Lithium diisopropylamide solvated by monodentate and bidentate ligands: Solution structures and ligand binding constants

Article abstract of DOI:10.1021/ja970029b

⁶Li and ¹⁵N NMR spectroscopic studies of lithium diisopropylamide ([⁶Li]LDA and [⁶Li,¹⁵N]LDA) in toluene/pentane solutions containing a variety of mono- and polydentate ligands are reported. LDA forms exclusively dimers in the presence of n-BuOMe, Et₂O, t-BuOMe, THF, 2- methyltetrahydrofuran, 2,2-dimethyltetrahydrofuran, tetrahydropyran, dimethoxyethane, N,N,N',N'-tetramethylethylenediamine, and MeOCH₂CH₂NR₂ (NR₂ = NMe₂, NEt₂, pyrrolidino). Addition of 1,2-dipyrrolidinoethane and (2-pyrrolidinoethyl)dimethylamine provides monomer-dimer mixtures. Treatment of LDA with trans-N,N,N',N'-tetramethylcyclohexanediamine (TMCDA) or trans- 1-(dimethyl-amino)-2-isopropoxycyclohexane in hydrocarbons afford exclusively monomers. Sparteine binds only reluctantly, giving a mixture of unsolvated oligomers and monomer. Competitions of the ethereal ligands vs TMCDA afford binding constants and associated free energies for dimer solvation which are correlated with those obtained previously for lithium hexamethyldisilazide.

Full text of DOI:10.1021/ja970029b

J. Am. Chem. Soc. 1997, 119, 5567-5572
5567
Lithium Diisopropylamide Solvated by Monodentate and
Bidentate Ligands: Solution Structures and Ligand Binding
Constants
Julius F. Remenar, Brett L. Lucht, and David B. Collum*
Contribution from the Department of Chemistry, Baker Laboratory, Cornell UniVersity,
Ithaca, New York 14853-1301
ReceiVed January 3, 1997X
Abstract: 6Li and 15N NMR spectroscopic studies of lithium diisopropylamide ([6Li]LDA and [6Li,15N]LDA) in
toluene/pentane solutions containing a variety of mono- and polydentate ligands are reported. LDA forms exclusively
dimers in the presence of n-BuOMe, Et2O, t-BuOMe, THF, 2-methyltetrahydrofuran, 2,2-dimethyltetrahydrofuran,
tetrahydropyran, dimethoxyethane, N,N,N′,N′-tetramethylethylenediamine, and MeOCH2CH2NR2 (NR2 ) NMe2, NEt2,
pyrrolidino). Addition of 1,2-dipyrrolidinoethane and (2-pyrrolidinoethyl)dimethylamine provides monomer-dimer
mixtures. Treatment of LDA with trans-N,N,N′,N′-tetramethylcyclohexanediamine (TMCDA) or trans-1-(dimethyl-
amino)-2-isopropoxycyclohexane in hydrocarbons afford exclusively monomers. Sparteine binds only reluctantly,
giving a mixture of unsolvated oligomers and monomer. Competitions of the ethereal ligands vs TMCDA afford
binding constants and associated free energies for dimer solvation which are correlated with those obtained previously
for lithium hexamethyldisilazide.
Introduction
investigation of mono- and polydentate ligands is a logical ex-
tension of previous investigations of lithium amide solvation
and aggregation. It also provides important structural and
thermochemical foundations for detailed rate studies of LDA-
mediated dehydrohalogenations described in the second portion
of the study.8
Despite the importance of organolithium reagents in organic
chemistry,1 our understanding of precisely how ligand structure
affects lithium ion coordination is still limited.2 Ligand-
dependent reactivities, selectivities, and other empirical observa-
tions are often suggested to reflect ligand binding constants
without adequate justification. We submit that an understanding
of reactivities requires a knowledge of (1) the reactant structures
and stabilities, (2) the aggregation and solvation events leading
up to the rate limiting transition structure, and (3) the influence
of organolithium reagent, solvent, and substrate on the stabilities
of the transition structures. Computational studies of the
possible transition structures become more important once
structure and rate studies establish the transition structure
stoichiometry.
(4) Solution structural studies of LDA: (a) Bernstein, M. P.; Romesberg,
F. E.; Fuller, D. J.; Harrison, A. T.; Williard, P. G.; Liu, Q. Y.; Collum, D.
B. J. Am. Chem. Soc. 1992, 114, 5100. (b) Kim, Y.-J.; Bernstein, M. P.;
Galiano-Roth, A. S.; Romesberg, F. E.; Williard, P. G.; Fuller, D. J.;
Harrison, A. T.; Collum, D. B. J. Org. Chem. 1991, 56, 4435. (c)
Romesberg, F. E.; Bernstein, M. P.; Gilchrist, J. H.; Harrison, A. T.; Fuller,
D. J.; Collum, D. B. J. Am. Chem. Soc. 1993, 115, 3475. (d) Galiano-
Roth, A. S.; Collum, D. B. J. Am. Chem. Soc. 1989, 111, 6772. (e) Galiano-
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. (f) Romesberg, F. E.;
Gilchrist, J. H.; Harrison, A. T.; Fuller, D. J.; Collum, D. B. J. Am. Chem.
Soc. 1991, 113, 5751.
(5) Solid state structural studies of LDA: Mair, R. S.; Clegg, W.; O’Neil,
P. A. J. Am. Chem. Soc. 1993, 115, 3388. Williard, P. G.; Hintze, M. J. J.
Am. Chem. Soc. 1987, 109, 5539. Williard, P. G.; Hintz, M. J. J. Am. Chem.
Soc. 1990, 112, 8602. Zarges, W.; Marsch, M.; Harmes, K.; Boche, G.
Angew. Chem., Int. Ed. Engl. 1989, 28, 1392. Barnett, N. D. R.; Mulvey,
R. E.; Clegg, W.; O’Neil, P. A. J. Am. Chem. Soc. 1991, 113, 8187. Williard,
P. G.; Salvino, J. M. J. Org. Chem. 1993, 58, 1. Henderson, K. W.; Dorigo,
A. E.; Liu, Q.-Y.; Williard, P. G.; Schleyer, P. v. R.; Bernstein, P. R. J.
Am. Chem. Soc. 1996, 118, 1339. Also, see ref 4a.
(6) Other physicochemical studies of LDA: (a) Romesberg, F. E.;
Collum, D. B. J. Am. Chem. Soc. 1992, 114, 2112. (b) Bernstein, M. P.;
Collum, D. B. J. Am. Chem. Soc. 1993, 115, 789. (c) Bernstein, M. P.;
Collum, D. B. J. Am. Chem. Soc. 1993, 115, 8008. (d) Romesberg, F. E.;
Collum, D. B. J. Am. Chem. Soc. 1994, 116, 9187. (e) Romesberg, F. E.;
Collum, D. B. J. Am. Chem. Soc. 1994, 116, 9198. Romesberg, F. E.;
Collum, D. B. J. Am. Chem. Soc. 1995, 117, 2166. (f) Renaud, P.; Fox,
M. A. J. Am. Chem. Soc. 1988, 110, 5702. (g) Xie, L.; Saunders, W. H.,
Jr. J. Am. Chem. Soc. 1991, 113, 3123. (h) Newcomb, M.; Varick, T. R.;
Goh, S.-H. J. Am. Chem. Soc. 1990, 112, 5186. (i) Fraser, R. R.; Mansour,
T. S. Tetrahedron Lett. 1986, 27, 331. (j) Kopka, I. E.; Fataftah, Z. A.;
Rathke, M. W. J. Org. Chem. 1987, 52, 448. (k) Newcomb, M. A.; Burchill,
M. T. J. Am. Chem. Soc. 1984, 106, 8276. (l) Mass, R. J.; Rickborn, B. J.
Org. Chem. 1986, 51, 1992. (m) Ahlbrecht, H.; Schneider, G. Tetrahedron
1986, 42, 4729. (n) Fraser, R. R.; Baignee, A.; Bresse, M.; Hata, K.
Tetrahedron Lett. 1982, 23, 4195. (o) Ashby, E. C.; Mehdizadeh, A.;
Deshpande, A. K. J. Org. Chem. 1996, 61, 1322 and references cited therein.
(7) Reviews of lithium amide structural studies: (a) Gregory, K.;
Schleyer, P. v. R.; Snaith, R. AdV. Inorg. Chem. 1991, 37, 47. (b) Mulvey,
R. E. Chem. Soc. ReV. 1991, 20, 167. (c) Collum, D. B. Acc. Chem. Res.
1993, 26, 227.
We describe a two-part investigation of lithium diisopropyl-
amide (LDA).3 In this paper we will present NMR spectro-
scopic investigations of LDA in the presence of a variety of
monodentate and bidentate ligands.4-7 While ethereal ligands
provide dimeric LDA, several diamines afford the first examples
of monomeric LDA. We will show how a strongly coordinating
diaminestrans-N,N,N′,N′-tetramethyl-1,2-cyclohexanedi-
aminescan be used to determine the relative binding constants
of the ethereal ligands on the LDA dimer fragment. This
X Abstract published in AdVance ACS Abstracts, May 1, 1997.
(1) (a) Klumpp, G. W. Recl. TraV. Chim. Pays-Bas 1986, 105, 1. (b)
Polyamine-Chelated Alkali Metal Compounds; Langer, A. W., Jr., Ed.;
American Chemical Society: Washington, 1974. (c) Ions and Ion Pairs
in Organic Reactions; Szwarc, M., Ed.; Wiley: New York, 1972; Vols. 1
and 2. (d) Jackman, L. M.; Bortiatynski, J. In AdVances in Carbanion
Chemistry; JAI: New York, 1992, Vol. 1, pp 45-87.
(2) Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624.
(3) LDA and related lithium dialkylamides are important reagents in
synthetic organic chemistry. For reviews, see: d’Angelo, J. Tetrahedron
1976, 32, 2979. Heathcock, C. H. In ComprehensiVe Carbanion Chemistry;
Buncel, E., Durst, T., Eds.; Elsevier: New York, 1980; Vol. B, Chapter 4.
Cox, P. J.; Simpkins, N. S. Tetrahedron: Asymmetry 1991, 2, 1. Asymmetric
Synthesis; Morrison, J. D., Ed.; Academic Press: New York, 1983; Vols.
2 and 3. Evans, D. A. In Asymmetric Synthesis; Morrison, J. D., Ed.;
Academic Press: New York, 1983; Vol. 3, Chapter 1. Snieckus, V. Chem.
ReV. 1990, 90, 879.
S0002-7863(97)00029-2 CCC: $14.00 © 1997 American Chemical Society

5568 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Remenar et al.
Chart 1
to 0.1 M solutions of [6Li,15N]LDA in 2:1 toluene-pentane
affords exclusively dimers 1-7. NMR spectra display 6Li
triplets and 15N quintets characteristic of cyclic dimers (Table
1).7c Although free and bound ethereal ligands could not be
observed in the slow exchange limit down to -125 °C,12
compelling spectroscopic,10 kinetic,4a,d,6c and computational4a,c,6a,6d
evidence indicates that the dimers are disolvated.
Table 1. 6Li and 15N NMR Spectroscopic Dataa
compd
δ 6Li (mult)b
δ
15N (mult)b
JLi-N (Hz)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1.98 (t)
1.88 (t)
1.84 (t)
1.51 (t)
1.89 (t)
1.93 (t)
1.76 (t)
1.62 (d)
1.35 (d)
1.49 (d)
1.66 (d)
1.47 (d)c
2.03 (t)
1.76 (t)
1.83 (t)
1.54 (t)
1.26 (t)
1.77 (t)
1.48 (t)
1.63 (d)
74.6 (q)
73.2 (q)
74.0 (q)
69.1 (q)
74.2 (q)
73.9 (q)
72.1 (q)
93.7 (t)
94.3 (t)
95.1 (t)
96.3 (t)
92.1 (t)c
73.4 (q)
71.2d
5.1
5.1
5.2
5.2
5.1
5.0
5.1
9.7
10.1
9.9
10.1
9.6c
4.6
4.9
5.0
5.0
4.3
5.0
5.0
9.5
70.9d
72.3 (q)
70.0 (q)
74.6 (q)
72.3 (-)d
94.2 (t)
a Samples contain 0.10 M [6Li,15N]LDA in 2:1 toluene-pentane
solution with variable (2-20 equiv) amounts of ligand at -90 °C.
Hydrocarbon solutions of [6Li,15N]LDA containing chelating
diamines and related polydentate ligands (Chart 1)9 form a
variety of structures depending upon the choice of ligand.
TMEDA (D) had been shown previously to afford exclusively
disolvated dimer 13.4a Sparteine (A) displays a reluctance to
solvate LDA, affording LDA monomer 8 along with substantial
concentrations of unsolvated oligomers characterized previously.4b
Monomer 8 becomes the sole observable form only upon
addition of >10 equiv of sparteine. Treatment of [6Li,15N]LDA
in 2:1 toluene-pentane at -125 °C with 2.0 equiv of dipyrro-
lidinoethane (F) affords a mixture of dimer 14 and monomer 9
(1:5) along with low concentrations of unsolvated oligomers.
Similarly, ethylenediamine G containing a dimethylamino and
a pyrrolidino group affords dimer 15 and monomer 10 (25:1).
In each case, increasing the ligand concentration does not change
the monomer-dimer proportions, consistent with equivalent per-
lithium solvation numbers. N,N,N′,N′-Tetraethylethylenedi-
amine (E) is the poorest ligand; treatment of LDA with 5 equiv
of E affords approximately 30% monomer 12 and 70%
unsolvated LDA.
6
b Chemical shifts are reported in ppm relative to 0.3 M LiCl/MeOH
at -100 °C (0.0 ppm) and neat Me2NEt (25.7 ppm). All J values are
reported in hertz. Multiplicities are listed as d ) doublet, t ) triplet,
q ) quartet, and br m ) broad multiplet. c Spectra recorded in neat
TMEDA at -55 °C as described previously.4a d The small 15N resonance
6
could only be observed with Li broad-band decoupling.
Results
Mixtures of LDA and potentially chelating ligands (Chart
1)9,10 were studied with use of variable-temperature 6Li and 15
N
NMR spectroscopy.7c The spectra were recorded on 0.1 M
solutions of [6Li,15N]LDA4b,f with 2:1 toluene-pentane mix-
tures. The Li-N connectivities stem directly from the spin 1
6Li and spin /2 15N coupling patterns (Table 1). Figure 1
1
6
includes Li and 15N NMR spectra displaying representative
multiplets. All other spectra are located in Supporting Informa-
tion. The experimental protocols are similar to those described
in greater detail elsewhere.11
LDA Solution Structures. Addition of 1.0 equiv of THF,
Et2O, n-BuOMe, t-BuOMe, THP, 2-MeTHF, and 2,2-Me2THF
We investigated DME (H) and related amino ethers I, J, and
K, several of which will assume a central role in the following
paper. Treating [6Li,15N]LDA with 2.0 equiv of H, I, J, or K
(8) Remenar, J. F.; Collum, D. B. J. Am. Chem. Soc. 1997, 119, 5573.
(9) For extensive bibliographies to each of the ligands depicted in Chart
1, see ref 10a.
(10) (a) Lucht, B. L.; Bernstein, M. P.; Remenar, J. F.; Collum, D. B. J.
Am. Chem. Soc. 1996, 118, 10707. (b) Lucht, B. L.; Collum, D. B. J. Am.
Chem. Soc. 1996, 118, 2217. (c) Lucht, B. L.; Collum, D. B. J. Am. Chem.
Soc. 1995, 117, 9863. (d) Lucht, B. L.; Collum, D. B. J. Am. Chem. Soc.
1994, 116, 6009.
(11) Hall, P.; Gilchrist, J. H.; Harrison, A. T.; Fuller, D. J.; Collum, D.
B. J. Am. Chem. Soc. 1991, 113, 9575.
(12) Addition of <1.0 equiv of ethers per lithium affords a complex
series of partially solvated oligomers in the slow exchange limit. Remenar,
J. F. Unpublished.

LDA SolVated by Mono- and Bidentate Ligands
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5569
Figure 2. Representative plot used to determine the relative binding
contant, Keq, of ethereal ligands to the LDA dimer. The function
represents a linear least-squares fit to eq 2.
Figure 1. Representative 6Li and 15N NMR spectra (A and B,
relative binding through competition of ethereal ligands with
respectively) showing coupling of spin 1 6Li and spin
/
15N. The
2
1
trans-TMCDA, B (eqs 1 and 2).10 Defining Keq(2) according to
spectra were recorded on a sample containing 0.10 M [6Li,15N]LDA in
2:1 toluene-pentane with 2 equiv of TMCDA and 20 equiv of Et2O at
-90 °C.
affords exclusively disolvated dimers 16-19 (respectively). The
assignments as η1 (non-chelated), oxygen-coordinated ligands
are based on previous investigations of LiHMDS10 as well as
the binding constant determinations described below. The
preference for oxygen rather than nitrogen coordination results
from the substantial steric demands of the trialkylamino
groups.4a,6a,10b,13,14 Amino ether C affords exclusively monomer
20.15 We note parenthetically that the ligands of 16-19 cause
high exchange rates when compared to simple monodentate
ligands as evidenced by the low probe temperatures required
6
to observe Li-15N coupling.
Relative Ethereal Ligand Binding Constants. The task of
determining ethereal ligand binding constants is notoriously
difficult.16 For example, we could not observe free and LDA-
bound ethereal ligands in the slow exchange limit, precluding
direct methods. Instead, we employed a protocol for measuring
(13) For a discussion of steric effects of amines in the context of transition
metal ligation see: Seligson, A. L.; Trogler, W. C. J. Am. Chem. Soc. 1991,
113, 2520. Choi, M.-G.; Brown, T. L. Inorg. Chem. 1993, 32, 1548.
Widenhoefer, R. A.; Buchwald, S. L. Organometallics 1996, 15, 3534.
(14) For early studies revealing poor coordination of trialkylamines to
organolithium aggregates, see: Settle, F. A.; Haggerty, M.; Eastham, J. F.
J. Am. Chem. Soc. 1964, 86, 2076. Lewis, H. L.; Brown, T. L. J. Am.
Chem. Soc. 1970, 92, 4664. Brown, T. L.; Gerteis, R. L.; Rafus, D. A.;
Ladd, J. A. J. Am. Chem. Soc. 1964, 86, 2135. Quirk, R. P.; Kester, D. E.
J. Organomet. Chem. 1977, 127, 111.
(15) The methoxy analog of ligand C appears to be base labile, affording
mixtures of some form of mixed dimer manifesting a 6Li doublet with a
small (4.6 Hz), dimer-like coupling constant along with cyclic dimer to the
exclusion of monomer.
(16) Lewis, H. L.; Brown, T. L. J. Am. Chem. Soc. 1970, 92, 4664.
Sergutin, V. M.; Zgonnik, V. N.; Kalninsh, K. K. J. Organomet. Chem.
1979, 170, 151. Quirk, R. P.; Kester, D. E. J. Organomet. Chem. 1977,
127, 111. Xu, W. Y.; Smid, J. J. Am. Chem. Soc. 1984, 106, 3790. Chan,
L.-L.; Wong, K. H.; Smid, J. J. Am. Chem. Soc. 1970, 92, 1955. Chan,
L.-L.; Smid, J. J. Am. Chem. Soc. 1967, 89, 4547. Smetana, A. J.; Popov,
A. I. J. Solution Chem. 1980, 9, 183. Gerhard, A.; Cobranchi, D. P.;
Garland, B. A.; Highley, A. M.; Huang, Y.-H.; Konya, G.; Zahl, A.; van
Eldik, R.; Petrucci, S.; Eyring, E. M. J. Phys. Chem. 1994, 98, 7923.
Spencer, J. N.; Mihalick, J. E.; Nicholson, T. J.; Cortina, P. A.; Rinehimer,
J. L.; Smith, J. E.; Ke, X.; He, Q.; Daniels, S. E. J. Phys. Chem. 1993, 97,
10509. Also, see ref 3.
eq 3, the relative dimer solvation energies (eqs 4 and 5) can be
calculated. A plot of [LDA-S]1/2/[11] vs [S]/[B] affords a line
of slope 1/Keq(1); an example is shown in Figure 2. The binding
constants for ethereal solvation of the dimers and affiliated free
energies are compiled in Table 2.17,18 Despite concerns that
mixed solvated monomers19 containing both ether and diamine
ligands might skew the results, previous investigations of
LiHMDS showed that the relative binding free energies
(17) Whereas a 1:2 toluene-pentane mixture affords a 6:1 mixture of
dimer 14 and monomer 9, a 2:1 toluene-pentane mixture affords a 2:1
mixture of 14 and 9. A similar hydrocarbon dependence of the LiHMDS
monomer-dimer equilibrium in the presence of monodentate trialkylamines10b
and chelating amines10a was traced to the stabilization of the disolVated
monomer by toluene.18
(18) Wu, S.; Lee, S.; Beak, P. J. Am. Chem. Soc. 1996, 118, 715.
Dougherty, D. A. Science 1996, 271, 163. Mills, N. S.; Ruud, C. C. J.
Chem. Soc., Perkin Trans. 2 1996, 2035. Pitchumani, K.; Ramamurthy,
V. Tetrahedron Lett. 1996, 37, 5297. For leading references to +Li-arene
solvates, see: Siemeling, U.; Redecker, T.; Neumann, B.; Stammler, H.-G.
J. Am. Chem. Soc. 1994, 116, 5507.
(19) Zarges, W.; Marsch, M.; Harms, K.; Boche, G. Chem. Ber. 1989,
122, 2303. Karsch, H. H.; Appelt, A.; Mueller, G. Organometallics 1985,
4, 1624.

5570 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Remenar et al.
Table 2. Binding Constants and Affiliated Free Energies for LDA
Dimer Solvation by Ethereal Ligands
to solVated dimer is indicative of the monomer-ligand interac-
tion energy. One might be tempted, for example, to infer from
the formation of predominantly monomer by dipyrrolidinoethane
(F) and dimer by TMEDA4a that dipyrrolidinoethane is a better
monomer ligand. However, sparteine is an even weaker
monomer ligand than F as evidenced by the substantial
concentrations required to consume the unsolvated LDA oli-
gomers. Yet sparteine affords monomer 11 rather than dimer
as the sole observable solvated form.
a
b
ligand
THF
THP
2-MeTHF
I
H (DME)
2,2-Me2THF
K
Keq(1)
Keq(2)
∆G° (kcal/mol)c
0.012
0.016
0.045
0.124
0.163
0.224
0.3
1.0
1.3
3.8
0.0
0.11
0.52
0.90
1.0
1.1
1.2
10
14
19
25
A complete description of solvent-dependent deaggregation
must include the influence of solvent on both the monomer
and dimer stabilities.23 Deaggregation of LDA (or any orga-
nolithium aggregate) will be maximized for those ligands
that coordinate strongly to the LDA monomer and coordinate
weakly (or not at all) to the dimer. Thus, sparteine affords
monomer rather than dimer due, at least in part, to its com-
plete failure to stabilize the dimer. Similarly, the failures of
DME (H) and bidentate amino ethers I, J, and K to deaggre-
gate LDA result as much from the relatively strong dimer
solvation by the unhindered η1 methoxy groups as from the
weak monomer chelation by the amino ether chelate.24 This
point is further underscored by amino ether C bearing a hin-
dered isopropoxy group. While ligand C affords exclusively
monomer 20, the corresponding methoxy analog affords sol-
vated dimer (and some decomposition), yet no detectable
monomer.15 Since it is unlikely that the isopropoxy group is
intrinsically monomer stabilizing when compared to a methoxy
group, the isopropoxy moiety must disproportionately destabilize
the dimer.
Ethereal Solvation of LDA Dimers: Relative Binding
Constants. Any effort to understand solvation effects within
organolithium chemistry must address solvation effects in the
organolithium reactant and transition structure. However,
quantitative determinations of lithium ion solvation energies are
notoriously elusive.16 In the rare instance where free and bound
solvent can be observed in the slow exchange limit, the relative
binding energies can be obtained by direct competition of the
ligands. Such a method afforded relative binding energies of a
wide range of ethereal solvents on LiHMDS dimers and
chelating amines on LiHMDS monomers.10c However, the
solvent exchanges on solvated LDA dimers are too fast.12
Accordingly, we turned to the less direct method. By comparing
ethereal solvation of the LDA dimers with TMCDA solvation
of the monomer (eqs 1 and 2), the relative ethereal solvation
energies can be determined (eqs 3-5). This method does not
rely upon slow ligand exchange on NMR time scales; simple
integration of the monomer-dimer proportions (along with the
free ligand concentrations) affords the relative ether binding
constants. Inspection of the results summarized in Table 2
reveals the expected decrease in solvation energy with increasing
steric demand. Furthermore, dimers 16-19 show binding
energies that are comparable to that of n-BuOMe, reinforcing
the assignment of 16-19 as methoxy-bound, non-chelated
forms. This conclusion will assume an added importance in
LDA-mediated dehydrobrominations described in the following
paper.
J
0.34
0.39
2.17
2.53
28
33
180
210
1.3
1.3
2.0
2.0
n-BuOMe
Et2O
t-BuOMe
a Equilibrium constants were calculated from 6Li NMR integrations
from samples containing 0.10 M [6Li]LDA in 2:1 toluene-pentane and
2-20 equiv each of ethereal ligands and TMCDA (B) at -80 °C (σ ≈
+10%). The values of Keq(1) were calculated according to eqs 1 and 2
with an estimated σ of +10%. b The values of Keq(2) were calculated
according to eqs 4 and 5 with an estimated σ of + 10% and are
normalized relative to THF. c The values of ∆G° were calculated from
Keq(2) and are reported on a per-lithium basis. The positive values of
∆G° indicate that THF is the most strongly coordinated ethereal ligand.
determined by this method concur with those determined by
direct competition.10 Moreover, increasing the ethereal solvent
concentrations shifts the equilibrium toward the dimer, consistent
with an ethereal solvation of the dimer but not the monomer.
During structural investigations of LDA in diamine/ether
mixtures, we noted an unusual temperature dependence in which
the ether-solvated dimers are favored at low temperature while
the TMCDA-solvated monomer (11) is favored at higher
temperature. Thus, monomer 11 is entropically favored while
the ether-solvated dimers are enthalpically favored. This sharply
contrasts with aggregate equilibria in ethers alone, in which the
lower aggregates are enthalpically favored and the higher
aggregates are entropically favored.20 We believe that the
difference stems from the relatively low solvation numbers of
chelated monomers. In ethereal solvents, deaggregation is
disfavored by the large negative translational entropy ac-
companying the coordination of additional solvents, yet is
enthalpically favored by the additional lithium-solvent contacts.
In the diamine-chelated, three-coordinate monomer 11, the low
ligand/Li ratio eliminates the adverse translational entropy and
attenuates the solvation enthalpy.
Discussion
Qualitative studies of LDA solvated by an assortment of
diamines reveal solvent-dependent mixtures of dimers, mono-
mers, and unsolvated oligomers (Table 1). The variable
aggregation states precluded systematic investigations of mono-
mer solvation energies.10 We did note, however, an 11-fold
greater monomer affinity of trans-N,N,N′,N′-tetramethylcyclo-
hexanediamine (TMCDA, B) than dipyrrolidinoethane (F).
Other comparisons lead to a qualitative sense of monomer
binding affinities. For example, sparteine (A) affords low
concentrations of monomer 8 along with substantial concentra-
tions of unsolvated oligomers, suggesting that sparteine is a poor
monomer ligand relative to B and F. Similar investigations of
LiHMDS,10 LiTMP,21 and phenyllithium22 suggest that sparteine
complexation is highly sensitive to steric congestion.
We can now take an important first step toward addressing
a persistent and fundamental question: Do lithium-solvent
interaction energies correlate for different organolithiums? A
plot of ether binding energies for LDA and LiHMDS determined
previously is shown in Figure 3. In general, there appears to
Origins of Deaggregation. The previous discussion lacks
any connotation that formation of solvated monomer relative
(20) Fraenkel, G.; Hsu, H.; Su, B. P. In Lithium: Current Applications
in Science, Medicine, and Technology; Bach, R. O., Ed.; Wiley: New York,
1985.
(21) Remenar, J. F.; Collum, D. B. Unpublished.
(22) Hoffmann, D.; Collum, D. B. Unpublished.
(23) Collum, D. B. Acc. Chem. Res. 1992, 25, 448.
(24) For a crystal structure of an LiHMDS-LiCl mixed aggregate
containing MeOCH2CH2NMe2 (l), see: Henderson, K. W.; Dorigo, A. E.;
Liu, Q.-Y.; Williard, P. G.; Schleyer, P. v. R.; Bernstein, P. R. J. Am. Chem.
Soc. 1996, 118, 1339.

LDA SolVated by Mono- and Bidentate Ligands
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5571
materials were manipulated under argon or nitrogen with use of standard
glovebox, vacuum line, and syringe techniques.
NMR Spectroscopic Analyses. Samples for spectroscopic analyses
were prepared and sealed in Vacuo following a sample preparation
6
protocol described in detail elsewhere.11 Standard Li, 15N, and 13C
NMR spectra were recorded on a Varian Unity 500 spectrometer
operating at 73.57, 58.84, and 125.76 MHz, respectively. The 6Li, 15N,
and 13C resonances are referenced to 0.3 M [6Li]LiCl/MeOH at -100
°C (0.0 ppm), neat Me2NEt at -100 °C (25.7 ppm), and the toluene
methyl resonance at -100 °C (20.4 ppm), respectively.
N,N,N′N′-Tetramethyl-trans-1,2-cyclohexanediamine (TMCDA,
B). TMCDA was prepared from commercially available trans-1,2-
cyclohexanediamine with use of the Eschweiler-Clark methylation of
amines as follows: trans-1,2-Cyclohexanediamine (100 mL, 0.82 mol)
was cooled to 0 °C in a 2-L round-bottom flask containing a stir bar.
Following dropwise addition of 88% aqueous formic acid (300 mL,
5.8 mol, 1.7 equiv), 37% aqueous formaldehyde (360 mL, 4.4 mol,
1.4 equiv) was added. Gradual heating to 60 °C initiates rapid gas
evolution. The reaction was allowed to proceed without further heating
until gas evolution subsided and was then heated to 80 °C for 24 h.
The reaction mixture was cooled, acidified with 20% aqueous HCl,
and extracted three times with 100-mL portions of ether. The aqueous
layer was stirred in a salt/ice bath and brought to pH 12 by dropwise
addition of 40% aqueous NaOH without allowing the internal temper-
ature to exceed 25 °C. Following separation of the resulting amine/
aqueous layers, the aqueous layer was further extracted three times
with 100-mL portions of ether. The combined organic layers were
dried over KOH pellets with stirring. Fractional distillation under
aspirator vacuum provided TMCDA >98% pure by GC. Further
purification was achieved by forming the bis-HCl salt from aqueous
HCl, drying under full vacuum, and recrystallizing from methanol/water
mixtures. The free base was liberated by distillation from KOH pellets.
Further drying was effected by adding additional KOH pellets (typically
5-10% by weight). Two layers sometimes form if the amine is
particularly wet. The amine is decanted and distilled from Na/
benzophenone (bp 85-87 °C) to provide 80 g (47% yield) of TMCDA
>99.9% pure by GC. 1H NMR (CDCl3) δ 1.11 (br m, 4H), 1.74 (br
m, 2H), 1.84 (br m, 2H), 2.27 (s, 12H), 2.38 (br m, 2H). 13C{1H} NMR
(CDCl3) δ 21.7, 24.5, 38.9, 62.7.
trans-N,N-Dimethyl-2-isopropoxycyclohexylamine (C). To a 500-
mL round-bottom flask fitted with a dry ice condenser was added
cyclohexene oxide (100 mL, 1.0 mol) and 40% aqueous dimethylamine
(200 mL, 1.6 mol). The mixture was refluxed with stirring for 5 h.
Cooling afforded two distinct layers. Separation of the layers and
fractional distillation of the organic layer under aspirator pressure (bp
78-80 °C) provided 110 g (77% yield) of trans-2-(N,N-dimethylami-
no)cyclohexanol, which was further dried by distillation from CaH2.
1H NMR (CDCl3) δ 1.10 (m, 1H), 1.23 (m, 3H), 1.71 (m, 1H), 1.78
(m, 2H), 2.10 (m, 1H), 2.17 (m, 1H), 2.26 (s, 6H), 3.32 (m, 1H), 3.94
(br s, 1H). 13C{1H} NMR (CDCl3) δ 20.1, 23.9, 25.1, 33.0, 39.9, 69.1,
69.3. A 250-mL round-bottom flask with a magnetic stir bar was
charged with trans-2-(N,N-dimethylamino)cyclohexanol (5 g, 0.035
mol) and a solution of triphenylphosphine (10.0 g, 0.038 mol) in 25-
mL of dry THF. The mixture was cooled to 25 °C and a solution of
CBr4 (12.5 g, 0.038 mol) in 10 mL of THF was added dropwise. The
precipitate (presumably the aziridinium salt) was filtered and dried under
full vacuum at 60 °C for 2 h. To a solution of 1.7 g of solid in 10 mL
of dry 2-propanol was added sodium metal (1.0 g, 1.2 mol). After the
initial reaction had subsided, the contents were heated to reflux until
all of the sodium dissolved. The solution was cooled to 0 °C, charged
with 10 mL of water, acidified with aqueous 10% aqueous HCl, and
extracted with three 25 mL portions of ether. The combined aqueous
layers were basicified with KOH pellets and extracted with ether. The
ether layer was dried over KOH pellets and the solvent removed in
Vacuo. Following addition of 10 mL of anhydrous ether, the ethereal
solution was treated with calcium hydride followed by potassium
hydride to scavenge any remaining alcohols. Following removal of
the ether, the product was distilled under full vacuum to yield 309 mg
of C shown to be pure by GC. 1H NMR (CDCl3) δ 1.15 (d, J ) 6.2
Hz, 3H), 1.16 (d, J ) 6.2 Hz, 3H), 1.22 (m, 5H), 1.66 (m, 2H), 1.76
(m, 1H), 2.04 (m, 1H), 2.37 (s, 6H), 3.33 (m, 1H), 3.75 (septet, J )
Figure 3. Plot of relative free energies of solvation for LDA dimer vs
LiHMDS dimer (determined previously).10a The values of ∆G° were
calculated relative to THF (in kcal/mol).
be a modest correlation.25 Binding to LiHMDS is generally
more sterically demanding; however, it is difficult to fully assess
the error deriving from distinctly different methods carried out
by different experimentalists.
Summary and Conclusions
We have shown for the first time that LDA can be deaggre-
gated by a select group of polyamines. The qualitative
investigations of ligand-dependent aggregation further illustrate
that both aggregate and monomer solvation are important
determinants.23 Quantitative studies of dimer solvation by
monodentate ethereal ligands as well as η1-oxygen-bound amino
ethers provide basic information about lithium amide dimer
solvation. The dimer solvation energies will prove generally
important as we attempt to unravel the complex relationships
of solvation, aggregation, and reactivity of lithium amides. Of
more immediate consequence, dimers 3, 16, 17, 18, and 19 are
related by nearly thermoneutral ligand substitution. The firmly
established reactant stabilities will allow us to explore how
chelation stabilizes the transition structures for LDA-mediated
dehydrohalogenations. These studies are described in the
following paper.
Experimental Section
Reagents and Solvents. All monodentate ethers, ligands, and
hydrocarbons were distilled by vacuum transfer from blue or purple
solutions containing sodium benzophenone ketyl. The hydrocarbon
stills contained 1% tetraglyme to dissolve the ketyl. Ligands A, D,
and H are available from Aldrich. Ligands B, C, E, F, G, I, J, and K
were prepared as described below.9,10a 6Li metal (95.5% enriched) was
obtained from Oak Ridge National Laboratory. The [6Li]ethyllithium
used to prepare the [6Li]LDA was prepared and purified by the standard
literature procedure.4b [6Li,15N]LDA was prepared and isolated as an
analytically pure solid as described previously.4b,f The diphenylacetic
acid used to check solution titers26 was recrystallized from methanol
and sublimed at 120 °C under full vacuum. Air- and moisture-sensitive
(25) Even the exhaustively investigated transition metal-phosphine
complexes have yielded quantitative binding constants very rarely: Li, C.;
Luo, L.; Nolan, S. P. Organometallics 1996, 15, 3456 and references cited
therein.
(26) Kofron, W. G.; Baclawski, L. M. J. Org. Chem. 1976, 41,
1879.

5572 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Remenar et al.
6.2 Hz, 1H). 13C{1H} NMR (CDCl3) δ 22.5, 23.7, 24.2, 24.7, 25.9,
31.8, 41.4, 66.9, 68.9, 76.0.
salt from methanol-water followed by liberation of the amine by
distillation from KOH pellets (bp 120-124 °C) afforded amine F in
26% overall yield. 1H NMR (CDCl3) δ 1.77 (br m, 8H), 2.52 (br m,
8H), 2.62 (s, 4H). 13C{1H} NMR (CDCl3) δ 22.7, 53.8, 54.9.
N,N-Dimethyl-2-pyrrolidinoethylamine (G) was prepared from
commercially available N,N-dimethyl-2-chloroethylamine hydrochloride
and pyrrolidine. After recrystallizing the amine hydrochloride from
methanol-water, amine G was liberated by vacuum transfer from KOH
pellets in 33% overall yield. 1H NMR (CDCl3) δ 1.77 (br m, 4H),
2.25 (s, 6H), 2.43 (t, J ) 7.0 Hz, 2H), 2.51 (br m, 4H), 2.57 (t, J ) 7.0
Hz, 2H). 13C{1H} NMR (CDCl3) δ 22.6, 45.1, 53.7, 58.0.
N,N-Diethyl-2-methoxyethylamine (J) was prepared from Et2NH
and 1,2-dichloroethane and recrystallized from THF-2-propanol.
Liberation of the amine by vacuum transfer from KOH pellets afforded
amine J in 35% overall yield. 1H NMR (CDCl3) δ 1.03 (t, J ) 7.4
Hz, 6H), 2.56 (quartet, J ) 7.4 Hz, 4H), 2.63 (t, J ) 6.2 Hz, 2H), 3.33
(s, 3H), 3.45 (t, J ) 6.2 Hz, 2H). 13C{1H} NMR (CDCl3) δ 10.8,
46.7, 51.5, 57.8, 70.4.
N,N-Dimethyl-2-methoxyethylamine (I) was prepared by the same
procedure as TMCDA (B) from commercially available 2-methoxy-
ethylamine. Following the aqueous workup, the material was dried,
dissolved in three times its volume of 10% 2-propanol-THF, and
acidified with HCl gas. Additional 2-propanol was added at reflux as
necessary to dissolve all of the salt. After crystallization and filtration,
the salt was dried under full vacuum at 60 °C for 2 h to ensure complete
removal of THF. Ligand H was liberated by vacuum transfer from
KOH (40% yield). 1H NMR (CDCl3) δ 2.25 (s, 6H), 2.47 (t, J ) 5.6
Hz, 2H), 3.34 (s, 3H), 3.45 (t, J ) 5.7Hz, 2H). 13C{1H} NMR (CDCl3)
δ 44.5, 57.4, 57.6, 69.3.
1-Methoxy-2-pyrrolidinoethane (K). Ligands F, G, J, and K were
prepared following very similar procedures; the preparation of K is
representative. To a 100-mL round-bottom flask fitted with a stir bar
and reflux condenser was added 2-(chloromethoxy)ethane (10.4 g, 0.11
mol), pyrrolidine (10.2 g, 0.14 mol), water (10 mL), and K2CO3 (10.0
g, 0.07 mol). After refluxing overnight, the organic layer was separated
and dried over KOH pellets. Toluene (10 mL) was added and the
solution was distilled, collecting fractions up to 115 °C. The product
was then distilled under aspirator vacuum, dried over CaH2, and
redistilled. Treating the distillate in 30 mL of approximately 4:1 THF-
2-propanol with gaseous HCl afforded the HCl salt as a white
precipitate, which was recrystallized directly from the THF-2-propanol.
The amine was liberated from the salt by treatment with excess KOH
pellets and fractionally distilled at aspirator pressure (bp 53-55 °C),
affording 3.6 g (25% overall yield) of pure product. 1H NMR (CDCl3)
δ 1.78 (br m, 4H), 2.54 (br m, 4H), 2.66 (t, J ) 5.8 Hz, 2H), 3.36 (s,
3H), 3.50 (t, J ) 5.8 Hz, 2H). 13C{1H} NMR (CDCl3) δ 22.8, 53.9,
55.1, 58.2, 71.0.
Acknowledgment. We thank the National Institutes of
Health for direct support of this work and for a predoctoral
training grant to B.L.L. We also acknowledge the National
Science Foundation Instrumentation Program (CHE 7904825
and PCM 8018643), the National Institutes of Health (RR02002),
and IBM for support of the Cornell Nuclear Magnetic Resonance
Facility.
Supporting Information Available: 6Li and 15N NMR
spectra of LDA (24 pages). See any current masthead for
ordering and Internet access instructions.
1,2-Dipyrrolidinoethane (F) was prepared from 1,2-dichloroethane
following the procedure desribed above. Recrystallization of the HCl
JA970029B