Notes Added in Proofs

INTERVALENCE CHARGE TRANSFER AND ELECTRON EXCHANGE STUDIES OF DINUCLEAR RUTHENIUM COMPLEXES 

The Ottawa-Carleton Chemistry Institute, Carleton University, 1125 Colonel By Dr., Ottawa, Ontario, K1S 5B6, Canada 

Studies of Exchange Coupled Dinuclear Ruthenium Complexes 

Dinuclear ruthenium complexes form the largest group by far of any mixed-valence system and are the exclusive subject of this chapter. Ruthenium is the transition metal of choice to study electron transfer or exchange because it is relatively inexpensive and forms stable Ru(III) and Ru(II) coordination complexes. In addition, the synthetic coordination chemistry of ruthenium is well developed (1). 

The intellectual push to study mixed-valence complexes was provided by the publication in 1967 of two review articles, by Allen and Hush (2) and Robin and Day (3), on the physical properties of mixed-valence systems. These were followed by Hush s publication (4) of his theoretical model of intervalence transitions, which provided a link between the properties of mixed-valence complexes in solution and the Marcus theory of intermolecular electron transfer (5, 6). The review by Robin and Day classified mixed-valence complexes into three types class I, 

Since the completion of this review, a number of papers on the subject have been published. 

Lightner and co-workers studied the behavior of various simple alkylpyrroles by irradiation in methanolic solution in the presence of a photosensitizer and a stream of oxygen. 

From 2-methylpyrrole, the hydroxylactam (19, R = R, = R2 = R3 = H R4 = Me) was obtained along with corresponding methyl ether. Similarly, 2,4-dimethylpyrrole gave the hydroxylactam and the methyl ether.121 

Quite surprisingly, in addition to forseeable products, photo-oxidation of 2,5-dimethyl122 and 2,5-dimethyl-3,4-diethyl-pyrrole123 afforded lactams like (19) with R4 = CH2OCH3, the formation of which is not easily rationalized. It was assumed that all these products can be derived from an endo-peroxide (20). 

Particularly interesting was the case of 3-methylpyrrole in addition to the expected hydroxy- and methoxylactams, compounds 97 and 98 were found.124 This fact was presented as the first evidence for the formation of a cyclic dioxetane intermediate (99) arising from 1,2-cycloaddition of singlet oxygen to the enamine-like 2,3 bond of the pyrrole. 

Class II Mixed-Valence Complexes Electron Exchange 

7-Methylenedioxy-3-quinazolinio)amidates (125) are zwitterions formed by the cyclization of the iV -acylhydrazones of o-amidophenyl ketones (124) with thionyl or phosphoryl chloride followed by basification. They readily dimerize and form the adducts (127) by attack of nucleophiles at C-4. These adducts are in equilibrium with the zwitterions 125 and their dimers 126. The relative proportions of these quinazolines depend on the solvent, temperature, and the substituents R, R, R, and The 

Lempert-Streter. K. Lempert, P. Bruck, and G. Toth, Acta Chim. Acad. Sci. Hung. 94, 391 (1977). 

Academy of Science of ihe GDR Central Institute of Organic Chemistry, Berlin-Adlershof, German Democratic Republic 

The present review is concerned with the chemistry of oxaziridines (1), diaziridines (2), and diazirines (3), three classes of compounds discovered after 1950 and widely investigated since then. 

More recent patents continue to report anti-inflammatory oxicams. Thus, Ferrini et a/.164 have found some N-(benzopyranyl)-4-hydroxy-2-alkyl-2H-l,2-benzothiazine-3-carboxamide 1,1-dioxides to be anti-inflammatory in animals. 

Ferrini, G. Haas, K. A. Jaeggi, and A. Rossi, European Patent 3-360 (1979). 

After completion of this article, a 21-page review on all possible benzothia-zinone dioxides, including 1,2- and 2,1-, as well as 1,3-, 2,3-, 1,4-, and 2,4-benzothiazinones, appeared.165 By restricting their subject to 3-one and 4-one derivatives, these authors found 71 literature references to the 1,2-and 2,1-benzothiazinone dioxides. The present chapter discusses these references and a significantly larger number of patents and scientific articles dealing with all 2,1- and 1,2-benzothiazines. 

The authors thank Dr. Beryl Dominy and Mr. John Hare of the Pfizer Technical Information Department for their help in computer-assisted searches of the scientific and patent literature and Mrs. Lynn Londregan, Mrs. Theresa D Amico, and Mrs. Linda Neilan for typing this manuscript. 

Isatoic Anhydrides and Their Uses in Heterocyclic Synthesis 

4-one derivatives, these authors found 71 literature references to the 1,2-and 2,1-benzothiazinone dioxides. The present chapter discusses these references and a significantly larger number of patents and scientific articles dealing with all 2,1- and 1,2-benzothiazines. 

After submission of the original manuscript, additional relevant papers were published. 

Buono and coworkers developed a platinum analog of their Pd-catalyzed cyclo-propanation method (Section 9.4.3). Their new catalyst, Pt(r] -acetato)- [(R)(Ph) P0]2H, was successfully applied to reactions of heteroatom-containing alkenes [57]. 

Chatani and coworkers published an efficient method for the Rh(I)-catalyzed anti-Markovnikov hydroamination of terminal alkynes using either primary or secondary amines [58]. This reactivity had been observed earlier in the course of their studies on hydrative alkyne dimerization (Equation 9.8). 

Two recent publications feature metal vinylidenes functioning as 1,3-dipole equivalents, as in the Cu-catalyzed Huisgen cyclization (Section 9.4.5). Fiirstner and coworkers described intramolecular Diels-Alder reactions of unactivated die-nynes catalyzed via a proposed [4+3]-diene/copper vinylidene cycloaddition [59]. 

Bertrand and coworkers invoke a retroiminoene-type mechanism for the formation of gold vinylidenes in an interesting crosscouphng of enamines and terminal alkynes to give allenes [60]. This reaction constitutes both a new reaction manifold for metal vinylidenes and a new retron for stereochemically defined allenes. 

Note Added in Proof. Since the completion of this manuscript a number of reports have appeared on compounds relevant to this topic. These are listed here briefly according to main group element type which corresponds to the section heading numbers used in this review. 

Section II. Haupt and Florke have reported the crystal structure of [In Re2(CO)6(/i-I) - Re2(CO)6(ji-I)2 ], which contains a spiro /i4-In bound to four rhenium atoms. This complex was obtained as one product from the reaction between Inl3 and [Re2(CO)10] at 437 K. [H. -J. Haupt and U. Florke, Acta Crystallogr., Sect. C. C45,1718 (1989)]. 

Leigh and K. H. Whitmire, Acta Crystallogr., Sect. C. C46,732 (1990)], which contain tetra-hedrally coordinated tin and lead atoms. Crystal structure data has been presented for [Sn Fe2-(CO)8) Fe3(CO)n ], which contains a tetrahedral, spiro, i/4-tin atom bridging the Fe—Fe bond in the Fe2(CO)8 fragment and one Fe—Fe vector in the triangular Fe3(CO)n unit [S. G. Anema, 

Mackay and B. K. Nicholson, J. Organomet. Chem. 372, 25 (1989)]. Structural data on the mixed metal complexes [Sn Fe(CO)2(CsHs) 2 Fe(CO)4 ] and [Sn Fe(CO)2(C3H3) 2 Cr-(CO), ] reveal tin atoms that are in a trigonal planar coordination environment, although weak axial isocarbonyl type contacts are also present [P. B. Hitchcock, M. F. Lappert and M. J. McGeary, Organometallics 9, 884 (1990)]. 

Organo-Transition Metal Compounds Containing Perfluorinated Ligands 

Note added in proof. While this article was being typeset, our laboratory obtained evidence that the activities synthesizing and degrading fructose-2,6-bisphos-phate in spinach leaves reside on different proteins (Macdonald, F.D., Cseke, C., Chou, Q. and Buchanan, B.B. (1987) Proc. Natl. Acad. Sci. USA, in press). 

Department of Chemistry and Biochemistry, Texas Tech University, Lubbock, 

Photolithotrophic growth [6] with reduced inorganic compounds (e.g., H2 and. 

Space limitations preclude a detailed treatment of all known pathways of substrate oxidation and the reader is referred to recent reviews [4,6,19] for such information. Instead of striving for a comprehensive treatment, we have thus opted to discuss only a few such pathways, focusing on those pathways for which some 

NAD photoreduction in chromatophores isolated from several purple non-sulfur bacteria [41,48-50] and from the purple sulfur bacterium Chromatium vinosum [51] but did not inhibit ATP-driven NAD reduction in the dark. 

The coefficients 8.10 and 0.010 in the second equation are usually ascribed to the reactivity ratios rj and rj (Table 19). This catalyst produces poly-propene consisting mainly of syndiotactic stereoblocks, together with short disordered blocks resulting from head-to-head (hh) and tail-to-tail (tt) pro-pene enchainment and occasional isolated isotactic units, and if these features apply to copolymers prepared with vanadium catalysts, the reaction is in effect a terpolymerization. Locatelli et al. [322] derive the equation for monomer/polymer composition ratios 

The polymerization of propene by VCI4/AlEtj Cl/anisole at —78°C shows unusu2il kinetic features in that the polymerization rate increases with time and with monomer concentration to a maximum at [M]/[A1]/ [V] = 2 X 10 /30/1 [323]. Evidently, in this system, possibly as a result of the low temperature employed, the monomer appears to occupy sites so as to preclude propagation and the rate is said to follow the relationship [322] 

For the rate to pass through a maximum with increase in [M] it follows that Kfji [M] 1, in accord with published values of for propene, since strong coordination of monomer to the sites would give rise to independence of rate on [M]. A fall in rate, however, implies that monomer is competing with active site formation it is difficult to see the mechzmism by which this could occur as exclusion of organometal from the catalyst surface by monomer would appear to be unlikely. 

Observations on this system may not apply to other catalyst systems but the fact that abnormal addition of propene has been demonstrated in elastomeric E/P copolymers prepared under very different conditions makes it clear that, at the least, the values of and rj are approximations. It would be of interest to calculate (assuming, for example, that the ratio of the fractions of head-to-head and head-to-tail units equals the ratio of the corresponding rate coefficients) what differences from the true rj and T2 ratios would result from the occurrence of 5—10% of abnormal addition. 

Paper to 15th Ann. Tech. Conf. Soc. Plastics Eng., New York, 27 January 1959, S.P.E. Journal, 5 (1959) 373. 

The variation of the mobile phase in liquid-solid chromatography (LSC) for controlling the retention and separation of the sample has been 

Copyright f by Academic Pre s. Inc. All rights of rcpKKjuclion in any form reserved. 

Previous treatments of the displacement model have for the most part ignored the effects of interactions between solute and solvent models in 

Division of Chemistry and Chemicai Engineering, California Institute of Technoiogy, Pasadena, California 91125 

Nitrogenase Molybdenum-Iron Protein Structural Description of the Nitrogenase Proteins 

Nitrogenase Molybdenum-Iron Protein Structure Structures of the Metal Centers of Nitrogenase 

Substrate Binding to the FeMo-Cofactor Future Outlook 

The potentiation of active AFGP glycoproteins by the smaller AFGP glycopeptides, however, does not seem to be inconsistent with any of the mechanisms discussed in this article. 

CSIRO Wheat Research Unit, North Ryde, New South Wales, Australia 

Reversibility of Conformational Changes During and Following Desorption. 300 

Effects of High Concentration and Orientation on Reaction Rate. . . 312 

Bonding and Crosslinking Following Exposure of Reactive Groups. . 313 

Despite the thermochemical accessibility of 02 in the reaction of 02 with [Ru(bpy)3]J+, it has been argued recently that 02 is not a major product (337). Closer reading of Ref. 324 reveals that this paper raises considerable doubts about our present understanding of Cl2 . The authors of Ref. 324 have also reinvestigated the chemistry of Br2, and here too, they find major discrepancies with prior work (338). The chemistry of S2082 has been reviewed, with extended discussion of the mechanism of hydrolysis (339). Our selection of data for the HN02/N02 system has been confirmed (340). Two comprehensive compilations of rate constants have appeared one summarizes the reactions of ejq, OH, and H (341), while the other treats reactions of inorganic free radicals (342). 

Reactions of the Difiuoroamino Radical with Other Radicals 

At first sight the direct fluorination of ammonia might appear to be a useful alternative method for obtaining the trifluoride, but this reaction, if uncontrolled, yields mainly nitrogen and hydrogen fluoride with only about 6% of NF3 (258). The yield is greatly improved when the fluorine is diluted with N2 and this gas is allowed to mix with ammonia in a reactor packed with copper turnings (212). Under these conditions 

When ammonia is in excess, yields of NF3 are 10-25% and the product contains up to 10% of N2F4 and 5% of N2F2, together with a small amount of difluoroamine, HNF2. 

The discovery of new calixpyrrole chemistry continues apace. Since this chapter was written, Eichen and coworkers have reported the synthesis of a calix[6]pyrrole and Aida 

Jacoby, D. Roriani, C. Chiesi-Villa. A. Rizzoli, C. J. Chem. Soc., Chem. Commun. 1991, 220. 

After the preparation of this review further progress in the physical chemical analysis of electric field effects in biological macromolecules and in the membranes of isolated cells and organelleshas been documented. A few additional references are selected. 

Neumann, Chapter 4 in this volume, Modern Bioelectrochemistry (F. Gutmann and H. Keyzer, eds.) Plenum Press, New York (1985). 

Neumann, K. Tsuji, and D. Schallreuter, in Biological Structures and Coupled Flows (A. Oplatka and M. Balaban, eds.), pp. 135-138, Academic, New York (1983). 

Neumann, in Ions in Macromolecular and Biological Systems (D. H. Everett and B. Vincent, eds.), pp. 170-191, Scientechnica, Bristol (1978). 

Sheppard, and G. P. South, Dielectric Behaviour of Biological Molecules in Solution, Clarendon, Oxford (1978). 

The EPR spectra of the molecular O ion have now been obtained by reacting M02 + 02 (M = Na, K, Rb, Cs) in rare gas and nitrogen matrices [D. M. Lindsay, D. R. Herschbach, and A. L. Kwiram, J. Phys. Chem. 87,2113 (1983)]. Both the g and alkali hyperfine tensors suggest a dominantly ionic product M+0. The EPR data are interpreted in terms of a model (02 - 02) structure in which a relatively weak bond connects two equivalent 02 moieties. The EPR spectra do not allow one to distinguish between cis- and trans-Ol, but symmetry restrictions may preclude formation of the cis isomer. The g tensor of O differs substantially from those observed for the isoelectronic 25-electron radicals S03, C103, and P03 (see Section V,B). 

Gorlaeus Laboratoria Rijksuniversiteit Leiden Leiden, The Netherlands 

Catalysis by metals and alloys plays an important role in industry as well as in laboratory-scale preparations. Catalyzed reactions are usually run at lower temperatures than the noncatalyzed ones and they are also more 

Research on alloy catalysts started in the 1950s with attempts to investigate the role in catalysis of the electronic structure of metals. This research was initiated by several papers of Dowden which, measured by their response in the literature, rank among the most important papers ever written on catalysis. However, it appeared later (for reviews, see 1-5) that two basic ideas, on which the so-called electronic theory of catalysis was built up, were not correct. These ideas were as follows  

After this article was completed, a large number of papers containing vibrational spectroscopic data on carbonyl complexes were published. As far as can be judged, however, none appear to demand modification of conclusions which have been reached. Some of the more important are listed here. Further, one or two specific comments appear to be appropriate. 

Raman data are beginning to appear more frequently (see for example 1, 3, 4, 8, 9, 12, 14, 16, 17, 18, 26) thus making assignments much more convincing. The value of polarized infrared spectra has been long realized but only recently (7) has the first example of the polarized infrared spectrum of a carbonyl complex appeared Mn2(CO)io and Re2(CO)io were studied in a nematic liquid crystal. The value of Cotton-Kraihanzel force constants has been further discussed (10,15) and a reasonably complete assignment of the vibrational spectrum of the cation Re(CO)e has been published (1). A number of isotopically substituted species have been studied (5, 6, 16, 23, 25). As a result of one of these studies Bor (5) reassigned the spectrum of Mn2(CO)io. 

Cotton, F. A., Discussion of Cotton-Kraihanzel force field, Inorg. Chem. 7, 1683 (1968). 

Griffith, W. P., and Wickham, A. J.. Vibrational spectra of some binuclear carbonyl complexes.. 7. Chem. Spc., A p. 834 (1909). 

Darensbourg, D. J., and Brown, T. L.. Intensity of CO-stretching bands of substituted Mo(CO)b complexes, Inorg. Ghem. 7, 959 (1968). 

Foffani, A., Poletti, A., and Cataliotti, R., Effect of solvent on CO frequencies, Spectrochim. Acta A 24, 1437 (1968). 

The present chapter on the topics of gas-liquid separation was written in 1975. Since then many results and data have been published in the sense proposed by the author. This complementary information will be obtained in the following papers  

Laurent, A., Fonteix, C., and Charpentier, J. C., Simulation of a pilot scale, liquid motivated, venturi jet scrubber by a laboratory scale model. AIChE J. 26, 282 (1980). 

Charpentier, J. C., Gas-liquid reactors. Chemical Reaction Engineering Reviews (D. Luss and V. W. Weekman, eds.), Amer. Chem. Soc. Symp. Ser. 72, 223-261 (1978). 

Schugerl, K., Lucke, J., and Dels, U., Bubble column bioreactors. Advan. Biochem. Eng. 7, 1 (1977). 

Gas-Liquid-Solid Reactor Design. McGraw-Hill, New York (1979). 

Kanatzidis and S.-P. Huang, Coord. Chem. Rev., 1994, 130, 509, and references therein (complexes of polyselenides and polytellurides). 

Muller, C. Grebe, U. Muller and K. Dehnicke, Z. Anorg. Allg. Chem., 1993, 619, 416, and references therein (polyselenides). 

Kaiber, W. Petter and L. Hulliger, J. Solid State Chem., 1983, 46, 112. 

The Monsanto Group has recently reported enantiomeric excesses of 95-96% for the hydrogenation of a-acylaminoacrylic acids using a chiral diphosphine [l,2-di-(0-anisyl-phenylphosphino) ethane] as a ligand (49). The chiral phosphine was prepared by oxidative coupling of chiral o-anisylmethylphenylphosphine oxide (50), followed by deoxygenation with trichlorosilane and tri-n-butylamine in acetonitrile. 

The Paris Group has reported studies of various chiral diphosphines related to DIOP (51). Enantiomeric excesses as high as 90% were obtained. Structural analogs in which the acetonide ring was replaced by a carbon ring were shown to be capable of high asymmetric induction, as high as that obtained with DIOP. 

The asymmetric reduction of enamides to produce chiral amine derivatives has also been examined by the Paris Group (52). Subsequent unpublished studies (53) have shown that the degree of asymmetric synthesis is much higher in benzene than it is in ethanol for such systems up to 92% enantiomeric excess was achieved in one case. 

A stereocorrelation model for DIOP hydrogenations has been proposed (54). 

Further results on asymmetric hydrogenations of activated carbonyl compounds catalyzed by bis(dimethylglyoximato) cobalt (Il)-chiral amine complexes have been reported (55,56). Some chiral reductive dimerizations were observed (55). 

The use of BNPS-skatole (113) for the selective chemical cleavage of tryptophanyl peptide bonds has been further reported 459-462, 467). The selectivity of the reagent and the good yields of peptide bond fission clearly indicate that BNPS-skatole is of general utility for fragmentation of polypeptides and proteins, especially for sequence analysis. 

The reaction of hexahydropyrroloindole (HPI) (46) with thiols to give the corresponding 2-thioether-tryptophan compounds has been further investigated (464). Reaction of cysteine with HPI (1.2 equiv) in 25% tri-fluoroacetic acid produces quantitatively tryptathionine, an amino acid contained in the toxic peptides of Amanita phalloides (see Section VI.2.4.). Reduced ribonuclease, a protein containing 8 cysteine residues per molecule, was treated with HPI, and the modified protein purified by gel filtration. The completeness of the reaction was confirmed by hydrolysis with /7-toluenesulfonic acid (233) and analysis of the hydrolyzate. A value of 7.6 (theory 8) residues per mole of protein of oxindolylalanine, the product of hydrolysis of the tryptathionine residues (431) (see Section III.4.2.), was obtained. This new reaction of cysteine residues should be of value in peptide synthesis, providing a simple method for linking tryptophan and cysteine as a basic step in the chemical synthesis of the peptides of Amanita phalloides. 

The reaction of the diazonium salt Fast Red B (376) (see also Section III. 12.) with N-acetyl-L-tryptophan ethyl ester has been investigated (466). The tricyclic compound (221) was isolated, together minor amounts of (222) and (223). In contrast to the reactions with 

Acheson, R. M. An Introduction to the Chemistry of Heterocyclic Compounds. Interscience Publ., J. Wiley (1973). 

Afghan, B. K., and J. Israeli Reaction of Metal Nitrilotriacetates with ( —)-Tryptophan. Bull. Soc. Chim. Fr. 4, 1393-1394 (1969). 

TRANSITION METAL AND ORGANIC REDOX-ACTIVE MACROCYCLES DESIGNED TO ELECTROCHEMICALLY RECOGNIZE CHARGED AND NEUTRAL GUEST SPECIES 

Inorganic Chemistry Laboratory, University of Oxford, Oxford 0X1 3QR, England 

Group IA, IIA Metal Cations by Organic Redox-Active Macrocycles 

Redox-Active Macrocycles Containing Transition Metal Organometallic Redox Centers 

Coordinated Transition Metal Redox-Active Macrocycles 

Criticisms have recently been made of the quantitative treatment of micellar rate effects (section 4.3) and alternative formulations have been discussed. [Reddy, 

My own work in the area has depended largely on the efforts of my coworkers, most of whom are noted in the references. This work has been supported by the National Science Foundation (Chemical Dynamics Program and Latin-American Program) and by the U.S. Army Office of Research. 

1 Fendler, J.H. and Fendler, E.J. (1975) Catalysis in Micellar and Macromolecular Systems, Academic Press, New York. 

3 Menger, F.M. (1977) in E.E. van Tamelen (Ed.), Bioorganic Chemistry, III. Macro- and Multicomponent Systems, Academic Press, new York, p. 137. 

4 Brown, J.M. (1979) in D.H. Everett (Senior Reporter), Colloid Science, A Specialist Periodical Report, Vol. 3, Chemical Society, London, p. 253. 

---

Note added in proof The compounds described as derivatives of 2,4-dichlorothiophene are derivatives of 2,3-dichlorothiophene/ Additional derivatives of 2,3-dichIorothiophene have been pre-pared. "" ... 

Note, Added in Proof-. In their study of the autoxidation of 2-butyl-isoindoline, Kochi and Singleton showed that 2-butylisoindole is formed and is converted by further oxidation to 2-butylphthalimide and 2-butylphthalimidine. The rate of oxidation of 2-butylisoindoline to the isoindole was found to be markedly dependent on hydrogen donor ability of the solvent and was shoivn to involve a free radical chain process. Autoxidation of 2-butylisoindole also appears to be a radical process since it can initiate autoxidation of 2-butylisoindoline. 

Note added in proof. J. C. Strijland and A. J. Nanassy (Physica 24, 935 (1958)) have shown recently that there are no changes of oscillator strength in argon + mercury. Increases in spectral intensity are entirely accounted for by increases in mercury concentration. 

Note added in proof A comprehensive review on gas absorption from bubbles has recently been published by P. H. Calderbank, Trans. Inst. Chem. Engrs. (London), Chem. Eng. 209 (1967). 

Note added in proof The calculated spin alignment spectra for diffusive motion plotted in Fig. 12 are incorrect, in particular the oscillations in the central part, due to a sign-error in the computer program. 

Note added in proof Trinuclear arsenic compounds, As—As=AsL, related to the triazene... 

NOTE ADDED IN PROOF This manuscript had been submitted shortly after the presentation of the paper at the Fourth Advanced Seminar on Pale-odiet, 1994. Ongoing research, especially stable isotope analysis of single amino acids from inoculated and non-inoculated marten bones (same specimens as in this paper) further and strongly support our conclusion that bacterial modifica-tion causes substantial shifts in collagen stable isotope ratios (Balzer et fl/. 1997). 

Note Added in Proof. Since this review was completed, two significant experiments have been done whose results bear strongly on the question of the mechanism of molecule formation following nuclear activation. 

Note Added in Proof This is nicely illustrated by the landmark publication of Protasiewicz et al. [94] who describe derivative (122), the first conjugated polymers featuring phosphorus-phosphorus multiple bonds (Scheme 33). The di-phosphene-PPV (122) exhibits an extended 71-conjugated system as shown by the optical HOMO-LUMO gap that is close to that of related PPV [94]. 

Note added in proof Since this paper was written, it has been found that the two lines supposed to constitute group 12 are due to impurities. With this change in the data, the lower limit of the heat of dissociation, D, of the hydrogen molecule becomes 4.10 volts or 94,600 calories, the upper limit 4.50 volts or 103,800 calories and the probable value of D 4.34 volts or 100,100 calories. 

---