Catalytic reduction of hydrazine to ammonia with MoFe3S4-polycarboxylate clusters. Possible relevance regarding the function of the molybdenum-coordinated homocitrate in nitrogenase

Article abstract of DOI:10.1021/ic960098b

The catalytic function of the previously synthesized and characterized [(L)MoFe₃S₄Cl₃]^2-,3- clusters (L = tetrachlorocatecholate, citrate, citramalate, methyliminodiacetate, nitrilotriacetate, thiodiglycolate) and of the [MoFe₃S₄Cl₃(thiolactate)]₂ ^4- and [(MoFe₃S₄Cl₄)₂(μ-oxalate)] ^4- clusters in the reduction of N₂H₄ to NH₃ is reported. In the catalytic reduction, which is carried out at ambient temperature and pressure, cobaltocene and 2,6-lutidinium chloride are supplied externally as electron and proton sources, respectively. In experiments where the N₂H₄ to the [(L)MoFe₃S₄Cl₃]^n- catalyst ratio is 100:1, and over a period of 30 min, the reduction proceeds to 92% completion for L = citrate, 66% completion for L = citramalate, and 34% completion for L = tetrachlorocatecholate. The [Fe₄S₄Cl₄]^2- cluster is totally inactive and gives only background ammonia measurements. Inhibition studies with PEt₃ and CO as inhibitors show a dramatic decrease in the catalytic efficiency. These results are consistent with results obtained previously in our laboratory and strongly suggest that N₂H₄ activation and reduction occur at the Mo site of the [(L)MoFe₃S₄Cl₃]^{2-, 3-} clusters. A possible pathway for the N₂H₄ reduction on a single metal site (Mo) and a possible role for the carboxylate ligand are proposed. The possibility that the Mo-bound polycarboxylate ligand acts as a proton delivery shuttle during hydrazine reduction is considered.

Full text of DOI:10.1021/ic960098b

4038
Inorg. Chem. 1996, 35, 4038-4046
Catalytic Reduction of Hydrazine to Ammonia with MoFe3S4-Polycarboxylate Clusters.
Possible Relevance Regarding the Function of the Molybdenum-Coordinated Homocitrate in
Nitrogenase
Konstantinos D. Demadis,† Steven M. Malinak, and Dimitri Coucouvanis*
Department of Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055
ReceiVed January 30, 1996X
The catalytic function of the previously synthesized and characterized [(L)MoFe3S4Cl3]2-,3- clusters (L ) tetra-
chlorocatecholate, citrate, citramalate, methyliminodiacetate, nitrilotriacetate, thiodiglycolate) and of the
[MoFe3S4Cl3(thiolactate)]24- and [(MoFe3S4Cl4)2(µ-oxalate)]4- clusters in the reduction of N2H4 to NH3 is reported.
In the catalytic reduction, which is carried out at ambient temperature and pressure, cobaltocene and 2,6-lutidinium
chloride are supplied externally as electron and proton sources, respectively. In experiments where the N2H4 to
the [(L)MoFe3S4Cl3]n- catalyst ratio is 100:1, and over a period of 30 min, the reduction proceeds to 92% completion
for L ) citrate, 66% completion for L ) citramalate, and 34% completion for L ) tetrachlorocatecholate. The
[Fe4S4Cl4]2- cluster is totally inactive and gives only background ammonia measurements. Inhibition studies
with PEt3 and CO as inhibitors show a dramatic decrease in the catalytic efficiency. These results are consistent
with results obtained previously in our laboratory and strongly suggest that N2H4 activation and reduction occur
at the Mo site of the [(L)MoFe3S4Cl3]2-, 3- clusters. A possible pathway for the N2H4 reduction on a single
metal site (Mo) and a possible role for the carboxylate ligand are proposed. The possibility that the Mo-bound
polycarboxylate ligand acts as a proton delivery “shuttle” during hydrazine reduction is considered.
Introduction
also known as the iron-molybdenum cofactor (FeMo-co)3 was
structurally outlined in its entirety in the crystal structure
determination of the MoFe protein of Azotobacter Vinelandii
and Clostridium pasteurianum by two individual groups.4,5 The
current model of the cofactor, based on the interpretation of
electron density maps, shows an elongated MoFe7S9 cluster
composed of MoFe3(µ3-S)3 and Fe4(µ3-S)3 cuboidal subunits
linked by three µ2-S bridges (Figure 1). The Mo atom occupies
a unique position at the pole of the cluster. In addition, it
The total amount of N2 fixed on a global basis is estimated
to be about 2.4 × 108 tons annually. Some 65% of this amount
is carried out by nitrogen fixing microorganisms, while the
industrial Haber process accounts for about 25%. The latter
occurs at high temperatures (300-400 °C) and pressures (350-
1000 atm), whereas the biological fixation proceeds under
ambient conditions. In biology, the enzyme nitrogenase in
symbiotic or free living bacteria is responsible for the fixation
1
of atmospheric N2 to NH3 with the concomitant reduction of
(3) (a) Shah, V. K.; Brill, W. J. Proc. Natl. Acad. Sci. U.S.A. 1977, 74,
3249. (b) Rees, D. C. Curr. Opin. Struct. Biol. 1993, 3, 921. (c)
Hoover, T. R.; Imperial, J.; Ludden, P. W.; Shah, V. K. Biofactors
1988, 1, 199. (d) Burgess, B. K. In AdVances in Nitrogen Fixation
Research; Veeger, C., Newton, W. E., Eds.; Nijhoff-Junk-Pudoc;
Dordrecht, The Netherlands, 1983; p 103. (e) Nelson, M. J.; Lindahl,
P. A.; Orme-Johnson, W. H. AdV. Inorg. Biochem. 1982, 4, 1. (f)
Burgess, B. K.; Newton, W. E. In Nitrogen Fixation: The Chemical-
Biochemical-Genetic Interface; Muller, A., Newton, W. E., Eds.;
Plenum Press: New York, 1983; p 83. (g) Burgess, B. K. In Nitrogen
Fixation Research Progress; Evans, H. J., Bottomley, P. J., Newton,
W. E., Eds.; Martinus Nijhoff Publishers: Dordrecht, The Netherlands,
1985; p 543. (h) Burgess, B. K. In Molybdenum Enzymes; Spiro, T.
G., Ed.; John Wiley and Sons: New York, 1985; p 161. (i) Burgess,
B. K. Chem. ReV. 1990, 90, 1377. (j) Huang, H. Q.; Kofford, M.;
Simpson, F. B.; Watt, G. D. J. Inorg. Biochem. 1993, 52, 59.
(4) (a) Kim, J.; Rees, D. C. Science 1992, 257, 1677. (b) Kim, J.; Rees,
D. C. Nature 1992, 360, 553. (c) Kim, J.; Woo, D.; Rees, D. C.
Biochemistry 1993, 32, 7104. (d) Chan, M. K.; Kim, J.; Rees, D. C.
Science 1993, 260, 792. (e) Kim, J.; Rees, D. C. Biochemistry 1994,
33, 389. (f) Rees, D. C.; Chan, M. K.; Kim, J. In AdVances in Inorganic
Chemistry; Vol. 40; Sykes, A. G., Ed.; Academic Press: New York,
1993; p 89. (g) Kim, J.; Georgiadis, M. M.; Chan, M. K.; Woo, D.;
Komiya, K.; Rees, D. C. J. Inorg. Biochem. 1993, 51, 2. (h) Reference
1a, p 170.
(5) (a) Bolin, J. T.; Ronco, A. E.; Morgan, T. V.; Mortenson, L. E.; Xuong,
N. H. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 1078. (b) Bolin, J. T.;
Campobasso, N.; Muchmore, S. W.; Minor, W.; Morgan, T. V.;
Mortenson, L. E. In New Horizons in Nitrogen Fixation; Palacios, P.,
Moura, J., Newton, W. W., Eds; Kluwer: Dordrecht, The Netherlands,
1993; p 83. (c) Dean, D. R.; Bolin, J. T.; Zheng, L. J. Bacteriol. 1993,
175, 6337. (d) Mortenson, L. E.; Seefeldt, L. C.; Morgan, T. V.; Bolin,
J. T. AdV. Enzymol. Relat. Areas Mol. Biol. 1993, 67, 299. (e) Bolin,
J. T.; Campobasso, N.; Muchmore, S. W.; Mortenson, L. E.; Morgan,
T. V. J. Inorg. Biochem. 1993, 51, 356. (f) Reference 1a, p 186.
2H+ to H2 (eq 1).
N2 + 8e- + 10H+ + 16MgATP f
2NH4+ + H2 + 16MgADP + 16PO4 (1)
2-
Spectroscopic data,2 accumulated over a period of two
decades, have revealed many structural and electronic features
of the Fe/Mo/S center essential for substrate binding and
activation. This remarkable polymetallic Mo/Fe/S aggregate,
† Present address: Venable and Kenan Laboratories, Department of
Chemistry, The University of North Carolina, Chapel Hill, NC 27599-3290.
X Abstract published in AdVance ACS Abstracts, June 1, 1996.
(1) (a) Molybdenum Enzymes, Cofactors and Model Systems; Stiefel, E.
I., Coucouvanis, D., Newton, W. E., Eds. ACS Symposium Series
535; American Chemical Society: Washington, DC, 1993; Chapters
10-23. (b) Biology and Biochemistry of Nitrogen Fixation; Dilworth,
M. J., Glenn, A. P., Eds.; Elsevier: Amsterdam, 1991. (c) Biological
Nitrogen Fixation; Stacey, G., Burris, R. H., Evans, H. J., Eds.;
Chapman and Hall: New York, 1992. (d) Smith, B. E.; Eady, R. R.
Eur. J. Biochem. 1992, 205, 1. (e) Eady, R. R.; Leigh, G. J. J. Chem.
Soc., Dalton Trans. 1994, 2739. (f) Orme-Johnson, W. H. Annu. ReV.
Biophys. Biophys. Chem. 1985, 14, 419. (g) Ludden, P. W. In
Encyclopedia of Inorganic Chemistry; King, B. R., Ed.; John Wiley
and Sons: New York, 1994; Vol. 5, p 2566.
(2) (a) Rawlings, J.; Shah, V. K.; Chisnell, J. R.; Brill, W. J.; Zimmermann,
R.; Mu¨nck, E.; Orme-Johnson, W. H. J. Biol. Chem. 1978, 253, 1001.
(b) Hawkes, T. R.; McLean, P. A.; Smith, B. E. Biochem. J. 1984,
217, 317. (c) Scott, D. J.; May, H. D.; Newton, W. E.; Brigle, K. E.;
Dean, D. R. Nature 1990, 343, 188.
S0020-1669(96)00098-5 CCC: $12.00 © 1996 American Chemical Society

Catalytic Reduction of Hydrazine to Ammonia
Inorganic Chemistry, Vol. 35, No. 13, 1996 4039
Mo/Fe/S aggregates many of which contain “cubane” type
structural subunits and a Fe:Mo ratio of 3:113 or 4:1.13j The
cubane [MoFe3S4]3+ structural units are partial structural models
for the nitrogenase FeMo-co13,14 and show very similar first and
second coordination spheres around the Mo atoms. The
3
characteristic EPR signature15 (S ) /2) of the single-cubane
clusters that contain the [MoFe3S4]3+ core also is very similar
to those obtained for the semireduced state of the Fe-Mo
protein of nitrogenase and of the extruded FeMo-co.16 Recently,
we reported the catalytic reduction of hydrazine to ammonia17
and of acetylene to ethylene18 using the [MoFe3S4Cl3(Cl4-
cat)(CH3CN)]2- and (Et4N)3[MoFe3S4Cl3(Hcit)]3- cubanes as
catalysts (L ) Cl4-cat ) tetrachlorocatecholate; Hcit ) the
citrate trianion). Similar studies have been carried out utilizing
the [VFe3S4]2+ cubanes19 for the catalytic reduction of hydrazine.
For these reactions evidence has been presented that strongly
supports the involvement of the heterometal in the catalytic
process. The catalytic reduction of hydrazine (eq 2) has been
Figure 1. Structure of the Fe/Mo/S center in nitrogenase.4,5
appears coordinatively saturated surrounded by three inorganic
sulfides, a histidine imidazole, and a bidentate homocitrate ((R)-
2-hydroxy-1,2,4-butanetricarboxylic acid) molecule in an ap-
proximately octahedral environment.
(10) (a) Chisnell, J. R.; Premakumar, R.; Bishop, P. E. J. Bacteriol. 1988,
170, 27. (b) Bishop, P. E.; Joerger, R. D. Annu. ReV. Plant Physiol.
Plant Mol. Biol. 1990, 41, 109. (c) Schneider, K.; Mu¨ller, A.;
Schramm, V.; Klipp, W. Eur. J. Biochem. 1991, 195, 653. (d)
Schu¨ddekopf, K.; Hennecke, S.; Liese, U.; Kutsch, M.; Klipp, E. W.
Mol. Microbiol. 1993, 8, 673. (e) Pau, R. N. In New Horizons in
Nitrogen Fixation; Palacios, P., Moura, J., Newton, W. W., Eds.;
Kluwer: Dordrecht,The Netherlands, 1993; pp 117-122. (f) Mu¨ller,
A.; Schneider, K.; Knu¨ttel, K.; Hagen, W. R. FEBS Lett. 1992, 303,
36.
(11) (a) Pombeiro, A. J. L.; Richards, R. L. Coord. Chem. ReV. 1990, 104,
13. (b) Leigh, G. J. Acc. Chem. Res. 1992, 25, 177. (c) Jime´nez-
Tenorio, M.; Puerta, M. C.; Valegra, P.; Hughes, D. L. J. Chem. Soc.,
Dalton Trans. 1994, 2431. (d) Gailus, H.; Woitha, C.; Rehder, D. J.
Chem. Soc., Dalton Trans. 1994, 3471. (e) Leigh, G. J. New J. Chem.
1994, 18, 157.
(12) (a) Shilov, A. E. J. Mol. Catal. 1987, 41, 221. (b) Shilov, A. E. In
Coordination Chemistry and Catalysis; Ziolkowski, J. J., Ed.; World
Scientific: Teaneck, NJ, 1988; p 393. (c) Henderson, R. A. Transition
Met. Chem. 1990, 15, 330 and references therein. (d) Edema, J. H.
H.; Meetsma, A.; Gambarotta, S. J. Am. Chem. Soc. 1989, 111, 6878.
(13) (a) Holm, R. H.; Simhon, E. D. In Molybdenum Emzymes; Spiro, T.
G., Ed. Wiley-Interscience: New York, 1985; p 1 and references
therein. (b) Holm, R. H. In AdVances in Inorganic Chemistry, Vol.
38; Sykes, A. G., Ed.; Academic Press: New York, 1992; p 1. (c)
Holm, R. H. Pure Appl. Chem. 1995, 67, 217. (d) Coucouvanis, D.
Acc. Chem. Res. 1991, 24, 1. (e) Coucouvanis, D. In AdVances in
Inorganic Biochemistry, Vol. 9; Eichorn, G. L., Marzilli, L. G., Eds.;
Elsevier: New York, 1994; p 75. (f) Coucouvanis, D. In ref 1a, p
304. (g) Liu, Q.; Huang, L.; Liu, H.; Lei, X.; Wu, D.; Kang, B.; Lu,
J. Inorg. Chem. 1990, 29, 4131. (h) Eldredge, P. A.; Bose, K. S.;
Barber, D. E.; Bryan, R. F.; Sinn, E.; Rheingold, A.; Averill, B. A.
Inorg. Chem. 1991, 30, 2365. (i) Coucouvanis, D. In Encyclopedia of
Inorganic Chemistry; King, B. R., Ed.; John Wiley and Sons: New
York, 1994; Vol. 5, p 2557. (j) Nordlander, E.; Lee, S. C.; Cen, Wei;
Wu, Z. Y.; Natoli, C. R.; Di Cicco, A.; Filliponi, A.; Hedman, B.;
Hodgson, K. O.; Holm, R. H. J. Am. Chem. Soc. 1993, 115, 5549-
5558.
The presence of homocitrate as an integral constituent of the
FeMo-co was detected and reported previously by Ludden et
al., who established its identity by a combination of tech-
niques.1ag,6 The exact site on the cofactor involved in N2
coordination and activation is not yet known although a variety
of possibilities have been proposed,1a,6j,7 some of them based
on theoretical calculations.7 Experimental results suggest, but
do not prove, that the Mo site may be involved directly or
indirectly in some stage(s) of substrate reduction.8 Two
“alternative” forms of nitrogenase, neither of which contains
Mo, have been reported. The first has V in the place of Mo
and shows similar substrate selectivity but much lower efficiency
than the Mo prototype.9 The second is an “all-Fe” nitrogenase,
and its functional details still are under investigation. The latter
shows even lower activity than the V nitrogenase.10
A large number of coordination compounds have been
proposed as possible structural or functional models for nitro-
genase. Included among these are: (a) mononuclear11 and
binuclear12 transition metal complexes, with reactivity features
relevant to the nitrogenase function,11e,12c and (b) polynuclear
(6) (a) Hoover, T. R.; Shah, V. K.; Roberts, G. P.; Ludden, P. W. J.
Bacteriol. 1986, 167, 999. (b) Hoover, T. R.; Robertson, A. D.; Cerny,
R. L.; Hayes, R. N.; Imperial, J.; Shah, V. K.; Ludden, P. W. Nature
1987, 329, 855. (c) Hoover, T. R.; Imperial, J.; Ludden, P. W.; Shah,
V. K. Biochemistry 1989, 28, 2768. (d) Hoover, T. R.; Imperial, J.;
Liang, J.; Ludden, P. W.; Shah, V. K. Biochemistry 1988, 27, 3647.
(e) Imperial, J.; Hoover, T. R.; Madden, M. S.; Ludden, P. W.; Shah,
V. K. Biochemistry 1989, 28, 7796. (f) Madden, M. S.; Kindon, N.
D.; Ludden, P. W.; Shah, V. K. Proc. Natl. Acad. Sci. U.S.A. 1990,
87, 6517. (g) Liang, J.; Madden, M. S.; Shah, V. K.; Burris, R. H.
Biochemistry 1990, 29, 8577. (h) Hoover, T. R.; Imperial. J.; Ludden,
P. W.; Shah, V. K. J. Bacteriol. 1988, 170, 1978. (i) Madden, M. S.;
Paustian, T. D.; Ludden, P. W.; Shah, V. K. J. Bacteriol. 1991, 173,
5403. (j) Reference 1a, p 196.
(14) (a) Coucouvanis, D.; Demadis, K. D.; Kim, C.-G.; Dunham, R. W.;
Kampf, J. W. J. Am. Chem. Soc. 1993, 115, 3344. (b) Demadis, K.
D.; Coucouvanis, D. Inorg. Chem. 1995, 34, 436.
(15) (a) Armstrong, W. H.; Holm, R. H. J. Am. Chem. Soc. 1981, 103,
6246. (b) Mascharak, P. K.; Papaefthymiou, G. C.; Armstrong, W.
H.; Foner, S.; Frankel, R. B.; Holm, R. H. Inorg. Chem. 1983, 22,
2851.
(7) (a) Deng, H.; Hoffman, R. Angew. Chem., Int. Ed. Engl. 1993, 32,
1062. (b) Dance, I. Aust. J. Chem. 1994, 47, 979.
(8) Smith, B. E. J. Less-Common Met. 1977, 54, 465 and references
therein.
(16) Electron Paramagnetic Resonance studies: (a) Hoffman, B. M.;
Venters, R. A.; Roberts, J. E.; Nelson, M.; Orme-Johnson, W. H. J.
Am. Chem. Soc. 1982, 104, 4711. (b) Hoffman, B. M.; Roberts, J. E.;
Orme-Johnson, W. H. J. Am. Chem. Soc. 1982, 104, 860. (c) Thomann,
H.; Morgan, T. V.; Jin, H.; Burgmayer, S. J. N.; Bare, R. E.; Stiefel,
E. I. J. Am. Chem. Soc. 1987, 109, 7913. (d) True, A. E.; Nelson, M.
J.; Venters, R. A.; Orme-Johnson, W. H.; Hoffman, B. M. J. Am.
Chem. Soc. 1988, 110, 1935. (e) Euler, W. B.; Martinsen, J.;
McDonald, J. W.; Watt, G. D.; Wang, Z.-C. Biochemistry 1984, 23,
3021. (f) George, G. N.; Bare, R. E.; Jin, H.; Stiefel, E. I.; Prince, R.
C. Biochem. J. 1989, 262, 349.
(17) Coucouvanis, D.; Mosier, P. E.; Demadis, K. D.; Patton, S.; Malinak,
S. M.; Kim, C. G.; Tyson, M. A. J. Am. Chem. Soc. 1993, 115, 12193.
(18) Laughlin, L. J.; Coucouvanis, D. J. Am. Chem. Soc. 1995, 117, 3118.
(19) Malinak, S. M.; Demadis, K. D.; Coucouvanis, D. J. Am. Chem. Soc.
1995, 117, 3126.
(9) (a) Robson, R. L.; Eady, R. R.; Richardson, T. H.; Miller, R. W.;
Hawkins, M.; Postgate, J. R. Nature 1986, 332, 388. (b) Hales, B. J.;
Case, E. E.; Morningstar, J. E.; Dzeda, M. F.; Mauterer, L. A.
Biochemistry 1986, 25, 7251. (c) Hales, B. J.; Langosch, D. J.; Case,
E. E. J. Biol. Chem. 1986, 261, 15301. (d) Eady, R. R.; Robson, R.
L.; Richardson, T. H.; Miller, R. W.; Hawkins, M. Biochem. J. 1987,
244, 197. (e) Arber, J. M.; Dobson, B. R.; Eady, R. R.; Stevens, P.;
Hasnain, S. S.; Garner, C. D.; Smith, B. E. Nature 1987, 325, 372. (f)
Smith, B. E.; Eady, R. R.; Lowe, D. J.; Gormal, C. Biochem. J. 1988,
250, 299. (g) Eady, R. R. in AdVances in Inorganic Chemistry, Vol.
36; Sykes, A. G., Ed.; Academic Press: New York, 1991; p 77. (h)
Eady, R. R. Polyhedron 1989, 8, 1696. (i) Hales, B. J. In AdVances in
Inorganic Biochemistry, Vol 8; Eichorn, G. L., Marzilli, L. G., Eds.;
Elsevier: New York, 1990; p 165 and references therein.

4040 Inorganic Chemistry, Vol. 35, No. 13, 1996
Demadis et al.
Figure 2. Schematic structures of the [(L)MoFe3S4 Cl3] cubanes used in this study with L ) carboxylate group containing ligands. The structures
of compounds 1, 3, 5, and 9-12 have been crystallographically determined.
+
N2H4 + 2e- + 4H+ f 2NH4
(2)
triacetate dianion, ox ) oxalate, im ) imidazole, dmpe ) dimeth-
ylphosphinoethane, Lut‚HCl ) 2,6-lutidine hydrochloride. Details on
the purification of solvents and instrumentation used for the spectro-
scopic and electrochemical studies have been given elsewhere.14b All
of the carboxylic acids, cobaltocene, hydrazine, and a 1.0 M solution
of HCl in Et2O were purchased from Aldrich and were used as received.
The clusters (Et4N)2[MoFe3S4Cl3(Cl4-cat)(CH3CN)] (1),24 (Et4N)2-
[MoFe3S4Cl3(mida)] (3),14b (Et4N)2[MoFe3S4Cl3(tdga)] (5),23 (Et4N)3-
[MoFe3S4Cl3(Hcit)] (6),14 (Et4N)2[MoFe3S4Cl3(H2cit] (7),14 (Et4N)2-
[MoFe3S4Cl3(Hcmal)] (8),14 (Et4N)4[MoFe3S4Cl3(tla)]2 (9),22 23 (Et4N)4-
[(MoFe3S4Cl4)2(µ-ox)] (10),14 (Et4N)[MoFe3S4Cl4(dmpe)] (11),25 and
(Bu4N)2(Fe4S4Cl4) (12)26 were synthesized by published procedures.
Also, Lut‚HCl was prepared as previously described.19 The structures
of the above clusters are depicted schematically in Figure 2.
(Et4N)2[MoFe3S4Cl3(Cl4-cat)(im)] (2). A 0.30 g amount of 1 (0.29
mmol) was mixed with imidazole (0.02 g, 0.29 mmol) in 30 mL of
CH3CN, and the homogeneous solution was stirred overnight. The color
turned to brown-orange. At this point, filtration (negligible amount of
black residue), addition of Et2O (100 mL) to the filtrate, and overnight
standing gave a highly crystalline black solid (yield 0.27 g, 90%). Anal.
Calcd for MoFe3S4Cl7O2C25H44N4 (2, MW 1072.57): C, 27.98; H, 4.10;
N, 5.22. Found: C, 28.31; H, 4.33; N, 5.25. Electronic spectrum
(CH3CN): lacking prominent features. Far-infrared spectrum (in CsI
disks, cm-1): 299 (m), 314 (w), 350 (vs), 373 (sh), 392 (w), 410 (m),
423 (m). Mid-infrared spectrum (in CsI disks, cm-1): peaks assigned
to the imidazole ligand ν(CdN + CdC) 1650 (m), 1600 (m); ν(N-
H) 3128 (m).
thoroughly studied.20 Hydrazine is a substrate21a-c as well as a
product of functioning nitrogenase and has been isolated by
quenching the enzyme.21d-g
In this paper, we report results of our studies on the behavior
of the (L)MoFe3S4 cubanes (L ) polycarboxylate anions) as
catalysts in the reduction of hydrazine to ammonia. Preliminary
reports of this study have appeared in the literature.17,22,23
Experimental Section
General Considerations. Abbreviations used: Cl4-cat ) tetrachlo-
rocatecholate, maa ) mercaptoacetate, tla ) thiolactate, tdga )
thiodiglycolate, Hcit ) citrate trianion, H2cit ) citrate dianion, Hcmal
) citramalate dianion, mida ) methyliminodiacetate, Hnta ) nitrilo-
(20) Examples of chemical reduction of N2H4 to NH3 by transition metal
complexes are ubiquitous in the literature: (a) Ramachadraiah, G. J.
Am. Chem. Soc. 1994, 116, 6733. (b) Cai, S.; Schrock, R. R. Inorg.
Chem. 1991, 30, 4105. (c) Schrock, R. R.; Glassman, T. E.; Vale, M.
G.; Kol, M. J. Am. Chem. Soc. 1993, 115, 1760. (d) Schrock, R. R.;
Glassman, T. E.; Vale, M. G. J. Am. Chem. Soc. 1991, 113, 725. (e)
Collman, J. P.; Wagenknecht, P. S.; Hutchison, J. E. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 1537. (f) Schrauzer, G. N.; Robinson, P. R.;
Moorehead E. L.; Vickrey, T. M. J. Am. Chem. Soc. 1976, 98, 2815.
(21) (a) Burgess, B. K.; Wherland, S.; Stiefel, E. I.; Newton, W. E.
Biochemistry 1981, 20, 5140. (b) Buler, W. A In Proceedings of the
First International Symposium on Nitrogen Fixation; Newton, W. E.,
Nyman, C. F., Eds.; Washington State University Press: Pullman, WA,
1976; p 177. (c) Davis, L. C. Arch. Biochem. Biophys. 1980, 204,
270. (d) Thorneley, R. N.; Lowe, D. J. Biochem. J. 1984, 224, 887.
(e) Evans, D. J.; Henderson, R. A.; Smith, B. E. In Bioinorganic
Catalysis; Reedijk, J., Ed.; Marcel Dekker Inc.: New York, 1993; p
89. (f) Thorneley, R.N.; Eady, R. R.; Lowe, D. J. Nature (London)
1978, 272, 557. (g) Dilworth, M. J.; Eady, R. R. Biochem. J. 1991,
277, 465.
(Et4N)2[MoFe3S4Cl3(Hnta)] (4). A 0.30 g amount of 1 (0.29 mmol)
was dissolved in 30 mL of CH3CN. Nitrilotriacetic acid (N(CH2-
(24) Palermo, R. E.; Singh, R.; Bashkin, J. K.; Holm, R. H. J. Am. Chem.
Soc. 1984, 106, 2600.
(25) Zhang, Y.-P.; Bashkin, J. K.; Holm, R. H. Inorg. Chem. 1987, 26,
694.
(26) (a) Bobrik, M. A.; Hodgson, K. O.; Holm, R. H. Inorg. Chem. 1977,
16, 1851. (b) Wong, G. B.; Bobrik, M. A.; Holm, R. H. Inorg. Chem.
1978, 17, 578. (c) Holm, R. H. Acc. Chem. Res. 1977, 10, 427.
(22) Demadis, K. D.; Coucouvanis, D. Inorg. Chem. 1994, 33, 4195.
(23) Demadis, K. D.; Coucouvanis, D. Inorg. Chem. 1995, 34, 3658.

Catalytic Reduction of Hydrazine to Ammonia
Inorganic Chemistry, Vol. 35, No. 13, 1996 4041
COOH)3, 0.083 g, 0.44 mmol, 1.5-fold excess) was partially dissolved
in 5 mL of MeOH, and this suspension was added dropwise to the
cluster solution under vigorous stirring. While the temperature was
kept at ∼50 °C, the mixture was stirred overnight, after which it was
allowed to cool at room temperature. A black insoluble residue was
removed by filtration and was discarded. Et2O (∼70 mL) was layered
onto the brown-reddish filtrate, and after 6 h of standing, a micro-
crystalline solid precipitated. It was isolated by filtration and washed
thoroughly with MeOH, acetone, and Et2O. Finally, it was recrystal-
lized from CH3CN/Et2O and dried in Vacuo to afford analytically pure
material (yield 0.15 g, 55%). Anal. Calcd for MoFe3S4Cl3O6N3C22H47
(4, MW 947.73): C, 27.90; H, 5.00; N, 4.43. Found: C, 28.25; H,
5.07; N, 4.19. Electronic spectrum (CH3CN, nm): 470, 314. Far-
infrared spectrum (cm-1): 324 (s), 353 (sh), 362 (vs), 397 (w), 414
(m). Mid-infrared spectrum (cm-1): νasym(CdO) 1625 (vs), 1658 (vs),
1746 (m, due to uncoordinated -COOH); νsym(C-O) 1393; ν(O-H)
3440 (m).
g, 2.00 mmol) were added under vigorous stirring. The solution was
stirred for ∼3 h, during which all of the cluster precipitated, leaving a
yellow supernatant liquid (due to Co(Cp)2Cl, identified from its
characteristic electronic spectrum). The brown-black residue was
isolated by filtration and tested for ammonia. The residue was washed
thoroughly with CH3CN, THF, EtOH, and Et2O (to remove the
byproducts) and finally dried in Vacuo. An 0.25 g amount of the final
product was isolated, which corresponds to 92% recovery of the catalyst.
This material is totally insoluble in all solvents, so solution studies
proved impossible. The far-infrared spectrum (302, 318, 346, 374, 383,
412 cm-1), is consistent with a cubane structural core.14 Mid-infrared
spectrum (cm-1): νasym(CdO) 1552, 1606, 1638, 1712. Vibrations
assigned to Co(Cp)2 :
+ 30 460, 866, 1414, 3100 cm-1. Anal. Calcd for
MoFe3S4Cl3Co2C26H26 (7, MW 1085.85), [Co(Cp)2]2[MoFe3S4Cl3-
(H2cit)]: C, 28.73; H, 2.39; N, 0.0. Found: C, 28.43; H, 2.59; N, 0.0.
It should be noted that actually compound 7 (the protonated form of
6) is isolated at the end of the reaction (Vide infra). Previous studies
on the identification of recovered catalyst 1 included elemental analyses,
Catalytic Reductions of Hydrazine. Reactions were performed
under strict anaerobic conditions in CH3CN. The final volume of the
reaction mixture was ∼40 mL, and the substrate:catalyst ratio was 100:
1. The concentration of the catalyst was 1.26 × 10-5 M, and all the
other reagent concentrations were scaled accordingly. The catalytic
reductions were carried out by two methods.
electronic spectroscopy, infrared spectroscopy, and quantification of
3
the characteristic S )
/ EPR signal. Those results unequivocally
2
demonstrated that there is little or no loss of the catalyst (by
decomposition) during catalytic cycles and the [MoFe3S4]3+ core
remains intact after the completion of the reaction.29
Method A. To an Erlenmeyer flask was added 0.5 mL of a 1 mM
stock solution of the cubane catalyst in CH3CN. Then a volume of
<40 mL of CH3CN was added. Hydrazine (0.25 mL of a 0.2 M stock
solution in CH3CN) was injected via syringe into the flask under
continuous stirring, and the the volume was adjusted to 40 mL. The
concentrations of the catalyst and hydrazine at this stage were 1.25 ×
10-5 and 1.25 × 10-3 M, respectively. Solid cobaltocene (0.02 g) and
Lut‚HCl (0.03 g) were added to the reaction mixture, which was
vigorously stirred. Samples (3.0 mL) were obtained at 5 min, 30 min,
1 h, and 12 h.
Method B. To a 125 mL flask was added 20.0 mL of a 0.25 M
solution of cobaltocene in CH3CN, followed by 20.0 mL of a 0.38 M
CH3CN solution of LutHX (X ) Cl, BPh4) and 0.50 mL of a 0.10 M
CH3CN solution of hydrazine. Immediately after the addition of
hydrazine, 0.10 mL of a 0.005 M solution of the catalyst in CH3CN
was added (t ) 0). Samples (3.0 mL) were obtained at 1, 2, 4, and 6
min (all (3 s).
Results and Discussion
The X-ray structure of the Fe/Mo/S aggregate of nitrogenase4,5
has revealed the Mo atom to be coordinatively saturated and
located on the periphery of the MoFe6S9 core. These observa-
tions have led to the suggestion that perhaps the Mo atom is
not directly involved in the activation and reduction of N2 or
other substrates.4 The merits of this conclusion may be explored
by reactivity studies on synthetic Mo/Fe/S clusters in which
the Mo coordination closely resembles the Mo environment in
the nitrogenase cofactor. The Fe/Mo/S “cubane” clusters that
contain the [MoFe3S4]3+ cores are well suited for such studies.
In these compounds the first and second coordination spheres
around the Mo atoms are nearly identical to those found in the
Fe/Mo/S center in nitrogenase.
Reactivity studies of the MoFe3S4 single cubanes have shown
these clusters to be unreactive toward dinitrogen in either the
oxidized (core charge ) +3) or the reduced (core charge )
+2) form.13a In contrast, coordination of hydrazine and
substituted hydrazines to the Mo site of the MoFe3S4 cubanes
occurs readily and has been demonstrated by spectroscopic
techniques and by X-ray crystallographic studies.31 The Fe sites,
mainly coordinated by thiolate or halide terminal ligands, do
not undergo ligand exchange with hydrazines, and to our
knowledge, no Fe-N2H4 adduct has ever been isolated or even
transiently detected. This lack of reactivity with hydrazines is
in contrast to what is observed with acetylene. The Fe sites in
the [MoFe3S4]3+ and [Fe4S4]2+ cubanes are capable of reducing
C2H2 to C2H4, as shown by results obtained in this18 and other
laboratories,32 and very likely the reduction is preceded by
acetylene coordination and activation by the Fe atoms.
Ammonia and Hydrazine Analysis. The addition of solid cobalto-
cene and Lut‚HCl (requiring some time to dissolve; method A) makes
it difficult to sample and analyze the reaction mixture at the very early
stages (<5 min). The rapid mixing of LutHX and N2H4 solutions
(method B) generates N2H5+X-, which at times precipitates (X ) Cl-)
or may serve as a N2H5+ source for the catalyst, which itself precipitates.
The latter method, however, allows for sampling in the earlier stages
of the reaction, often prior to the onset of precipitation.
Samples (3.0 mL) were removed from the reaction solution at
predetermined times, were immediately quenched with an excess of
HCl in Et2O (0.2 mL), and were then taken to dryness in Vacuo. The
water-soluble species were extracted with 25 mL of deionized water.
Aliquots (1 mL) of the aqueous extracts were used for ammonia
determination with the indophenol test.27 Most determinations were
at least triplicate runs with a spread of approximately 7%. Periodic
tests for hydrazine with the PDMAB method28 ensured complete N
atom balance (100 ( 5%).
Recovery and Identification of the (Et4N)2[MoFe3S4Cl3(H2cit)],
7, Cluster Catalyst. In previous publications we have described in
detail protocols for identifying a variety of MFe3S4 (M ) Mo, V) cluster
catalysts at the end of catalytic cycles.19,29 Here, we briefly present
analogous studies on the (Et4N)2[MoFe3S4Cl3(Hcit], 6, cluster. For
practical reasons we scaled up the amount of the cluster and performed
a “pseudo-catalytic” experiment (N2H4:cluster 2:1). An 0.27 g sample
of 6 (0.25 mmol) was dissolved in 40 mL of CH3CN. Then, hydrazine
(0.25 mL of a 0.2 M stock solution in CH3CN, 0.50 mmol) was added
dropwise, and finally Co(Cp)2 (0.02 g, 1.00 mmol) and Lut‚HCl (0.03
The recently reported catalytic reduction of hydrazine by the
MoFe3S4 single cubanes is of some relevance to the function
of nitrogenase considering that HNdNH and H2N-NH2 are
possible intermediates in biological dinitrogen fixation. Indeed,
Thorneley, Lowe, and others21d-g have detected hydrazine upon
quenching nitrogenase under turnover conditions. Furthermore,
(30) Materikova, R. B.; Babin, V. N.; Lyatifov, I. R.; Salimov, R. M.; Fedin,
E. I.; Petrovskii, P. V. J. Organomet. Chem. 1981, 219, 259.
(31) (a) Challen, P. R.; Koo, S. M.; Kim. C. G.; Dunham, W. R.;
Coucouvanis, D. J. Am. Chem. Soc. 1990, 112, 8606. (b) Mosier, P.
E.; Kim, C. G.; Coucouvanis, D. Inorg. Chem. 1993, 32, 3620. (c)
Challen, P. R. Ph.D. Thesis, The University of Michigan, 1990.
(32) (a) Tanaka, K.; Honjo, M.; Tanaka, T. J. Inorg. Biochem. 1984, 22,
187. (b) Tanaka, T.; Nakamoto, M.; Tashiro, Y. Bull. Chem. Soc. Jpn.
1985, 58, 316. (c) McMillan, R. S.; Renaud, J.; Reynolds, J. G.; Holm,
R. H. J. Inorg. Biochem. 1979, 11, 213.
(27) (a) Chaney, A. L.; Marbach, E. P. Clin. Chem. 1962, 8, 130. (b)
Weatherburn, M. W. Anal. Chem. 1974, 39, 971.
(28) p-(Dimethylamino)benzaldehyde (PDMAB) method: Watt, G. W.;
Chrisp, J. D. Anal. Chem. 1952, 24, 2006.
(29) Details on these studies have been deposited as supplementary material
with ref 17. Also, see ref 23.

4042 Inorganic Chemistry, Vol. 35, No. 13, 1996
Demadis et al.
a
Table 1. Production (%) of NH3 from the Catalytic Reduction of
b
NH2NH2 by Various MoFe3-Carboxylate Cubanes Using Co(Cp)2
as the Reducing Agent and Lut‚HCl as the Proton Source (Method
A; [NH2NH2]:[Catalyst] ) 100:1)c
no. of
cluster catalystd
trials 5 min 30 min 1 h 12 h
[MoFe3S4Cl3(Cl4-cat)(CH3CN)]2- (1)
[MoFe3S4Cl3(Cl4-cat)(im)]2- (2)
[MoFe3S4Cl3(mida)]2- (3)
[MoFe3S4Cl3(Hnta)]2- (4)
[MoFe3S4Cl3(tdga)]2- (5)
[MoFe3S4Cl3(Hcit)]3- (6)
[MoFe3S4Cl3(H2cit)]2- (7)
[MoFe3S4Cl3(Hcmal)]2- (8)
[MoFe3S4Cl3(tla)]24- (9)e
[(MoFe3S4Cl4)2(µ-ox)]4- (10)e
[MoFe3S4Cl4(dmpe)]- (11)
(Fe4S4Cl4)2- (12)
3
1
2
3
3
4
3
3
2
2
1
2
30
31
55
48
38
83
80
62
58
63
12
0
34
34
64
53
45
92
86
66
58
71
13
0
38 61
35 48
70 79
59 65
47 65
96 98
94 95
68 78
66 71
77 85
14 14
0
7
a The experiments were performed according to a protocol described
in the Experimental Section. b Periodically, N2H4 quantification with
the p-(dimethylamino)benzaldehyde method28 confirmed complete N
atom balance. In all cases, there was a balance of 100 ( 5%.
c Sampling of the reaction mixture and treatment of the samples were
performed as described previously.19 The results were reproducible to
within 5% from the mean. d Reactions of catalyst with Co(Cp)2 and
Lut‚HCl (absence of N2H4) or of N2H4 with Co(Cp)2 and Lut‚HCl
(absence of catalyst) gave NH3 within background limits (<3%). e For
clusters 9 and 10 two molecules of N2H4 per double cubane (or one
N2H4 per Mo) were used in the experiments.
the presence of N2H4 has also been detected by Dilworth and
Eady33 in the reduction of N2 to NH3 by the alternative V
nitrogenase.9
A cofactor essential for nitrogenase function is the homoci-
trate molecule coordinated to the Mo atom of the FeMo-co. It
has been reported that the nifV gene which encodes for the
synthesis of homocitrate is needed for the biosynthesis of all
three (FeMo-co, FeV-co, and FeFe-co) cofactors.34 The dis-
covery of the Mo-bound homocitrate in the FeMo-co led to
exhaustive biochemical studies.6 The catalytic competence of
the FeMo-co was investigated in the presence of various
polycarboxylate ligands in experiments designed to delineate
the possible role of the homocitrate ligand in N2 fixation. A
number of interesting results have derived from this sructure/
function study: (a) a minimum of two -COOH groups and
one -OH (in some cases a keto) group is required for activity;
(b) the D acids are preferred over the L acids; (c) small changes
on various parts of the homocitrate ligand cause dramatic effects
on the N2 fixing ability of the FeMo-co. Citrate, the acid anion
structurally closest to homocitrate, effected the biosynthesis of
a FeMo-co which could fix N2, albeit at very low yields and at
much higher concentrations than homocitrate.6g The apparent
reactivity differences tunable by slight modifications of the Mo
coordination environment and/or alterations in the polycarbox-
ylate chain6f,j prompted us to explore the reactivity characteristics
of the synthetic MoFe3S4-polycarboxylate cubanes. The
methodology for the synthesis of these compounds has been
developed in our laboratory14,22,23 and has made them available
for the studies reported herein.
Figure 3. (A) Ammonia production (%) Vs time in CH3CN solvent,
at ambient temperature, from the reduction of N2H4 in the presence of
Co(Cp)2 and Lut‚HCl under initially heterogeneous conditions using
various catalysts, as indicated. This graph was based on data taken
from Table 1 using method A (see Experimental Section). (B) Ammonia
production (%) Vs time in CH3CN solvent, at ambient temperature, from
the reduction of N2H4 in the presence of Co(Cp)2 and Lut‚HBPh4 under
initially homogeneous conditions using various catalysts, as indicated
(method B, Experimental Section).
protonation, facile protonation and partial dissociation of the
bound carboxylate functions in compounds 3-8 are expected
to occur. The yields of ammonia, obtained for all clusters after
5 min of reaction time (Table 1, Figure 3), represent 80%-
90% of the ammonia produced after 1 h. This rapid leveling
off of the amount of NH3 produced indicates that the anionic
cluster catalysts rapidly precipitate as insoluble [Co(Cp)2]+,
N2H5+, or NH4+salts. A reaction of 8 with 2 equiv of hydrazine
in the presence of [Co(Cp)2] and Lut‚HCl has resulted in the
formation of the very insoluble [Co(Cp)2]2[(CitH2)MoFe3S4-
(Cl)3] product (see Experimental). The onset of precipitation
for each cluster is difficult to ascertain, and consequently
meaningful initial rates (at <5% of substrate consumption) are
nearly impossible to measure. As a result, the amount of NH3
obtained after 5 min (Figure 3) for each of the clusters cannot
be used as an indicator of relative reactivity.
The catalytic reduction of hydrazine was attempted with a
variety of these clusters, and the results are compiled in Table
1 and displayed in Figure 3. All of the polycarboxylate clusters
have a coordinatively saturated Mo site, and the coordination
and activation of N2H4 must be preceded by at least partial
dissociation of the carboxylate ligand. With the high proton
concentrations of acid that prevail under the reaction conditions
(100-fold excess of LutH‚HCl), in addition to hydrazine
In an attempt to obtain meaningful kinetic data, conditions
were adopted that made it possible to sample the reaction
mixture at times less than 5 min in the “initial rate” linear region.
These conditions consist of the initial dissolution of the
[Co(Cp)2] and Lut‚HX reagents in CH3CN followed by the
addition of hydrazine and finally a solution of the catalyst that
(33) Dilworth, M. J.; Eady, R. R. Biochem. J. 1991, 277, 46.
(34) Kennedy, C.; Dean, D. Mol. Gen. Genet. 1992, 231, 494.

Catalytic Reduction of Hydrazine to Ammonia
Inorganic Chemistry, Vol. 35, No. 13, 1996 4043
defines t ) 0. The choice of counterion in Lut‚HX had a
profound influence in the results. For X ) Cl-, upon addition
of hydrazine, a white precipitate was observed, presumably
N2H5+Cl-. The percent conversion of hydrazine to ammonia
was low after 6 min (∼16%) for a selected group of clusters
(Cl4-cat, mida) probably limited by the rate of dissolution of
N2H5+Cl-. Reactions with Lut‚HCl were therefore discontin-
ued. For X ) BPh4-, the reactions were considerably faster
and the results are shown in Figure 3B for four cubanes (Cl4-
cat, mida, Hcit, tdga). These results indicate little or no
difference between the relative rates of hydrazine reduction
between the clusters. All show an initial burst to about 25%
which then slowly levels off to about 50% in 6 min. Due to
the lack of precision in aqcuiring data in times less than 1 min,
initial rates cannot be obtained reliably. Nevertheless, the
curvature in the lines (Figure 3B) may indicate relative rates in
addition to the onset of catalyst precipitation. It appears likely
that, by having all components soluble initially, the “true” rate
of reaction can be witnessed. Unfortunately reliable data cannot
be obtained in times less than 1 min from t ) 0. An
examination of Figure 3 shows that the carboxylate cubanes
are more efficient catalysts than the Cl4-cat cubane, 1. The rapid
leveling off of the rate for the citrate cubane, 6, after only 1
min of reaction time indicates that this cluster precipitates very
rapidly and may have a faster initial rate than indicated. The
overall greater yields of ammonia over the same period of time
obtained with method A (Figure 3A) very likely reflect a control
of the rate (slowing down) of precipitation of the catalyst by
the slow release of the reagents in solution and the smaller ionic
strength of the solution.
coordination saturation of the Mo atom in these clusters does
not interfere with the catalytic process.
Cluster 2 has the same core structure and Mo coligands as 1
except that it has an imidazole coordinated to the Mo instead
of an CH3CN (a histidine imidazole has been found coordinated
to the Mo of the FeMo-co). Its catalytic efficiency is almost
identical to that 1 and indicates that the imidazole ligand
dissociates from the Mo (very likely as a result of protona-
tion37,38 ), thus generating a vacant site on the Mo for N2H4
coordination and reduction.
In the MoFe3S4 cubanes, the Mo-O (catecholate or carbox-
ylate) moieties are located in positions that allow for hydrogen-
bonding interactions with the Mo-coordinated hydrazine mol-
ecule. Such interactions are evident in the structures of the
[MoFe3S4Cl3(Cl4-cat)(NH2NHPh)]2- 39 and {[MoFe3S4Cl2-
(Cl4-cat)]2(µ-N2H4)(µ-S)}4- clusters.13d,31a Although the H
atoms of the PhNHNH2 or N2H4 molecules were not located,
the distances between the tetrachlorocatecholate oxygen atoms
and the hydrazine nitrogens (2.73-3.85 Å) are indicative of
hydrogen-bonding interactions. Not unlike the possible involve-
ment of the carboxylate ligands in H+ transfer, the catecholate
oxygen atoms also may be important in the protonation of the
substrate during the catalytic process.
The lack of reactivity of cluster 11 (Figures 2 and 3) may be
attributed to the following factors: (a) the nonlability of the
Mo-bound Cl, which already has been established25 for this
cluster, and (b) the inability of the dmpe ligand to serve as an
effective proton transfer site. The “all-Fe” cluster 12 was found
totally inactive in the reduction of N2H4 to NH3, showing only
a 7% conversion after 12 h.40 A possible explanation for this
lack of reactivity could be the inability of N2H4 to displace the
Fe-bound Cl- ligands.41
In addition to generating a coordination site on the Mo atom,
protonation causes anodic shifts of the electrochemical reduction
potentials of the clusters. For example, addition of a 5-fold
excess of Lut‚HCl to 5 shifts the reduction potential from -750
to -670 mV.35 Such shifts are at times needed to bring the
reduction potential of the clusters within the reducing capacity
of cobaltocene (under identical conditions,35 E1/2 of cobaltocene
) -940 mV). In catalytic studies with the [VFe3S4]2+
cubanes,19 Lut‚HCl, effectively used in the catalytic reaction,
is sufficiently acidic (pKa(CH3CN) ) 14.1) to protonate
Inhibition Studies. To further explore the importance of
the direct Mo involvement in the catalytic process, attempts were
made to inhibit hydrazine reduction with ligands that are known
to bind exclusively and irreversibly to the Mo atom in the
[MoFe3S4]3+ cubanes. Triethylphosphine is one such ligand,43
and in its presence the reduction of hydrazine is strongly
inhibited (Figure 4). The observed low (above background)
levels of NH3 production can be explained by (a) the dissociation
of some of the bound PEt3 in the presence of a large excess of
N2H4 or (b) the H+-promoted dissociation of one “arm” of the
Cl4-cat ligand that generates a vacant coordination site on the
Mo atom.
+
hydrazine. Acids weaker than N2H5 (pKa(CH3CN) ) 16.6)
such as Et3NH+ (pKa(CH3CN) ) 18.3) however are not
effective, and as a consequence the use of Et3NH+ as a source
of protons shuts off the catalytic process.
Hypotheses concerning the possible role of the unique
homocitrate ligand have been presented in the literature and
include possible modulation of the electrochemical potentials
or participation in proton transfer steps.1e,g,6 The latter seems
more reasonable, taking into account the fact that the bidentate
coordination mode of the homocitrate ligand allows two
uncoordinated -COOH (or -COO-) groups to become in-
volved in protonation/deprotonation steps.36
(37) In dimethylformamide, DMF, for Lut‚HCl, pKa ≈ 5.0; for citric acid,
pKa1 ) 7.2, pKa2 ) 10.9, and pKa3 ) 12.2: Izutsu, K. Acid Base
Dissociation Constants in Dipolar Aprotic SolVents; IUPAC Chemical
Data Series, No. 35; Blackwell Scientific Publications: Oxford, U.K.,
1990.
(38) (a) Benoit, R. L.; Lefebvre, D.; Frechette, M. Can. J. Chem. 1987,
65, 996. (b) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.
(39) Coucouvanis, D.; Patton, S.; Kim, C. K. Unpublished results.
(40) This cluster has been found effective in the reduction of acetylene at
a rate much slower than that of the Mo cubanes.18
(41) Tanaka et al. have reported N2H4 reductions to NH3 using electro-
chemically generated forms of the [Fe4S4(SR)4]2- cluster (R ) SPh,
SCH2CH2OH) as the active center (Hozumi, Y.; Imasaka, Y.; Tanaka,
K.; Tanaka, T. Chem. Lett. 1983, 897). These systems differ from
ours in that they are substoichiometric (noncatalytic) and employ
thiolates as Fe ligands. It is well established that Fe-bound thiolates
cathodically shift the electrochemical potentials by least 200 mV;42
therefore, the reduced cluster is “charged” with higher reducing power.
(42) (a) Cambray, J.; Lane, R. W.; Wedd, A. G.; Johnston, R. W.; Holm,
R. H. Inorg. Chem. 1977, 16, 2565. (b) Hill, C. L.; Renaud, J.; Holm,
R. H.; Mortenson, L. E. J. Am. Chem. Soc. 1977, 99, 2549.
(43) (a) Holm, R. H. In Biomimetic Chemistry; Yoshida, Z., Ise, N., Eds.;
Elsevier: New York, 1983; p 79-99. (b) Mascharak, P. K.; Armstrong,
W. H.; Mizobe, Y.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 475.
The qualitative results obtained with the majority of the
carboxylate MoFe3S4 clusters 3-10 show that the catalytic
reduction of hydrazine is occurring generally with a higher
efficiency than that of either 1 or 2. Furthermore, the apparent
(35) Voltammetric measurements in CH3CN solution vs a Ag/AgCl
reference electrode using Pt working and auxiliary electrodes and
Bu4NClO4 as supporting electrolyte.
(36) It is very interesting to notice that the closely related citrate binds to
an Fe of an Fe4S4 cluster in the active site of aconitase, in the same
bidentate fashion: (a) Glusker, J. P. In The Enzymes; Boyer, P. D.,
Ed.; Academic Press: New York, 1971; p 413. (b) Lauble, H.;
Kennedy, M. C.; Beinert, H.; Stout, C. D. Biochemistry 1992, 31, 2735.
(c) Kennedy, M. C.; Stout, C. D. In AdVances in Inorganic Chemistry,
Vol. 38; Cammack, R., Sykes, A. G., Eds.; Academic Press: New
York, 1992; p 413.

4044 Inorganic Chemistry, Vol. 35, No. 13, 1996
Demadis et al.
initial protonation and subsequent reduction. These data are
consistent with results reported for many other binuclear
transition metal complexes that contain end-on bridging N2.
These complexes upon reduction/protonation give high yields
of N2H445 but little or no NH3.46 In the case of the MoFe3S4-
carboxylate cubanes, steric constraints preclude dimerization that
would result in an end-on bridging N2H4 ligand and the
activation/reduction of hydrazine must occur on a single metal
site.
The catalytic reduction and disproportionation of N2H4 by a
binuclear Mo complex with pyridinethiolate coligands has been
reported.47 The actual nature of the active catalyst and the
coordination mode of the hydrazine ligand are not clear. In
this system, N2H4 may be activated either by an end-on bridging
fashion or (more likely) with the two Mo centers acting
independently in the activation of two terminally bound N2H4
molecules. Kuwata et al. recently reported the disproportiona-
tion of N2H4 and substituted N2H4 with a binuclear Ru
complex.48 In this case, binuclear activation of N2H4 seems
attractive due to the coordination unsaturation of both Ru
centers.
Figure 4. Ammonia production (%) Vs time in CH3CN solvent, at
ambient temperature, from the reduction of N2H4 in the presence of
PEt3 or CO as inhibitor, Co(Cp)2, and Lut‚HCl using various catalysts,
as indicated. This graph was based on data taken from Table 2.
Table 2. Production (%) of NH3 by the Catalytic Reduction of
NH2NH2 by Various MoFe3-Carboxylate Cubanes in the Presence
of Inhibitor PEt3 or CO (Experimental Conditions as in Table 1;
[NH2NH2]:[Catalyst] ) 100:1)
Possible Role of the Mo-Bound Polycarboxylate Ligands
in the Function of the [MoFe3S4]3+ Cubanes and Their
Relevance to Homocitrate in the FeMo-co. A possible
pathway for the reduction of N2H4 to NH3 by the polycarbox-
ylate cubanes shown in Figure 5 is similar to that published for
the reduction effected by the VFe3S4 cubanes19 but also includes
the possible involvement of the carboxylate ligands. The
starting point (A) (Figure 5) is the proton-induced dissociation
of a carboxylate (-COO-) “arm” from the Mo atom, generating
a vacant site for N2H4 coordination (B). The subsequent Mo-
N2H4 adduct may be stabilized Via hydrogen-bonding interac-
tions ((hydrazine)N-H‚‚‚O or (carboxylate)O-H‚‚‚N). An
initial protonation to the â-N of N2H4 occurs (C) which is
followed by 1e- reduction (D). An additional protonation
releases the first equivalent of NH4+, leaving a Mo-NH2 moiety
(E) (or alternatively a ModNH (F) group). At this point,
protonation by the uncoordinated -COOH groups and subse-
quent reduction will cause the release of the second equivalent
no. of
5
30
cluster catalyst
trials min min 1 h 12 h
[MoFe3S4Cl3(Cl4-cat)(CH3CN)]2- (1)a
[MoFe3S4Cl3(Cl4-cat)(CH3CN)]2- (1) + PEt3
[MoFe3S4Cl3(Cl4-cat)(CH3CN)]2- (1)/COb
[MoFe3S4Cl3(Hcmal)]2- (8)a
3
2
2
3
1
1
30 34 38 61
18 20 23 37
7
62 66 68 78
54 55 57 66
17 18 18 19
7
8
9
[MoFe3S4Cl3(Hcmal)]2- (8) + PEt3 (1:10)
[MoFe3S4Cl3(Hcmal)]2- (8) + PEt3 (1:200)
a For comparison reasons NH3 yields (%) from Table 1 using clusters
1 and 8 as catalysts have been also included. b In this experiment, the
reaction flask was evacuated and immediately refilled with CO (Johnson
& Mathey) as soon as all the reagents were mixed.
Carbon monoxide is known to inhibit nitrogenase reactivity;44
furthermore, CO can bind to the MoFe3S4 cubanes only at their
reduced (2+) oxidation level.43b In both cases, the precise site
of CO coordination (Fe, Mo, or both) is not entirely clear. A
hydrazine reduction experiment using cluster 1 as a catalyst was
performed under a CO atmosphere. The ammonia production
yields (Table 2), although slightly above background, were much
lower than the corresponding yields under a N2 atmosphere
(Figure 4). This result suggests that the [MoFe3S4]3+ core
undergoes reduction to the 2+ oxidation level during the
catalytic process, where it is attacked and irreversibly inactivated
by CO (very likely at the Mo site). The fact that hydrazine
protonation precedes cluster reduction further suggests that the
CO molecules compete with the N2H5+ ligand for binding sites
in the Mo coordination sphere. The PEt3 and CO ligands
previously have been used successfully in our laboratory as
inhibitors in N2H4 reduction by the [VFe3S4]2+ cores19 and in
C2H2 reduction by the [MoFe3S4]3+ cores.18
+
of NH4 and the regeneration of the starting cluster A.
At this stage, it is difficult to predict in which protonation
step the -COOH group participates. Furthermore, it is uncertain
whether the H+ ions originate from a -COOH group, a (Mo)-
OH group, or a combination of both. In contrast to intermediate
C (Figure 5), which has been isolated and characterized with
Cl4-cat in place of citrate on the Mo,49 D, E, and F have not
been isolated or detected.
Proposed Pathway for Nitrogen Fixation by Nitrogenase.
A plethora of binuclear complexes with bridging N2 have been
synthesized, and many of them have been stucturally character-
Mononuclear Ws Binuclear Activation of N2H4. Recently
we reported that the {[MoFe3S4Cl2(Cl4-cat)]2(µ-N2H4)(µ-S)}4-
doubly bridged double cubane31a did not reduce the end-on
bridging hydrazine to ammonia in the presence of Co(Cp)2 and
Lut‚HCl.17 This result was explained by the unavailability of
lone pairs on the bridging hydrazine molecule for the essential
(45) (a) Manriquez, J. M.; Sanner, R. D.; Marsh, R. E.; Bercaw, J. E. J.
Am. Chem. Soc. 1976, 98, 3042. (b) Churchill, M. R.; Li, Y. J.;
Theopold, K. H.; Schrock, R. R. Inorg. Chem. 1984, 23, 4472. (c)
Leigh, G. J. J. Mol. Catal. 1988, 47, 363. (d) Anderson, S. N.;
Richards, R. L.; Hughes, D. L. J. Chem. Soc., Chem. Commun. 1982,
1291.
(46) (a) Teuben, J. H. J. Organomet. Chem. 1973, 57, 159. (b) Leigh, G.
J.; Prieto-Alco´n, R.; Sanders, J. R. J. Chem. Soc., Chem. Commun.
1991, 921. (c) Ferguson, R.; Solari, E.; Floriani, C.; Chiesi-Villa, A.;
Rizzoli, C. Angew. Chem., Int. Ed. Engl. 1993, 32, 396. (d) Dilworth,
J. R.; Henderson, R. A.; Hills, A.; Hughes, D. L.; Macdonald, C.;
Stephens, A. N.; Walton, D. R. M. J. Chem. Soc., Dalton Trans. 1990,
1077. (e) Henderson, R. A.; Morgan, S. H.; Stephens, A. N. Chem.
Soc., Dalton Trans. 1990, 1101.
(44) (a) Smith, B. E.; Lowe, D. J.; Bray, R. C. Biochem. J. 1973, 135,
331. (b) Hawkes, T. R.; Lowe, D. J.; Smith, B. E. Biochem. J. 1983,
211, 495. (c) Hwang, J. C.; Chen, C. H.; Burris, R. H. Biochim.
Biophys. Acta 1973, 292, 256. (d) Davis, L. C.; Shah, V. K.; Brill,
W. J. Biochim. Biophys. Acta 1975, 403, 67. (e) Orme-Johnson, W.
H.; Davis, L. C.; Henzl, M. T.; Averill, B. A.; Orme-Johnson, N. R.;
Mu¨nck, E.; Zimmerman, R. In Recent DeVelopments in Nitrogen
Fixation; Newton, W. E., Postgate, J. R., Rodriguez-Barrueco, C., Eds.;
Academic Press: London, 1977; p 131.
(47) Block, E.; Ofori-Okai, G.; Kang, H.; Zubieta, J. J. Am. Chem. Soc.
1992, 114, 758.
(48) Kuwata, S.; Mizobe, Y.; Hidai, M. Inorg. Chem. 1994, 33, 3619.
(49) Mosier, P. E.; Coucouvanis, D. Results to be published.

Catalytic Reduction of Hydrazine to Ammonia
Inorganic Chemistry, Vol. 35, No. 13, 1996 4045
Figure 5. Proposed pathway for the catalytic reduction of N2H4 to NH3 by cluster [(H2cit)MoFe3S4Cl3]2-, 7, with emphasis on a possible H+
delivery role of the carboxylate function. The intra- and intermolecular electron transfer steps arbitrarily are depicted by changes in the formal
oxidation level of the Mo atom.
ized.50 Many show exceptional reactivity upon reaction with
strong acids, e.g. HX (X ) Cl, Br), with N2H4 being the
product,45 but only a small number of them yield NH3, usually
in very low yields.46,50 Proton attack on the µ-N2 (or on the
µ-HNdNH at a subsequent step51,52) ligand can be easily
visualized given the π-electron density of the bridging unit. Once
the µ-N2H4 level is reached however, protonation is impossible
because neither π-electron density nor lone pairs on the N atoms
are available any longer. As a consequence, reduction of µ-N2
ceases at the hydrazine level.
intermediates have been isolated and some have been structurally
characterized.11c,20b,54
A proposed pathway (Figure 6) for the reduction of N2 to
NH3 consistent with the available data on the reduction of N2
and N2H4 in model complexes, the structure of the nitrogenase
cofactor, and previously suggested mechanisms7 is as follows:
(a) The N2 molecule initially coordinates to one of the Fe4 square
faces7b of the Fe6 prismatic unit of the cofactor (step 1). (b)
The bridging N2 undergoes proton attack followed by a two-
electron reduction to diazene. Subsequent protonation followed
by reduction leads to hydrazine (or hydrazide).45 Protonation
steps during the first two steps of reduction (N2 f HNdNH f
NHx-NHy55 ) may be facilitated by the µ2-S ligands in the center
of the FeMo-co, perhaps forming µ-S‚‚‚H species (steps 2 and
3). (A N-H‚‚‚S hydrogen bond between a histidine-195
imidazole and one of the central, doubly bridging, S ligands of
the FeMo-co was recently reported.56). (c) At the hydrazine
stage, the bridging N2H4 cannot be protonated and further
reduced due to the unavailability of lone pairs on the N atoms.
It is likely, then, that the hydrazine molecule either changes to
a monodentate binding mode or migrates from the Fe site(s) to
the Mo site and coordinates to it in a terminal fashion (step 4).
In mononuclear complexes, terminal N2 ligands are reduc-
tively transformed to terminally bound hydrazines with lone
pairs on the unbound nitrogen atoms available for protonation
and subsequent reduction. As a result, the N2 ligands in these
complexes are reduced readily to NH3.21e,50d,53 In these reduc-
tions, several metal-hydrazido (N2Hxn-; x ) 1-3; n ) 1-3)
(50) (a) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. ReV. 1978, 589.
(b) Anderson, S. N.; Richards, R. L.; Hughes, D. L. J. Chem. Soc.,
Chem. Commun. 1982, 921. (c) Chatt, J.; Leigh, G. J. Q. ReV. Chem.
Soc. 1972, 1, 121. (d) Henderson, R. A.; Leigh, G. J.; Pickett, C. J.
AdV. Inorg. Chem. Radiochem. 1983, 27, 197.
(51) In fact, such a pathway has been proposed to take place upon
protonation of the thermally unstable [Fe(iPr)(PPh3)2]2(µ-N2) complex.
The final product of this transformation is N2H4: Yu, G.; Borodko,
M. O.; Boitman, L. M.; Kachapina, A. E.; Shilov, A. E.; Yu, L. Chem.
Commun. 1971, 1185.
(52) Metal-diazene complexes are described in the literature: (a) Smith,
M. R., III; Cheng, T.-Y.; Hillhouse, G. L. J. Am. Chem. Soc. 1993,
115, 8638. (b) Cheng, T.-Y.; Peters, J. C.; Hillhouse, G. L. J. Am.
Chem. Soc. 1994, 116, 204. (c) This article reports the crystal structure
of a binuclear Fe complex having an end-on µ-NHNH: Sellman, D.;
Soglowek, W.; Knock, F.; Moll, M. Angew. Chem., Int. Ed. Engl.
1989, 28, 1271.
(53) (a) Sellman, D. In ref 1a, p 332. (b) Hidai, M.; Mizobe, Y. In ref 1a,
p 346. (c) George, T. A.; DeBord, J. R. D. In ref 1a, p 363. (d) George,
T. A.; Rose, D. J.; Chang, Y.; Chen, Q.; Zubieta, J. Inorg. Chem.
1995, 34, 1295 and references therein.
(54) (a) George, T. A.; Kaul, B. B.; Chen, Q.; Zubieta, J. Inorg. Chem.
1993, 32, 1706. (b) Vogel. S.; Huttner, G.; Zsolnai, L.; Emmerich, C.
Naturforsch. 1993, 48B, 353.
(55) For a study on protonation sites in metal-hydrazido complexes, see:
Wagenknecht, P. D.; Norton, J. R. J. Am. Chem. Soc. 1995, 1167,
1841.

4046 Inorganic Chemistry, Vol. 35, No. 13, 1996
Demadis et al.
Figure 6. Possible mechanism for the binding, activation, and reduction of N2 to NH3 by the FeMo-co based on a combination of available
literature data and the present study. Diazene (HNdNH) and hydrazine (NH2NH2) as possible intermediates are also shown.
These events possibly are accompanied by positional, confor-
mational, and electronic changes in the vicinity of and/or within
the FeMo-co. The protein environment surrounding the FeMo-
co plays an important role in the substrate reduction process,
since the isolated FeMo-co does not reduce N2. Other substrates
such as acetylene and cyclopropene are reduced by the cofactor,
albeit at a small fraction (8%) of the enzymatic rate.57,58 (d)
During the last reduction step the Mo-bound homocitrate ligand
may participate in proton transfers (see also the proposed
pathway in Figure 5) to the substrate (hydrazine).
Abiological nitrogen fixation systems that involve carboxylate
coligands have been reported. Early reports include studies by
Folkesson and Larsson59 on N2 reduction to N2H4 utilizing VII
species in the presence of R,ω-dicarboxylates (-OOC(CH2)n-
COO-, 3 e n e 10) and the photocatalytic reduction of N2 to
NH3 by Ru(EDTA) complexes.60 Similarly, RuIII(EDTA)(H2O)
compounds catalyze the reduction of N2H4 to NH3 in aqueous
acidic solution.20a The above studies emphasized the positive
effects of the polycarboxylate ligands in the process. However,
their possible role as proton transfer devices was not invoked.
In a recent report, N2 binding to [MH2(η2-OOCCH3)(dppe)]+
(M ) Mo, W) was reported to require initial dissociation of
the acetate ligand.61
bind N2, and dihydrogen, H2, is displaced when N2 binds to the
enzyme.62 The participation of the Mo-bound core µ3-S ligands
in the FeMo-co during the last protonation steps is possible and
may play an important role. Indeed, the presence of hydrogen
bonds between the µ3-S ligands in the [Fe4S4]2+ centers and
N-H groups (from neighboring amino acids) is well docu-
mented.63
Summary
The principal findings and conclusions of this investigation
are the following: (a) The [MoFe3S4]3+ cores catalytically
reduce N2H4 to NH3 with Lut‚HCl as the proton source and
Co(Cp)2 as the reducing agent. (b) The Mo-bound carboxylate
ligands under acidic conditions partially dissociate and allow
coordination of hydrazine to the Mo atom. They appear to
promote rather than hinder the catalytic reduction of hydrazine
to ammonia. (c) Control experiments with the Fe4S4 cubanes
and inhibition studies, with ligands that are known to coordinate
to the Mo atom, have shown total or pronounced loss of catalytic
activity. These results support the suggestion that reduction
occurs at the Mo site. (d) The possible involvement of the
carboxylate functions as “built-in”, local, H+ delivery agents
may play an important role in catalysis. The homocitrate
molecule in the FeMo-co may very well function in a similar
manner, liberating a site on the coordinatively saturated Mo and
subsequently providing the protons to the substrate(s).
The pathway proposed in Figure 6 does not include or attempt
to explain the H2 evolution concomitant with N2 reduction (eq
1). It has been proposed62 that binding of a H+ to the FeMo-
co (possibly involving µ3-S ligands) and its 2e- reduction to
H-, can be envisioned as a step that prepares the enzyme to
Acknowledgment. Support of this research by a grant from
the National Institutes of Health (GM-33080) is gratefully
acknowledged.
(56) (a) Kim, C.-H.; Newton, W. E.; Dean, D. R. Biochemistry 1995, 34,
2798. (b) DeRose, V. J.; Kim, C.-H.; Newton, W. E.; Dean, D. R.;
Hoffman, B. M. Biochemistry 1995, 34, 2809. (c) Thomann, H.;
Bernardo, M.; Newton, W. E.; Dean, D. R. Proc. Natl. Acad. Sci.
U.S.A. 1991, 88, 6620.
(57) Shah, V. K.; Chisnell, J. R.; Brill, W. J. Biochim. Biophys. Res.
Commun. 1978, 81, 232.
(58) McKenna, C. E.; McKenna, M. C.; Huang, C. W. Proc. Natl. Acad.
Sci. U.S.A. 1979, 76, 4773.
(59) Folkesson, B.; Larsson, R. Acta Chem. Scand. Ser. A 1979, 33, 347.
(60) Rao, N. N. J. Mol. Catal. 1994, 93, 23.
IC960098B
(62) Thorneley, R. N.; Lowe, D. J. Biochem. J. 1984, 224, 895.
(63) (a) Valentine, J. S.; Sheridan, R. P.; Allen, L. C.; Kahn, P. C. Proc.
Natl. Acad. Sci. U.S.A. 1979, 76, 1009. (b) Adman, E. T.; Watenpaugh,
K. D.; Jensen, L. H. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 4854. (c)
Smith, W. W.; Burnett, R. M.; Darling, G. D.; Ludwig, M. L. J. Mol.
Biol. 1977, 117, 195. (d) Sheridan, R. P.; Allen, L. C. Chem. Phys.
Lett. 1980, 69, 600. (e) Merritt, E. A.; Stout, G. H.; Turley, S.; Sieker,
L. C.; Jensen, L. H.; Orme-Johnson, W. H. Acta Crystallogr. 1993,
B49, 272.
(61) Hughes, D. L.; Ibrahim, S. K.; Pickett, C. J.; Querme, G.; Laueman,
A; Talarmin, J.; Queiros, A.; Fonseca, A. Polyhedron 1994, 13, 3341.