Reactivity of individual organolithium aggregates: A RINMR study of n-butyllithium and 2-methoxy-6-(methoxymethyl)phenyllithium

Article abstract of DOI:10.1021/ja0689334

Low-temperature rapid injection NMR (RINMR) experiments were performed on two lithium reagents, n-butyllithium and 2-methoxy-6-(methoxymethyl)phenyllithium (5), with the goal of measuring the relative reactivity of the different aggregates (dimer, mixed dimer, and tetramer for n-BuLi, monomer and tetramer for 5) toward typical electrophiles. The reaction of the n-BuLi dimer with (trimethylsilyl)acetylene first forms the mixed dimer n-BuLi·Me₃SiC≡CLi, which is about 1/60 as reactive as the n-BuLi homodimer. The tetramer does not react. In the deprotonation of (phenylthio)acetylene, the n-BuLi dimer was found to be 3.5 × 10⁸ as reactive as the tetramer, and in the addition to p-diethylaminobenzaldehyde, the relative reactivity was at least 2 × 10⁴. In the deprotonation of (p-tolylsulfonyl)acetylene, the monomer of 5 was at least 10¹⁴ times as reactive as the tetramer. These measurements show that the difference in reactivity between the lower and higher aggregates of organolithium reagents can be many orders of magnitude higher than all previous estimates. Copyright

Full text of DOI:10.1021/ja0689334

Published on Web 03/07/2007
Reactivity of Individual Organolithium Aggregates: A RINMR Study of
n-Butyllithium and 2-Methoxy-6-(methoxymethyl)phenyllithium
Amanda C. Jones, Aaron W. Sanders, Martin J. Bevan, and Hans J. Reich*
Department of Chemistry, UniVersity of Wisconsin, Madison, Wisconsin 54706
Received December 13, 2006; E-mail: reich@chem.wisc.edu
The high sensitivity of organolithium rates and selectivities to
solvation is often attributed to variation in aggregate concentrations.
Although lower aggregates appear to be more reactive in most cases,
the relative reactivity of aggregates has rarely been determined1a,2,3,4a
since kinetic studies are normally performed under Curtin-
Hammett conditions, where contributions to the rate by different
aggregates are inferred from nonidealized reaction orders in the
organolithium reagent. More is known about N- and O-lithiated
species such as amides and enolates.5,6a
We have developed a rapid injection NMR (RINMR) apparatus
which operates at low enough temperatures that reaction rates faster
than the rate of interconversion of aggregates can be measured.
Our apparatus has injection syringes mounted above the spectrom-
eter, making room for mechanical stirring of the NMR sample, so
that efficient (<1 s) mixing can be achieved at any temperature in
the liquid range of the solvent. All actions (lowering and raising
of the stirrer paddle, injection of one or two substrates) are
controlled manually or by the pulse program of the spectrometer.7
The temperature jump that occurs when the warm reactant is
injected into the cold sample can be measured in real time using
our 13C chemical shift thermometer.4b Nearly constant temperatures
are achieved by raising the temperature setting the appropriate
amount (typically 5-6 degrees) 10 s before an injection.7 At starting
temperatures below -135 °C, freezing (in mixtures of Me2O and
THF) becomes problematic; therefore, the fastest process we can
study has a ∆Gq ) 8.6 kcal/mol at -130 °C (t1/2 ) 2.7 s). We
Figure 1. Selected spectra, concentration versus time plot, and kinetic
scheme derived from a RINMR experiment of 1a (0.20 mL in THF/ether,
2 equiv) injected into 4 mL of n-BuLi (0.026 M) in 1:3 THF/Me2O at -135
°C. The lines are simulations using the kinetic scheme shown.
report here on the individual aggregate reactivity of n-butyllithium
dimers and tetramers (interconversion barrier of 10.3 kcal/mol1b,8a,b
)
and the tetramer and monomer/dimer of 2-methoxy-6-(methoxy-
methyl)phenyllithium (5, barrier of 11.7 kcal/mol). The reactivity
of n-BuLi aggregates has been previously studied using a RINMR
apparatus of a different design.1a,c,d
Figure 1 shows selected 7Li NMR spectra from a RINMR experi-
ment of the deprotonation of (trimethylsilyl)acetylene (1a) by the
dimer and tetramer of n-BuLi. Under standard conditions (1:3 THF/
Me2O at -129 ( 1 °C), the n-BuLi dimer signal disappears during
that the reaction was approaching first-order dependence on the
acetylene (k2
) 1.3 × 10-3 M-1 s-1). The reaction with (p-
T-1c
tolylsulfonyl)acetylene (1d) was also briefly studied; it reacted with
the n-BuLi tetramer too fast to measure (k2
least 40 000 times as reactive as 1c).
g 50 M-1 s-1, at
T-1d
With a rate constant for the n-BuLi tetramer reacting with 1c in
hand, we needed to estimate the rate constant for the dimer. The
rate of reaction of deuterated (trimethylsilyl)acetylene (d-1a) with
the n-BuLi dimer was readily measurable and first order in acetylene
the 1 s mixing period (k2
> 20 M-1 s-1). An intermediate
D-1a
species, identified as the mixed dimer 3a,7,8c forms and then decays
in a second-order process (k2MD-1a ) 2.6 M-1 s-1). The final product
is the acetylide homodimer 4a.5d,8c,d,10,11 The n-BuLi tetramer does
not react with 1a but must first dissociate to the dimer since its
(k2
) 2.1 M-1 s-1). A deuterium isotope effect of 38 ( 10
D-d-1a
was measured by injection of n-BuLi into a mixture of 1a and d-1a
(this is an average of the isotope effects for deprotonation by n-BuLi
dimer and by 3a). We were also able to measure by RINMR the
rate of disappearance is independent of acetylene concentration (k1
isotope effect for reaction of 3a with 1a and d-1a (k2MD-1a/k2
T
MD-d-1a
) 1.9 × 10-4 s-1, k2T-1a < 5 × 10-4 M-1 s-1).9 We can calculate
k2D-1a/k2T-1a > 40 000 as a conservative lower limit since it includes
neither the actual rate of the tetramer nor that of the dimer.
To measure the actual rate ratio, we examined more reactive
acetylenes to find one sufficiently acidic to react directly with the
n-BuLi tetramer. (Triphenylsilyl)acetylene (1b) was unreactive, but
for (phenylthio)acetylene (1c), reaction with the tetramer could be
detected. At low concentrations, the tetramer disappearance was
marginally faster than reactions with 1a and 1b, where dissociation
is the sole pathway. Increasing the concentration of 1c confirmed
) 65). From these values, we can calculate k2D-1a ) 52k2
)
D-d-1a
110 M-1 s-1
.
Competition experiments were then performed between 1a, 1b,
and 1c in reactions with n-BuLi, giving relative rates of 1:63:3800.
Thus k2
) 3800k2
) 4.2 × 105 M-1 s-1. To circumvent
D-1c
D-1a
the mixing problems in competition experiments involving very
fast reactions,12 n-BuLi in hexane was injected as its unreactive
tetrameric form into a solution of two acetylenes at -129 °C in
THF/Me2O, where its slow dissociation to dimers ensures homo-
geneity throughout the reaction. Comparison of k2
and k2
D-1c
T-1c
9
3492
J. AM. CHEM. SOC. 2007, 129, 3492-3493
10.1021/ja0689334 CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
diethylaminobenzaldehyde, where the n-BuLi dimer still reacts on
the time scale of mixing, the tetramer is unreactive (zero order in
aldehyde). This allowed a relative rate estimate of at least 20 000.
The tetramer was also unreactive with a variety of other substrates,
including benzoyl chloride, 3-methoxyacetophenone, 3-methoxy-
N,N-dimethylbenzamide, acetone, and methyl benzoate. The rate-
limiting step when n-BuLi in hexane is added to a THF solution of
such substrates is dissociation of the tetramer to dimer.
In conclusion, we have demonstrated the utility of a newly
developed RINMR apparatus by showing that the differences in
reactivity between the higher and lower aggregates of n-BuLi and
of the ArLi reagent 5 are enormously larger than previous estimates.
This has particular significance for rationales of regio-, stereo-, and
chemoselectivities of organolithium reactions which invoke com-
petition between aggregation states. The RINMR apparatus opens
up exciting possibilities for directly determining aggregate and
mixed aggregate contributions to organometallic reactions.
Figure 2. Summary of relative rates for reaction of n-BuLi aggregates
with acetylenes.
shows that n-BuLi dimer is 3.2 × 108 times as reactiVe as the
tetramer toward 1c! This value is 5 orders of magnitude larger than
any previous estimates of organolithium aggregate relative reactiv-
ity, which, with two exceptions,1a,4a were made under Curtin-
Hammett conditions and hence had limited dynamic range.
The importance of understanding mixed aggregate reactivity and
formation is highlighted by their ubiquity in asymmetric organo-
lithium-mediated reactions,6b,13 as well as the potential for reaction
products to be incorporated into the parent aggregate.6c,d Still,
quantitative information is scarce.5a We note that the acetylide ligand
in 3a has a rate-retarding effect on the reactivity of the remaining
butyl fragment (k2D-1a/k2MD-1a ) 42), which is consistent with the
weaker coordinating ability of an acetylide anion; the butyl fragment
will thus coordinate more tightly to lithium (see Figure 2 for a rate
summary). This contrasts with the higher reactivity found for
n-BuLi/alkoxide mixed tetramers1a and an enolate/amide mixed
dimer5b compared to that of the homoaggregates.
Acknowledgment. The authors wish to thank Jeremiah J. Wilke
and Kristin L. Jansen for syntheses, and Dr. Bob Shanks and Dr.
Charlie Fry for NMR assistance.
Supporting Information Available: Detailed description of the
RINMR apparatus and its use, sample spectra, kinetic runs, and charac-
terization of 3, 4, 5, and the reaction and quenching products. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) McGarrity, J. F.; Ogle, C. A.; Brich, Z.; Loosli, H.-R. J. Am. Chem.
Soc. 1985, 107, 1810-1815. (b) McGarrity, J. F.; Ogle, C. A. J. Am.
Chem. Soc. 1985, 107, 1805-1810. (c) McGarrity, J. F.; Prodolliet, J.;
Smyth, T. Org. Magn. Reson. 1981, 17, 59-65. (d) Ogle, C. A.; Johnson,
H. C.; Wang, X. L.; Strickler, F. H.; Bucca, D.; Gordon, B., III.
Macromolecules 1995, 28, 5184-5191.
(2) West, P.; Waack, R.; Purmort, J. I. J. Am Chem. Soc. 1970, 92, 840-845.
(3) Bates, T. F.; Clarke, M. T.; Thomas, R. D. J. Am. Chem. Soc. 1988, 110,
5109-5112.
(4) (a) Reich, H. J.; Whipple, W. L. Can. J. Chem. 2005, 83, 1577-1587.
(b) Sikorski, W. S.; Sanders, A. W.; Reich, H. J. Magn. Reson. Chem.
1998, 36, S118-S124.
(5) (a) Abu-Hasanayn, F.; Streitwieser, A. J. Am. Chem. Soc. 1996, 118,
8136-8137. (b) Kim, Y.-J.; Streitweiser, A. Org. Lett. 2002, 4, 573-
575. (c) Leung, S. S.; Streitwieser, A. J. Am. Chem. Soc. 1998, 120,
10557-10558. (d) Gareyev, R.; Streitwieser, A. J. Org. Chem. 1996, 61,
1742-1747.
We have performed a similar analysis on 2-methoxy-6-(meth-
oxymethyl)phenyllithium (5), unique since three coexisting ag-
gregates are present: monomer, dimer, and tetramer in a 1.0:0.4:
0.3 ratio at 0.04 M in 3:2 THF/ether at -130 °C. The lithium signals
of the monomer and dimer coalesce at -132 °C (∆Gq-132°C ca. 7.2
kcal/mol), whereas the tetramer signal coalesces with the other two
at -29 °C (∆Gq-29°C ca. 11.7 kcal/mol).7 Thus, the monomer/dimer
reactions will be under Curtin-Hammett conditions, but those of
the tetramer can be easily measured relative to the monomer/dimer
pair. The tetramer of 5 is unreactive in the deprotonation of the
highly reactive 1d (k1T ) 2.2 × 10-5 s-1, k2T-1d < 6 × 10-5 M-1
s-1 at -127 °C). Even with the much less reactive 1a, the coalesced
monomer/dimer signal of 5 disappears too quickly to measure, but
(6) (a) DePue, J. S.; Collum, D. B. J. Am. Chem. Soc. 1988, 110, 5524-
5533. (b) 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. (c) Hall, P. L.; Gilchrist, J. H.; Collum, D. B. J. Am. Chem.
Soc. 1991, 113, 9571-9574. (d) Qu, B.; Collum, D. B. J. Org. Chem.
2006, 71, 7117-7119.
the reaction with d-1a is measurable (k2
) 0.6 M-1 s-1
,
M-d-1a
assuming the monomer is the reactive species). Thus k2M-1a ) 67
M-1 s-1 using the measured isotope effect of 66. Competition
experiments such as those for n-BuLi show that the monomer of 5
reacts 2900 times as fast with 1c as with 1a. The extrapolation
from 1c to 1d cannot be made precisely, but we know that 1d is at
least 40 000 times as reactive as 1c toward the n-BuLi tetramer. If
we assume that the monomer of 5 has a similar 1d to 1c rate ratio
(other relative rates are very similar for n-BuLi dimer and 5M),
(7) For a detailed description, see Supporting Information.
(8) (a) Seebach, D.; Haessig, R.; Gabriel, J. HelV. Chim. Acta 1983, 66, 308-
337. (b) Heinzer, J.; Oth, J. F. M.; Seebach, D. HelV. Chim. Acta 1985,
68, 1848-1862. (c) Haessig, R.; Seebach, D. HelV. Chim. Acta 1983, 66,
2269-2273. (d) Bauer, W.; Seebach, D. HelV. Chim. Acta 1984, 67,
1972-1988. (e) Seebach, D.; Amstutz, R.; Dunitz, J. D. HelV. Chim. Acta
1981, 64, 2622-2626.
(9) We measure ∆Gq
) 10.3 kcal/mol for the n-BuLi tetramer
dissociation in 3:1 -M13e52°CO/THF. Barriers of ∆Gq
) 10.81b and 8.1
then we can estimate k2
) 40 000 × 2900 × k2
) 5 ×
-135°C
M-1d
M-1a
kcal/mol8b can be calculated for tetramer-dimer interchange from DNMR
activation parameters measured in pure THF.
109 M-1 s-1. Comparison of k2
and k2
shows that the
T-1d
M-1d
(10) Lithium acetylide products were characterized by 13C NMR spectros-
copy: 4a, 4b, and 4c are dimeric in 1:3 THF/Me2O,5c,d and 4d is
monomeric, with a 1:1:1:1 quartet for the C-Li carbon from coupling to
one 7Li.8a This is the first reported characterization by NMR of a
monomeric lithium acetylide.5d,14
monomer of 5 is at least 1014 times as reactiVe as the tetramer.
Such large rate effects are not necessarily general since basic
substrates could coordinate to lithium and gain access to the
relatively inaccessible anion of the higher aggregates.5c,8e McGarrity,
Ogle, Brich, and Loosli studied the reaction of n-BuLi aggregates
with benzaldehyde at -85 °C.1a While we confirmed their observa-
tion that the tetramer reacts directly with benzaldehyde (k2T ) 7.6
× 10-3 M-1 s-1), we find that the dimer reacts faster than mixing
(k2D > 32 M-1 s-1), with an estimated k2D/k2T of >4200. Thus, the
added dynamic range provided by our RINMR setup makes it
immediately clear that the dimer/tetramer rate ratio is much larger
than the factor of 10 they reported. For the less reactive p-
(11) Fraenkel, G.; Pramanik, P. J. Chem. Soc., Chem. Commun. 1983, 1527-
1529.
(12) For a discussion of mixing problems in fast competitive reactions, see:
(a) Francis, A. W. J. Am Chem. Soc. 1926, 48, 655-672. (b) Tolgyesi,
W. S. Can. J. Chem. 1965, 43, 343-355. (c) Holm, T. J. Org. Chem.
2000, 65, 1188-1192.
(13) Sugasawa, K.; Shindo, M.; Noguchi, H.; Koga, K. Tetrahedron Lett. 1996,
37, 7377-7380.
(14) Tanaka, Y.; Arakawa, M.; Yamaguchi, Y.; Hori, C.; Ueno, M.; Tanaka,
T.; Imahori, T.; Kondo, Y. Chem. Asian J. 2006, 1, 581-585.
JA0689334
9
J. AM. CHEM. SOC. VOL. 129, NO. 12, 2007 3493