UPDATE 12/4: Since this was posted much has happened. Eli has added a whole bunch of stuff to the bottom of this post in roughly chronological order, including a response from Black Light Power, and some discussion of it. A fair amount of it has been taken from the
hydrino forum discussion entitled
five years, but since clicking through is not a habit many have, it will be reproduced here.
12/7: This also includes a comment by Scarmani speculating that monovalent aluminum plays a role who points out that there is enough Al in Raney Ni to account for the energy generated. Some additional comments about papers that show Raney-Ni is very active and explodes when heated under various conditions can be found at the bottom, with references for the interested
*******************************
Black Light Power is an interesting construct. Operating at the fringe of quantum mechanics, Black Light's entrepreneurial leader, Randell Mills has successfully raised a ton of money (~$60 million, not in the class of your average hedge fund, but better than a typical NIH grant). BLP is one of the few things that
Eli and Lumo agree about and comments about
its business model have appeared before in this humble blog. It starts but does not end with a refusal to believe in quantum mechanics, followed by a blizzard of hand waving and algebra from which, as Venus from the quantum foam, arises the hydrino, a hydrogen orbit with fractional quantum numbers, very strongly bound, much more so than the ground state of hydrogen (E ~ -R/n
2). One of the occupants of the Lumidek,
Nigel, summed it up succinctly
This is sustained by the facts, which contradict Mills, who is basically doing for QED what Ptolemy did for ancient cosmology, against Aristarchus’ solar system.
So what now young hares? After claiming that hydrinos are the answer to the energy problem, discovering antigravity, and a few other things, BLP has
brought forth a new "energy source"
Our experiments on the BlackLight technology have demonstrated that within the range of measurement errors the significant energy generated, which is 100 times the energy that could be attributed to measurement error, cannot be explained by other known sources like combustion or nuclear energy,” says Dr. Jansson, professor of engineering at Rowan University. “The ability to generate such tremendous power in this controlled process demonstrates that the claim by BlackLight Power that it is able to demonstrate repeatable heat experiments based on their technology can be replicated by independent scientists.”
which has been
"validated" by Rowan University.
“Our experiments on the BlackLight technology have demonstrated that within the range of measurement errors the significant energy generated, which is 100 times the energy that could be attributed to measurement error, cannot be explained by other known sources like combustion or nuclear energy,” says Dr. Jansson, professor of engineering at Rowan University. “The ability to generate such tremendous power in this controlled process demonstrates that the claim by BlackLight Power that it is able to demonstrate repeatable heat experiments based on their technology can be replicated by independent scientists.”
This flight of fancy has made its way into the mainstream, the
New York Times,
CNN, the
Wall Street Journal,
NPR and a
ton of blogs and other media. A lot of these sources play some variation of "no one else has been able to explain this". Recently a small electrical coop in NM
contracted for one of the magic reactors
Cranbury, NJ (December 11, 2008)—BlackLight Power (BLP) Inc. today announced its first commercial license agreement with Estacado Energy Services, Inc. in New Mexico, a wholly-owned subsidiary of Roosevelt County Electric Cooperative, (Estacado). In a non-exclusive agreement, BLP has licensed Estacado to use the BlackLight Process and certain BLP energy technology for the production of thermal or electric power. Estacado may produce gross thermal power up to a maximum continuous capacity of 250 MW or convert this thermal power to corresponding electricity.
That's why you hired Eli and his merry band. Eli didn't put the last brick onto this sarcophagus but he did start the thing rolling down hill and since Rabett Run is read by many (maybe two or three) involved in energy issues, it is a decent place to put the answer and let's face it the 116,985th most popular blog could use some links.
How to describe the "Black Light Process" You can find some details on the
Black Light Power site and
on a couple of
discussion groups, and yes,
there is a video and a
bunch of animations. The reactor is a tube that can be heated. The cell contains a mixture of Raney nickel and sodium hydride and sodium hydroxide processed who knows how and certainly containing some majic ingridients. As we shall see you don't need anything but the Raney nickel.
Raney nickel is an old, well know catalyst for hydrogenation. It is produced
. . . .when a block of nickel-aluminium alloy is treated with concentrated NaOH. This treatment, called "activation", dissolves most of the aluminium out of the alloy. The porous structure left behind has a large surface area, which gives high catalytic activity. A typical catalyst is around 85-percent nickel by mass, corresponding to about two atoms of nickel for every atom of aluminium. The aluminium which remains helps to preserve the pore structure of the overall catalyst.
This is usually done in the presence of hydrogen gas.
BLP claims that in their reactor shown above, an initial heating decomposes sodium hydride yielding hydrogen and sodium atoms. The sodium atoms are said to (look, it's BLP, not Eli) catalyze a transition in the hydrogen atoms from the ground state to one of the lower lying "hydrinos" states. The claimed yield from 1.5 kg of Raney nickel is about a megajoule. The Rowan experiments may or may not have recycled the evolved hydrogen. BLP and their friends at Rowan University say there is no possible chemical explanation for this.
Well, there are some problems of course, if there is a net gain of energy, the hydrinos can't transition back to the hydrogen ground state cause if they did you would have to put in energy to bring them from the deep hydrino basement back up to the hydrogen ground state, so they are said to whisk off as dihydrinos, but, as Fermi said about space aliens, where are they cause they would have been seen everywhere? But the real question is what is happening.
We don't know how BLP is preparing their majic mix except that it has a lot of Raney nickel and sodium hydride, NaH. There is also some aluminum hanging about. Fortunately we do have the chemical literature. Raney nickel is used for hydrogination (adding hydrogen to molecules). Hydrogen atoms chemisorbed (that means they actually form a bond rather than just sitting there) on the Ni surface are easily attached to other molecules. Raney nickel used for such experiments is known to explode or catch on fire later and must be disposed of with care. An old paper, "The Role of Hydrogen in Raney Nickel Catalysts by Hilton Smith, Andrew Chadwell and S.S. Kirslis J. Phys. Chem. 1955 59 820-22 tells the observant pretty much everything
The hydrogen content of Raney nickel has been found by direct analysis to vary between one-half and one atom of hydrogen per atom of nickel. The activity of a sample of this catalyst has been shown to be proportional to its hydrogen content. The surface area decreases linearly with loss of hydrogen until about 70% is removed; then it decreases more rapidly. If the hydrogen is released rapidly, heat is evolved which results in an explosion of the catalyst. The best explanation for these phenomena appears to be based on the assumption that the hydrogen is in the form of atoms attached to the nickel in a metastable state. If desorption of the hydrogen is rapid, the highly exothermic recombination of the hydrogen atoms becomes explosive.
The reaction of the NaOH with the metals to form the catalyst produces a lot of hydrogen which can be absorbed on the surface. The JPC paper found that the maximum amount of H2 that could be desorbed was between ~100 and ~150 ml STP per gram. The aluminum oxide content was ~20%. A lot of surface science experiments tell us that the hydrogen chemisorbed on the surface is in the form of H atoms and can be driven off by heating. The mechanism involves recombination of two H atoms on the surface of the Raney nickel to form H2. The H2 is not bound strongly to the surface and desorbs.
At this point, we KNOW that the sodium hydride has nothing to do with the case, that the Raney nickel alone can evolve a large amount of heat when the H atoms are driven off and that the hydrogen atoms are loosely bound to the nickel (metastable enough that we can drive them of by mild heating)
Grabbing one of the discarded Christmas card envelops that the gentle readers were so kind as to send and turning it to the back, the chemical reaction would be
2 Ni-H(s) --> Ni-Ni(s) + H2(g)
Scaling the 100 to 150 ml/g H2 for Raney nickel up to the 1500 g Rowan/BLP reactor we get (100-150 ml/g)*(1500g) = 150 - 225 liters of H2 or .
(150-225 liter)/(22.4 l/mole) = 6.7 - 10 moles (a serious amount of Hydrogen)
To do this right we would have to know the heat of formation of a mole of H atoms on the Raney nickel and a whole lot of details, Raney nickel is a very nano material, where structure is everything. However, for the back of our envelope we can use bond strengths from the table in the Chem Rubber Bible (aka Chemical Rubber Company Handbook of Chemistry and Physics, 9-64 (2008)).
Ni-Ni: 204 kJ/mol
Ni-H: 240 kJ/mol
H-H: 436 kJ/mol
So we break two Ni-H bonds, that costs us 480 kJ/mol and we make one Ni-Ni bond getting back 204 kJ/mol and one H-H bond, getting back 436 kJ/mol
Net heat of reaction is estimated by adding the energies for the bonds broken and subtracting the energies for the bonds formed. (A negative number means the reaction will be exothermic or give off energy in the form of heat)
Net heat of reaction per mole of H2 generated= 2*240 kJ/mol - 436 kJ/mol - 204 kJ/mol = -160 kJ/mol (an exothermic reaction)
Net heat evolved from 1.5 kg of Raney nickel = (6.7-10.0 mol) x -160 kJ/mol = - 1072 kJ to -1600 kJ = -1.1 to -1.6 MJ!!
If the lab bunnies follow the bouncing ball they will see that the BLP and Rowan folk say that reaction with 1.5 kg of their mix, which is mostly Raney nickel releases 1 MJ of energy. This is in the Table on the bottom of page 7 of the Rowan report. They report heats of 1.015 MJ (average of two runs) for the runs with 1.5 kg. Page 4 gives the results with 30 g of Raney-Ni. In these runs, they got 20 kJ. Lets check the reproducibility: (1500 g/ 30 g) * 20 kJ = 1000 kJ = 1 MJ. Pretty good linearity. For standard chemistry with a reaction going to completion, this makes sense.
UPDATES: From Randell Mills via a yahoo discussion group,
------------
The heat of formation of nickel hydride is negative and small (-2.1 kcal mole/H2). It is referenced in my paper R. L. Mills, G. Zhao, K. Akhar, R. Chang, J. He, Y. Lu, W. Good, G. Chu, "Commercializable Power Source from Forming New States of Hydrogen", in press (Reference 78 at Eq. (45)).
[78. B. Baranowski, S. M. Filipek, "45 years of nickel hydride‹history and perspectives", Journal of Alloys and Compounds, 404-406, (2005), pp. 2-6.]
http://www.blacklightpower.com/papers/WFC112108WebS.pdf
Thus, nickel hydride decomposition is endothermic, not exothermic. Furthermore, his main mistake is that he has incorrectly calculated the heat of formation of nickel hydride using bond energies. Specifically, the bond energies regard gaseous atoms,and the energy to vaporize Ni metal to atomic nickel is +429.7 kJ/mole Ni (CRC) or +4.45 eV/Ni atom.
D. R. Lide, CRC Handbook of Chemistry and Physics, 86th Edition, CRC Press, Taylor & Francis, Boca Raton, (2005-6), p. 5-16.
Also, nickel hydride is H dissolved in a Ni metal lattice. It does not comprise covalent Ni-H bonds.
Eli's response here (see the comments for complete details):
Folks, BoE means you grab stuff. Using the chemical reaction
2Ni-H --> Ni-Ni + H2
is a cartoon of what is happening, and using the diatomic bond strengths is a really rough estimate, but it is rough on both sides of the arrow, AND is very much in keeping with how bond strengths are used to estimate heats of reaction. (see your tattered GChem book) . . . . . .
Part II: As Eli understands it, BLP assumes the Raney nickel is a passive support. Raney-Ni is known to be a very active catalyst for hydrogenation and a pyrophoric material, indeed a dangerous one to be handled with care that has on occasion exploded. Scarmani and Oakthicket on the hydrino forum and Eli have pointed out that the metals themselves in Raney-Ni, principally Al and Ni, can participate in energetic chemistry, as could H atom recombination. And, let us be frank about it, and Na metal or NaH could also take part. We also know from the development of hydrogen storage devices that the addition of small amounts of other metals can significantly change the storage capacity of Ni for atomized hydrogen. It is very easy to see that there are enough possible chemical reactions around in 1.5 kg of doped Raney-Ni to provide 1 MJ of exothermicity. . . .
Scarmani at the
hydrino forum;
[[ edit: in a response to Rabett's blog entry posted to SocietyforClassicalPhysics mailing list, Mills has provided a reference for this: doi:10.1016/0021-9517(81)90102-0
The paper states "A commercial Raney nickel evacuated at 25 °C evolved about 23 cm3 (STP)/g H2, as measured by volumetric and temperature-programmed-desorption (TPD) methods... Without H2O addition hydrogen desorption was only slightly exothermic, but when H2O was added, it was highly exothermic. The latter process appears to be the reaction of H2O plus zero-valent aluminum; this reaction did not proceed to a large extent in the usual desorption tests. Thus, most of the hydrogen was present as chemisorbed and interstitial hydrogen."/edit ]]
This seemingly contradicts the description in the old 1955 paper "The Role of Hydrogen in Raney Nickel Catalyst" (J. Phys. Chem., 1955, 59 (9), doi: 10.1021/j150531a005) which states "The hydrogen atoms are slowly desorbed on standing... The process may be greatly accelerated by increasing the temperature. In fact, the desorption may be so rapid that the highly exothermic recombination of the hydrogen atoms becomes explosive."
The difference may be that with sufficiently high concentrations of atomic H, recombination of atomic H can proceed rapidly; under high vacuum or low surface loading of hydrogen, recombination of atomic H does not occur at a significant rate because the concentrations of atomic H do not reach sufficient levels. http://hal.archives-ouvertes.fr/docs/00 ... 41C707.pdf . . . .
Recombination of atomic H to molecular H on the surface of solids can occur more quickly, due to other mechanisms (http://en.wikipedia.org/wiki/Reactions_on_surfaces).
A friend I'd been trying to persuade to look closely into Blacklight mentioned in passing that rapid recombination of built-up atomic hydrogen to molecular hydrogen might be behind the observation of excess heat generally in systems containing hydrogen where energy was input (e.g. cold fusion, Mills electrolysis and Blacklight gas plasma cells). He pointed out the atomic hydrogen torch (http://en.wikipedia.org/wiki/Atomic_hydrogen_welding). At the time I dismissed this as a possible explanation, but it is starting to seem more plausible.
Oakthicket makes two points:
If you want to show that 'hydrinos' result in large amounts of excess heat, why would you not run a control sample of undoctored Raney-nickel catalyst under identical conditions and measure excess energy? You then compare the two sets of results. Using control samples is a standard scientific approach. That wasn't done by Rowan University. They simply took doctored Raney-nickel, heated this highly-exothermic substance, measured the energy output and declared that most of the energy generated cannot be explained by normal chemistry.
Scarmani
continues
I completely agree. In Mills' main paper (http://www.blacklightpower.com/papers/WFC112108WebS.pdf, pp 21-24) there is a control - but the control is "3 wt% Al(OH)3 doped R-Ni/Al alloy"
In other words, the control Mills selects is not actual Raney Nickel - it is the non-porous, non-hydrogen containing starting alloy powder of 50/50 Ni/Al, laced with a small amount of Bayerite. Why did Mills chose a solid metal alloy powder as the control, rather than use the obvious, meaningful, directly comparable control (identical, porous, hydrogen-containing R-Ni, sans vital catalyst)?
and brings another idea about
a conventional chemical reaction
In the R-Ni experiments, if you take the conventional view, the original driving source for the energy is not to do with NaH, but rather the oxidation of Aluminum metal. To illustrate how much energy is stored in the Aluminum, the amount of energy released in preparing R-Ni from the starting alloy is 300x the amount of energy released from the same weight of R-Ni in the Mills / Rowan experiments. (Another illustration is http://www.fuelcelltoday.com/media/pdf/ ... 0Cells.pdf)
The 15 wt% or so residual Aluminum metal present in the R-Ni is used to drive the formation of hydrogen from NaOH (or if hydride formed, will likely give up hydrogen very easily with heat in the presence of R-Ni due to lower activation energy, doi: 10.1016/j.ijhydene.2007.06.030). Because of the surface doping, most of the resulting hydrogen is directly absorbed in atomic form on the R-Ni surface. At a threshold temperature, activation energy is reached and there is a runaway recombination / desorption of the hydrogen from the R-Ni. The recombination to molecular hydrogen is highly exothermic; this very exothermic runaway recombination does not occur in earlier thermal desorption studies of R-Ni because there is not a great enough concentration of H atoms in those studies (under full vacuum rather than in a sealed pressurized container, surface loading of H lower than critical threshold since the R-Ni is not doped in a way that can directly generate hydrogen on the surface of the Nickel).
12/7 The paper by Nicolau and Anderson, J. Catal. 68 (1981) 339 cited in the comments by one of BLPs supporters reports that a small amount of water vapor causes Raney-Ni to release more than double the amount of hydrogen and significant energy on heating (they had a small sample, about a half a gram, which could not heat the mount it was on much, not 1.5 kg). They ascribe this to the reaction of water with zero-valent aluminum.
An interesting paper from Hotta, et al, in Kagaku to Kogyo (Osaka) 41 (1967) 269 reports that Raney-Ni heated with small amounts of organics, such as alcohols exploded on heating to about 200 oC. Another paper by Mikhailenko, et al, J. Catal 141 (1993) 688 notes that
However, usage of Raney catalysts is often limited by their propensity for self-heating and self ignition in air. The nature of these phenomina is influenced by the presence of a large amount of hydrogen in active form (5, 8-10)
We do not know exactly what was in the 1.5 kg furnished by BLP to Rowan, but we do know that Raney nickel is a very energetic material and not a passive support. Eli, Scarmani and others have pointed to several possibilities for reactions that would furnish enough energy to produce 1 MJ from a 1.5 kg sample of Raney - Ni, and we have pointed to several reports of occasions on which Raney-nickel was reported to explode. It is not a passive support. There are several chemical possibilities to explain the observed heating.
Comments?