 48 h or longer. The surface morphology and the bulk structure of the codeposited Pd/D film, shown in Fig. 2a, undergoes substantial changes when the operating cell is placed in an external electrostatic field. This is illustrated in a series of SEM photographs taken from various runs. In the absence of an electric field, the electrode structure consists of globules, 3-7 /zm in diameter, arranged in short columns. Each of the individual globules is an aggregate of much smaller, almost spherical units, having a diameter in a submicron range. This structure is uniform throughout the electrode. 2.1. Morphological

Changes - Minor

Deformations

The first noticeable effect, seen shortly after placing the cell in an electric field, is swelling of the codeposited material followed by displacement toward the negative capacitor plate. The reorientation without substantial change in their size is shown in Fig. 2b. We selected examples of various structures to emphasize the complexity of the system as well as to indicate the impossibility of a quantitative analysis. The selected examples include minor deformation of the original structure, definitive shape change, unusual structures to a deformation associated with, what appears to be, a localized catastrophic event. Another example of the disintegration of the Pd/D structure is shown in Fig. 2c. This figure illustrates the breaking of the bonds holding together the individual globules. The breaking of the bonding and the separation of globules may be due

364

£3X£

(a)

5

Mi IB Mi

A^

(d)

(e) %&.

B

Figure 2. An illustration of minor morphological changes, (a) Reference morphology (no field), (b) reorientation, (c) disintegration, (d) branches (fractals), and (e) dendritic growth.

to action of the electric field alone or may be due to combined action of electrical and mechanical forces arising from the bulk material response to the changing magnitude of the surface forces. A different set of processes appears to be responsible for the structural changes, viz (i) formation of branches (fractals) (Fig. 2d) and (ii) the production of dendritic growth (Fig. 2e). In what follows, we argue that these two very different forms may have a common origin, namely that they are the result of a combined action of the current flow through a porous structure, the presence of evolving deuterium, and the electric field on the separated microglobules suspended in the electrolyte and restricted by the porous structure. The observed morphological and structural changes occur during the reduction of D + / D 2 0 ions/molecules at the porous electrode. Thus, at least three factors should be considered: (i) the external field, (ii) the distribution of the cell current, and (iii) the presence of gaseous deuterium within the confines of the structure. Since the depth of current penetration (for a given electrode kinetics, current density, etc.)

365

into electrode depends on pore size and assuming that all factors are involved, a different response to the field is expected at different sites of the Pd/D material. At sites of a relatively large pore size, the microglobules are acted upon by two factors, the electric field and the convective flow due to mixing by the evolving deuterium. The electric field redistributes the surface charges while the evolving gas brings microglobules in contact with each other. Viewing Fig. 2d we identify three areas having distinct features: area A with high density of branches and unattached microglobules, area B which is sparsely populated by microglobules and area C where the unattached microglobules are absent and where branches are well denned. The latter indicates that the growth of branches by addition of microglobules leading to an apparent reversal of the action of an electric field. Entirely different situations exist in small pore sizes; the pore wall may be covered by gaseous deuterium, thus shifting the cell current deeper into the porous structure. If a microglobule is placed into the current path, and if the potential drop over the length of the microglobule in the electrolyte is greater than the sum of cathodic and anodic overpotentials needed to dissolve Pd and deposit the P d 2 + ions, then the dendritic growth is possible. 11 2.2. Morphological

Changes - Shape

Changes

The transition from the "cauliflower-like" morphology to other forms is expected due to an interaction of the electric field and the response of a solid to the action of surface forces. While the morphologies shown in Fig. 2b-e can be accounted for, those in Fig. 3b-e suggest that additional factors are involved in producing the observed shape changes. Of the great variety of forms, we selected those illustrating the reshaping of the spherical globules into (i) rods (circular and square), (ii) long wire, (iii) folded thin film, and (iv) a crater, the latter suggesting the presence of a violent event. These structural changes require substantial energy expenditure, far in excess of that can be extracted from the electric field. One such source is of nuclear origin as first suggested by Fleischmann et al.12 and supported by the emission of soft X-rays, 13 charged particles, 14 and tritium 15 and helium 16 production. The SEM photographs, Fig. 3b-e, show well defined areas that seem to represent solidification of molten metal occurring under a liquid. If, as suggested earlier (vide supra) that the energy needed to melt a metal is of nuclear origin, then the chemical analysis of these distinctly different areas should reflect it. To illustrate, we selected three morphologically different cases, viz (i) a thin layer detached from the Au substrate (Fig. 4), (ii) boulder-like region (Fig. 5), and (iii) a place where, most likely, a catastrophic event has occurred (Fig. 6). Analysis of the detached thin film (Fig. 4), showed the presence of Ca, Al, Si, Mg, Zn, Au, O, and CI. Since the latter three elements are present in the cell components, they cannot be attributed to nuclear events (further experiments are needed for verification). The presence of these new elements on the electrode surface, after a long-term polarization, was claimed to be due to the concentration of impurities. 2 ' 3 While we cannot express an opinion as to the validity of such

366

SEE

Condensed Matter Nuclear Science: Proceedings of the 11th International Conference on Cold Fusion: Marseilles, France, 31 October-5 November 2004

Condensed Matter Nuclear Science: Proceedings of the 12th International Conference on Cold Fusion, Yokohama, Japan 27 November-2 December 2005

Neural Information Processing: 11th International Conference, ICONIP 2004, Calcutta, India, November 22-25, 2004, Proceedings

Condensed Matter Theories: Proceedings of the 31st International Workshop

Condensed Matter Theories: Proceedings of the 33rd International Workshop

Proceedings of the International Conference on Colloid and Surface Science

Proceedings of International Science Education Conference 2009

Automated technology for verification and analysis: second international conference, ATVA 2004, Taipei, Taiwan, ROC, October 31-November 3, 2004: proceedings

Software Process Improvement: 11th European Conference, EuroSPI 2004, Trondheim, Norway, November 10-12, 2004. Proceedings (Lecture Notes in Computer Science)

Quantum Theory of Condensed Matter: Proceedings of the 24th Solvay Conference on Physics

Quantum Theory of Condensed Matter: Proceedings of the 24th Solvay Conference on Physics