Discovery and mechanistic study of AlIII-catalyzed transamidation of tertiary amides

Article abstract of DOI:10.1021/ja0762994

Cleavage of the C-N bond of carboxamides generally requires harsh conditions. This study reveals that tris(amido)Al^III catalysts, such as Al₂(NMe₂)₆, promote facile equilibrium-controlled transamidation of tertiary carboxamides with secondary amines. The mechanism of these reactions was investigated by kinetic, spectroscopic, and density functional theory (DFT) computational methods. The catalyst resting state consists of an equilibrium mixture of a tris(amido)Al^III dimer and a monomeric tris(amido)Al ^III-carboxamide adduct, and the turnover-limiting step involves intramolecular nucleophilic attack of an amido ligand on the coordinated carboxamide or subsequent rearrangement (intramolecular ligand substitution) of the tetrahedral intermediate. Fundamental mechanistic differences between these tertiary transamidation reactions and previously characterized transamidations involving secondary amides and primary amines suggest that tertiary amide/secondary amine systems are particularly promising for future development of metal-catalyzed amide metathesis reactions that proceed via transamidation.

Full text of DOI:10.1021/ja0762994

Published on Web 12/20/2007
Discovery and Mechanistic Study of AlIII-Catalyzed
Transamidation of Tertiary Amides
Justin M. Hoerter, Karin M. Otte, Samuel H. Gellman,* Qiang Cui,* and
Shannon S. Stahl*
Department of Chemistry, UniVersity of WisconsinsMadison, 1101 UniVersity AVenue,
Madison, Wisconsin 53706
Received August 24, 2007; E-mail: gellman@chem.wisc.edu; cui@chem.wisc.edu; stahl@chem.wisc.edu
Abstract: Cleavage of the C-N bond of carboxamides generally requires harsh conditions. This study
reveals that tris(amido)AlIII catalysts, such as Al2(NMe2)6, promote facile equilibrium-controlled transamidation
of tertiary carboxamides with secondary amines. The mechanism of these reactions was investigated by
kinetic, spectroscopic, and density functional theory (DFT) computational methods. The catalyst resting
state consists of an equilibrium mixture of a tris(amido)AlIII dimer and a monomeric tris(amido)AlIII-
carboxamide adduct, and the turnover-limiting step involves intramolecular nucleophilic attack of an amido
ligand on the coordinated carboxamide or subsequent rearrangement (intramolecular ligand substitution)
of the tetrahedral intermediate. Fundamental mechanistic differences between these tertiary transamidation
reactions and previously characterized transamidations involving secondary amides and primary amines
suggest that tertiary amide/secondary amine systems are particularly promising for future development of
metal-catalyzed amide metathesis reactions that proceed via transamidation.
Introduction
appropriate catalyst that promotes facile formation and cleavage
of the respective covalent bonds (Figure 1).
“Dynamic covalent chemistry” (DCC) has gained widespread
attention as a means of preparing organic molecules and
materials under thermodynamic, rather than kinetic, control.1
Successful applications of DCC have been demonstrated with
a variety of functional groups,2 including esters, thioesters,
imines, disulfides, and alkenes.3 The exchange reactions involv-
ing these functional groups are achieved through the use of an
The diversity and significance of carboxamide-containing
molecules in chemistry and biology suggest that amide-exchange
reactions, including transamidation and amide metathesis (eqs
1 and 2), could have broad utility in DCC applications.
(1) Reviews of dynamic covalent chemistry: (a) Lehn, J.-M. Chem. Eur. J.
1999, 5, 2455-2463. (b) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.;
Sanders, J. K. M.; Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898-
952. (c) Corbett, P. T.; Leclaire, J.; Vial, L.; West, K. R.; Wietor, J.-L.;
Sanders, J. K. M.; Otto, S. Chem. ReV. 2006, 106, 3652-3711.
(2) For selected recent applications of dynamic covalent chemistry, see: (a)
Brady, P. A.; Bonar-Law, R. P.; Rowan, S. J.; Suckling, C. J.; Sanders, J.
K. M. Chem. Commun. 1996, 319-320. (b) Oh, K.; Jeong, K.-S.; Moore,
J. S. Nature 2001, 414, 889-893. (c) Otto, S.; Furlan, R. L. E.; Sanders,
J. K. M. Science 2002, 297, 590-593. (d) Kilbinger, A. F. M.; Cantrill, S.
J.; Waltman, A. W.; Day, M. W.; Grubbs, R. H. Angew. Chem., Int. Ed.
2003, 42, 3281-3285; Angew. Chem. 2003, 115, 3403-3407. (e) Vignon,
S. A.; Thibaut, J.; Iijima, T.; Tseng, H.-R; Sanders, J. K. M.; Stoddart, J.
F. J. Am. Chem. Soc. 2004, 126, 9884-9885. (f) Cantrill, S. J.; Grubbs, R.
H.; Lanari, D.; Leung, K. C.-F.; Nelson, A.; Poulin-Kerstien, K. G.; Smidt,
S. P.; Stoddart, J. F.; Tirrell, D. A. Org. Lett. 2005, 7, 4213-4216. (g)
Cacciapaglia, R.; Di Stefano, S.; Mandolini, L. J. Am. Chem. Soc. 2005,
127, 13666-13671. (h) Shi, B.; Stevenson, R.; Campopiano, D. J.; Greaney,
M. F. J. Am. Chem. Soc. 2006, 128, 8459-8467. (i) Vial, L.; Ludlow, R.
F.; Leclaire, J.; Pe´rez-Ferna´ndez, R.; Otto, S. J. Am. Chem. Soc. 2006,
128, 10253-10257. (j) Dirksen, A.; Dirksen, S.; Hackeng, T. M.; Dawson,
P. E. J. Am. Chem. Soc. 2006, 128, 15602-15603.
(3) For representative fundamental studies of equilibrium-controlled exchange
of covalent bonds, see ref 1 and the following: (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) Nyce, G. W.; Lamboy, J. A.; Connor, E. F.;
Waymouth, R. M.; Hedrick, J. L. Org. Lett. 2002, 4, 3587-3590. (c) Ong,
T.-G.; Yap, G. P. A.; Richeson, D. S. Chem. Commun. 2003, 2612-2613.
(d) Singh, R.; Kissling, R. M.; Letellier, M.-A.; Nolan, S. P. J. Org. Chem.
2004, 69, 209-212. (e) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001,
34, 18-29. (f) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Angew. Chem.,
Int. Ed. 2006, 45, 7581-7584.
In contrast to the examples in Figure 1, however, methods for
catalytic exchange of carboxamide C-N bonds suitable for DCC
applications are not readily available.4-6 This deficiency
underlies our recent efforts to develop catalysts that promote
these reactions.7
We recently reported the first catalysts capable of promoting
transamidation of secondary amides with primary amines under
(4) Transamidation at very high temperatures (>250 °C), typically with
polyamides, has been reported: (a) Smith, M. E.; Adkins, H. J. Am. Chem.
Soc. 1938, 60, 657-663. (b) Beste, L. F.; Houtz, R. C. J. Polym. Sci. 1952,
8, 395-407. (c) Ogata, N., Makromol. Chem. 1959, 30, 212-224. (d)
Miller, I. K. J. Polym. Sci., Part A: Polym. Chem. 1976, 14, 1403-1417.
(e) McKinney, R. J. U.S. Patent 5,302,756, 1994. (f) McKinney, R. J. U.S.
Patent 5,395,974, 1995.
(5) For enzymatic approaches to secondary amide exchange reactions, see: (a)
Sergeeva, M. V.; Mozhaev, V. V.; Rich, J. O.; Khmelnitsky, Y. L.
Biotechnol. Lett. 2000, 22, 1419-1422. (b) Swann, P. G.; Casanova, R.
A.; Desai, A.; Frauenhoff, M. M.; Urbancic, M.; Slomczynska, U.;
Hopfinger, A. J.; Le Breton, G. C.; Venton, D. L. Biopolymers 1996, 40,
617-625.
9
10.1021/ja0762994 CCC: $40.75 © 2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 647-654
647

A R T I C L E S
Hoerter et al.
Table 1. Catalytic Transamidation of Tertiary Amides with
Secondary Amidesa
Figure 1. Self-exchange reactions that have been applied to dynamic
covalent chemistry (DCC) applications.
Scheme 1. Proposed Mechanism for AlIII-Catalyzed
Transamidation of Secondary Amides
a Reaction conditions: [Al2(NMe2)6] ) 4.3 mM, [carboxamide] ) 0.17
M, [amine] ) 0.17 M, 2 mL of toluene, 90 °C, 16 h. b Determined by GC
(internal standard ) triphenylmethane); data represent the average of two
reactions. Formation of N,N-dimethylcarboxamide product was observed
in ∼5% yield; the remaining amide products were I and II in each case.
secondary carboxamide to form a tris(amidate)AlIII species (2)
(eq 3).8
Kinetic and NMR-spectroscopic studies revealed that complexes
of this type represent the resting state of AlIII during catalytic
turnover and that the tris(κ2-amidate)AlIII complexes enter the
catalytic cycle via a bimolecular reaction with a primary amine
substrate (2 + RNH2 f A, Scheme 1).
The insights from this mechanistic study suggested that more
efficient catalysis could be achieved by destabilizing the tris(κ2-
amidate)AlIII species, thereby making it more reactive, or by
circumventing this species altogether. Tertiary carboxamides
lack an acidic N-H group and cannot form the amidate-alumi-
num species. Therefore, tertiary amides could react directly with
tris(amido)AlIII complexes to form a species analogous to inter-
mediate B in Scheme 1 and undergo efficient transamidation
(eq 4).
moderate conditions (eq 1).7a One of the most effective catalysts
for these reactions proved to be the homoleptic amidoaluminum
complex Al2(NMe2)6 (1). Subsequent mechanistic studies re-
vealed that this dimeric Al complex is actually a “precatalyst”.
Under the reaction conditions, it reacts rapidly with 3 equiv of
(6) Precedents for transamidation under synthetically practical conditions
are typically limited to intramolecular reactions or require a stoichio-
metric reagent: (a) Galat, A.; Elion, G. J. Am. Chem. Soc. 1943, 65, 1566-
1567. (b) Martin, R. B.; Parcell, A.; Hedrick, R. I. J. Am. Chem. Soc.
1964, 86, 2406-2413. (c) Crombie, L.; Jones, R. C. F.; Haigh, D.
Tetrahedron Lett. 1986, 27, 5151-5154. (d) Zaragoza-Do¨rwald, F.;
von Kiedrowski, G. Synthesis 1988, 11, 917-918. (e) Gotor, V.; Brieva,
R.; Gonza´lez, C.; Rebolledo, F. Tetrahedron 1991, 47, 9207-9214.
(f) Bon, E.; Bigg, D. C. H.; Bertrand, G. J. Org. Chem. 1994, 59, 4035-
4036. (g) Suggs, J. W.; Pires, R. M. Tetrahedron Lett. 1997, 38,
2227-2230. (h) Langlois, N. Tetrahedron Lett. 2002, 43, 9531-9533. (i)
Lasri, J.; Gonza´lez-Rosende, M. E.; Sepu´lveda-Arques, J. Org. Lett.
2003, 5, 3851-3853. (j) Klapars, A.; Parris, S.; Anderson, K. W.;
Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 3529-3533. (k) Alajar´ın,
M.; Vidal, A.; Tovar, F. Tetrahedron 2005, 61, 1531-1537. (l) C¸ alimsiz,
S.; Lipton, M. A. J. Org. Chem. 2005, 70, 6218-6221. (m) Dineen,
T. A.; Zajac, M. A.; Myers, A. G. J. Am. Chem. Soc. 2006, 128, 16406-
16409.
Here, we validate this hypothesis by demonstrating the first
examples of tertiary amide transamidation catalysis. The mech-
(8) Hoerter, J. M.; Otte, K. M.; Gellman, S. H.; Stahl, S. S. J. Am. Chem. Soc.
2006, 128, 5177-5183.
(7) (a) Eldred, S. E.; Stone, D. A.; Gellman, S. H.; Stahl, S. S. J. Am.
Chem. Soc. 2003, 125, 3422-3423. (b) Bell, C. M.; Kissounko, D. A.;
Gellman, S. H.; Stahl, S. S. Angew. Chem., Int. Ed. 2007, 46, 761-
763. See also: (c) Shi, M.; Cui, S.-C. Synth. Commun. 2005, 35, 2847-
2858.
(9) The stoichiometric formylation of amines with N,N-dimethylformamide has
been reported: (a) Pettit, G. R.; Thomas, E. G. J. Org. Chem. 1959, 24,
895-896. (b) Pettit, G. R.; Kalnins, M. V.; Liu, T. M. H.; Thomas, E. G.;
Parent, K. J. Org. Chem. 1961, 26, 2563-2566. (c) Kraus, M. A. Synthesis
1973, 361-362.
9
648 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

AlIII-Catalyzed Transamidation of Tertiary Amides
A R T I C L E S
Figure 2. Representative time-courses for Al2(NMe2)6-catalyzed transamidation between 3a and 4 (A) and 3b and 4 (B) in both the forward and reverse
directions. Conditions: [Al2(NMe2)6] ) 4.3 mM, [carboxamide] ) 0.17 M, [amine] ) 0.17 M, 5 mL of toluene, 90 °C.
anism of these reactions is investigated by experimental and
computational methods. The results highlight similarities and
differences between AlIII-catalyzed transamidation of secondary
and tertiary carboxamides and suggest opportunities for the
development of improved catalysts.
(3a)/piperidine (4) and N-benzyl-N-methyl-p-toluamide (3b)/
piperidine (4) proved suitable for this purpose (eq 5).
Results and Discussion
AlIII-Catalyzed Tertiary Amide Transamidation. This
study was initiated by evaluating whether the amidoaluminum
complex Al2(NMe2)6 (1) could catalyze the transamidation
of tertiary amides. We selected tertiary amide/secondary amine
pairs that should be approximately thermoneutral (Table 1).
Conditions analogous to those used in the transamidation of
secondary amides were employed: 5 mol % [Al] (2.5 mol %
[1]) in toluene as the solvent at 90 °C. The reactions were
performed in both forward and reverse directions, as shown in
Table 1, and achievement of equilibrium was demonstrated when
an identical product ratio was obtained for the forward and
reverse reactions. The data in Table 1 indicate that equilibrium
was attained for each of the six substrate pairs examined. Small
amounts of the N,N-dimethylcarboxamide (∼5%) were observed
in each reaction mixture, resulting from incorporation of the
dimethylamino fragment of 1 into the amide products. To our
knowledge, these results represent the first examples of catalytic
equilibration of tertiary amide-secondary amine mixtures.9
Following these initial demonstrations of catalytic reactivity,
we undertook mechanistic studies to probe the similarities and
differences between AlIII-catalyzed transamidation of secondary
versus tertiary amides.
In the presence of 2.5 mol % Al2(NMe2)6 (1) in toluene at 90 °C,
these substrate pairs react to produce an equilibrium mixture
of carboxamides 3 and 5 and amines 4 and 6 (Figure 2). Identical
product ratios were obtained when the reactions were performed
in the forward or reverse directions, which establishes that the
reactions reached equilibrium. A small amount of N,N-dimeth-
ylcarboxamide product (7) was detected in each of the reaction
mixtures.
Initial-rate kinetic methods were used to examine the
dependence of the reaction rate on each of the components, Al2-
(NMe2)6, amine, and carboxamide. The reaction rate was
obtained by monitoring the carboxamide concentrations by gas
chromatography during the first 5% conversion of the starting
materials. The initial-rate dependence on [Al2(NMe2)6] was
monitored over a range of 1-10 mol % of 1 (2-20 mol %
monomeric AlIII), and the rates were measured at different
concentrations of the substrates, carboxamide 3a and amine 4
(Figure 3). A saturation dependence of the rate on [1] and [3a]
was observed, but the rate was unaffected by changes in the
Kinetic Studies of AlIII-Catalyzed Transamidation of
Tertiary Toluamides. Mechanistic studies of AlIII-catalyzed
transamidation between secondary amines and tertiary carboxa-
mides were initiated by identifying substrates that could be
monitored readily by NMR spectroscopy and/or gas chroma-
tography. The substrate pairs N-benzyl-N-methylheptanamide
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 649

A R T I C L E S
Hoerter et al.
Figure 3. Kinetic data for Al2(NMe2)6-catalyzed transamidation of heptanamide 3a with piperidine 4. The [Al]- and [amide]-dependence data were fit to
a generic hyperbolic function (i.e., saturation dependence), and the line in the [amine]-dependence plot reflects a zero-order fit. Standard reaction conditions:
[1] ) 2.1 mM; [3a] ) [4] ) 0.17 M, 5 mL of toluene, 90 °C.
Figure 4. Kinetic data for Al2(NMe2)6-catalyzed transamidation of toluamide 3b with piperidine 4. The Al-dependence data was fit to a generic hyperbolic
function (i.e., saturation dependence), and the lines in the amide- and amine-dependence plots reflect a zero-order fit. Standard reaction conditions: [1] )
4.3 mM; [3b] ) [4] ) 0.17 M, 5 mL of toluene, 90 °C.
Figure 5. Kinetic data for Al2(NMe2)6-catalyzed transamidation of toluamide 3b with piperidine 4. (The quantity of piperidine (4) used in each reaction was
adjusted to account for the aluminum-bound piperidido groups in order to maintain a 1:1 ratio of amino groups derived from piperidine and benzylmethylamine.
For example, with a 2.5 mol % loading of Al2(NC5H10)6, the quantity of piperidine was reduced by 15% relative to 3.) The curve-fit of the Al-dependence
data reflects a linear least-squares fit of the data, and the lines in the amide- and amine-dependence plots reflect a zero-order fit. Standard reaction
conditions: [8] ) 4.3 mM; [3b] ) [4] ) 0.17 M; 5 mL of toluene, 90 °C.
amine concentration. No deuterium kinetic isotope effect was
observed when the N-deuterated amine was used as the substrate.
transamidation rate exhibits a linear, first-order dependence on
[8] (Figure 5). No rate dependence on [amine] or [carboxamide]
was observed (Figure 5).
Kinetic studies of transamidation of the toluamide substrate
3b (Figure 4) revealed some differences relative to those with
heptamide substrate 3a. The overall rates of toluamide transa-
midation were 5-10-fold slower than those with 3a, depending
on the reaction conditions. The kinetic data revealed a saturation
dependence on the catalyst concentration, [1], but changes to
the amide and amine concentrations had no effect on the rate.
In order to avoid the formation of 7, we prepared the tris-
(piperidido)-AlIII complex Al2(NC5H10)6 (8)10 and investigated
its use as a catalyst for transamidation between 3b and 4
(eq 6). Kinetic studies of this reaction revealed that the
Spectroscopic Studies of AlIII-Catalyzed Transamidation.
1H NMR spectroscopy was employed to probe the degenerate
transamidation (self-exchange) of N,N-dimethyltoluamide (7b)
with dimethylamine under typical catalytic conditions (toluene-
d8, 90 °C).11 Spectra of the individual reaction components, Al2-
(NMe2)6 (1), 7b, and Me2NH, are shown in Figure 6A-C. The
catalyst, 1, exists as a dimer at 90 °C, as revealed by the separate
resonances for the bridging and terminal dimethylamido ligands
(2.39 and 2.67 ppm, respectively; Figure 6A). Addition of 5
equiv of Me2NH to a solution of 1 results in broadening of the
resonances associated with both species. The resonances for the
terminal and bridging dimethylamido ligands of 1 remain
separate (Figure 6D). In contrast, addition of carboxamide 7b
to 1 results in coalescence of the resonances for terminal and
bridging dimethylamido ligands (Figure 6E); the chemical shift
(11) 1H NMR spectra of cross-exchange reaction solutions, such as that initially
containing 3b and 4, proved to be too complex to identify all of the various
species present.
(10) Passarelli, V.; Carta, G.; Rossetto, G.; Zanella, P. Dalton Trans. 2003,
1284-1291.
9
650 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

AlIII-Catalyzed Transamidation of Tertiary Amides
A R T I C L E S
Figure 6. NMR spectra relevant to AlIII-catalyzed transamidation of tertiary amides. All spectra were acquired at 90 °C in toluene-d8. (IS ) internal
standard ) trimethoxybenzene; S ) solvent ) toluene-d8.)
of the broad dimethylamido resonance at 2.57 ppm corresponds
to the weighted average of the chemical shifts for the terminal
and bridging amido ligands. This result implies that the
dimethylamido ligands of 1 undergo facile exchange between
terminal and bridging coordination sites in the presence of 7b.
The peaks associated with the methyl groups of 7b, (CH3)2N-
and Ar-CH3 (2.67 and 2.08 ppm, respectively), remain sharp,
and their chemical shifts are essentially unchanged. When all
three reaction components (1, 7b, and Me2NH) are present in
solution the CH3- resonances of dimethylamine are broad, but
the chemical shift is unchanged relative to the two-component
mixture (1 + Me2NH; Figure 6D). Sharp peaks are observed
for 7b (as in the mixture of 1 and 7b; Figure 6E), but the single
broad peak associated with dimethylamido ligands coordinated
to Al is shifted upfield (2.51 ppm) relative to the peak observed
in the two-component mixture of 1 and 7b (2.57 ppm).
the Al centers of 1 (eqs 7 and 8). Broadening of the resonances
of 1 and dimethylamine in Figure 6D can arise from coordina-
tion of Me2NH to the Al centers to form the monomeric
(Me2N)3Al-NHMe2 species (9) (eq 7), a process that will pro-
mote exchange between the terminal and bridging dimethyla-
mido ligands in 1. The observation of separate resonances for
the terminal and bridging dimethylamido ligands in this spec-
trum suggests that equilibrium coordination of Me2NH to 1
These spectroscopic data suggest that the amine and car-
boxamide substrates are capable of reversible coordination to
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 651

A R T I C L E S
Hoerter et al.
Figure 8. Relative free energies (kcal/mol) of Al2(NMe2)6 (1) and the
adducts arising from coordination of carboxamides 7a and 7b to Al(NMe2)3
derived from DFT calculations. See the Experimental Section for compu-
tational details.
Figure 7. Relative free energies (kcal/mol) of Al2(NMe2)6 (1) and the
corresponding Al-amine and Al-amide adducts (9 and 10, respectively)
derived from DFT calculations.
used to probe the reaction profile for Al-catalyzed transamidation
of tertiary amides. Coordination of N,N-dimethylacetamide to
Al provides a means of activating the carboxamide substrate,
and the acetamide adduct (Me2N)3Al-OdC(NMe2)Me (A) is
calculated to have a free energy 2.7 kcal/mol higher than the
dimeric Al complex 1.13 Nucleophilic attack of a coordinated
dimethylamido ligand on the bound acetamide yields metalla-
cyclic intermediate 12 via transition state TS1. Transamidation
occurs by interchange of the Me2N- fragments in 12 via
transition state TS2, which is the highest-energy species on the
reaction coordinate. No interaction is detected between the Al
center, and the two N atoms in TS2; the Al‚‚‚N bond distances
in this structure are 3.05 Å.
The 28.4 kcal/mol calculated activation barrier for Al-
catalyzed transamidation (Figure 9) compares favorably with
activation barriers determined experimentally. For example, Al2-
(NMe2)6-catalyzed transamidation of 3a and 4 exhibits a free
energy of activation (∆Gq) of 31.0 kcal/mol ([1] ) 4.3 mM,
Figure 3). The 2.6 kcal/mol difference between the calculated
and experimental activation energies is quite reasonable in light
of our observation that substrate structure can have a significant
influence on the rate of transamidation (e.g., compare data in
Figures 3 and 4).
Proposed Catalytic Mechanism and Analysis of Mecha-
nistic Data. A catalytic cycle consistent with the data presented
above is shown in Scheme 2. Coordination of the carboxamide
substrate to Al results in cleavage of dimer A and formation of
the monomeric Al-carboxamide adduct B. Nucleophilic attack
of an amido ligand on the coordinated carboxamide in B yields
the Al-stabilized tetrahedral intermediate C, and transamidation
occurs by interchange of the two dialkylamino fragments (C
f C′) via transition state D. Subsequent formation of the
alternate carboxamide adduct B′, followed by substitution of
the coordinated carboxamide and exchange of amido ligands,
completes the catalytic cycle.
The catalyst resting state is species A, B, or a mixture of the
two. These complexes appear to be nearly isoenergetic, with
their relative stability depending upon the identity of the
carboxamide, the amido ligands coordinated to Al, and the
reaction conditions (e.g., the Al concentration). The equilibrium
between A and B, as well as the ligand substitution steps
favors the separated reagents (1 + Me2NH, eq 7) and is
relatively slow on the NMR time scale. In the mixture of
carboxamide 7b and 1, coalescence of the resonances for
terminal and bridging amido ligands of 1 suggests that coordina-
tion of 7b to 1 is rapid on the NMR time scale (eq 8). Moreover,
in the three-component mixture (1 + 7b + Me2NH; Figure 6F),
the upfield chemical shift of the Al-bound dimethylamido
ligands relative to the weighted average of the chemical shifts
for the terminal and bridging dimethylamido ligands is consistent
with population of a new Al species, presumably the Al-
carboxamide adduct (Me2N)3Al-OdC(tolyl)(NMe2) (10) (eq
8). The latter assignment remains tentative because no changes
are evident to the N-CH3 resonances of the carboxamide, and
it is not obvious how the presence of amine shifts the position
of the equilibrium involving coordination of the carboxamide
to 1.12 The experimental assignment is supported, however, by
the computational studies described below.
Computational Studies of AlIII-Catalyzed Transamidation.
Density functional theory (DFT) calculations were used to
evaluate the energies of proposed intermediates as well as the
reaction coordinate for Al-catalyzed transamidation of tertiary
amides. The free energies of dimeric Al complex 1 and the
monomeric substrate adducts 9 and 10 were calculated to probe
the origin of the NMR spectroscopic data described above
(Figure 7). The results indicate that coordination of Me2NH to
the Al center is slightly disfavored energetically, ∆G ) +2.8
kcal/mol, relative to 1 and free amine, whereas coordination of
carboxamide 7b to Al is slightly favored, ∆G ) -1.5 kcal/
mol. These results support the conclusions derived from the
spectroscopic data described above.
The identity of the carboxamide influences the stability of
the Al-coordinated adduct. Coordination of N-piperidinyltolua-
mide to Al(NMe2)3 is calculated to be slightly favored thermo-
dynamically with respect to the dimeric Al complex 1 (∆G )
-1.3 kcal, Figure 8). In contrast, coordination of N-piperidi-
nylheptanamide to Al is slightly disfavored (∆G ) +0.3 kcal/
mol). The relative stability of these two carboxamide adducts
probably accounts for the kinetic results in Figures 3 and 4,
which indicate that the rate of transamidation exhibits a
saturation dependence on the heptanamide 3a but a zero-order
dependence on the toluamide 3b (see further discussion below).
A simplified transamidation model system, consisting of N,N-
dimethylacetamide and dimethylamine as the substrates, was
(13) Dissociation of 1 into monomeric Al(NMe2)3 is calculated to be uphill by
only 5.2 kcal/mol. Therefore, formation of the Al-carboxamide adduct 11
probably occurs by a dissociation of 1 into 2 equiv of Al(NMe2)3 followed
by coordination of the carboxamide to the three-coordinate Al center.
(12) Additional spectroscopic data is provided in the Supporting Information.
9
652 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

AlIII-Catalyzed Transamidation of Tertiary Amides
A R T I C L E S
Figure 9. Calculated free energy profile (kcal/mol) for degenerate transamidation of N,N-dimethylacetamide mediated by Al(NMe2)3 derived from DFT
calculations.
Scheme 2. Proposed Mechanism for AlIII-Catalyzed
Transamidation
ligands (relative to dimethylamido) could favor formation of a
monomeric Al complex.
The [carboxamide]-dependence data can be analyzed simi-
larly. The rate should exhibit a first-order dependence on
[carboxamide] if the catalyst resting state is the Al dimer and
a zero-order dependence if the resting state is the Al-
carboxamide adduct. In one of the three reactions studied (Figure
3), the transamidation rate exhibits a saturation dependence on
[carboxamide], and in the other two reactions, a zero-order
dependence is observed (Figures 4 and 5). The saturation
dependence reflects pre-equilibrium formation Al-carboxamide
adduct, which favors of the carboxamide adduct at high
[carboxamide].
The zero-order dependence of the transamidation rates on
[amine] (Figures 3-5) reflects the fact that free amine does
not participate in the turnover-limiting steps (B f C f D) of
the catalytic reaction. An amido ligand coordinated to Al serves
as the reactive nucleophile in the transamidation reaction, and
amido ligand exchange with free amine is fast relative to the
turnover rate.
Conclusion and Consideration of Broader Implications.
Analysis of the mechanism of Al-catalyzed transamidation of
involved in the interconversion between B′ and B, proceed
rapidly relative to formation of the tetrahedral intermediate, B
f C, and interchange of the dialkylamino fragments of
tetrahedral intermediate, C f C′.
secondary amides8 suggested that tertiary amides would be
effective transamidation substrates because they would prevent
formation of stable amidate complexes with AlIII. In ad-
dition to validating this initial hypothesis by demonstrating the
The proposed mechanism accounts for the different kinetic
first examples of transamidation of tertiary amides, we have
dependence on [Al] in different transamidation reactions
performed kinetic, spectroscopic, and computational studies to
(Figures 3-5). The rate should exhibit a first-order dependence
elucidate the mechanism of Al-catalyzed transamidation of
on [Al] if monomeric species B is the catalyst resting state but
tertiary amides. The results reveal both similarities and differ-
a half-order dependence if A is the resting state. If the resting
ences between secondary and tertiary carboxamides in catalytic
state consists of a mixture of monomeric and dimeric Al species,
transamidation reactions. The similarities are evident in the
an intermediate dependence on [Al] is expected. The nonlinear
catalytic mechanisms (Schemes
1 and 2), which fea-
rate dependence on [Al] in the transamidation reactions with
Al2(NMe2)3 as the catalyst (Figures 3 and 4) is intermediate
between first- and half-order in [Al], implying that a mixture
of monomeric and dimeric Al species is present.14 In contrast,
the transamidation reaction with piperidido-Al complex 8 as
the catalyst (Figure 5) exhibits a first-order dependence on [Al],
consistent with a monomeric catalyst resting state. These
differences in the nature of the catalyst resting state for different
reactions presumably reflect the different ligands coordinated
to Al. For example, the sterically more-demanding piperidido
ture closely related fundamental steps and intermediates. For
example, one of the key steps in both transformations is
nucleophilic attack of an amido ligand on an Al-coordinated
carboxamide. Perhaps the most important difference is the
identity of the catalyst resting state, which leads to sig-
nificant differences in the kinetic properties of the two reactions
(Table 2).
The differences between secondary and tertiary transamidation
reactions have important implications for the development of
amide metathesis methods. Amide metathesis can potentially
be achieved via successive transamidation reactions (Scheme
3). Efforts to promote secondary amide metathesis with Al
(14) A log-log plot of the [Al] concentration data for these reactions exhibits
a slope of ∼0.6, which reflects the approximate kinetic order in [Al].
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 653

A R T I C L E S
Hoerter et al.
MHz spectrometer. 1H NMR spectra were referenced to residual C6D5-
CD2H (2.09 ppm). Gas chromatography was performed with a
Shimadzu GC-17A gas chromatograph equipped with a 15 m RTX-5
capillary column (Restek). Al2(NMe2)6 was purchased from Strem and
used as received, and Al2(NC5H6)6 was prepared by using the literature
procedure for the Al-pyrrolidine complex;10 characterization data
matched the known compound.17
Table 2. Comparison of Secondary and Tertiary Transamidation
Reactions
secondary amide
transamidation
tertiary amide
transamidation
property
catalyst resting state:
General Procedure for Kinetics Studies. The following representa-
tive procedure was used to obtain initial-rate data that are contained
in Figures 3-5. Toluene stock solutions of aluminum catalyst 1 (41
mM, 0.66 g in 5.0 mL of toluene), amide 3b (0.83 M, 1.99 g in 10.0
mL of toluene), amine 4 (0.83 M, 0.71 g 10.0 mL of toluene), and
internal standard triphenylmethane (0.42 M, 1.02 g in 10.0 mL of
toluene) were prepared in the glovebox. In a 12-well temperature-
controlled reactor (IKA RCT fitted with an IKA ETS-04 controller),
1.0 mL of each of the stock solutions for 3b, 4, and internal standard
was premixed with 1.5 mL of toluene. The reaction was initiated by
addition of 0.5 mL of the catalyst stock solution (5 mol % catalyst).
The first 100 µL aliquot of the reaction mixture was withdrawn
immediately upon mixing. Samples were collected every 2 min for
initial-rate studies, and the reaction in each 100 µL aliquot was quenched
by dilution of the aliquot in 2 mL of toluene at room temperature. The
samples were analyzed by gas chromatography, and concentrations were
determined by integration relative to the internal standard, triphenyl-
methane.
kinetic order of catalytic
rate law:
[Al]
[carboxamide]
[amine]
first-order
zero-order
first-order
half-to-first-order
zero-order or saturation
zero-order
Scheme 3. Transamidation-Based Strategy to Achieve the
Metathesis of 2° and 3° Carboxamides
Representative NMR Spectroscopic Study. Stock solutions of 7b
(167 mM, 0.136 g in 2.0 mL of toluene-d8), 1 (33 mM, 0.026 g in 2.0
mL of toluene-d8), and internal standard1,3,5-trimethoxybenzene (42
mM, 0.046 g in 2.0 mL of toluene-d8) were prepared in a glovebox.
Three NMR tubes were prepared with 40, 120, or 200 µL of 7b (2, 6,
and 10 equiv relative to 1) and 200 µL each of solutions of 1 and
internal standard. The solvent volume of each NMR tube was brought
up to 0.5 mL with toluene-d8 (360, 180, and 100 µL, respectively),
and NMR spectra were acquired at 28 and 90 °C on a 500 MHz
spectrometer.
DFT Calculations. Calculations were performed with the Gaussian
03 program18 and the Becke3LYP functional.19 Initial geometries were
obtained using the lanl2dz basis set followed by optimization and
frequency analysis using the 6-31G(d,p) basis set for all atoms except
Al, for which the lanl2dz basis set with an additional a polarization
function (ú ) 0.325) was used. Final total energies were obtained with
the 6-311+G(d,p) on all atoms. The nature of all stationary points was
confirmed by performing frequency analyses. For transition states,
subsequent intrinsic reaction coordinate (IRC) calculations were
performed to demonstrate their connection to the adjacent ground states.
All quoted free energies include a correction for zero-point energies,
and thermal contributions (within the rigid-rotor-harmonic-oscillator
approximation) are adjusted to 363.15 K (90 °C).
transamidation catalysts were unsuccessful, however.15,16 Mecha-
nistic studies later revealed the probable origin of this failure:
the rate law for transamidation of 2° amides is first-order in
[Al] and [amine], both of which will be low (or negligible, in
the case of [amine]) in amide metathesis reactions. Furthermore,
the only amide metathesis reagent present in high concentration
is the carboxamide, but the rate law for secondary transamida-
tions is zero-order in [carboxamide].
The kinetic properties of Al-catalyzed tertiary transamidation
(Table 2) are quite distinct from those for secondary transami-
dation, and they appear compatible with a transamidation-based
strategy for tertiary amide metathesis. Most importantly, the rate
law for tertiary transamidation is zero-order in [amine]. The
dependence of the rate on [Al] ranges from half- to first-order,
and the dependence on [carboxamide] varies from zero-order
to saturation behavior. On the basis of these observations, we
are currently pursuing the development of tertiary amide
metathesis reactions. An important goal of this ongoing work
is the identification of improved transamidation catalysts.
Most DCC applications utilize noncovalent interactions to
bias the thermodynamic preference for a particular product
among an equilibrium mixture, and optimal implementation of
amide-exchange catalysis in DCC applications will require
the identification of catalysts that operate at lower temper-
atures.
Acknowledgment. We are grateful to the NSF Collaborative
Research in Chemistry program (CHE-0404704) for financial
support of this work.
Supporting Information Available: Complete ref 16, ad-
ditional 1H NMR spectroscopic data for mixtures of Al2(NMe2)6,
carboxamide, and amine, and Cartesian coordinates and energies
for computed structures. This material is available free of charge
via the Internet at http://pubs.acs.org.
Experimental Section
General. All syntheses and sample preparations were performed
using Schlenk techniques or in a nitrogen atmosphere glovebox.
Toluene-d8 was dried over sodium and benzophenone and distilled prior
to use. Other solvents were dried with activated alumina purification
columns. NMR spectral data were obtained using a Varian Unity 500
JA0762994
(17) Andrianarison, M. M.; Ellerby, M. C.; Gorrel, I. B.; Hitchcock, P. B.; Smith,
J. D.; Stanley, D. R. J. Chem. Soc., Dalton Trans. 1996, 211-217.
(18) Frisch, M. J.; et al. Gaussian 03, revision B.04; Gaussian, Inc.: Pittsburgh,
PA, 2004.
(15) Lee, S. E.; Gellman, S. H.; Stahl, S. S. Unpublished results.
(16) Successful development of secondary amide metathesis was achieved by
employing an alternative, transacylation-based mechanism. See ref 7b.
(19) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372-1377. (b) Lee, C. T.;
Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
9
654 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008