Factors influencing the thermodynamics of zinc alkoxide formation by alcoholysis of the terminal hydroxide complex, [TpBut,Me]ZnOH: An experimental and theoretical study relevant to the mechanism of action of liver alcohol dehydrogenase

Article abstract of DOI:10.1021/ja002286d

The factors that influence the formation of a tetrahedral alkoxide complex related to a critical intermediate of the catalytic cycle of liver alcohol dehydrogenase have been probed by a combined experimental and computational investigation of the reactions of the tris(pyrazolyl)hydroborato zinc hydroxide complexes [Tp^RR′]ZnOH with alcohols. The study demonstrates that zinc alkoxide formation is electronically favored by incorporation of electron-withdrawing substituents in the alcohol but is sterically disfavored for bulky alkoxides. A computational analysis indicates that these trends are a result of homolytic Zn-OR and Zn-OAr BDEs being more sensitive to the nature of R and Ar than are the corresponding H-OR and H-OAr BDEs. Thus, electron-withdrawing substituents increase Zn-OAr bond energies to a greater extent than H-OAr bond energies, while bulky substituents decrease Zn-OR bond energies to a greater extent than H-OR bond energies. With the exception of derivatives of acidic alcohols (e.g., nitrophenol), the zinc alkoxide complexes [Tp^RR′]ZnOR are very unstable toward hydrolysis. This hydrolytic instability of simple zinc alkoxide complexes suggests that the active site environment of LADH plays an important role in stabilizing the alkoxide intermediate, possibly via hydrogen-bonding interactions.

Full text of DOI:10.1021/ja002286d

J. Am. Chem. Soc. 2000, 122, 12651-12658
12651
Factors Influencing the Thermodynamics of Zinc Alkoxide Formation
t
by Alcoholysis of the Terminal Hydroxide Complex, [TpBu ,Me]ZnOH:
An Experimental and Theoretical Study Relevant to the Mechanism
of Action of Liver Alcohol Dehydrogenase
Catherine Bergquist, Hannah Storrie, Lawrence Koutcher, Brian M. Bridgewater,
Richard A. Friesner,* and Gerard Parkin*
Contribution from the Department of Chemistry, Columbia UniVersity, New York, New York 10027
ReceiVed June 26, 2000
Abstract: The factors that influence the formation of a tetrahedral alkoxide complex related to a critical
intermediate of the catalytic cycle of liver alcohol dehydrogenase have been probed by a combined experimental
and computational investigation of the reactions of the tris(pyrazolyl)hydroborato zinc hydroxide complexes
[TpRR′]ZnOH with alcohols. The study demonstrates that zinc alkoxide formation is electronically favored by
incorporation of electron-withdrawing substituents in the alcohol but is sterically disfavored for bulky alkoxides.
A computational analysis indicates that these trends are a result of homolytic Zn-OR and Zn-OAr BDEs
being more sensitive to the nature of R and Ar than are the corresponding H-OR and H-OAr BDEs. Thus,
electron-withdrawing substituents increase Zn-OAr bond energies to a greater extent than H-OAr bond
energies, while bulky substituents decrease Zn-OR bond energies to a greater extent than H-OR bond energies.
With the exception of derivatives of acidic alcohols (e.g., nitrophenol), the zinc alkoxide complexes [TpRR′]-
ZnOR are very unstable toward hydrolysis. This hydrolytic instability of simple zinc alkoxide complexes suggests
that the active site environment of LADH plays an important role in stabilizing the alkoxide intermediate,
possibly via hydrogen-bonding interactions.
Introduction
Scheme 1
Zinc alkoxide species are proposed to be essential intermedi-
ates in the oxidation of alcohols catalyzed by liver alcohol
dehydrogenase (LADH).1-3 Such species are generated by
displacement of the water molecule at the tetrahedral active site,
accompanied by proton transfer (Scheme 1). An understanding
of the factors that influence the stability of tetrahedral zinc
alkoxide complexes with respect to hydrolysis is, therefore,
critical to understanding the mechanism of action of LADH. In
this paper, we report a combined experimental and computa-
tional study to quantify the effect of varying the alkoxide group
on equilibria involving alcoholysis of a tetrahedral zinc hy-
droxide complex. Specifically, the results demonstrate that
incorporation of electron-withdrawing substituents promotes the
formation of a tetrahedral alkoxide species.
(1) (a) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. ReV. 1996,
96, 2239-2314. (b) Lipscomb, W. N.; Stra¨ter, N. Chem. ReV. 1996, 96,
2375-2433. (c) Kimura, E.; Koike, T.; Shionoya, M. Struct. Bond. 1997,
89, 1-28.
(2) For studies modeling aspects of LADH structure and chemistry, see:
(a) Bergquist, C.; Parkin, G. Inorg. Chem. 1999, 38, 422-423. (b) Kimblin,
C.; Hascall, T.; Parkin, G. Inorg. Chem. 1997, 36, 5680-5681. (c) Shoner,
S. C.; Humphreys, K. J.; Barnhart, D.; Kovacs, J. A. Inorg. Chem. 1995,
34, 5933-5934. (d) Kimura, E.; Shionoya, M.; Hoshino, A.; Ikeda, T.;
Yamada, Y. J. Am. Chem. Soc. 1992, 114, 10134-10137. (e) Engbersen,
J. F. J.; Koudijs, A.; van der Plas, H. C. J. Org. Chem. 1990, 55, 3647-
3654. (f) Kellogg, R. M.; Hof, R. P. J. Chem. Soc., Perkin Trans. 1 1996,
1651-1657. (g) Mu¨ller, B.; Schneider, A.; Tesmer, M.; Vahrenkamp, H.
Inorg. Chem. 1999, 38, 1900-1907.
(3) For computational studies concerned with LADH structure and
mechanism, see: (a) Ryde, U. Int. J. Quantum Chem. 1994, 52, 1229-
1243. (b) Ryde, U. J. Comput. Aid. Mol. Des. 1996, 10, 153-164. (c) Cini,
R. J. Biomol. Struct. Dyn. 1999, 16, 1225-1237. (d) Ryde, U. Eur. Biophys.
J. 1996, 24, 213-221. (e) Agarwal, P. K.; Webb, S. P.; Hammes-Schiffer,
S. J. Am. Chem. Soc. 2000, 122, 4803-4812.
Results and Discussion
The active site of LADH is composed of a tetrahedral zinc
center which is attached to the protein backbone by the nitrogen
and sulfur donors of one histidine and two cysteine residues
(Scheme 1). Synthetic analogues with this coordination environ-
ment that also feature the catalytically important aqua (or
hydroxide) entity are, however, unknown.2b Furthermore, re-
gardless of the nature of the supporting ligands, the formation
of mononuclear tetrahedral zinc alkoxide complexes from zinc
hydroxide derivatives is not well documented. Two notable
examples, however, are (i) the synthesis of alkoxide complexes
by the reactions of tris(pyrazolyl)hydroborato zinc hydroxide
10.1021/ja002286d CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/02/2000

12652 J. Am. Chem. Soc., Vol. 122, No. 51, 2000
Bergquist et al.
Scheme 2
t
Figure 1. Molecular structure of [TpBu ,Me]ZnOC6H4NO2.
The equilibrium constants (KAr) for the alcoholysis reactions
t
of [TpBu ,Me]ZnOH with p-XC6H4OH (Scheme 2) have been
1
determined by H NMR spectroscopy. For the reaction with
p-MeC6H4OH, the equilibrium constant in THF-d8 was deter-
mined directly, but for all other derivatives, the equilibrium
constants were obtained via a thermodynamic cycle involving
the experimentally determined equilibrium constant for alkoxide
t
exchange between [TpBu ,Me]ZnOC6H4Me and p-XC6H4OH. The
equilibrium constant data are listed in Table 2.
A Hammett plot6 of log K versus σ gives a good linear
correlation with a F value of 2.8 (Figure 2), indicating that zinc
aryloxide formation is strongly favored by electron-withdrawing
substituents.7 For comparison, the equilibrium is substantially
more sensitive to electronic effects than is the exchange of
(Me2C2)Re(O)(OPh) with ArOH, which is characterized by a F
value of only 0.71 for a plot of log K versus σ.8 To gain more
insight into the factors influencing the formation of the aryl-
oxide complexes, we have studied the thermodynamics of the
alcoholysis reactions using computational methods.
Significantly, DFT calculations (B3LYP) performed using
Jaguar9 are in excellent agreement with the experimental results.
Thus, the calculated electronic energy differences are only
∼2.5 kcal mol-1 less exothermic than the experimental enthalpy
changes (Table 2). More importantly than the difference being
small, it is also roughly constant throughout the series, such
that the correlation between the experimental and calculated
results is excellent (Figure 3).
derivatives [TpRR′]ZnOH with ROH,2a,4 and (ii) the intramo-
lecular generation of alkoxide complexes in which the alcohol
functionality is pendant to the macrocyclic ligand that binds
the zinc.5
Since mononuclear zinc hydroxide complexes with a [SSN]-
t
ZnOH composition are unknown, we have selected [TpBu ,Me]-
ZnOH for studies designed to evaluate the factors responsible
for influencing alcoholysis equilibria, recognizing that the
coordination motif is not identical to that of LADH. We have
previously reported that the equilibrium constants for the
t
formation of the alkoxide complexes [TpBu ,Me]ZnOR from
t
[TpBu ,Me]ZnOH and ROH (R ) Me, Et, Pri, But) are highly
dependent upon the nature of R, decreasing markedly in the
following sequence: Me > Et > Pri > But.2a Since this trend
is presumably the result of a composite steric and electronic
influence, we have sought to separate these components by
studying a series of para-substituted phenols, p-XC6H4OH, for
which electronic substituent parameters (e.g., Hammett σ
constants) are available.
The effect of a substituent on the energetics of the reactions
t
between [TpBu ,Me]ZnOH and ArOH is dictated by its influence
t
[TpBu ,Me]ZnOH reacts with a variety of p-XC6H4OH deriva-
t
on the respective [TpBu ,Me]Zn-OAr and H-OAr BDEs. In this
t
tives to give the phenoxide complexes [TpBu ,Me]ZnOC6H4X as
regard, the thermodynamic cycle illustrated in Scheme 3
demonstrates that the substituent effect can be conceptually
discussed equally well in terms of the influence of substituents
components in an equilibrium mixture. The latter complexes
may be independently synthesized by the irreversible reactions
t
of the hydride complex [TpBu ,Me]ZnH with p-XC6H4OH (Scheme
t
on either homolytic or heterolytic [TpBu ,Me]Zn-OAr and
t
2). The molecular structures of [TpBu ,Me]ZnOC6H4X (X ) But,
H-OAr BDEs.10 However, since homolytic ArO-H BDEs have
been discussed to a greater extent in the literature,11 and also
the effect of solvation on homolytic BDEs is less than that on
heterolytic BDEs, it is more useful to focus attention on the
effects of the alcoholysis reactions in terms of homolytic BDEs.
The experimental and computational studies indicate that
electron-withdrawing substituents favor the formation of the zinc
C(O)Me, NH2, NO2) have been determined by X-ray diffraction,
thereby confirming their mononuclear nature, as illustrated for
t
[TpBu ,Me]ZnOC6H4NO2 in Figure 1. Selected bond lengths and
t
angles for the [TpBu ,Me]ZnOC6H4X derivatives are summarized
in Table 1.
(4) Walz, R.; Weis, K.; Ruf, M.; Vahrenkamp, H. Chem. Ber./Recl. 1997,
130, 975-980.
(5) (a) Kimura, E.; Kodama, Y.; Koike, T.; Shiro, M. J. Am. Chem. Soc.
1995, 117, 8304-8311. (b) Kimura, E.; Nakamura, I.; Koike, T.; Shionoya,
M.; Kodama, Y.; Ikeda, T.; Shiro, M. J. Am. Chem. Soc. 1994, 116, 4764-
4771. (c) Koike, T.; Kajitani, S.; Nakamura, I.; Kimura, E.; Shiro, M. J.
Am. Chem. Soc. 1995, 117, 1210-1219.
(6) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165-195.
(7) A plot based on σ- has a slope of 2.0.
(8) Erikson, T. K. G.; Bryan, J. C.; Mayer, J. M. Organometallics 1988,
7, 1930-1938.
(9) Jaguar 3.5 and 4.0, Schro¨dinger, Inc., Portland, OR, 1998.

Thermodynamics of Zinc Alkoxide Formation
J. Am. Chem. Soc., Vol. 122, No. 51, 2000 12653
Table 1. Selected Bond Lengths and Angles
t
t
t
t
[TpBu ,Me]ZnOC6H4C(O)Me
[TpBu ,Me]ZnOC6H4NO2
[TpBu ,Me]ZnOC6H4NH2
[TpBu ,Me]ZnOC6H4But
Zn-Nav/Å
Zn-O/Å
Zn-O-C/deg
2.04
1.827(2)
168.4(2)
2.04
1.849(3)
157.5(3)
2.05
2.05
1.845(2), 1.849(2)a
146.6(2), 147.5(3)a
1.846(2), 1.842(2)a
145.0(2), 152.7(2)a
a Data for two crystallographically independent molecules.
Table 2. Equilibrium Constant and Enthalpy Data for Alcoholoysis
t
Reactions of [TpBu ,Me]ZnOH
∆Hexpt
/
∆Ecalc/
ROH
K (300 K)
kcal mol-1 a
kcal mol-1
But b
∼10-8
∼8
8.86
7.14
4.37
4.03
1.47
1.06
1.64
1.17
-0.44
-1.04
-1.42
-3.57
Pri b
3(1) × 10-5
9(2) × 10-4
1.4(2) × 10-3
4.2(9)
3.5(2)
Etb
1.5(2)
1.2(1)
Meb
C6H4OMe
C6H4But
C6H4Me
C6H5
-0.9(1)
-0.9(1)
-1.0(1)
-1.4(1)
-2.7(1)
-3.3(1)
-3.4(1)
-4.9(1)
4.8(10)
4.9(10)
1.0(2) × 101
9(2) × 101
2.7(6) × 102
3.1(6) × 102
3.5(8) × 103
C6H4I
C6H4CO2Me
C6H4COMe
C6H4NO2
a ∆H values are calculated by assuming values of -9 eu for reac-
tions involving ROH and 0 eu for reactions involving ArOH. These
values are based on measured values for the reactions between (i)
Figure 3. Correlation of experimental ∆H and calculated ∆E values
for the various alcoholysis equilibria.
t
t
[TpBu ,Me]ZnOH and MeOH in MeOH solvent and (ii) [TpBu ,Me]ZnOH
Scheme 3
and p-MeC6H4OH in THF. b Data taken from ref 2a.
t,
Figure 2. Hammett plot of log K versus σ for the reaction of [TpBu Me]-
ZnOH with XC6H4OH.
aryloxide compound because they increase the homolytic
ization of the electron density from the oxygen atom; destabi-
lization of the product radical ArO• by electron-withdrawing
substituents serves to reinforce this effect (Figure 5).11,12,16
Electron-donating para substituents correspondingly decrease
the H-OAr BDE, but the effect is principally due to stabilization
t
Zn-OAr BDEs in [TpBu ,Me]ZnOAr to a greater extent than the
corresponding H-OAr BDEs.11,12 Thus, rather than exhibiting
the 1:1 correlation between M-X and H-X bond energies that
has been reported for certain other systems,13 the Zn-OAr BDE
is substantially more sensitive to the para substituent than is
the H-OAr BDE, by a factor of 1.4:1 (Figure 4).14
Previous studies indicate that electron-withdrawing para
substituents increase homolytic ArO-H BDEs as a result of
preferential stabilization15 of ArOH due to increased delocal-
(13) See, for example: (a) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.;
Tam, W.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 1444-1456. (b)
Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163-1188. (c) Bryndza,
H. E.; Domaille, P. J.; Tam, W.; Fong, L. K.; Paciello, R. A.; Bercaw, J.
E. Polyhedron 1988, 7, 1441-1452.
(14) The data in Figure 4 correspond to the calculated values. The
experimental ratio is also 1.4:1, but for two fewer compounds since data
are not available. Specifically, D(Zn-OAr) ) 1.40D(H-OAr) - D(Zn-
(10) Heterolytic M-OAr values are related to the homolytic values by
the electron affinity of ArO• and the ionization potential of M• (M ) H,
t
[TpBu ,Me]Zn); i.e., D(M-OAr)Het ) D(M-OAr)Homo - EA(OAr•) +
OH) - 152 kcal mol-1
.
IP(M•).
(15) It is recognized that the word “stabilization” is being used loosely
in this context since it is inappropriate to compare energies of molecules
with different compositions. See ref 11.
(11) For a recent review of homolytic ArO-H BDEs, see: Borges dos
Santos, R. M.; Martinho Simo˜es, J. A. J. Phys. Chem. Ref. Data 1998, 27,
707-739.
(12) For other calculations on ArO-H BDEs, see: (a) Wu, Y.-D.; Lai,
D. K. W. J. Org. Chem. 1996, 61, 7904-7910. (b) Brinck, T.; Haeberlein,
M.; Jonsson, M. J. Am. Chem. Soc. 1997, 119, 4239-4244. (c) Bordwell,
F. G.; Zhang, X.-M.; Satish, A. V.; Cheng, J.-P. J. Am. Chem. Soc. 1994,
116, 6605-6610.
(16) For calculations on ROH, see: (a) Damrauer R. J. Am. Chem. Soc.
2000, 122, 6739-6745. (b) Safi, B.; Choho, K.; De Proft, F.; Geerlings, P.
J. Phys. Chem. A 1998, 102, 5253-5259. (c) De Proft, F.; Langenaeker,
W.; Geerlings, P. Tetrahedron 1995, 51, 4021-4032. (d) Tupper, K. J.;
Gajewski, J. J.; Counts, R. W. J. Mol. Struct (THEOCHEM) 1991, 236,
211-217.

12654 J. Am. Chem. Soc., Vol. 122, No. 51, 2000
Bergquist et al.
Figure 6. Variation of calculated homolytic Zn-OAr and H-OAr
BDEs as a function of the σ value of the para substituent.
Figure 4. Correlation of calculated homolytic Zn-OAr and H-OAr
BDEs.
differential in relative Zn-OAr and H-OAr bond energies than
do electron-donating substituents. Thus, electron-donating sub-
stituents (OMe, But, Me) influence both Zn-OAr and H-OAr
bond energies to a similar degree (Figures 6 and 7); NH2, the
most strongly electron-donating substituent, is the only one for
which a notable differential between relative Zn-OAr and
H-OAr BDEs exists (Figures 6 and 7). For the homolytic bond
dissociation energies, this result is in accord with the afore-
mentioned notion that an electron-donating substituent princi-
pally modifies the H-OAr bond energy by influencing the
energy of the ArO• radical as opposed to exerting a ground-
state effect on ArOH.11,12 Thus, for the extreme case that a
substituent influences only the radical and not the ground state,
its influence on a M-OAr BDE would be independent of the
nature of M (M ) Zn, H).
Figure 5. Conceptual influence of electron-donating and electron-
withdrawing substituents on XC6H4O-H homolytic BDEs (see refs 11
and 12a).
The above results suggest that the influence of electron-
withdrawing substituents on alcoholysis equilibria is associated
with preferential stabilization of the partial negative charge on
of the ArO• radical, with destabilization of ArOH contributing
only to a small degree.
t
the oxygen atom of [TpBu ,Me]ZnOAr as compared to that of
ArOH. This interpretation is in accord with that provided by
Bergman, Andersen, and Holland to rationalize the influence
of electron-withdrawing substituents on equilibria involving
transition metal amido, aryloxo, and alkoxo complexes.17,18
However, the possibility that electron-withdrawing substituents
increase the homolytic Zn-OAr bond dissociation energy by
minimizing “filled-filled” π repulsions between the oxygen
lone pairs and the electrons associated with the metal center
should also be considered.19 Bergman, Andersen, and Holland,
however, invoked “Ockham’s razor”20 and concluded that if
polarization effects alone may explain the data, there was no
reason to incorporate “filled-filled” repulsions into the argu-
ment. In this regard, it should be noted that “filled-filled” π
repulsions are clearly not possible for ArOH, but yet the H-OAr
BDE is still strongly influenced by the para substituent.
Support for the proposal that the greater sensitivity of the
Zn-OAr versus H-OAr bond is due to the greater polarity of
the Zn-OAr bond is provided by application of the dual-
parameter E-C model.21 Specifically, this model separates a
substituent effect into electrostatic and covalent components via
Within the construct that rationalizes the increase in homolytic
H-OAr BDE in terms of the ability of an electron-withdrawing
substituent to stabilize electron density on the oxygen atom,
the greater sensitivity of the Zn-OAr BDE may be attributed
to the Znδ+-OArδ- bond being more polar than the Hδ+
-
OArδ- bond. For example, the Mulliken charges on the oxygen
t
atom in [TpBu ,Me]ZnOAr are in the range -0.73 to -0.77,
whereas those in ArOH are in the range -0.39 to -0.41. An
electron-withdrawing substituent would, therefore, exert a
greater influence in stabilizing the partial negative charge on
t
the oxygen atom in [TpBu ,Me]ZnOAr than that in ArOH, with
the result that the Zn-OAr BDE is more sensitive to the
substituent than is the H-OAr BDE, as illustrated in Figure 6.
Alternatively, in terms of arguments based on heterolytic bond
dissociation energies, the experimental and computational
studies indicate that electron-withdrawing substituents decrease
the heterolytic Zn-OAr bond dissociation energies to a lesser
extent than those for the H-OAr bond energies (Table 3 and
Figure 7). Essentially, electron-withdrawing substituents de-
crease both heterolytic Zn-OAr and H-OAr bond energies
because they stabilize the ArO- anion to a greater extent than
M-OAr (M ) Zn, H), but the effect is less for the Zn-OAr
bond because it is more polar (i.e., the aryloxide moiety is
closer to OAr-); indeed, the heterolytic Zn-OAr BDEs are
∼200 kcal mol-1 lower than the corresponding heterolytic
H-OAr BDEs (Table 3).
(17) Holland, P. L.; Andersen, R. A.; Bergman, R. G. Comments Inorg.
Chem. 1999, 21, 115-129.
(18) Holland, P. L.; Andersen, R. A.; Bergman, R. G.; Huang, J.; Nolan,
S. P. J. Am. Chem. Soc. 1997, 119, 12800-12814.
(19) Caulton, K. G. New. J. Chem. 1994, 18, 25-41.
(20) Rodr´ıguez-Ferna´ndez, J. L. EndeaVour 1999, 23, 121-125.
(21) (a) Drago, R. S. Applications of Electrostatic-CoValent Models in
Chemistry; Surfside Scientific Publishers: Gainesville, FL, 1994. (b) Vogel,
G. C.; Drago, R. S. J. Chem. Educ. 1996, 73, 701-707. (c) Drago, R. S.;
Dadmun, A. P. J. Am. Chem. Soc. 1993, 115, 8592-8602.
It is also important to emphasize that electron-withdrawing
substituents (relative to hydrogen) create a more significant

Thermodynamics of Zinc Alkoxide Formation
J. Am. Chem. Soc., Vol. 122, No. 51, 2000 12655
Table 3. Homolytic and [Heterolytic] Bond Dissociation Energies
t
D(RO-H)expt
/
D(RO-H)calc
/
D([TpBu ,Me]Zn-OR)calc
/
D([Tp]Zn-OR)calc/
R
kcal mol-1
kcal mol-1
kcal mol-1
kcal mol-1
But
Pri
Et
Me
H
C6H4NH2
C6H4OMe
C6H4But
C6H4Me
C6H5
105.1a [375.9]d
104.7a [376.7]d
104.2a [378.6]d
104.4a [381.7]d
119b [390.7]b
79.2c [352.5]e
83.5c [350.4]e
87.1c [348.5]e
86.8c [350.4]e
88.7c [349.2]e
88.5c [-]
105.9 [385.6]
109.0 [389.9]
108.1 [392.5]
108.3 [396.8]
122.4 [419.8]
83.2 [364.7]
86.3 [362.1]
90.9 [359.4]
90.5 [360.9]
92.8 [360.1]
91.8 [35.7]
80.1 [149.3]
84.9 [155.3]
86.7 [160.7]
87.3 [165.4]
105.2 [192.2]
63.5 [134.4]
67.8 [133.2]
72.9 [130.9]
71.9 [131.8]
74.7 [131.5]
75.3 [124.7]
78.6 [119.1]
78.9 [117.9]
83.6 [111.0]
83.4 [172.7]
83.5 [174.0]
82.8 [176.7]
82.9 [181.0]
99.7 [206.7]
62.3 [187.0]
66.9 [191.5]
71.8 [196.5]
71.4 [196.1]
74.2 [198.8]
74.7 [199.3]
78.0 [202.7]
78.6 [203.3]
82.8 [207.4]
C6H4I
C6H4CO2Me
C6H4COMe
C6H4NO2
-[337.2]e
94.5 [345.5]
94.4 [343.9]
97.0 [334.9]
90.9c [335.6]e
94.7c [327.9]e
a McMillen, D. F.; Golden, D. M. Ann. ReV. Phys. Chem. 1982, 33, 493-532. b Berkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys. Chem. 1994,
98, 2744-2765. c Reference 11. d Ervin, K. M.; Gronert, S.; Barlow, S. E.; Gilles, M. K.; Harrison, A. G.; Bierbaum, V. M.; DePuy, C. H.; Lineberger,
W. C.; Ellison, G. B. J. Am. Chem. Soc. 1990, 112, 5750-5759. e Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D.; Mallard,
W. G. J. Phys. Chem. Ref. Data 1998, 17, Suppl. 1.
Table 4. Comparison of the Calculated Energetics of the
t
Alcoholysis Reactions of [TpBu ,Me]ZnOH and [Tp]ZnOH
t
∆Ecalc{[TpBu ,Me]Zn}/ ∆Ecalc{[Tp]Zn}/
∆∆Ecalc/
ROH
kcal mol-1
kcal mol-1
kcal mol-1
But
Pri
Et
8.86
7.14
4.37
4.03
2.77
1.47
1.06
1.64
1.17
-0.44
-1.04
-1.42
-3.57
0.06
2.99
2.83
8.80
4.15
1.54
1.13
4.33
4.56
4.44
5.06
4.99
4.90
4.92
5.23
4.69
Me
2.90
C6H4NH2
C6H4OMe
C6H4But
C6H4Me
C6H5
-1.56
-3.09
-3.38
-3.42
-3.82
-5.34
-5.96
-6.65
-8.26
C6H4I
C6H4CO2Me
C6H4COMe
C6H4NO2
Figure 7. Variation of calculated heterlytic Zn-OAr and H-OAr
BDEs as a function of the σ value of the para substituent.
the expression log(Kx/KH) ) dE∆Ex + dC∆Cx, where ∆Ex and
∆Cx are the electrostatic and covalent substituent constants, and
dE and dC are the electrostatic and covalent counterparts of the
Hammett F value. The derived values, dE ) -8.5 and dC
)
1.3, indicate that the electrostatic component dominates the
equilibrium and thereby suggest that electron-withdrawing
substituents increase the Zn-OAr BDE to a greater extent than
the H-OAr BDE (see Supporting Information) due to the
greater polarity of the former bond. For comparison, the
equilibrium between Cp*Ni(PEt3)NHTol and XC6H4NH2 is
characterized by the values dE ) -18.4 and dC ) 1.8.17,18,22
We have also carried out calculations on the energetics of
t
the reactions of [TpBu ,Me]ZnOH with aliphatic alcohols (MeOH,
EtOH, PriOH, ButOH), and the results are in close agreement
with the experimental results, with zinc alkoxide formation
becoming increasingly disfavored in the following sequence:
Figure 8. Influence of the tris(pyrazolyl)hydroborato ligand on the
calculated energetics of the alcoholysis equilibria.
KMe > KEt > KPr > KBu (see Figure 3 and Table 2).2a The
calculations indicate that this trend is a result of the homolytic
Zn-OR BDE decreasing more rapidly upon increasing the bulk
of R than does the corresponding H-OR BDE (see Table 3).
Presumably, the greater sensitivity of the Zn-OR BDEs is a
reflection of enhanced steric interactions between R and the
groups. Not only do the calculations indicate that reducing
the steric demands of the tris(pyrazolyl)hydroborato ligand
favors the formation of the alkoxide complex (Table 4), but
the calculations also indicate that the difference between the
i
t
t
[TpBu ,Me]ZnOR and [Tp]ZnOR systems becomes more pro-
nounced as the alkoxide group becomes bulkier (Figure 8). This
t
t
tert-butyl substituents of the [TpBu ,Me] ligand. Support for this
trend is a consequence of the [TpBu ,Me]Zn-OR BDE decreasing
suggestion is provided by calculations on the parent [Tp]ZnOR
system, which is devoid of bulky substituents on the pyrazolyl
more rapidly than the [Tp]Zn-OR BDE upon increasing the
size of R.
Reducing the steric demands of the tris(pyrazolyl)hydroborato
ligand also favors the formation of the aryloxide derivatives,
but the effect is relatively insensitive to the nature of the para
(22) For the application of the dual-parameter E-C model to the pKa
values of phenols, see: Bosch, E.; Rived, F.; Rose´s, M.; Sales, J. J. Chem.
Soc., Perkin Trans. 2 1999, 1953-1958.

12656 J. Am. Chem. Soc., Vol. 122, No. 51, 2000
Bergquist et al.
substituent. For example, ∆∆E for the various para-substituted
between the zinc system described here and the vanadium
system is that the vanadium complexes (dipic)VO(OR) are
electronically unsaturated; i.e., in the absence of π-donation,
the vanadium centers possess 12-electron configurations. As
such, the thermodynamics for the vanadium system is strongly
influenced by favorable π-donor interactions.
aryl oxide derivatives span a range of only 0.9 kcal mol-1
,
compared to a range of 7.7 kcal mol-1 for the alkoxide
derivatives (Table 4).
Interestingly, while steric effects clearly play an important
role in promoting the reaction of an alcohol with [Tp]ZnOH
t
over that of [TpBu ,Me]ZnOH, it is important to note that the
t
intrinsic electronic effect of the [TpBu ,Me] ligand versus the [Tp]
Experimental Section
ligand is actually to increase the strength of the Zn-OR bond.
As an illustration for the hydroxide complexes, where steric
effects are minimal, the calculated homolytic Zn-OH BDE of
General Considerations. All manipulations were performed using
a combination of glovebox, high-vacuum, or Schlenk techniques.28
Solvents were purified and degassed by standard procedures. NMR
spectra were recorded on Bruker Avance 300wb DRX, Bruker Avance
300 DRX, Bruker Avance 400 DRX, and Bruker Avance 500 DMX
spectrometers. 1H and 13C chemical shifts are reported in ppm relative
to SiMe4 (δ ) 0) and were referenced internally with respect to the
protio solvent impurity or the 13C resonances, respectively. All coupling
constants are reported in hertz. IR spectra were recorded as KBr pellets
on a Perkin-Elmer Spectrum 2000 spectrophotometer and are reported
in reciprocal centimeters. C, H, and N elemental analyses were measured
t
[TpBu ,Me]Zn-OH (105.2 kcal mol-1) is greater than that of [Tp]-
Zn-OH (99.7 kcal mol-1); in fact, with the exception of the
tert-butoxide derivatives, the Zn-OR BDEs of all other
t
[TpBu ,Me]Zn-OR derivatives are greater than those of the [Tp]-
Zn-OR counterparts (Table 3). The greater strength of the
t
[TpBu ,Me]Zn-OR bonds is presumably due to the ability of the
t
more strongly electron-donating [TpBu ,Me] ligand23,24 to stabilize
the partial positive charge on zinc to a greater degree. Increasing
the steric demands of the substituent on oxygen reduces the
t
using a Perkin-Elmer 2400 CHN elemental analyzer. [TpBu ,Me]ZnH was
prepared by the literature method.2a
t
Zn-OR BDEs of both [TpBu ,Me]Zn-OR and [Tp]Zn-OR, but
t
Synthesis of [TpBu ,Me]ZnOC6H4NH2. A solution of 4-aminophenol
t
(49 mg, 0.45 mmol) and [TpBu ,Me]ZnH (200 mg, 0.41 mmol) in THF
since steric interactions influence the former to a greater degree,
t
the formation of [TpBu ,Me]Zn-OR is inhibited more than that
(10 mL) was heated at 120 °C for 3 days. The reaction mixture was
of [Tp]Zn-OR from the respective hydroxide complexes.
Recent theoretical studies have re-emphasized that alcohol
deprotonation and zinc alkoxide formation takes place prior to
hydride transfer to NAD+; furthermore, the formation of the
alkoxide is important because it facilitates the hydride-transfer
step.3e Thus, zinc alkoxide species play a central role in the
mechanism of action of LADH, with two critical issues being
the kinetic and thermodynamic stability of the zinc alkoxide
with respect to (a) hydrolysis and (b) hydride transfer. In this
paper, we have addressed specifically the issue of the thermo-
dynamic stability of a zinc alkoxide species with respect to
hydrolysis. The study demonstrates that tetrahedral zinc alkox-
ides are very susceptible to hydrolysis, an observation that
potentially mitigates against their viability as reactive intermedi-
ates. Nevertheless, the investigation also reveals that the stability
of the zinc alkoxide complex is very sensitive to both steric
and electronic influences and thereby demonstrates that the
active site environment most likely plays an important role in
promoting the formation of a zinc alkoxide species. For example,
hydrogen bonding may provide a mechanism for stabilizing zinc
alkoxide species, as illustrated by recent studies which suggest
that the benzyl alkoxide ligand forms a hydrogen bond with
the hydroxyl group of Ser48 in horse liver alcohol dehydroge-
nase.3e,25
filtered, and the volatile components were removed from the filtrate in
t
vacuo to give [TpBu ,Me]ZnOC6H4NH2 as a white solid (127 mg, 51%).
Anal. Calcd for C30H46BN7OZn: C, 60.4; H, 7.8; N, 16.4. Found: C,
1
60.7; H, 7.9; N 16.5. H NMR (C6D6): 1.50 [s, 3(C(CH3)3)], 2.08 [s,
3
3(CH3)], 2.73 [s, NH2], 5.63 [s, 3(C3N2H)], 6.71 [d, JH-H ) 9, C6H4
3
(2H)], 7.20 [d, JH-H ) 9, C6H4 (2H)], HB not observed. 13C NMR
1
1
(C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.7 [q, JC-H ) 126,
1
3(C(CH3)3)], 32.0 [s, 3(C(CH3)3)], 103.2 [d, JC-H ) 174, 3(C3N2H)
(1C)], 117.0 [d, 1JC-H ) 146, C6H4 (2C)], 121.5 [d, 1JC-H ) 158, C6H4
(2C)], 136.2 [s, C6H4 (para C)], 144.2 [s, 3(C3N2H) (1C)], 158.2 [s,
C6H4 (ipso C)], 163.9 [s, 3(C3N2H) (1C)]. IR (KBr, cm-1): 3416 (w)
[ν(N-H)], 3345 (w) [ν(N-H)], 2962 (s), 2551 (w) [ν(B-H)], 1609
(m), 1542 (s), 1507 (vs), 1433 (s), 1365 (s), 1342 (m), 1304 (vs), 1250
(s), 1187 (vs), 1069 (s), 1032 (m), 988 (w), 830 (s), 794 (s), 767 (s),
732 (m), 696 (m), 651 (s), 519 (m).
t
Synthesis of [TpBu ,Me]ZnOC6H4OMe. A solution of 4-methoxy-
t
phenol (56 mg, 0.45 mmol) and [TpBu ,Me]ZnH (200 mg, 0.41 mmol)
in benzene (10 mL) was heated at 120 °C for 2 days. The reaction
t
mixture was filtered, and the filtrate was lyophilized giving [TpBu ,Me]-
ZnOC6H4OMe as a white solid (206 mg, 82%). Anal. Calcd for C31H47-
BN6O2Zn: C, 60.8; H, 7.7; N, 13.7. Found: C, 60.7; H, 7.7; N 13.1.
1H NMR (C6D6): 1.49 [s, 3(C(CH3)3)], 2.08 [s, 3(CH3)], 3.58, [s,
(OCH3)], 5.63 [s, 3(C3N2H)], 7.10 [d, 3JH-H ) 9, C6H4 (2H)], 7.28 [d,
3JH-H ) 9, C6H4 (2H)], HB not observed. 13C NMR (C6D6): 12.6 [q,
1
1JC-H ) 128, 3(CH3)], 30.7 [q, JC-H ) 126, 3(C(CH3)3)], 32.0 [s,
1
1
3(C(CH3)3)], 55.6 [q, JC-H ) 142, OCH3], 103.2 [d, JC-H ) 171,
3(C3N2H) (1C)], 115.0 [dd, 1JC-H ) 156, 2JC-H ) 6, C6H4 (2C)], 121.5
Finally, it is important to emphasize that the tendency to form
alkoxide compounds by alcoholysis reactions does not always
follow the trends described in this paper (i.e., Ar > Me > Et >
Pri > But). For example, quantitative measurements of equilibria
involving (dipic)VO(OR)26 and R′OH indicate that the alkoxides
that preferentially bind to vanadium follow the opposite
sequence, namely, But > Pri > Et > p-ClC6H4, an observation
that has been rationalized in terms of the trend being dictated
by the π-donor capabilities of the ligands.27 The clear distinction
1
2
[dd, JC-H ) 150, JC-H ) 6, C6H4 (2C)], 144.2 [s, 3(C3N2H) (1C)],
151.0 [s, C6H4 (para C)], 159.4 [s, C6H4 (ipso C)], 163.9 [s, 3(C3N2H)
(1C)]. IR (KBr, cm-1): 2961 (s), 2829 (w), 2554 (w) [ν(B-H)], 1542
(s), 1503 (vs), 1464 (s), 1436 (s), 1385 (m), 1366 (s), 1343 (m), 1302
(s), 1263 (s), 1231 (vs), 1186 (vs), 1126 (w), 1097 (w), 1068 (s), 1033
(m), 985 (w), 857 (w), 827 (s), 789 (s), 765 (vs), 732 (w), 680 (w),
646 (s), 521 (w).
t
Synthesis of [TpBu ,Me]ZnOC6H4But. A solution of 4-tert-butylphe-
t
nol (21 mg, 0.14 mmol) and [TpBu ,Me]ZnH (70 mg, 0.14 mmol) in
benzene (3.5 mL) was heated at 120 °C for 3 days and then lyophilized
t
to give [TpBu ,Me]ZnOC6H4But as a white solid (57 mg, 73%). Anal.
(23) Kitajima, N.; Tolman, W. B. Prog. Inorg. Chem. 1995, 43, 419-
531.
(24) In support of this statement, the ionization potential of the frag-
(28) (a) McNally, J. P.; Leong, V. S.; Cooper, N. J. In Experimental
Organometallic Chemistry; Wayda, A. L., Darensbourg, M. Y., Eds.;
American Chemical Society: Washington, DC, 1987; Chapter 2, pp 6-23.
(b) Burger, B. J.; Bercaw, J. E. In Experimental Organometallic Chemistry;
Wayda, A. L., Darensbourg, M. Y., Eds.; American Chemical Society:
Washington, DC, 1987; Chapter 4, pp 79-98. (c) Shriver, D. F.; Drezdzon,
M. A. The Manipulation of Air-SensitiVe Compounds, 2nd ed.; Wiley-
Interscience: New York, 1986.
t
ment {[TpBu ,Me]Zn} is calculated to be 20.0 kcal mol-1 lower than that for
{[Tp]Zn}.
(25) Ramaswamy, S.; Park, D.-H.; Plapp, B. V. Biochemistry 1999, 38,
13951-13959.
(26) dipicH2 ) pyridine-2,6-dicarboxylic acid.
(27) Thorn, D. L.; Harlow, R. L.; Herron, N. Inorg. Chem. 1996, 35,
547-548.

Thermodynamics of Zinc Alkoxide Formation
J. Am. Chem. Soc., Vol. 122, No. 51, 2000 12657
Calcd for C34H53BN6OZn: C, 64.0; H, 8.4; N, 13.2. Found: C, 64.1;
0.20 mmol) in benzene (5 mL) was heated at 120 °C for 2 days and
t
1
H, 8.6; N 12.6. H NMR (C6D6): 1.45 [s, (C6H4)C(CH3)3], 1.50 [s,
then lyophilized to give [TpBu ,Me]ZnOC6H4CO2Me as a white solid (79
3(C(CH3)3)], 2.08 [s, 3(CH3)], 5.63 [s, 3(C3N2H)], 7.33 [d, 3JH-H ) 9,
mg, 62%). Anal. Calcd for C32H47BN6O3Zn: C, 60.1; H, 7.4; N, 13.1.
3
C6H4 (2H)], 7.50 [d, JH-H ) 9, C6H4 (2H)], HB not observed. 13C
Found: C, 60.3; H, 7.2; N 13.3. 1H NMR (C6D6): 1.40 [s, 3(C(CH3)3)],
1
1
NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.7 [q, JC-H ) 124,
2.06 [s, 3(CH3)], 3.64 [s, OCH3], 5.61 [s, 3(C3N2H)], 7.29 [d, 3JH-H
)
1
3
3(C(CH3)3)], 32.0 [s, 3(C(CH3)3)], 32.2 [q, JC-H ) 122, (C6H4)C-
9, C6H4 (2H)], 8.51 [d, JH-H ) 9, C6H4 (2H)], HB not observed. 13C
1
1
1
(CH3)3], 34.0 [s, (C6H4)C(CH3)3 ], 103.2 [d, JC-H ) 171, 3(C3N2H)
NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.6 [q, JC-H ) 126,
(1C)], 120.9 [dd, 1JC-H ) 155, 2JC-H ) 5, C6H4 (2C)], 126.0 [dd, 1JC-H
3(C(CH3)3)], 31.9 [s, 3(C(CH3)3)], 51.0 [q, JC-H ) 146, CO2CH3],
1
2
) 152, JC-H ) 8, C6H4 (2C)], 137.5 [s, C6H4 (para C)], 144.2 [s,
103.3 [d, 1JC-H ) 174, 3(C3N2H) (1C)], 118.1 [s, C6H4 (para C)], 121.3
1
2
1
3(C3N2H) (1C)], 162.7 [s, C6H4 (ipso C)], 163.9 [s, 3(C3N2H) (1C)].
IR (KBr, cm-1): 2922 (vs), 2855 (vs), 2547 (w) [ν(B-H)], 1605 (m),
1542 (m), 1511 (s), 1465 (s), 1365 (s), 1340 (w), 1315 (s), 1263 (w),
1248 (w), 1187 (s), 1107 (w), 1068 (s), 1028 (m), 988 (w), 878 (w),
811 (w), 858 (w), 829 (m), 811 (m), 792 (m), 767 (m), 730 (w), 694
(w), 681 (w), 646 (w), 551 (w), 521 (w).
[dd, JC-H ) 157, JC-H ) 4, C6H4 (2C)], 132.2 [dd, JC-H ) 159,
2JC-H ) 7, C6H4 (2C)], 144.5 [s, 3(C3N2H) (1C)], 163.9 [s, 3(C3N2H)
(1C)], 167.6 [s, C6H4 (ipso C)], 169.9 [s, CO2CH3]. IR (KBr, cm-1):
2965 (s), 2871 (m), 2560 (w) [ν(B-H)], 1712 (s) [ν(CO)], 1597 (vs),
1544 (s), 1513 (s), 1476 (m), 1433 (s), 1366 (m), 1331 (vs), 1276 (vs),
1189 (s), 1159 (s), 1111 (m), 1096 (m), 1068 (m), 1030 (w), 988 (w),
852 (w), 796 (w), 772 (m), 705 (w), 680 (w), 655 (w), 520 (w), 455
(w).
t
Synthesis of [TpBu ,Me]ZnOC6H4Me. p-Cresol was purified before
use by storing a solution in benzene over molecular sieves for 2 days,
followed by filtration and removal of the solvent in vacuo. A solution
t
Synthesis of [TpBu ,Me]ZnOC6H4COMe. A solution of 4-hydroxy-
t
t
of p-cresol (52 mg, 0.48 mmol) and [TpBu ,Me]ZnH (200 mg, 0.41 mmol)
acetophenone (61 mg, 0.45 mmol) and [TpBu ,Me]ZnH (200 mg, 0.41
in benzene (10 mL) was heated at 120 °C for 1.5 days. The reaction
mmol) in benzene (10 mL) was heated at 120 °C for 2 days. The
mixture was filtered, and the filtrate was lyophilized to give
reaction mixture was filtered, and the filtrate was lyophilized to give
t
t
[TpBu ,Me]ZnOC6H4Me as a white solid (149 mg, 61%). Anal. Calcd
[TpBu ,Me]ZnOC6H4COMe as a white solid (182 mg, 71%). Anal. Calcd
for C31H47BN6OZn: C, 62.5; H, 7.9; N, 14.1. Found: C, 62.2; H, 7.9;
N 15.0. 1H NMR (C6D6): 1.49 [s, 3(C(CH3)3)], 2.08 [s, 3(CH3)], 2.40
[s, (C6H4)CH3], 5.63 [s, 3(C3N2H)], 7.29 [m, C6H4], HB not observed.
for C32H47BN6O2Zn: C, 61.6; H, 7.6; N, 13.5. Found: C, 61.7; H, 7.8;
N 13.0. 1H NMR (C6D6): 1.42 [s, 3(C(CH3)3)], 2.06 [s, 3(CH3)], 2.37
[s, COCH3], 5.61 [s, 3(C3N2H)], 7.24 [d, 3JH-H ) 9, C6H4 (2H)], 8.22
1
1
3
13C NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 20.9 [q, JC-H
)
[d, JH-H ) 9, C6H4 (2H)], HB not observed. 13C NMR (C6D6): 12.6
1
1
1
125, (C6H4)CH3], 30.7 [q, JC-H ) 126, 3(C(CH3)3)], 32.0 [s,
[q, JC-H ) 128, 3(CH3)], 25.9 [q, JC-H ) 126, COCH3], 30.6 [q,
1JC-H ) 126, 3(C(CH3)3)], 31.9 [s, 3(C(CH3)3)], 103.3 [d, 1JC-H ) 174,
3(C3N2H) (1C)], 121.1 [dd, 1JC-H ) 157, 2JC-H ) 4, C6H4 (2C)], 126.8
1
1
3(C(CH3)3)], 103.2 [d, JC-H ) 174, 3(C3N2H) (1C)], 121.4 [d, JC-H
) 154, C6H4 (2C)], 123.8 [s, C6H4 (para C)], 129.9 [d, JC-H ) 152,
1
1
2
C6H4 (2C)], 144.3 [s, 3(C3N2H) (1C)], 162.9 [s, C6H4 (ipso C)], 164.0
[s, 3(C3N2H) (1C)]. IR (KBr, cm-1): 2963 (s), 2553, (w) [ν(B-H)],
1609 (s), 1544 (s), 1509 (vs), 1477 (s), 1434 (s), 1366 (s), 1341 (m),
1305 (vs), 1246 (m), 1188 (vs), 1128 (w), 1103 (w), 1068 (s), 1030
(m), 987 (w), 869 (w), 824 (s), 789 (s), 769 (s), 680 (w), 648 (m), 517
(m).
[s, C6H4 (para C)], 131.2 [dd, JC-H ) 156, JC-H ) 7, C6H4 (2C)],
144.6 [s, 3(C3N2H) (1C)], 163.9 [s, 3(C3N2H) (1C)], 170.0 [s, C6H4
(ipso C)], 195.0 [s, COCH3]. IR (KBr, cm-1): 2960 (s), 2553 (w) [ν-
(BH)], 1668 (s) [ν(CO)], 1587 (vs), 1544 (s), 1515 (s), 1466 (m), 1424
(s), 1336 (vs), 1274 (vs), 1188 (vs), 1163 (s), 1067 (s), 1031 (m), 952
(w), 873 (m), 838 (s), 795 (s), 767 (s), 719 (w), 679 (w), 650 (m), 586
(s), 521 (w).
t
Synthesis of [TpBu ,Me]ZnOC6H5. A solution of phenol (31 mg, 0.33
t
t
mmol) and [TpBu ,Me]ZnH (150 mg, 0.31 mmol) in benzene (5 mL) was
Synthesis of [TpBu ,Me]ZnOC6H4NO2. A solution of 4-nitrophenol
t
heated at 120 °C for 1.5 days. The reaction mixture was filtered, and
(64 mg, 0.46 mmol) and [TpBu ,Me]ZnH (200 mg, 0.41 mmol) in benzene
t
the filtrate was lyophilized to give [TpBu ,Me]ZnOC6H5 as a white solid
(5 mL) was heated at 120 °C for 1 day. The reaction mixture was
t
(95 mg, 53%). Anal. Calcd for C30H45BN6OZn: C, 61.9; H, 7.8; N,
14.4. Found: C, 62.2; H, 7.8; N 14.2. 1H NMR (C6D6): 1.47 [s,
3(C(CH3)3)], 2.07 [s, 3(CH3)], 5.62 [s, 3(C3N2H)], 6.95 [tt, 3JH-H ) 7,
4JH-H ) 1, C6H5 (1H)], 7.37 [m, C6H5 (2H)], 7.48 [m, C6H5 (2H)], HB
filtered, and the filtrate was lyophilized to give [TpBu ,Me]ZnOC6H4-
NO2 as a yellow solid (149 mg, 58%). Anal. Calcd for C30H44BN7O3-
1
Zn: C, 57.5; H, 7.1; N, 15.6. Found: C, 57.8; H, 6.9; N, 15.7. H
NMR (C6D6): 1.35 [s, 3(C(CH3)3)], 2.06 [s, 3(CH3)], 5.61 [s, 3(C3N2H)],
1
3
3
not observed. 13C NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.7
7.02 [d, JH-H ) 9, C6H4 (2H)], 8.41 [d, JH-H ) 9, C6H4 (2H)], HB
[q, 1JC-H ) 126, 3(C(CH3)3)], 32.0 [s, 3(C(CH3)3)], 103.2 [d, 1JC-H
)
not observed. 13C NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.5
1
1
171, 3(C3N2H) (1C)], 115.8 [d, JC-H ) 158, C6H4 (para C)], 121.8
[dt, 1JC-H ) 152, 2JC-H ) 7, C6H5 (2C)], 129.4 [dd, 1JC-H ) 155, 2JC-H
) 9, C6H4 (2C)], 144.3 [s, 3(C3N2H) (1C)], 164.0 [s, 3(C3N2H) (1C)],
165.2 [s, C6H4 (ipso C)]. IR (KBr, cm-1): 2968 (s), 2551 (w) [ν(B-
H)], 1590 (s), 1544 (s), 1495 (vs), 1435 (s), 1365 (s), 1343 (m), 1303
(vs), 1246 (m), 1188 (vs), 1069 (s), 1030 (m), 995 (w), 859 (m), 790
(s), 768 (s), 757 (s), 695 (m), 648 (m), 520 (w).
[q, 1JC-H ) 126, 3(C(CH3)3)], 31.9 [s, 3(C(CH3)3)], 103.4 [d, 1JC-H
)
174, 3(C3N2H) (1C)], 120.8 [dd, 1JC-H ) 160, 2JC-H ) 5, C6H4 (2C)],
1
2
126.5 [dd, JC-H ) 162, JC-H ) 5, C6H4 (2C)], 138.2 [s, C6H4 (para
C)], 144.8 [s, 3(C3N2H) (1C)], 163.8 [s, 3(C3N2H) (1C)], 171.4 [s, C6H4
(ipso C)]. IR (KBr, cm-1): 2965 (s), 2555 (w) [ν(BH)], 1585 (vs),
1543 (s), 1503 (vs), 1428 (s), 1366 (s), 1320 (vs), 1247 (m), 1187 (s),
1111 (s), 1069 (s), 1032 (m), 989 (w), 872 (m), 848 (m), 793 (s), 768
(s), 681 (m), 667 (s), 649 (m), 523 (w).
t
Synthesis of [TpBu ,Me]ZnOC6H4I. A solution of 4-iodophenol (74
t
mg, 0.34 mmol) and [TpBu ,Me]ZnH (100 mg, 0.20 mmol) in benzene
Measurement of the Equilibrium Constant for Alcoholysis of
t
t
(10 mL) was heated at 120 °C for 1.5 days. The reaction mixture was
[TpBu ,Me]ZnOH with p-Cresol. A solution of [TpBu ,Me]ZnOC6H4Me
(9.3 mg, 0.016 mmol) in THF-d8 (600 µL) was treated with 12 5-µL
portions of a solution of D2O in THF-d8 (6.37 M). After each addition,
the solution was allowed to equilibrate for at least 5 min and the
t
filtered, and the filtrate was lyophilized to give [TpBu ,Me]ZnOC6H4I as
a white solid (90 mg, 64%). Anal. Calcd for C30H44BIN6OZn: C, 50.9;
1
H, 6.3; N, 11.9. Found: C, 50.8; H, 6.2; N 11.4. H NMR (C6D6):
t
t
1.40 [s, 3(C(CH3)3)], 2.06 [s, 3(CH3)], 5.61 [s, 3(C3N2H)], 7.03 [d,
concentration ratio of [TpBu ,Me]ZnOC6H4Me/[TpBu ,Me]ZnOH was de-
3
1
3JH-H ) 9, C6H4 (2H)], 7.70 [d, JH-H ) 9, C6H4 (2H)], HB not
termined by H NMR spectroscopy. The equilibrium constant Kp-Me
t
t
1
observed. 13C NMR (C6D6): 12.6 [q, JC-H ) 128, 3(CH3)], 30.6 [q,
) [{[TpBu ,Me]ZnOC6H4Me}][H2O]/[{[TpBu ,Me]ZnOH}][p-XC6H4OH]
was obtained by computer fitting of the data. The experiment was
repeated three times, and an average value of the equilibrium constant
was obtained.
1JC-H ) 126, 3(C(CH3)3)], 31.9 [s, 3(C(CH3)3)], 76.2 [s, C6H4 (para
1
1
C)], 103.3 [d, JC-H ) 174, 3(C3N2H) (1C)], 124.4 [dd, JC-H ) 162,
2JC-H ) 5, C6H4 (2C)], 138.2 [dd, 1JC-H ) 161, 2JC-H ) 7, C6H4 (2C)],
144.4 [s, 3(C3N2H) (1C)], 163.9 [s, 3(C3N2H) (1C)], 164.9 [s, C6H4
(ipso C)]. IR (KBr, cm-1): 2961 (s), 2557 (w) [ν(B-H)], 1576 (s),
1541 (s), 1485 (vs), 1433 (s), 1384 (m), 1365 (s), 1342 (m), 1307 (vs),
1246 (m), 1186 (vs), 1068 (s), 1032 (m), 986 (w), 857 (m), 825 (s),
787 (s), 766 (s), 680 (w), 645 (s), 521 (w).
Measurement of the Equilibrium Constant for Alcohol Exchange
t
of [TpBu ,Me]ZnOC6H4Me with ArOH. A solution was prepared of
t
∼10 mg of [TpBu ,Me]ZnOC6H4X, (X ) OMe, But, H, I, CO2Me, COMe,
NO2) and p-cresol (∼2 mg) in C6D6. The concentration ratios of
t
t
[TpBu ,Me]ZnOC6H4Me/[TpBu ,Me]ZnOC6H4X and HOC6H4X/HOC6H4Me
t
Synthesis of [TpBu ,Me]ZnOC6H4CO2Me. A solution of 4-hy-
were obtained by H NMR spectroscopy. The relative concentrations
1
t
droxymethylbenzoate (31 mg, 0.20 mmol) and [TpBu ,Me]ZnH (100 mg,
were varied, and six to eight independent measurements were made

12658 J. Am. Chem. Soc., Vol. 122, No. 51, 2000
Bergquist et al.
Table 5. Crystal, Intensity Collection, and Refinement Data
t
t
t
t
[TpBu ,Me]ZnOC6H4C(O)Me
[TpBu ,Me]ZnOC6H4NO2‚C6H6
[TpBu ,Me]ZnOC6H4NH2
[TpBu ,Me]ZnOC6H4But
lattice
monoclinic
C32H47BN6O2Zn
623.94
P21/c (No. 14)
11.0272(7)
14.7628(9)
22.756(2)
90
114.180(1)
90
3379.4(4)
4
monoclinic
C36H50BN7O3Zn
705.01
P21 (No. 4)
11.912(6)
11.194(5)
14.893(7)
90
106.766(8)
90
1901(1)
2
monoclinic
C30H46BN7OZn
596.92
P21/n (No. 14)
21.9967(12)
15.4016(7)
22.1563(11)
90
119.396(1)
90
6539.8(6)
8
monoclinic
C34H53BN6OZn
638.00
P21/c (No. 14)
21.9346(11)
18.9262(9)
18.4830(9)
90
107.275(1)
90
7326.9(6)
8
formula
formula wt
space group
a/Å
b/Å
c/Å
R/deg
â/deg
γ/deg
V/Å3
Z
temp/K
µ (Mo KR)/mm-1
no. of data
no. of params
R1
213
0.763
7547
396
0.0346
0.0844
233
0.689
8231
435
0.0484
0.1110
223
0.784
15117
770
0.0555
0.1059
203
0.703
16595
814
0.0651
0.1149
wR2
for each substituent to give an average value for the equilibrium
derivative from a hydroxide complex is influenced by the nature
of the alcohol and ligands attached to zinc. Thus, zinc alkoxide
formation is favored electronically by incorporation of electron-
withdrawing substituents in the alcohol but is disfavored
sterically for bulky alcohols. Furthermore, alkoxide formation
is more favored for [Tp]ZnOR derivatives than for their
t
t
constant, Kexch ) [{[TpBu ,Me]ZnOC6H4Me}][p-XC6H4OH]/[{[TpBu ,Me]-
ZnOC6H4X}][p-MeC6H4OH]. The equilibrium constants for the reaction
t
between [TpBu ,Me]ZnOH and p-XC6H4OH were determined by the
t
expression Kp-X ) Kp-Me/Kexch. ∆H values for the reactions of [TpBu ,Me]-
ZnOH with XC6H4OH are estimated from the experimental ∆G values
using the assumption that ∆S ≈ 0, an approximation that was confirmed
t
[TpBu ,Me]ZnOR counterparts. These trends are a result of
t
for the reaction of [TpBu ,Me]ZnOH with MeC6H4OH (∆H ) -1.3(1)
kcal mol-1, ∆S ) -0.3(2) eu over the range 250-330 K). The
enthalpies for X ) OMe, But, Me, H, I, CO2Me, and C(O)Me have
also been determined by titration calorimetry. These values are in
reasonable agreement with those determined by the equilibrium studies,
but are ∼1-4 kcal mol-1 more exothermic. However, the calorimetry
values were not averaged over many experiments and so are considered
to be less reliable than those obtained by the equilibrium study.
homolytic Zn-OR BDEs being more sensitive to the nature of
R than are the corresponding H-OR bond energies. Thus,
electron-withdrawing substituents increase Zn-OAR bond
energies to a greater extent than H-OAr bond energies, while
bulky substituents decrease Zn-OR bond energies to a greater
extent than H-OR bond energies. The trends reported for this
zinc system are of relevance to the tetrahedral zinc alkoxide
intermediate in the catalytic cycle of liver alcohol dehydrogenases
such information is of importance since other systems, e.g.,
(dipic)VO(OR),27 exhibit a trend opposite to that reported here.
With the exception of derivatives of acidic alcohols (e.g.,
nitrophenol), the zinc alkoxide complexes [TpRR′]ZnOR are very
unstable toward hydrolysis. This hydrolytic instability of simple
zinc alkoxide complexes suggests that the active site environ-
ment of LADH plays an important role in stabilizing the
alkoxide intermediate, possibly via hydrogen-bonding interac-
tions. Future studies are intended to investigate the influence
of substituents on another step of the catalytic cycle, namely,
hydride transfer from the alkoxide.
t
X-ray Structure Determinations. X-ray diffraction data for [TpBu ,Me]-
ZnOC6H4X (X ) But, C(O)Me, NH2, NO2) were collected on a Bruker
P4 diffractometer equipped with a SMART CCD detector. Crystal data,
data collection, and refinement parameters are summarized in Table 5.
The structures were solved using direct methods and standard difference
map techniques and were refined by full-matrix least-squares procedures
on F2 with SHELXTL (Version 5.03).29 Hydrogen atoms on carbon
were included in calculated positions.
Computational Details. All calculations were performed using
Jaguar.9 Initial geometries were obtained from crystal structures where
available, and in other cases, the desired molecule was built through
modification of the coordinates of a similar compound with known
structure. DFT geometry optimizations were performed on all com-
plexes at the B3LYP level using the LACVP** basis set. Single-point
energies were calculated for the optimized structures at the B3LYP
level using the triple ú basis set CC-PVTZ (-f) for all elements except
boron, for which the 6-31G** basis set was used, and zinc and iodine,
for which the LACV3P** basis set was used.
Acknowledgment. We thank the National Institutes of
Health (GM46502 to G.P. and GM40526 to R.A.F.) for support
of this research, Professor Ronald Breslow for the use of his
titration calorimeter, and Professor Sally Chapman and Dr. Barry
Dunietz for helpful discussions.
Conclusion
In summary, the present study indicates the extent to which
the thermodynamics of the formation of a zinc alkoxide
Supporting Information Available: Tables of crystal-
lographic data (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
(29) Sheldrick, G. M. SHELXTL, An Integrated System for Solving,
Refining and Displaying Crystal Structures from Diffration Data. University
of Go¨ttingen, Go¨ttingen, Federal Republic of Germany, 1981.
JA002286D