Super Abundant Vacancies, or Fukai Vacancies

This study has been split into two pages. A list of all papers cited in a number of sources, with abstracts, is on the subpage, Abstracts.This page will then become a summary review.

  • Tools for visualization

Super Abundant Vacancies refers to phases of metal hydrides where the crystal structure incorporates hydrogen at very high loading, above 100% atom ratio. The term can be misleading. Without hydrogen, palladium has a Face-Centered Cubic structure, which is cubic, but with an additional lattice position at the center of each face of the cube. FCC palladium soaks up hydrogen, at lower loading the loading of hydrogen is exothermic. However, as loading reaches a certain point, the loading becomes endothermic, it takes energy to force more hydrogen in. The limit with electrochemistry is somewhere under 100 atom percent. To reach a higher percentage, there are two basic approaches: the first to be recognized was loading under high pressure. Fukai reported in 1993 that he had placed palladium and a hydrogen source (Lithal or lithium aluminum hydride, LiAlH4) in a diamond-anvil press, subjecting it to a pressure of 5 GPa, which is 10,000 bar (1 bar is approximately one atmosphere, so 5 GPa is about 5 kilograms per square millimeter, the pressure cell is an 8 mm cube). He had a beam of highly collimated X-rays passing through the cell, the X-rays are scattered and the technique, X-ray diffraction, is used to determine crystal structure. He also had a heater in the cell, a graphite tube heater, which could handle the pressure.

What he reported was that, at 400 C., when the Lithal decomposed, releasing the hydrogen, the palladium, a disc 1 mm in diameter and 0.2 mm thick, expanded at first, as predicted and known. However, when the temperature was raised to 800 C., and over three hours, the crystal structure shrank. (Palladium rapidly anneals at 890 C, according to Johnson-Matthey. I would expect slower annealing at 800 C.)

The X-ray crystallography was consistent with an FCC structure with one of the four sublattices being replaced by vacancies (but filled with hydrogen, up to six hydrogen atoms can occupy a vacancy, calculations and studies have shown). This is represented, with the material created in the Fukai pressure process, as Pd3VacH4. The loading has been measured confirming this, at least roughly.

When the material was quenched and returned to atmospheric pressure, the material remained in this state. Ordinary vacancy rates for palladium vary with temperature and other conditions, but may be on the order of 0.1%, off the top of my head, this material has vacancy rates (referring to the FCC structure) on the order of 20-25%. The material remained in this high-defect structure even when the hydrogen was removed by heating at low pressure, but staying well below 800 C. At 800 C., the material would anneal back to pure Pd FCC structure.

It is then speculated, and there is some evidence with similar metal hydrides, that SAV material may form at lower pressures and temperatures, even at room temperature, if palladium hydride is built up atom by atom, as in co-deposition, or if the material is stressed heavily and exposed to hydrogen.

So, then, we notice that if SAV is the Nuclear Active Environment, many of the mysterious and frustrating characteristics of the Fleischmann-Pons effect can be explained. Further, what is being noticed with some excitement is that SAV material can be deliberately created.

If SAV palladium is the NAE, it could be highly active, dangerous, even, if loaded with deuterium. My sense is that experiments are under way.

We have a draft of the Michael Staker ICCF-21 paper here, see below. We also have video of Staker’s presentation and a transcript, with time-links to the video and integrated with the abstract and slides.

See also our subpage, McKubre and Staker (2018), “NAE = SAV = SPD + D”.  

One of the major unresolved issues in research into the Anomalous Heat Effect is the location of the reactions. Much early research assumed that the reaction would be taking place generally in the bulk of the palladium metal, loaded with deuterium, used in the Fleischmann-Pons experiment. Eventually, it became quite clear, with many evidences, that the heat, and associated helium, was coming from a surface effect. Where on the surface, the whole surface? Not likely. Different theories propose different sites. Storms’ hydroton theory proposes nanocracks, formed when repeated loading and deloading stresses the metal. There have been other proposals, most notably vacancies, empty locations within the lattice, which might then host reactions. However, ordinary vacancies are just that, ordinary. Whatever is causing the FP Heat Effect is not ordinary. Something special must happen first.

Recently, I became aware of concepts that were first published in the 1990s, but which were not necessarily widely noticed and understood. As there was some discussion arising (much of it private), I collected sources on what are called “Fukai phases” or “super abundant vacancies.” The “vacancies” name could be seen as misleading, because these are not simply vacancies, they are phases of an alloy, palladium hydride/deuteride, phases not ordinarily seen because they do not form under ordinary conditions. This is getting quite interesting, and I will be able to write more about this soon.

Then I have a document from a LENR researcher critical of SAV theory. Permission to quote was given, but there were unclear restrictions. I will report the ideas here, absent clear permission to quote with attribution.

(I have just received copies of many of these papers and have uploaded them. It may take a little time before Google responds, but I’m entering the search terms now.)

Much of material originally here has been replaced by the Abstracts page. I have left in place coverage of two papers, Nazarov (2014), and Fukada (2016) as being recent reviews.

Nazarov (2014)

This paper begins with a review of SAV studies. The introductory paragraph here I want to copy.

The desired or unwanted presence of hydrogen in metals is a long-standing research topic in materials science. One of the astonishing implications of this presence can be an increase of the vacancy concentration in a material by several orders of magnitude, the so-called superabundant vacancy (SAV) formation. The physical picture behind this effect is a trapping of hydrogen in vacancies, yielding an overall reduction of the vacancy energy of formation. Despite the straightforwardness of such an explanation, it took until 1993 before the SAV phenomenon was first discovered experimentally by Fukai and co-workers in Pd [1] and Ni [2]. Since then, however, it has been observed in many metallic systems such as Cu [3], Ti [4], Pd and Pd alloys [5–8], Al [9], Mn [10], Fe [10,11], Mo [12], Cr [13], Co [10], Ni [14], Ni-Fe alloy [15], Nb [16–18], some hydrogen storage alloys [19], some metal hydrides [20], and stainless steels [21,22]. There are now examples available that the concentration of vacancies can become as large as 10% and more [10,23,24]. Even vacancy-ordered phases have been detected in Pd [1,5,25–27], Mn [28], Ni [14], and Fe [29]. These high-vacancy concentrations are typically not formed immediately, but only after hydrogen loading for several hours at sufficiently high temperatures [23]. However, recent investigations [10] have shown that large concentrations of vacancies (up to 10−4) can be generated at internal sources of pure metals in less than 1 s. Further, hydrogen-induced vacancy formation can be substantially promoted by deformation in mechanical loading experiments [9,30–32] or strain due to phase transformations [8,33,34].

The references:

[1] Y. Fukai and N. Okuma, Phys. Rev. Lett. 73, 1640 (1994). 
[2] Y. Fukai and N. Okuma, Jpn. J. Appl. Phys., Part 2 32, L1256 (1993). 
[3] Y. Fukai, M. Mizutani, S. Yokota, M. Kanazawa, Y. Miura, and T. Watanabe, J. Alloys Compd. 356, 270 (2003)
[4] K. Nakamura and Y. Fukai, J. Alloys Compd. 231, 46 (1995). [paper needed]
[5] D. dos Santos, S. Miraglia, and D. Fruchart, J. Alloys Compd. 291, L1 (1999)
[6] Y. Fukai, Y. Ishii, T. Goto, and K. Watanabe, J. Alloys Compd. 313, 121 (2000). [paper needed]
[7] K. Watanabe, N. Okuma, Y. Fukai, Y. Sakamoto, and Y. Hayashi, Scr. Mater. 34, 551 (1996)[paper needed]
[8] K. Sakaki, R. Date, M. Mizuno, H. Araki, and Y. Shirai, Acta Mater. 54, 4641 (2006)[paper needed]
[9] H. Birnbaum, C. Buckley, F. Zaides, E. Sirois, P. Rosenak, S. Spooner, and J. Lin, J. Alloys Compd. 253, 260 (1997)[paper needed]
[10] Y. Fukai, T. Haraguchi, E. Hayashi, Y. Ishii, Y. Kurokawa, and J. Yanagawa, Defect Diffus. Forum 194, 1063 (2001)[paper needed]
[11] Y. Fukai, K. Mori, and H. Shinomiya, J. Alloys Compd. 348, 105 (2003)[paper needed]
[12] Y. Fukai, Y. Kurokawa, and H. Hiraoka, J. Jpn. Inst. Met. 61, 663 (1997). [paper needed][reference obscure, no vol 61, paper not at page in 1997. About Mo, see this 2003 paper]
[13] Y. Fukai and M. Mizutani, Mater. Trans. 43, 1079 (2002). (copy)  Britz Fukai2003b
[14] Y. Fukai, Y. Shizuku, and Y. Kurokawa, J. Alloys Compd. 329, 195 (2001).  Britz Fukai2001
[15] Y. Fukai, T. Hiroi, N. Mukaibo, and Y. Shimizu, J. Jpn. Inst. Met. 71, 388 (2007)[paper needed]
[16] H. Koike, Y. Shizuku, A. Yazaki, and Y. Fukai, J. Phys.: Condens. Matter 16, 1335 (2004)[paper needed]
[17] T. Iida, Y. Yamazaki, T. Kobayashi, Y. Iijima, and Y. Fukai, Acta Mater. 53, 3083 (2005)[paper needed]
[18] J. Cızek, I. Prochazka, F. Becvar, R. Kuzel, M. Cieslar, G. Brauer, W. Anwand, R. Kirchheim, and A. Pundt, Phys. Rev. B 69, 224106 (2004)[paper needed]
[19] Y. Shirai, H. Araki, T. Mori, W. Nakamura, and K. Sakaki, J. Alloys Compd. 330, 125 (2002).  [paper needed]
[20] Y. Fukai and H. Sugimoto, J. Phys.: Condens. Matter 19, 436201 (2007).  [paper needed]
[21] V. Gavriljuk, V. Bugaev, Y. Petrov, A. Tarasenko, and B. Yanchitski, Scr. Mater. 34, 903 (1996).  [paper needed]
[22] Y. Yagodzinskyy, T. Saukkonen, S. Kilpelinen, F. Tuomisto, and H. Hnninen, Scr. Mater. 62, 155 (2010).  [paper needed]
[23] Y. Fukai, J. Alloys Compd. 356, 263 (2003)
[24] Y. Fukai, Phys. Scr. T103, 11 (2003)[paper needed]
[25] S. Semiletov, R. Baranova, Y. Khodyrev, and R. Imamov, Kristallografiya 25, 1162 (1980) ,[Sov. Phys.–Crystallogr. 25, 665 (1980)]. [paper needed]
[26] Y. Fukai, J. Alloys Compd. 231, 35 (1995) [paper needed]
[27] S. Miraglia, D. Fruchart, E. Hlil, S. Tavares, and D. D. Santos, J. Alloys Compd. 317-318, 77 (2001).
[28] Y. Fukai, Computer Aided Innovation of New Materials (Elsevier, Amsterdam, 1993), Vol. II, pp. 451–456. [the Fukai paper appears to be in Vol I?] [paper needed]
[29] Y. Fukai, M. Yamakata, and T. Yagi, Z. Phys. Chem. 179, 119 (1993). [paper needed] bad doi, corrected:
[30] M. Nagumo, M. Takamura, and K. Takai, Metall. Mater. Trans. A 32, 339 (2001)[paper needed]
[31] K. Sakaki, T. Kawase, M. Hirato, M. Mizuno, H. Araki, Y. Shirai, and M. Nagumo, Scr. Mater. 55, 1031 (2006)[paper needed]
[32] Y. Z. Chen, G. Csiszar, J. Cizek, C. Borchers, T. Ung ´ ar, S. Goto, and R. Kirchheim, Scr. Mater. 64, 390 (2011)[paper needed]
[33] Y. Shirai, F. Nakamura, M. Takeuchi, K. Watanabe, and M. Yamaguchi, in Eighth International Conference on Positron Annihilation, edited by V. Dorikens, M. Drikens, and D. Seegers (World Scientific, Singapore, 1989), p. 488. [paper needed]
[34] P. Chalermkarnnon, H. Araki, and Y. Shirai, Mater. Trans. JIM 43, 1486 (2002). [copy

Another SAV paper has been pointed out to me:

Yoshiki FukadaTatsumi HiokiTomoyoshi MotohirobMultiple phase separation of super-abundant-vacancies in Pd hydrides by all solid-state electrolysis in moderate temperatures around 300 °C, Journal of Alloys and Compounds, Volume 688, Part B, 15 December 2016, Pages 404-412. DOI * ResearchGate


The dynamics of hydrogen-induced vacancies are the key for understanding various phenomena in metal–hydrogen systems under a high hydrogen chemical potential. In this study, a novel dry-electrolysis experiment was performed in which a hydrogen isotope was injected into a Pd cathode and time-resolved in situ monochromatic X-ray diffraction measurement was carried out at the Pd cathode. It was found that palladium-hydride containing vacancies forms multiple phases depending on the hydrogen chemical potential. Phase separation into vacancy-rich, vacancy-poor, and moderate-vacancy-concentration phases was observed when the input voltage was relatively low, i.e., ∼0.5 V. The moderate-vacancy-concentration phase may be attributed to Ca7Ge or another type of super-lattice Pd7VacH(D)8. Transition from the vacancy-rich to the moderate-vacancy-concentration phase explains the sub-micron void formations without high temperature treatment that were observed at the Pd cathode but have never been reported in previous anvil experiments.

“Graphical Abstract”|

The researchers are also working on Leading the Japanese Gvt NEDO project on anomalous heat effect of nano-metal and hydrogen gas interaction

For comparison, the sources from Fukada:

(all these sources have links in the paper found on ResearchGate)

[1] Y. Fukai, N. Okuma, Evidence of Copious Vacancy Formation in Ni and Pd under a High Hydrogen Pressure, Jpn.J.Appl.Phys., 32 (1993) L1256–L1259.
[2] Y. Fukai, N.Okuma, Formation of Superabundant Vacancies in Pd Hydride under High Hydrogen Pressures, Phys.Rev.Lett., 73 (1994) 1640.
[3] S. Harada, S. Yokota, Y. Ishii, Y. Shizuku, M. Kanazawa, Y. Fukai, A relation between the vacancy concentration and hydrogen concentration in the Ni–H, Co–H and Pd–H systems, J. Alloys Compd., 404–406 (2005) 247–251.
[4] Y. Tateyama and T. Ohno, Stability and clusterization of hydrogen–vacancy complexes in α-Fe: An ab initio study, Phys. Rev., B67 (2003) 174105.
[5] M. Nagumo, M. Nakamura, K. Takai, Hydrogen thermal desorption relevant to delayed-fracture susceptibility of high-strength steels, Metal.Mater.Trans. A, 32A (2001) 339–347.
[6] M. Nagumo, Hydrogen related failure of steels—a new aspect, Mater.Sci.Tech., 20 (2004) 940–950.
[7] Y. Fukai, M. Mizutani, S. Yokota, M. Kanazawa, Y. Miura, T. Watanabe, Superabundant vacancy–hydrogen clusters in electrodeposited Ni and Cu, J. Alloys Compd., 356–357 (2003) 270–273.
[8] N. Fukumuro, T. Adachi, S. Yae, H. Matsuda, Y. Fukai, Influence of hydrogen on room temperature recrystallisation of electrodeposited Cu films: thermal desorption spectroscopy, Trans. Inst. Met. Finish., 89 (2011) 198–201.
[9] N. Hisanaga, N. Fukumuro, S. Yae, H. Matsuda, Hydrogen in Platinum Films Electrodeposited from Dinitrosulfatoplatinate(II) Solution, ECS Trans., 50(48) (2013) 77–82.
[10] K. Watanabe, N. Okuma, Y. Fukai, Y. Sakamoto and Y. Hayashi, Superabundant vacancies and enhanced diffusion in Pd–Rh alloys under high hydrogen pressures, Scripta Materialia, 34(4) (1996) 551–557.
[11] E. Hayashi Y. Kurokawa and Y. Fukai, Hydrogen-Induced Enhancement of Interdiffusion in Cu–Ni Diffusion Couples, Phys.Rev.Lett., 80(25) (1998) 5588.
[12] N. Fukumuro, M. Yokota, S. Yae, H. Matsuda, Y.Fukai, Hydrogeninduced enhancement of atomic diffusion in electrodeposited Pd films, J. Alloys Compd., 580 (2013) s55–s57.
[13] Y. Fukai, Formation of superabundant vacancies in metal hydrides at high temperatures, J. Alloys Compd., 231 (1995) 35–40.
[14] Y. Fukai, Y. Kurokawa, H. Hiraoka, Superabundant Vacancy Formation and Its Consequences in Metal–Hydrogen Alloys, J. Japan Inst. Metals, 61 (1997) 663–670 (in Japanese).
[15] Y. Fukai, Y. Shizuku, Y. Kurokawa, Superabundant vacancy formation  in Ni–H alloys, J. Alloys Compd., 329 (2001) 195–201.
[16] Y. Fukai, Y. Ishii, Y. Goto, K. Watanabe, Formation of superabundant vacancies in Pd–H alloys, J. Alloys Compd., 313 (2000) 121–132.
[17] Y. Fukai, H. Sugimoto, Formation mechanism of defect metal hydrides containing superabundant vacancies, J. Phys.: Condens. Matter, 19 365 (2007) 436201. [paper needed]
[18] Y. Fukai, H. Sugimoto, The defect structure with superabundant vacancies to be formed from fcc binary metal hydrides: Experiments and simulations, J. Alloys Compd., 446–447 (2007) 474–478. [paper needed] Defective citation, lead author is Harada. Harada2007.
[19] C. Zhang, Ali Alavi, First-Principles Study of Superabundant Vacancy Formation in Metal Hydrides, J. Am. Chem. Soc., 127(27) (2005) 9808–9817.
[20] S.Yu. Zaginaichenko, Z.A. Matysina, D.V. Schur, L.O. Teslenko, A. Veziroglu, The structural vacancies in palladium hydride. Phase diagram, Int. J. Hydrogen Energy, 36 (2011) 1152–1158.
375 [21] R. Nazarov, T. Hickel, and J. Neugebauer, Ab initio study of H–vacancy interactions in fcc metals: Implications for the formation of superabundant vacancies, Phys. Rev. B89 (2014) 144108
[22] R. Felici, L. Bertalot, A. DeNinno, A. LaBarbera and V. Violante, In situ measurement of the deuterium (hydrogen) charging of a palladium 380 electrode during electrolysis by energy dispersive x-ray diffraction, Rev. Sci. Instrum., 66(5) (1995) 3344.
[23] E.F. Skelton, P.L. Hagans, S.B. Qadri, D.D. Dominguez, A.C. Ehrlich and J.Z. Hu, In situ monitoring of crystallographic changes in Pd induced by diffusion of D, Phys. Rev., B58 (1998) 14775.
[24] D.L. Knies, V.Violante, K.S. Grabowski, J.Z. Hu, D.D. Dominguez, J.H. He, S.B. Qadri and G.K. Hubler, In-situ synchrotron energy-dispersive x-ray diffraction study of thin Pd foils with Pd:D and Pd:H concentrations up to 1:1, J. Appl. Phys., 112 (2012) 083510.
[25] C.E. Buckley, H.K. Birnbaum, D. Bellmann, P. Staron, Calculation of the radial distribution function of bubbles in the aluminum hydrogen system, J. Alloys Compd., 293–295 (1999) 231–236.
[26] H. Wulff, M. Quaas, H. Deutsch, H. Ahrens, M. Fr¨ohlichc, C.A. Helm, Formation of palladium hydrides in low temperature Ar/H2-plasma, Thin Solid Films, 596 (2015) 185–189.
[27] Y. Fukada, T. Hioki, T. Motohiro, S. Ohshima, In situ x-ray diffraction study of crystal structure of Pd during hydrogen isotope loading by solidstate electrolysis at moderate temperatures 250–300◦, J. Alloys Compd., 647 (2015) 221–230.
[28] H. Osono, T. Kino, Y. Kurokawa, Y. Fukai, Agglomeration of hydrogen induced vacancies in nickel, J. Alloys Compd., 231 (1995) 41–45.
[29] D.S dos Santos, S. Miraglia, D. Fruchart, A high pressure investigation of Pd and the Pd–H system, J. Alloys Compd., 291 (1999) L1–L5.
[30] D. S. dos Santos, S. S. M. Tavares, S. Miraglia, D. Fruchart, D. R. dos Santos, Analysis of the nanopores produced in nickel and palladium by high hydrogen pressure, J. Alloys Compd., 356–357 (2003) 258–262.
[31] O.Yu. Vekilova, D.I. Bazhanov, S.I. Simak, I.A. Abrikosov, First-principles study of vacancy–hydrogen interaction in Pd, Phys.Rev., B80 (2009) 024101.
[32] I.A. Supryadkina, D.I. Bazhanov, and A.S. Ilyushin, Ab Initio Study of the Formation of Vacancy and Hydrogen–Vacancy Complexes in Palladium and Its Hydride, Journal of Experimental and Theoretical Physics, 118 (2014) 80–86.
[33] Daisuke Kyoi, Toyoto Sato, Ewa R¨onnebro, Yasufumi Tsuji, Naoyuki Kitamura, Atsushi Ueda, Mikio Ito, Shigeru Katsuyama, Shigeta Hara, Dag Nor´eus, Tetsuo Sakai, A novel  magnesium–vanadium hydride synthesized by a gigapascal-high-pressure technique, J. Alloys  Compd.,  (2004) 253–258.
[34] S. Tavares, S. Miraglia, D. Frucharta, D.Dos Santos, L. Ortega and A. Lacoste, Evidence for a superstructure in hydrogen-implanted palladium, J. Alloys Compd., 372 (2004) L6–L8.
[35] H. Araki, M. Nakamura, S. Harada, T. Obata, N. Mikhin, V. Syvokon, M. Kubota, Phase Diagram of Hydrogen in Palladium, J. Low Temp. Phys., 134 (2004) 1145–1151.
[36] Y. Fukai, The Metal–Hydrogen System, Second edition, Springer-Verlag, (2005).
[37] O. Blaschko, Structural features occurring in PdDx within the 50 K anomaly region, J. Less-Comm. Met., 100 (1984) 307–320
[38] Atsushi Yabuuchi, Teruo Kihara, Daichi Kubo, Masataka Mizuno, Hideki Araki, Takashi Onishi and Yasuharu Shirai, Effect of Hydrogen on Vacancy Formation in Sputtered Cu Films Studied by Positron Annihilation Spectroscopy, Jpn.J.Appl.Phys., 52 (2013) 046501.
[39] Y. Fukai, The structure and phase diagram of M–H systems at high chemical potentials—High pressure and electrochemical synthesis, J. Alloys Compd., 404–406 (2005) 7–15.

Commentary and confusion

The Fukai discoveries and implications are upsetting some apple-carts, and they are rather easily misinterpreted. (I’ve certainly made a fair share of errors in learning about this, and I write while I am learning).

I had more or less dismissed the Staker 2018 ICCF-18 presentation, as Yet Another Fleischmann Pons Replication with unimpressive heat, accompanied with some theory . . . ah, a polite term would be “stuff.” Goes to show about knee-jerk impressions! Fortunately, McKubre noticed, and was willing to put his 29 years of experience on the shelf as . . . having missed something important in the early 1990s, so he co-authored, with Staker, a presentation at the Greccio conference, and my review of that was requested before it was presented.

Heh! That’s a way to get me to read something!

And then some arguments against SAV as NAE, McKubre’s hypothesis presented as a theme, began to appear, contradicting the idea with arguments that were . . . off, often misrepresenting what has actually  been claimed. So, here, I will present some of these arguments. Comments are open here and my views are just that, my views. However, they are informed, and I do take some offense at misrepresentation of sources, because it causes and spreads unnecessary conflict and confusion. There is lots of room for disagreement, but, please, no fake news, which is close to lying, even if merely incautious and superficial.

Some of these comments are questions, which I answer, AFAIK, here.

What is a simple definition of SAV?

SAV refers to material with a vacancy rate far higher than normal isolated vacancies; the normal rate depends on temperature. As I recall, normal vacancy rate, missing atoms in the normal crystal structure of Pd, is on the order of 10-4.  Super Abundant Vacancies are on the order of 25% (for Pd3VH4, δ phase) or 14% (for Pd7VH6-8, γ phase).

The SAV phases are apparently crystal structures of PdH/D that incorporate vacancy locations. Not simply a Face-Centered-Cubic metal structure with some H/D stuffed in (that’s the α and β phases.)

The SAV phases are new phases in the PdH phase diagram, unknown before 1993. They may start to form at loadings of about 85%, where γ begins to separate from the beta phase. However, due to kinetics, the actual transformation, with ordinary loaded Pd, does not occur until the material reaches annealing temperature. There is evidence that the SAV phases, for sufficiently loaded metal hydrides, can be formed at lower temperatures under some conditions, co-deposition being one of them, and repeated stress in the metal from loading and deloading.

What is the mechanical limit for vacancy concentration?

I’ve seen no limit. The delta phase appears to be stable, with nominal vacancy concentration of 25%. At 50%, I doubt that the material would be stable, it would likely disintegrate. Lower hydrogen concentration stabilizes the SAV Pd structure, allowing high-vacancy phases to form, and, then, at lower temperatures, even if the hydrogen is removed, kinetics prevents the structure from annealing. So my guess would be the limit is somewhere south of 50%, possibly quite close to 25%. With much more than 25% vacancies, which implies more than one vacancy per cubic cell, the material would be seriously weakened. The Pd3V structure has all the vacancies separated by at least one metal atom, it is still a lattice. I’d expect Pd2V2 would fall apart. A Pd3VH4 structure may have some level of vacancies and still have some coherence, but these vacancies could not be common.

Can the SAV argument be defeated on the basis of lattice/metal failure when exposed to highly concentrated 24 MeV reactions?

This would have to be the argument for SAV as NAE. No. There is no evidence of nuclear activity from melted lattice. High temperature, above, say, 400 C, will decompose PdH, and if the temperature goes above 700 C., the metal will anneal out the vacancies. However, there are rate considerations. Those effects take time. The higher the temperature the shorter the time. Obviously, if metal melts in a region, SAV conditions cannot be maintained in that metal, but may remain in the rest of the metal.

(as a response to the above question:) … the proposed high concentration of active sites implied by the SAV idea would suggest that most sites would melt.

This assumes a particular mechanism, i.e., an idea that the SAV would immediately generate reactions, and that the rate of these reactions would be high enough to melt themselves. In fact, there is some evidence that such melts take place. The method used to initiate LENR have not, to date, methods what would generate large amounts of SAV material. Vacancies, highly loaded with hydrogen (It seems it can be up to nine hydrogen atoms per vacancy), may be a necessary but not sufficient condition for the reaction.

It is also possible that with existing LENR approaches, only a small amount of SAV material is created, and, indeed, it melts. Because pure SAV has never been tested, I suggest treating such material with high caution, particularly with deuterium. Treat it as if it is highly reactive, so start with small quantities, perhaps very small. Be prepared for the entire batch to melt. At an extreme, to vaporize, but if the hydrogen pressure is slowly raised, there is no reason to expect, with SAV material, a sharp threshold. So XP should kick in slowly.

There is a big lack of substantial and recent review papers regarding SAVs. The last review paper (in 3 parts) I’m aware of, written by Fukai, was in Japanese and was made in 2011 / 2012.

You can see more recent abstracts here. However, those papers are not full reviews of the SAV concept. Staker (2108) includes substantial review. One of the issues would be that the SAV concept does not appear to be controversial among metallurgists. This 2018 article in Chemical Science treats superabundant vacancies as a known fact.

It is known that hydrogen molecules invade inside metal or alloy lattices as hydrogen atoms and generate defect structures with superabundant vacancies, promoting atomic diffusion and structural change of alloys.

ref 46 is Fukumuro (2013)
ref 47 is Mukaibo (2008) (which is a Fukai paper but missing from our Abstract list.)
ref 48 is Hayashi (1998)

Bukonte (2017) is a recent theoretical study which amounts to a review.


In 1993, Fukai and Okuma (F-O) (1, 2) heated PdH to 800° C while applying 5 GPa of physical pressure

Small point, but it was 700 C for Pd. (The same work was done with Nickel at 800 C.) The author of this makes a point of calling the pressure “physical,” and had a concept of the machine anvils pressing against the palladium. No, the pressure became hydrogen gas pressure, on a package of solid materials. The anvils were not in contact with the palladium at all. The sequence was

  1. apply 5 GPa pressure to the package, which included LiAlH4′
  2. raise the temperature so that Lithal decomposes. (This could create a pressure of 1 GPa if not already under higher pressure).
  3. continue to raise the temperature to 700 C.

The material was observed to decrease in physical volume and the lattice parameter to decrease in value, based on a face-centered cubic (fcc) structure.

The material, from the full treatment, did not “decrease in volume.” The loading at high pressure caused the material to expand (which was expected). The lattice parameter was determined from X-Ray Diffraction analysis. The lattice parameter is based on XRD and is not “based on a . . . [ presumed] structure.” The lattice parameter remained increased over the normal FCC structure and untreated Pd.

This reduction in volume is proposed to result from formation of what they call super abundant vacancies (SAV).

This, as it was understood, leads to patent nonsense. There is, first of all, no reduction in volume, properly considered. There is a reduction in lattice parameter. But this is with an alloy. The density of Pd in atoms per unit volume actually decreases from the process. The lattice parameter decreases, but some of the lattice positions are vacant.

Changes in the X-ray diffraction pattern while PdH is being held at high pressure, which are retained after the sample had been returned to ambient conditions, are used to support the idea.

In the experiment, Britz Fukai2003, the palladium starts as ordinary palladium, with a lattice parameter of about 3.855 Å. Pressure has little effect, apparently. The normal, reported lattice parameter for Pd is 3.859  Å(Wikipedia). As the temperature increased, the lattice parameter,measured by in-situ XRD analysis, increased, and when the Lithal decomposed, beginning below 400 C, it rapidly increased, to about 4.100 Å at 700 C. However, they then waited, and within three hours, the XRD was showing two phases, with lattice parameters of 4.100 and 4.055. By about 6 hours, this had settled into a single phase, lattice parameter about 4.070. When the temperature was lowered, by 600 C, there were again two phases, lattice parameters 4.025 and 4. 055. Back at room temperature, the parameters were 3.975 and 4.010.

The behavior with nickel was similar, except nickel only shows a single phase, and they used 800 C for the nickel.




The critic’s description is implies that the idea came first and then the evidence was used to support it. No, the conclusions of Fukai are rather obvious from the data. Just unexpected, because for a long time, nobody had been able to find evidence for phases beyond  with PdH/D, and they had looked. But nobody had, before, taken loaded PdH to 700 C. Normally, if you raise the temperature of that alloy above 400 C, it will decompose (just like the Lithal, though Lithal is quite unstable, compared to PdH. To keep the PdD from decomposing takes high pressure. High pressure had been used, but never combined with high temperature.

The authors propose that high physical pressure causes the Pd atoms to be removed from sites where the gold atoms are located and these sites remain vacant of Pd atoms, thereby justifying the concept of SAV. Hydrogen atoms are proposed to fill these vacant sites, thereby creating what they call the Pd3VacH4 compound.

This is, again, backwards. Fukai et al do not propose that “high physical pressure” causes the Pd atoms to be removed from sites (the reference to “gold atoms” is to a diagram for Cu3Au; the proposed δ phase would be Pd3VacH4. This is a method of describing the crystalline phase by comparing it with a known structure. However, the Vac is not actually “vacant,” or empty. The delta phase structure has more hydrogen atoms in it than metal (loading 1.33). They are occupying the “vacancy.”

However, when the material is quenched and returned to 1 bar, and then the H is removed by heating to 350 C., what is left is Pd, but it is Pd with vacancies in the structure, that are now really vacant. From δ material, this is Pd3V, a fluffy form of Pd, 25% vacancy rate.  It’s ordered, not a foam. From γ material, it is Pd7V.

The evidence shows that it is not pressure that causes the shift. It is not that “Pd atoms” are being “removed.” Rather, the new structure for PdD is more stable and when the temperature is sufficient to provide mobility for Pd atoms, they migrate to more stable positions, relieving stress. Loading Pd with H stresses it, as the loading becomes high. That expanded Pd is stressed, and can, if the kinetics allow, readjust itself to a more efficient, less stressful packing of the two substances.

Normal annealing will remove vacancies from an alloy. This is no exception, because the “vacancies” aren’t really vacancies; they do become vacancies, though, when the material is deloaded.

The hydrogen is there, before any change in structure. It is not, then, that vacancies are first created and then hydrogen atoms fill them. The metal is first filled with hydrogen and then, when kinetics allow, it anneals into the new structure. This change is not reversible. When the material is cooled and depressurized, it remains a new material, with an increased lattice constant over raw Pd.

Application of large physical pressure is known to cause the crystal lattice of a material to form a new arrangement of atoms having a greater packing efficiency.

The concept of physical pressure is not “physical,” when it is distinguished from gas pressure. That is, pressure is force per unit area, and all pressure is really electromagnetic, when solids are pressed against each other, at the atomic scale, they do not actually “touch.” Rather the electron shells repel each other. Most solids are effectively incompressible.  The author here cites Brittanica on high pressure, apparently referring to this:

The principal effect of high pressure, observed in all materials, is a reduction in volume and a corresponding shortening of mean interatomic distances.

Yes. Notice: observed in all materials. This is quite obvious, actually. A crystallized material with no voids in the structure is maintained in its shape, with characteristic interatomic distances by the repulsive forces between the atoms, balanced by attraction. If the material is placed under high external pressure (which is what the author must mean by “physical pressure,” there must be a balancing force, which is supplied by increased interatomic repulsion, which is supplied by a reduction in the interatomic distance. But this reduction is quite small, and is irrelevant when the pressure is applied, not to the surface of a crystal, but by a gas that can penetrate the crystal, allowing pressure to equalize.

The actual experiment shows that the reduction does not occur from pressure. Pressure with hydrogen, in fact, causes the lattice parameter to increase, not decrease.

In the case of the face-centered-cubic structure (fcc) of PdH, high physical pressure along with high H2 pressure is proposed to cause a variation of the body-centered-cubic (bcc) structure to form, as is discussed below.

This is the author’s proposal. It asserts “physical pressure,” creating confusion. High pressure does not cause the new phase to form. (that had been done before with PdH.) The H2 pressure in the Fukai experiment will be equal to the pressure on the entire experimental package, neglecting other trapped gases. (That is, the H2 pressure may be a little lower if there is some air from assembly included. There would not be much, compared to the large volume of H2 released, enough to raise, by itself, the pressure in the cell by 1 GPa, if merely confined in the space taken up by itself.)

The new phase is produced when the temperature is increased to a level sufficient to allow PdH to anneal.

First, let’s summarize the conclusions proposed by Fukai in his various papers. Applied high physical pressure combined with high pressure H2 at high temperature is said to cause removal of certain identified atoms from the Cu3Au-type structure, producing what is call super abundant vacancies (SAV) in the atom arrangement.

It should be understood that this is not just Fukai. This is widely confirmed, and supported with theory, but the conclusions are not as stated by this author. He has reworded and reinterpreted it. The sequence is that high loading causes stress on the material, which is well-known. In the Fukai experiment, high loading is caused by high pressure in a hydrogen atmosphere. Under those conditions, palladium will load to high values, and Fukai apparently loaded to above 1.0 atom ratio H/Pd. Without the pressure, the loading would decline, but pressure maintains the loading, even if the temperature is raised to a temperature that would normally cause rapid deloading.

The pressure does not cause a “removal of atoms.” Rather, at annealing temperature, vacancies can propagate and materials can rearrange themselves. This is normal annealing process, not at all unusual. What is unusual is setting up conditions so that PdH can anneal! Normally, at lower temperatures, the crystal structure is too rigid, it takes too much energy to move a Pd atom. But the atoms don’t disappear, they move, and in that process, the material is purified as to crystal structure. This is why annealed metals are generally stronger. Every vacancy creates a weakness in the structure.

This vacancy structure is proposed to be the true stable structure in many metal-H2 systems that forms even in the absence of high pressure when such compounds are electrodeposited. This structure is claimed to be the stable hydride such that the presently accepted phase diagrams need to be modified because they only describe a metastable condition.

This is true, it is proposed, based on experimental evidence and theory. The SAV phases are true stable hydrides for their composition ratio. The presently accepted phase diagrams are not “wrong,” because this is how palladium hydride behaves when the material being loaded with hydrogen is already formed into a crystal structure. But the present diagrams are not complete, and that is fairly obvious. The phase formed and that remains, apparently stable, actually depends on the history of the material.

Leaving aside how atoms can be physically removed from their locations in a structure . . .

The process is well-understood. They are not “removed. ” Rather, lattice imperfections can propagate when the material is at a sufficient temperature (generally well below the melting point, I think.)

“What change would be expected to result from removal of atoms located where the Au atoms are located in the Cu3Au structure (Fig.1) as F-O propose”?

The assumption here is that Fukai et al propose that. It is not the sequence that metallurgists have been pointing to. This is my explanation of it. Loading Pd with H, a pressure is created on the Pd atoms, weakening the bonds. If a Pd atom can move to an adjacent vacant position, it will relieve some of the pressure. That is a form of annealing. Annealing in general removes stress. So the atoms are not “removed.” Rather, they move, to an adjacent site. When such an atom moves, stress may be relieved. If the new structure that forms this way is more stable — which must consider how much hydrogen is in the metal –, that structure may grow, as a crystal, within the older structure, until the entire structure is changed.

The Cu3Au structure would be the δ phase. There is a phase before δ, the gamma phase. The gamma phase is proposed by Staker (Slide 65 et seq) as the stable phase at room temperature and a loading of 0.85 to 1.15 atom ratio H/Pd. The gamma phase would start to form as a mixture with beta phase (which has the Pd in normal FCC lattice) at about loading 0.8.

What happens to PdD at loading 80%? That is the level at which LENR effects start being reported. Just a coincidence? Perhaps. The problem is that these phases do not normally form with existing Pd lattice, but they may form on the surface or where the material has been stressed, and particularly if stress is combined with loading. The gamma phase may form at the surface, relatively easily, when hydrogen fugacity is high, and repeated loading and deloading may increase the gamma phase material, because, once formed, gamma phase is claimed to be metastable, it will normally remain even if deloaded.

The delta phase starts, according to the Staker diagram, at about about loading 1.15. It is possible that delta phase material can be created with high pressure (Fukai saw both gamma and delta phase, as I read him). We do not know if any of these phases affect the level of nuclear activity.

What this author has described is delta phase. That is, Pd3Vac. But Pd3Vac is not stable against annealing. It anneals back to ordinary FCC palladium, without vacancies, probably at 890 C., the normal palladium annealing temperature. The SAV phases are metastable if the hydrogen is removed.

Fukai et al.(4) used X-ray diffraction to identify the structure formed when pressure was applied. The patterns were complex, contained lines from several known materials present in the high pressure cell and had no clear relationship to a characteristic crystal structure. What appeared to be arbitrary assumptions were made about the effect of the proposed vacancies on the patterns, with many of the lines being assumed to result from vacancy ordering.

I am not going to second-guess the metallurgists who did this work, they are expert with XRD, as far as I can tell. And Fukai literally wrote the book on the Metal-Hydrogen System. As part of this work, I found an inexpensive copy of Lewis, The Palladium/Hydrogen System (1966), and I’ve read dozens of papers on this specific topic, and hundreds of abstracts. “Appeared to be arbitrary” is subjective. Rather, they took experimental data and looked for ordinary explanations. Nothing they came up with was actually extraordinary, it does not overturn prior knowledge, only some assumptions. And someone who does not like their own assumptions being challenged will call the ideas of others “arbitrary assumptions.” The author here does not actually present evidence for the idea of a simple, no-vacancy structure.

Before that, this author wrote:

Given that Nature favors the most compact structure when high pressure is applied, why would a less compact structure containing extra volume called vacancies be [p 3] retained when a slight shift in atom locations would result in the expected compact structure?

Sure. Nature will favor the most compact structure, but sometimes the kinetics does not allow that. What is completely missed here is that the new structure is not just palladium. It is loaded heavily with hydrogen. Collapsing to a bcc structure, as the author proposes, simply does not happen with palladium. The XRD data does not support this at all. Remember, this is a standard error of this author: the idea that pressure causes a collapse to a more compact structure. The author is really trying hard to avoid what is now well-established in metallurgy, the existence of ordered-vacancy materials. Why? Speculating on that is beyond the scope of this commentary.

There is no “extra volume.” It is filled to the gills with hydrogen!

However, when the material is quenched, returned to STP, the material remains in the new phase, and if the material is heated to, say, 400 C., the material loses its hydrogen but not the new phase. It has a higher lattice constant than ordinary Pd. It would be metastable, as shown by it annealing if the temperature is raised.

Removal of the atoms from the eight positions identified by the large golden spheres in [a diagram where they represent Cu in the Cu3Au alloy] would create one vacancy/unit cell because each of these atoms is shared with eight other unit cells. Each unit cell of the fcc structure contains a total of four atoms of Pd. Consequently, removal of these atoms from the unit cell of fcc would result in a value for Xv of 1/4 = 0.25. Fukai and Okuma calculate a value of 0.18.

Pd3VH4 is the proposed δ phase. The γ phase is suggested by Staker as Pd7VH6-8, not Pd3VH4.  That indicate a vacancy rate of 0.14. So 0.18 could seem to be from a mixture of  γ and δ.

On cooling, the original report shows the formation of two phases, during annealing, settling on one phase, then again as the temperature was lowered from 700 C. Both have a smaller lattice parameter than before annealing. This is direct evidence of the existence of two phases beyond β.

The calculation of 0.18 is from Fukai (1994). For the method, Fukai and Okuma refer to Simmons (1960)

The measurement and calculation of 0.18 is from material that was annealed at 700 C, then quenched, then then heated at 350 C to remove the hydrogen, then subjected to density and XRD measurements of lattice parameter. From those the vacancy level is calculated.

By using Equation (1), the authors are not actually calculating the fraction of
vacancies as they claim. Instead, the equation would provide the fractional difference
between the measured physical volume change, using the density, and the fraction change
in volume of a perfect unit cell based on the fcc structure. That expectation would result
only when the volume of the unit cell is correctly calculated using a3 rather [than] 3a as used
in their equation.

The author is ignoring the Simmons method, which does not use the cell volume, though it is possible that volume is mentioned in the full Simmons paper (which I don’t yet have). The Fukai formula (after Simmons) does not use “3a,” but rather 3(a – a0), in the equation

xv = -∆ρρ0 – 3∆a/a0

which is from Simmons. However, if we expand the terms and assume a cubic lattice, the volume of a cell would be, then, a3, . Simmons explicitly claims:

This result [the formula] is independent of the detailed nature of the defects, for example, the lattice relaxation or degree of association. The nature of the defects is considered and it is concluded that they are predominantly lattice vacancies.

In addition, the claimed atom fraction of vacancies cannot be calculated
this way because the volume of a typical vacancy is not known.

The author is right. It cannot be calculated using the volume of a vacancy. However, that is not what Fukai, after Simmons did.

The author argues with the Simmons method, without referencing Simmons, I would guess not even looking at Simmons. Google Scholar on the two authors shows that Simmons (1960) was cited in 635 papers. Indications are, then, that this is a widely-accepted result, and an error of this magnitude (a multiplier in place of a power) is unlikely.

The error, if an error, converts the Simmons method, which was experimentally based,  into utter nonsense, which they then would have repeated, exactly, not only with another paper on with silver in 1960, but also with gold in 1962. And again, in 1963 for copper, and 1964 for an aluminum-silver alloy. I am not here attempting to justify the Simmons method. There appear to be papers with theoretical support.

To be sure, those prior findings were all at quite low vacancy concentrations. Fukai uses the method at far higher apparent concentrations (but this author claims that they could, in fact, still be very low, but if they are very low, then why does the Simmons relationship not hold?)

Fukai used this measurement to calculate the atom fraction of vacancy clusters, Xcl, by using the equation Xcl/(1+Xcl)=∆V/V. Solving for Xcl, the equation becomes Xcl= (∆V/V)/(1-∆V/V), where V is the initial sample volume and ∆V is the change in volume [p 7] resulting from hydrogen removal. The form of this equation suggests several unjustified conclusions. First, at small values of ∆V, Xcl is nearly equal to ∆V but as the value for ∆V increases, the volume fraction of vacancy clusters becomes greater than ∆V. Why such an effect should occur is not obvious. Second, this equation assumes that loss of H causes the proposed vacancies to move to other locations, i.e. to cluster, which results in the reduction of cell volume. In order for vacancies to move, their sites would have to again be filled by Pd atoms. As a consequence, the original fcc structure would reform and would be expected to produce the fcc X-ray lines. These lines are not detected. Instead, the Pd atoms in the proposed bcc structure can be assumed to simply come closer together as H atoms are removed from the sites between the Pd atoms, as is known to happen when H is removed from the fcc structure.

The actual equation used by Fukai for xv, the vacancy concentration, is:

xv = -∆ρρ0 – 3∆a/a0

where ∆ρ and a are the changes of the density ρ and the lattice parameter a, respectively, from their original values, ρ0 and a0: ∆ρ = ρ – ρ0 and ∆a = a – a0.

What this author reports is the author’s own reinterpretation, assuming that a gross error has been correctly identified in old and widely-accepted work.

Rather than being identified as Pd vacancies, this extra volume can be better attributed to a combination of error and physical voids resulting from the removal of hydrogen. In fact, such voids are frequently seen as pits in the surface or as excess volume within the physical structure that form without application of high pressure.

Pits are indeed seen in when palladium has been deloaded and annealed. This is understood as resulting from the migration of Pd atoms to fill vacancies. The formation of pits will improve the kinetics of annealing. Pits are not present in SAV material merely upon deloading.

Without vacancies, there would be no reason for the BCC structure that the author proposes, that structure would readily convert to FCC at annealing temperatures, even if it were metastable. It would not form pits. Pits will form from the local amalgamation of vacancies.

The results reported by Fukai can be interpreted several different ways without vacancy formation being involved. To start this reinterpretation, several facts need to be acknowledged.

First, atoms cannot simply be removed from a structure as if they had been dematerialized, as Fukai et al. describe. The atoms must go elsewhere and form additional unit cells.

It is a fact that atoms are not dematerialized. It is not a fact that Fukai et al described the formation of vacancies that way.

Vacancies may propagate in a crystal until they find an edge. The rate of propagation varies with temperature, because movement of a Pd atom out of a lattice location requires energy, which is supplied by temperature. (With a high enough temperature, the entire lattice melts). When the temperature is adequate, vacancies will anneal out, and in the reverse direction, when the formation of a vacancy relieves stress, the atom will move to an adjacent cell, and this pushes an atom from that cell to another cell. At low temperatures, the kinetics does not allow this. The suggestion from Fukai is that the vacancy phases are the Gibbs Free Energy preferred structures for loaded PdH. The β phase (above 85% or so, per Staker) is metastable, due to stress created by high loading. That is why it disappears when PdH is taken up to 700 C.

What would this stress do to an alleged pure-BCC phase if loaded? What PdH does,with a high enough temperature, it becomes obvious, is to anneal into a more stable form. It is not pressure that causes that, nor is there any evidence that I have seen that pressure moves Pd into a BCC structure.  5 GPa pressure caused no changes in lattice parameter, other than the normal expansion from loading. For pure PdD, FCC is the stable phase. It takes high hydrogen loading to change that. High vacancy Pd, the Fukai material after deloading, anneals to FCC palladium. Not BCC.

. . . the Cu3Au structure identified by Fukai et al. is not a description of the final crystal form because once Pd leaves the lattice sites, this crystal form would no longer exist. Apparently, this crystal form was used by Fukai et al. only to identify which lattice sites are vacated, not as the final structure resulting from applied pressure. The true structure produced by applied pressure needs to be identified.

The Cu3Au structure is proposed and used in explanations as a way to describe the vacancy phases. It is merely a way of describing the delta phase (not the gamma phase). This author does not consider the clear finding from Fukai of the existence of two phases beyond β, that form and merge at 700 C., that then crystallize into two regions, at loading in the range of 1.15 – 1.23 (per Staker’s phase diagram, Slide 70). (See Fukai’s figure 2, shown above).

There is no new phase reported “produced by applied pressure.” Pressure created no phase shift, only normal lattice expansion as H2 loaded. The phase shift occurred after dwell at 700 C. Previously, high pressure had been applied to PdH/D under observation with XRD, and it caused no phase shift. The phase shift is clearly a result of temperature; below annealing temperature, the kinetics do not allow spontaneous annealing for PdH.

Tools for visualization

Cubic close packing, also called FCC.

Cu3Au, proposed for SAV metal hydrides. (with gold being “vacant,” vacancy referring to metal atoms, not to an empty site, rather, the gold corner sites would be filled with hydrogen and also the tetrahedral sites.)

Leave a Reply