Structure and reactivity of mixed alkali metal alkoxide/aryloxide catalysts

Article abstract of DOI:10.1021/jo016105h

The average solution aggregation state of the ester interchange catalyst 1, obtained by mixing 1 equiv of NaOt-Bu and 3 equiv of NaOC₆H₄-4-t-Bu, was determined to be 4.0 by vapor pressure osmometry (VPO) in THF. Low-temperature ¹H NMR spectra of 1 indicated that the THF solution contained a mixture of tetrameric clusters. On the basis of symmetry arguments and the sensitivity of the different species to the alkoxide/aryloxide ratio, the compounds were determined to be mixed clusters with 0:4, 1:3, 2:2, and 3:1 mixtures of the NaOt-Bu and NaOC₆H₄-4-t-Bu components. On a per -Ot-Bu basis, each cluster has a similar absolute activity, though the aryloxide-rich catalysts are significantly longer-lived. Unlike 1, catalysts containing ortho-substituted aryloxides, 2, do not give a strictly statistical distribution of clusters, and the activities of these catalysts depend on steric and electronic factors, though the absolute rate differences are not large.

Full text of DOI:10.1021/jo016105h

J . Org. Chem. 2001, 66, 9005-9010
9005
Str u ctu r e a n d Rea ctivity of Mixed Alk a li Meta l Alk oxid e/Ar yloxid e
Ca ta lysts
Rebecca M. Kissling and Michel R. Gagne´*
Department of Chemistry, CB 3290, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
mgagne@unc.edu
Received September 10, 2001
The average solution aggregation state of the ester interchange catalyst 1, obtained by mixing 1
equiv of NaOt-Bu and 3 equiv of NaOC6H4-4-t-Bu, was determined to be 4.0 by vapor pressure
1
osmometry (VPO) in THF. Low-temperature H NMR spectra of 1 indicated that the THF solution
contained a mixture of tetrameric clusters. On the basis of symmetry arguments and the sensitivity
of the different species to the alkoxide/aryloxide ratio, the compounds were determined to be mixed
clusters with 0:4, 1:3, 2:2, and 3:1 mixtures of the NaOt-Bu and NaOC6H4-4-t-Bu components. On
a per -Ot-Bu basis, each cluster has a similar absolute activity, though the aryloxide-rich catalysts
are significantly longer-lived. Unlike 1, catalysts containing ortho-substituted aryloxides, 2, do not
give a strictly statistical distribution of clusters, and the activities of these catalysts depend on
steric and electronic factors, though the absolute rate differences are not large.
Sch em e 1
In tr od u ction
Alkali metal alkoxides such as KOt-Bu and NaOt-Bu
are excellent catalysts for the ester interchange reaction,
with some turnover frequencies (TOF) approaching 107
h-1 for eq 1.1 In aprotic donating solvents (typical reaction
conditions), KOt-Bu and NaOt-Bu form cubelike tetram-
ers2 that provide the framework for kinetically facilitating
alkoxide/ester interchange while solubilizing a form of
“NaOMe”, an otherwise insoluble material,3 Scheme 1.
Although these catalysts effectively convert methyl to
more easily hydrolyzed tert-butyl esters (5-8 mol % KOt-
Bu), the catalyst eventually precipitates from solution as
methoxide builds up in the catalyst cluster, Scheme 2.3c
Sch em e 2
The problem of over-methoxylation and precipitation
of these catalysts was solved by the addition of an
aryloxide salt cocatalyst, which was presumed to form a
soluble mixed NaOt-Bu/NaOAr catalyst, Scheme 3;4
aryloxide additives were chosen as they do not produc-
tively interchange with alkyl esters. Our optimized
catalyst combination was a 1:3 NaOt-Bu/NaOC6H4-4-t-
Bu mixture, 1, which had a significantly longer lifetime
and was milder and more robust than its alkoxide-only
parents. This new catalyst formulation enabled one-step
Sch em e 3
(1) (a) Stanton, M. G.; Allen, C. B.; Kissling, R. M.; Lincoln, A. L.;
Gagne´, M. R. J . Am. Chem. Soc. 1998, 120, 5981-5989. (b) Stanton,
M. G.; Gagne´, M. R. J . Am. Chem. Soc. 1997, 119, 5075-5076.
(2) (a) Halaska, V.; Lochmann, L.; Lim, D. Collect. Czech. Chem.
Commun. 1968, 33, 3245-3253. (b) Schmidt, P.; Lochmann, L.;
Schneider B. J . Mol. Struct. 1971, 403-411.
transformations with lower catalyst loadings (0.5-3 mol
% 1), enhanced functional group tolerance (e.g., nitro
groups), and interchange of epimerizable R-amino esters.4
During catalyst development, experiments were de-
signed under the reasonable assumption that, like [KOt-
Bu]4 and [NaOt-Bu]4, these mixed cluster catalysts were
also tetrameric, though little data were available to
(3) NaOMe forms infinite two-dimensional sheets and is virtually
insoluble in aprotic solvents. See: (a) Weiss, E.; Aldsorf, H. Z. Anorg.
Allg. Chem. 1970, 572, 206-213. (b) Weiss, E.; Aldsorf, H. Z. Anorg.
Allg. Chem. 1964, 332, 197-203. (c) Vasin, V. A.; Razin, V. V. Synlett
2001, 658-660.
(4) Kissling, R. M.; Gagne´, M. R. Org. Lett. 2000, 2, 4209-4212.
10.1021/jo016105h CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/30/2001

9006 J . Org. Chem., Vol. 66, No. 26, 2001
Kissling and Gagne´
directly support this notion. While many lithium5 and
sodium6 aryloxide clusters have been characterized, to
our knowledge, no mixed alkoxide/aryloxide clusters of
sodium have been reported. However, a Cambridge
Structural Database search revealed several relevant
structure types that result from the aggregation of
mixtures of anionic components, and these include a
mixed lithium bis-aryloxide/alkoxide (hexamer),7 an aryl-
oxide/alkyl (tetramer),8 a mixed alkoxy/chloride (dimer
of tetramers),9 an aryloxy/iodide (tetramer),10 an enolate/
halide (dimers),11 an enolate/amido (dimers),12 an alkox-
ide/alkyl (tetramers),13 and a Li/K enolate/alkoxide (oc-
tamer).14 As a general rule, sodium aryloxides preferen-
tially crystallize as tetramers in the presence of a highly
polar aprotic solvent6a,e,k,m or coordinating aryloxide
substituent.6a,f,l As a result, the catalyst was similarly
expected to be a tetramer under reaction conditions.15
Ta ble 1. Va p or P r essu r e Osm om etr y Resu lts for 1 a n d
Its Com p on en ts
molarity
molarity
(M)
by VPO
aggregation
state
(M)a
NaOt-Bu
NaOC6H4-4-t-Bu
(NaOt-Bu)(NaOAr)3 (1)
0.080
0.120
0.120
0.020
0.028
0.029
4.0
4.3
4.0
a
Dry THF at 23 °C.
Since little relevant data were available, we initiated and
report herein experiments that probe the structure,
longevity, and activity of mixed sodium aryloxy/alkoxy
ester interchange catalysts.
(5) Representative examples include: (a) Stauch, J .; Warren, T. H.;
Erker, G.; Fro¨lich, R.; Saarenketo, P. Inorg. Chim. Acta 2000, 300-
302, 810-821. (b) Boyle, T. J .; Pedrotty, D. M.; Alam, T. M.; Vick, S.
C.; Rodriguez, M. A. Inorg. Chem. 2000, 39, 5133-5146. (c) Khanjin,
N. A.; Menger, F. M. J . Org. Chem. 1997, 62, 8923-8927. (d) J ackman,
L. M.; C¸ ismeciyan, D.; Williard, P. G.; Nichols, M. A. J . Am. Chem.
Soc. 1993, 115, 6262-6267. (e) J ackman, L. M.; Chen, X. J . Am. Chem.
Soc. 1992, 114, 403-401. (f) Watson, K. A.; Fortier, S.; Murchie, M.
P.; Bovenkamp, J . W.; Rodrigue, A.; Buchanan, G. W.; Ratcliffe, C. I.
Can. J . Chem. 1990, 68, 1201-1207. (g) J ackman, L. M.; Smith, B. D.
J . Am. Chem. Soc. 1988, 110, 3829-3835. (h) Murchie, N. P.;
Bovenkamp, J . W.; Rodrigue, A.; Watson, K. A.; Fortier, S. Can. J .
Chem. 1988, 66, 2515-2523. (i) Huffman, J . C.; Geerts, R. L.; Caulton,
K. G. J . Crystallogr. Spectrosc. Res. 1984, 14, 541-547. (j) J ackman,
L. M.; DeBrosse, C. W. J . Am. Chem. Soc., 1983, 105, 4177. (k)
Shobotake, K.; Nakamoto, K. Inorg. Chim. Acta 1970, 4, 485-487.
(6) Representative examples include: (a) Sobota, P.; Klimowicz, M.;
Utko, J .; J erzykiewicz, L. B. New J . Chem. 2000, 24, 523-526. (b)
Czado, W.; Mu¨ller, U. Z. Krystallogr. 1999, 214, 63-64. (c) Dinnebier,
R. E.; Pink, M.; Sieler, J .; Stephens, P. W. Inorg. Chem. 1997, 36,
3398-3401. (d) Bock, H.; Dienelt, R.; Na¨ther, C.; Havlas, Z. Chem.
Ber. Recueil. 1997, 130, 1533-1537. (e) Kunert, M.; Dinjus, E.; Nauc,
M.; Seiler, J . Chem. Ber. Recueil. 1997, 130, 1461-1465. (f) Hogerheide,
M. P.; Ringelberg, S. N.; J anssen, M. F.; Boersma, J .; Spek, A. L.; van
Koten, G. Inorg. Chem. 1996, 35, 1195-1200. (g) Matilainen, L.;
Leskela¨, M.; Klinga, M. J . Chem. Soc., Chem. Commun. 1995, 421-
422. (h) Kunert, M.; Seiler, J . Z. Anorg. Allg. Chem. 1995, 621, 1597-
1599. (i) J o¨rchel, P.; Seiler, J . Z. Anorg. Allg. Chem. 1995, 1058-1062.
(j) Seiler, J .; Pink, M.; Zahn, G. Z. Anorg. Allg. Chem. 1994, 620, 743-
748. (k) Walther, D.; Ritter, U.; Gessler, S.; Sieler, J .; Kunert, M. Z.
Anorg. Allg. Chem. 1994, 620, 101-106. (l) van der Schaaf, P. A.;
J astrzebski, J . T. B. H.; Hogerheide, M. P.; Smeets, W. J . J .; Spek, A.
L.; Boersma, J .; van Koten, G. Inorg. Chem. 1993, 32, 4111-4118. (m)
Evans, W. J .; Golden, R. E.; Ziller, J . W. Inorg. Chem. 1993, 32, 3041-
3051. (n) Cragg-Hine, I.; Davidson, M. G.; Kocian, O.; Kottke, T.; Mair,
F. S.; Snaith, R.; Stoddart, J . F. J . Chem. Soc., Chem. Commun. 1993,
1355-1357. (o) Brooker, S.; Edelmann, F. T.; Kottke, T.; Roesky, H.
W.; Scheldrick, G. M.; Stalke, D.; Whitmire, K. H. J . Chem. Soc., Chem.
Commun. 1991, 144-146. (p) Buchanan, G. W.; Khan, M. Z.; Rip-
meester, J . A.; Bovenkamp, J . W.; Rodrigue, A. Can. J . Chem. 1987,
65, 2564-2567. (q) Fraser, M. E.; Fortier, S.; Markewicz, M. K.;
Rodrigue, A.; Bovenkamp, J . W. Can. J . Chem. 1987, 65, 2558-2563.
(7) Ko, B.-T.; Lin, C.-C. J . Am. Chem. Soc. 2001, 123, 7973-7977.
(8) (a) Harder, S.; Lutz, M.; Streitweiser, A. J . Am. Chem. Soc. 1995,
117, 2361-2362. (b) Harder, S.; Streitweiser, A. Angew. Chem., Int.
Ed. Engl. 1993, 32, 1066-1068.
Resu lts a n d Discu ssion
P r ep a r a tion of Mixed Alk oxid e/Ar yloxid e Ca ta -
lysts. Mixed cluster catalyst 1 was prepared for each
experiment by dissolving 1 equiv of NaOt-Bu and 3 equiv
of NaOC6H4-4-t-Bu in THF. This protocol provided a
catalyst with optimum activity and reproducibility for
kinetic studies. However, 1 could also be prepared and
stored for later use by dissolving the alkoxide and
aryloxide salts in THF as described above. Solvent
removal provided 1 as a white powder that was stable
in the glovebox for several months. Alternatively, a third,
somewhat more convenient method for the large-scale
preparation or storage of 1 was to combine 4 equiv of
NaOt-Bu and 3 equiv of 4-t-Bu-phenol in THF. This
solution could be used as prepared or concentrated in
vacuo to remove the THF and the t-BuOH; although for
careful kinetic studies we preferred the first method, the
activity of 1 was independent of the manner of prepara-
tion. For screening purposes, the generic catalysts 2,
[NaOt-Bu/3NaOAr], were prepared in situ using the
NaOt-Bu and aryloxide salt method.
Va p or P r essu r e Osm om etr y (VP O). As a first step
to determining the structure of catalyst 1, we measured
its solution aggregation state by vapor pressure osmom-
etry. VPO enables average molecular weights to be
measured, from which average aggregation states can be
calculated. Table 1 outlines the results for this study. By
(9) Pauls, J .; Neumu¨ller, B. Z. Anorg. Allg. Chem. 2000, 626, 270-
279.
(10) (a) van der Schaaf, P. A.; Hogerheide, M. P.; Grove, D. M.; Spek,
A. L.; van Koten, G. J . Chem. Soc., Chem. Commun. 1992, 1703-1704.
(b) Raithby, P. R.; Reed, J .; Snaith, R.; Wright, D. S.; Angew. Chem.,
Int. Ed. Engl. 1991, 30, 1011-1013.
(11) (a) Henderson, K. W.; Dorigo, A. E.; Williard, P. G.; Bernstein,
P. R. Angew. Chem., Int. Ed. Engl. 1996, 35, 1322-1324. (b) Hender-
son, K. W.; Dorigo, A. E.; Liu, Q.-L.; Williard, P. G.; von Rague´-
Schleyer, P.; Bernstein, P. R. J . Am. Chem. Soc. 1996, 118, 1339-
1347. (c) Hall, P. L.; Gilchrist, J . H.; Harrison, A. T.; Fuller, D. J .;
Collum, D. B. J . Am. Chem. Soc. 1991, 113, 9575-9585.
(12) (a) Sun, C.; Williard, P. G. J . Am. Chem. Soc. 2000, 122, 7829-
7830. (b) Sun, X.; Collum, D. B. J . Am. Chem. Soc. 2000, 122, 2459-
2463. (c) Gailano-Roth, A. S.; Kim, Y. J .; Gilchrist, J . H.; Harrison, A.
T.; Fuller, D. J .; Collum, D. B. J . Am. Chem. Soc. 1991, 113, 5053-
5055. (d) Williard, P. G.; Hintze, M. J . J . Am. Chem. Soc. 1990, 112,
8602-8604. (e) Williard, P. G.; Hintze, M. J . J . Am. Chem. Soc. 1987,
109, 5539-5541.
(13) (a) Briggs, T. F.; Winemiller, M. D.; Xiang, B.; Collum, D. B. J .
Org. Chem. 2001, 66, 6291-6298. (b) Parsons, R. L., J r.; Fortunak, J .
M.; Dorow, R. L.; Harris, G. D.; Kauffman, G. S.; Nugent, W. A.;
Winemiller, M. D.; Briggs, T. F.; Xiang, B.; Collum, D. B. J . Am. Chem.
Soc. 2001, 123, 9135-9143. (c) Sun, X.; Winemiller, M. D.; Xiang, B.;
Collum; D. B. J . Am. Chem. Soc. 2001, 123, 8039-8046. (d) Thompson,
A.; Corley, E. G.; Huntington, M. F.; Grabowski, E. J . J .; Remenar, J .
F.; Collum, D. B. J . Am. Chem. Soc. 1998, 120, 2028-2038.
(14) Williard, P. G.; MacEwan, G. J . J . Am. Chem. Soc. 1989, 111,
7671-7672.
(15) Gutmann, V. Electrochim. Acta 1976, 21, 661-667.

Mixed Alkali Metal Alkoxide/Aryloxide Catalysts
J . Org. Chem., Vol. 66, No. 26, 2001 9007
F igu r e 1. Room-temperature 1H NMR (400 MHz, THF-d8):
(a) NaOC6H4-4-t-Bu; (b) NaOC6H4-2-CH3; and (c) 1 (vertical
scale in aryl region has been increased for clarity).
VPO, the molecular weight of a 0.080 M solution of NaOt-
Bu is 383 g/mol (nominal MW ) 96 g/mol), i.e., the
average aggregation state is 4.0, a value consistent with
previous reports.2 Similarly, a 0.12 M solution of NaOC6H4-
4-t-Bu (172.2 g/mol) has a molecular weight of 738 g/mol
(0.028 M) and an aggregation of 4.3.5k,16 A 0.030 M
solution of catalyst 1 that was predicted to have a
composition corresponding to (NaOt-Bu)(NaOC6H4-4-t-
Bu)3 (612.6 g/mol) was then tested and found to have a
molecular weight of 633 g/mol (0.029 M). While not an
unambiguous structural identification, the VPO meas-
urements support the assertion that mixed clusters such
as 1 are tetramers in THF.
1
F igu r e 2. Variable-temperature H NMR (400 MHz) of 1 (0.4
M in THF-d8).
intensities were invariant, indicating that clusters of a
single nuclearity are equilibrating rather than a tet-
ramer/hexamer-type equilibria. J ackman previously ob-
served that the transesterification activity of (LiOAr)x
clusters was described by an equilibrating mixture of
tetramers (x ) 4) and hexamers (x ) 6), which react at
different rates.17
Additional evidence for the identity of the compounds
observed in the 198 K 1H NMR spectrum of 1 was
obtained by examining different NaOt-Bu/NaOC6H4-4-t-
Bu ratios, Figure 3. The ortho hydrogen of the aryloxide-
only cluster [NaOC6H4-4-t-Bu]4 resonates at 6.37 ppm,
the chemical shift of the most downfield of the resolved
signals in the mixture, Figure 3a. Increasing the alkoxide
content of the solution generates new species more
upfield of the parent, presumably the expected mixed
alkoxy/aryloxy clusters. To test the hypothesis that these
were compositional isomers of a cubic tetramer, we
systematically varied the NaOt-Bu/NaOAr ratio from
aryloxide rich (1:7) to aryloxide poor (7:1). Low-temper-
Clu ster Ch a r a cter iza tion by 1H NMR. The aryl
region of the room-temperature 1H NMR spectra of
NaOC6H4-4-t-Bu (0.40 M, THF-d8, Figure 1a) revealed
well-resolved signals, with a characteristic upfield 2-H
doublet (6.95 ppm, J ) 8.4 Hz) and a tert-butyl signal
(1.25 ppm) that is slightly downfield of NaOt-Bu (1.08
ppm). Ortho-substituted aryloxides had similar spectral
patterns in the aryl region (NaOC6H4-2-CH3 is provided
1
as an example, Figure 1b). In the room-temperature H
NMR spectrum of 1 (Figure 1c), the signals for the ortho
(LWHH ) 20 Hz, at 400 MHz) and meta to O- hydrogens
are broadened and shifted somewhat upfield (∆δ∼0.06
ppm) from the all-aryloxide case; a similar trend was
observed with catalysts 2. Broadening of the aryl signals
suggested a dynamic process that was investigated by
1
variable-temperature H NMR experiments.
As shown in Figure 2, cooling a THF-d8 solution of 1
resulted in further broadening of the aryl signals, but
more quickly in the ortho position, which ultimately loses
coupling information at 268 K (Figure 2b), broadening
further (LWHH ) 40 Hz) and becoming unsymmetrical
by 253 K, Figure 2c. Further cooling to 223 K decoalesces
this signal into a set of four sharp doublets in a 46:39:
13:2 ratio at 6.37, 6.30, 6.25, and 6.21 ppm, respectively
(Figure 2d). The 3-H signal is less informative, and
further cooling does not help to resolve the overlap,
Figure 2e. Clearly, several environments are accessible,
most reasonably suggesting that (1) multiple species are
present in solution, (2) these equilibrate on the NMR time
scale, and (3) that more than one catalyst may be
involved in turnover.
1
ature H NMR spectra, Figure 3, were consistent with
this hypothesis, as the isomer ratio was sensitive to the
salt ratio, though not on the absolute concentration. For
example, increasing the alkoxide/aryloxide ratio from 1:3
to 2:2 shifts the peak populations upfield from pure
[NaOC6H4-4-t-Bu]4, Figure 3d. An aryloxide-rich mixture,
on the other hand, shifts this distribution downfield
18
toward the all-aryloxide component, Figure 3b.
The mixture of compounds that give rise to four distinct
2-H proton signals is most easily explained by a mixture
of tetrameric cubes made up of 0:4, 1:3, 2:2, 3:1, and 4:0
combinations of NaOt-Bu and NaOC6H4-4-t-Bu. When the
integrated ratio of clusters is corrected for the number
of aryloxides in each cluster type, the normalized ratio
When differing concentrations of 1 (0.08-4.0 M) were
cooled to 198 K (THF-d8), the signal pattern and relative
(16) (a) Reichle, W. T. J . Org. Chem. 1972, 37, 4. (b) J ackman, L.
M.; Lange, B. C. Tetrahedron 1977, 33, 2737-2769.
(17) J ackman, L. M.; Petrei, M. M.; Smith, B. D. J . Am. Chem. Soc.
1991, 113, 3451-3458.

9008 J . Org. Chem., Vol. 66, No. 26, 2001
Kissling and Gagne´
F igu r e 5. Low-temperature 1H NMR (THF-d8, 198 K) of
different mixtures of NaOt-Bu and NaOC6H4-2-Me: (a) pure
NaOC6H4-2-Me (NaOAr); (b) 1:3 NaOt-Bu/NaOAr; (c) 2:2
NaOt-Bu/NaOAr; (d) 3:1 NaOt-Bu/NaOAr. The ratio of uncor-
rected ortho signals for b and c are 27:59:14:0 and 3:39:46:12,
respectively.
F igu r e 3. 1H NMR spectra for different ratios of NaOt-Bu
and NaOC6H4-4-t-Bu (THF-d8, 198 K): (a) pure NaOC6H4-4-
t-Bu (NaOAr); (b) 1:7 NaOt-Bu/NaOAr (ratio of ortho protons
) 74:23:3:0); (c) 1:3 NaOt-Bu/NaOAr (catalyst 1); (d) 2:2 NaOt-
Bu/NaOAr (ratio of ortho protons ) 14:38:35:13); (e) 3:1 NaOt-
Bu/NaOAr; (f) 7:1 NaOt-Bu/NaOAr.
Sch em e 4
the intercluster thermodynamics. On the other hand,
cluster distributions in 2-substituted aryloxides do not
follow a statistical model. For example, rather than the
expected 32:42:21:5 mixture, a 1:3 NaOt-Bu/NaOC6H4-
2-Me mixture had a 20:58:22:<1 distribution of clusters
(Figure 5b), which is disproportionately populated in the
1:3 cluster. This departure from a statistically controlled
distribution is apparently due to heightened steric in-
teractions between the ortho substituents in the all-
aryloxide cube that are relieved in the alkoxide-diluted
cases (Figure 5c,d). Like the 2-Me case, other 2-substi-
tuted aryloxides (e.g., 2-Et, 2-iPr, and 2-Cl phenoxide)
have a slight thermodynamic bias for the 1:3 complex.
The combination of VPO measurements, the statistical
analysis of low-temperature NMR for variable alkoxide/
aryloxide ratios, and the concentration-independent NMR
spectra, together, suggest that the observed equilibrating
species are tetramers, a proposition that is also supported
by symmetry arguments. Consider the four possible
mixed tetramers in Scheme 4a; the symmetry of each
cluster is such that only one b or O environment results.
Use of an aryloxide reporter ligand would then lead to
four unique clusters, the ratio of which should be sensi-
tive to the ratio of components in a predictable way. On
the other hand, the inherently lower symmetry of the
hexameric structure (another a priori reasonable struc-
ture)7 leads to many more combinations of two compo-
nents, e.g., 1:5, 2:4, 3:3, 4:2, 5:1, and 0:6. Moreover, most
of these combinations also have multiple configurational
isomers in addition to multiple chemically distinct envi-
ronments per cluster. Scheme 4b shows the possible
isomers for the 0:6, 1:5, and 2:4 combinations along with
the number of chemically distinct b environments in
F igu r e 4. Assignment of clusters present in 1 from the ortho
-OAr proton signals.
of tetramers is 35:40:19:6 for A, B, C, and D respectively
in 1 (Figure 4). These values closely match those expected
by a purely statistical analysis (32:42:21:5).19 The tet-
ramer corrected integral values for the 2:2 and 3:1
mixtures also agree with a statistical treatment and
further support the mixed tetramer hypothesis.
Ar yloxid e Effects on Clu ster Distr ibu tion s. Like
the 4-tert-butylphenyl-substituted case discussed above,
the measured and calculated cluster ratios were similar
for different ratios of NaOt-Bu with NaOC6H5, NaOC6H4-
4-Cl, and NaOC6H4-4-OMe, indicating that electronic
perturbations in the aryloxide component do not affect
(18) Similar signal patterns and shifts in signal populations were
observed in the 31P NMR of HMPA-solvated lithium phenoxide clusters
and have been assigned to mono-, di-, tri-, and tetrasolvated cubes.
See ref 5e.
(19) For statistical methods and treatment of raw NMR data, see
Supporting Information.

Mixed Alkali Metal Alkoxide/Aryloxide Catalysts
J . Org. Chem., Vol. 66, No. 26, 2001 9009
Ta ble 2. In itia l Ra tes (νin itia l) a n d 2 h Con ver sion s to
DIMP (Eq 2) for Differ en t Ra tios of Na Ot-Bu a n d
Na OC6H4-4-t-Bu
NaOt-Bu/NaOAra
νinitial (rel to 1)b
[DIMP]2h, Mc
4:0
7:1
3:1
1:1
1:3 (1)
1:7
1.13
1.02
0.87
0.93
1.00
0.80
0.05
0.07
0.085
0.13
0.17
0.17
a
b
Ar
)
-C6H4-4-t-Bu. νinitial for
1
)
0.0155
M
s-1
.
c [DIMP]equilibrium ) 0.19 M.
each. Since the isomer ratio is statistically controlled (a
flat isomer surface) and the number of observable isomers
is described by a distribution of tetrameric mixed clus-
ters, we consider these symmetry considerations to be
additional evidence for describing catalysts 1 and 2 as
equilibrating tetramers, although, as ususal, the ob-
served isomers may not be the catalytically active species.
In flu en ce of Ar yloxid e on Ca ta lyst Activity. In
addition to structural changes, we wished to know how
the catalyst activity responded to the ratio of components
or to the structure of the aryloxide. The phosphonyl-
carbonyl interchange reaction in eq 220 is sufficiently
challenging (does not reach equilibrium with 5 mol %
NaOt-Bu) that it can be utilized to measure two perfor-
mance parameters: initial rate (νinitial, M sec-1) and
catalyst longevity21 as expressed by [DIMP]2h.22 Under
the reaction conditions, the equilibrium concentration of
DIMP is 0.19 M and is typically not achievable with only
1 mol % of NaOtBu, measuring [DIMP] after 2 h provides
a convenient measure of longevity. The progress of the
reaction can be reliably assayed by aliquoting techniques
and GC analysis, and [DMMP] decreases linearly to 0.2
M DMMP (10-35 s, ∼120 turnovers). The negative slope
of the plot of [DMMP] vs time thus yields the initial rate
(νinitial) of the reaction.
F igu r e 6. Initial loss of DMMP for the reaction in eq 2 with
different 2. Initial reaction conditions: 0.50 M DMMP, 1.00
M iPrOAc, 1 mol % NaOt-Bu, 3 mol % NaOAr, in 1:25 v/v THF/
pentane.
Ta ble 3. In itia l Rea ction Ra tes (r ela tive to 1) a n d
[DIMP ]2h for Ca ta lysis of Eq 2 w ith Va r iou s 2
catalyst,
1, 2
(Ar )) (rel to 1)
catalyst,
1, 2
(Ar ))
a
b
a
b
νinitial
[DIMP]2h
(M)
νinitial
[DIMP]2h
(M)
(rel to 1)
1
1
0.17
0.14
0.14
0.12
0.07
0.05
0.09
0.11
0.15
0.13
3-Cl
3-CF3
3-Br
1.25
1.16
1.13
1.06
0.84
1.19
1.08
1.11
1.05
0.17
0.17
0.16
0.17
0.14
0.16
0.17
0.18
0.16
2-Me
2-H
0.47
0.84
0.4
0.31
0.35
0.99
0.99
0.81
0.56
2-Et
2-iPr
2-t-Bu
2-OMe
2-F
3-F
2-Npc
6-Br-2-Np
4-OMe
4-F
2-NH2
2-Cl
4-Cl
a
b
ν
initial for 1 ) 0.0155 M sec-1
.
Concentration measured after
2 h; [DIMP]equilibrium ) 0.19 M. c Np ) naphthyl.
was found to be modest (4-fold range in initial rates),
though several stereoelectronic trends emerged. Figure
6 tracks DMMP loss for several 2-substituted catalysts.
Not surprisingly, ortho substituents tended to slow the
catalyst (H > Me > Et > i-Pr ∼ t-Bu), although hetero-
substituents such as OMe, F, or NH2 partially or com-
pletely compensated for the size effect. Electron-donating
substituents at the m- or p-position modestly improved
the activity, while electron-withdrawing groups (e.g.,
3-Cl, 3-CF3, 3-Br) were more strongly activating. The
most active catalysts were 3-Cl, 3-CF3 phenol, and 6-Br-
2-naphthol, each of these also achieving near-equilibrium
conditions after 2 h. While the steric effect can most
easily be ascribed to an inhibition of substrate prebind-
ing, the electronic effect is less easily rationalized.The
rate acceleration observed with 2-heterosubstituents is
curious; for example, 2-OMe, 2-NH2, and 2-F each have
a νinitial that is comparable to 1 (only 2-OMe is shown in
Figure 6).23 The initial rate of 2-OMe is about twice that
of the 2-ethyl and 2-Cl despite having an A-value24 that
First, we determined the νinitial and longevity of differ-
ent ratios of NaOt-Bu and NaOC6H4-4-t-Bu, with the goal
of assessing the relative reactivity of the clusters that
were observed in Figure 4 (Table 2). While the differences
in initial rates are subtle and ultimately do not allow this
distinction to be made, the differences in catalyst longev-
ity are striking, as only the 1:3 and 1:7 alkoxide/aryloxide
mixtures provide catalysts that significantly approach
equilibrium (eq 2). On the basis of an analysis previously
put forward,4 we ascribe this increased longevity to a
shifting of the clusters to more soluble aryloxide-rich
catalysts. Under aryloxide-poor conditions, the catalysts
precipitate prior to establishing equilibrium.
We also wished to determine if electronic and steric
variations in the aromatic additive led to systematic
trends in catalyst activity. Overall, the aryloxide effect
(23) While it is tempting to speculate that R-heteroatom chelation
to the sodium center is the source of the rate enhancement, a
chemically convincing kinetic role for such an event is speculative at
best. Examples of ground-state chelation of the alkali metal can be
found in refs 5c and 6f,l,n.
(24) (a) J ensen, F. R.; Bushweller, C. H. Adv. Alicyclic Chem. 1971,
3, 139-194. (b) Buchanan, G. W.; Webb, V. L. Tetrahedron Lett. 1983,
24, 4519. (c) Schneider, H.; Hoppen, V. Tetrahedron Lett. 1974, 579.
(20) Initial reaction conditions are 0.5 M DMMP, 1.0 M isopropyl
acetate in a 1:40 v/v mixture of tetrahydrofuran/pentane, and 1 mol %
catalyst.
(21) Kissling, R. M.; Gagne´, M. R. J . Org. Chem. 1999, 64, 1585-
1590.
(22) The concentration of diisopropyl methylphosphonate, DIMP,
after 2 h of reaction time.

9010 J . Org. Chem., Vol. 66, No. 26, 2001
Kissling and Gagne´
rameric clusters under reaction conditions as evidenced
by VPO and low-temperature NMR studies. For para-
and meta-substituted aryloxides, the mixed cluster popu-
lations fit a statistical model of the possible two-
component tetramers, while mixed catalysts with bulky
ortho substituents do not. The activities of catalyst
mixtures as described by νinitial, both from the standpoint
of alkoxide/aryloxide ratio and aryloxide structure, are
similar, as the more aryloxide-rich catalyst mixtures are
significantly more robust. Although mechanistic inter-
pretation of electronic effects is complicated by a non-
linear free energy relationship, a screening of aryloxide
additives has identified several new catalysts with
slightly improved performance.
Ackn owledgm en t. Financial support from the Army
Research Office (DAAD190110526) and Union Carbide
(Innovation Recognition) is gratefully acknowledged.
M.R.G. is a Camille Dreyfus Teacher Scholar (2000).
R.M.K. is a GAANN Fellow (2000/2001).
F igu r e 7. Initial loss of DMMP for the reaction in eq 2 for
chloro-substituted 2.
is intermediate between the two. Further evidence that
the position of a given substituent impacts the intial rates
is well illustrated by the 2-, 3-, and 4-chloro substituted
catalysts (Figure 7).
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails, statistical methods, and treatment of NMR data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Con clu sion s
Mixed alkali metal alkoxide/aryloxide catalysts for the
ester interchange reaction exist as a mixture of tet-
J O016105H