Mechanism of AlIII-catalyzed transamidation of unactivated secondary carboxamides

Article abstract of DOI:10.1021/ja060331x

The carbon-nitrogen bond of secondary carboxamides is generally thermodynamically and kinetically unreactive; however, we recently discovered that the trisamidoaluminum(III) dimer Al₂(NMe₂) ₆ catalyzes facile transamidation between simple secondary carboxamides and primary amines under moderate conditions. The present report describes kinetic and spectroscopic studies that illuminate the mechanism of this unusual transformation. The catalytic reaction exhibits a bimolecular rate law with a first-order dependence on the Al^III and amine concentrations. No rate dependence on the carboxamide concentration is observed. Spectroscopic studies (¹H and ¹³C NMR, FTIR) support a catalyst resting state that consists of a mixture of tris-(κ²- amidate)aluminum(III) complexes. These results, together with the presence of a significant kinetic isotope effect when deuterated amine substrate (RND ₂) is used, implicate a mechanism in which the amine undergoes preequilibrium coordination to aluminum and proton transfer to a κ²-amidate ligand to yield an Al(κ²-amidate) ₂(κ¹-carboxamide)(NHR) complex, followed by rate-limiting intramolecular delivery of the amido ligand (NHR) to the neutral Al^III-activated κ¹-carboxamide. Noteworthy in this mechanism is the bifunctional character of Al^III, which is capable of activating both the amine nucleophile and the carboxamide electrophile in the reaction.

Full text of DOI:10.1021/ja060331x

Published on Web 03/24/2006
Mechanism of AlIII-Catalyzed Transamidation of Unactivated
Secondary Carboxamides
Justin M. Hoerter, Karin M. Otte, Samuel H. Gellman,* and Shannon S. Stahl*
Contribution from the Department of Chemistry, UniVersity of Wisconsin-Madison, 1101
UniVersity AVenue, Madison, Wisconsin 53706
Received January 17, 2006; E-mail: gellman@chem.wisc.edu; stahl@chem.wisc.edu
Abstract: The carbon-nitrogen bond of secondary carboxamides is generally thermodynamically and
kinetically unreactive; however, we recently discovered that the trisamidoaluminum(III) dimer Al2(NMe2)6
catalyzes facile transamidation between simple secondary carboxamides and primary amines under
moderate conditions. The present report describes kinetic and spectroscopic studies that illuminate the
mechanism of this unusual transformation. The catalytic reaction exhibits a bimolecular rate law with a
first-order dependence on the AlIII and amine concentrations. No rate dependence on the carboxamide
concentration is observed. Spectroscopic studies (1H and 13C NMR, FTIR) support a catalyst resting state
that consists of a mixture of tris-(κ2-amidate)aluminum(III) complexes. These results, together with the
presence of a significant kinetic isotope effect when deuterated amine substrate (RND2) is used, implicate
a mechanism in which the amine undergoes preequilibrium coordination to aluminum and proton transfer
to a κ2-amidate ligand to yield an Al(κ2-amidate)2(κ1-carboxamide)(NHR) complex, followed by rate-limiting
intramolecular delivery of the amido ligand (NHR) to the neutral AlIII-activated κ1-carboxamide. Noteworthy
in this mechanism is the bifunctional character of AlIII, which is capable of activating both the amine
nucleophile and the carboxamide electrophile in the reaction.
Introduction
alcohol exchange reactions can be promoted by acids, bases,
nucleophilic catalysts, and metal alkoxides.2,3 These reactions
The secondary carboxamide group is a fundamental compo-
nent of biological and synthetic polymers (i.e., proteins and
nylons) and constitutes a ubiquitous and important functional
group in organic chemistry. Recent developments in dynamic
covalent chemistry1 suggest that facile amide exchange reactions
would enable the synthesis of important new amide-based
molecules and polyamide materials under equilibrium-controlled
conditions. Toward this end, we have recently initiated efforts
to identify catalysts that mediate efficient amide exchange
reactions, including transamidation and amide metathesis (eqs
1 and 2).
provide a powerful strategy for the manipulation of organic
molecules that possess the carboxylic ester functional group.
Catalysts for ester amidation have also been reported recently.4
In contrast, examples of transamidation are scarce, and the
limited synthetic applications to date generally consist of
intramolecular reactions.5 The difficulty of this reaction arises
from the intrinsic strength of the amide C-N bond together
with the presence of an acidic N-H bond in secondary amides.
In a neutral aqueous solution, for example, hydrolysis of
carboxylic esters proceeds more than 2 orders of magnitude
faster than hydrolysis of carboxamides.6
Transamidation generally requires harsh conditions to cleave
the chemically robust amide bond. For example, high temper-
(3) For important recent leading references, see: (a) Stanton, M. G.; Gagne´,
M. R. J. Am. Chem. Soc. 1997, 119, 5075. (b) Stanton, M. G.; Allen, C.
B.; Kissling, R. M.; Lincoln, A. L.; Gagne´, M. R. J. Am. Chem. Soc. 1998,
120, 5981. (c) Grasa, G. A.; Kissling, R. M.; Nolan, S. P. Org. Lett. 2002,
4, 3583. (d) Nyce, G. W.; Lamboy, J. A.; Connor, E. F.; Waymouth, R.
M.; Hedrick, J. L. Org. Lett. 2002, 4, 3587.
(4) (a) Han, C.; Lee, J. P.; Lobkovsky, E.; Porco, J. A., Jr. J. Am. Chem. Soc.
2005, 127, 10039. (b) Movassaghi, M.; Schmidt, M. A. Org. Lett. 2005, 7,
2453.
(5) (a) C¸ alimsiz, S.; Lipton, M. A. J. Org. Chem. 2005, 70, 6218. (b) Alajar´ın,
M.; Vidal, A.; Tovar, F. Tetrahedron 2005, 61, 1531. (c) Klapars, A.; Parris,
S.; Anderson, K. W.; Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 3529.
(d) Lasri, J.; Gonza´lez-Rosende, M. E.; Sepu´lveda-Arques, J. Org. Lett.
2003, 5, 3851. (e) Langlois, N. Tetrahedron Lett. 2002, 43, 9531. (f) Bon,
E.; Bigg, D. C. H.; Bertrand, G. J. Org. Chem. 1994, 59, 4035. (g) Gotor,
V.; Brieva, R.; Gonza´lez, C.; Rebolledo, F. Tetrahedron 1991, 47, 9207.
(h) Zaragoza-Do¨rwald, F.; Von Kiedrowski, G. Synthesis 1988, 11, 917.
(i) Crombie, L.; Jones, R. C. F.; Haigh, D. Tetrahedron Lett. 1986, 27,
5151. (j) Martin, R. B.; Parcell, A.; Hedrick, R. I. J. Am. Chem. Soc. 1964,
86, 2406.
Transesterification (eq 3) is a relatively facile process that
has been employed in numerous disciplines including organic
synthesis, food science, and biodiesel applications.2 These ester-
(1) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart,
J. F. Angew. Chem., Int. Ed. 2002, 41, 898.
(2) For a review, see: Otera, J. Chem. ReV. 1993, 93, 1449.
(6) Wolfenden, R.; Snider, M. J. Acc. Chem. Res. 2001, 34, 938.
9
10.1021/ja060331x CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 5177-5183
5177

A R T I C L E S
Hoerter et al.
Figure 1. (A) Representative time-course for Al2(NMe2)6-catalyzed transamidation between 2 and 3 (eq 4). The calculated curve fits reflect a nonlinear
least-squares fit of the data to an equation associated with a bimolecular approach to equilibrium ([A]t ) [A]∞ + ([A]0 - [A]∞) e-kt, k ) k1 + k-1).
Conditions: [Al2(NMe2)6] ) 4.2 mM, [carboxamide] ) 0.17 M, [amine] ) 0.17 M, 5 mL toluene, 90 °C. (B) Kinetic plot demonstrating a linear correlation
between the rate and the square of the “reaction concentration” (linear least-squares curve fit). The “reaction concentration” (defined in the text) is arbitrarily
assigned as the initial carboxamide concentration.
atures (>250 °C) have been employed to effect exchange in
amide-containing polymers and polymer/amine mixtures.7 A
comparatively mild intermolecular transamidation has been
achieved with stoichiometric or excess AlCl3.5f Intramolecular
examples have been reported that use stoichiometric AlMe3 in
combination with catalytic quantities of lanthanide reagents,5a
and Brønsted acids are effective in intramolecular transamidation
reactions that feature opening of a strained lactam ring.5c
We recently discovered that the homoleptic amidoaluminum
complex Al2(NMe2)6 catalyzes intermolecular transamidation
between primary alkylamines and secondary carboxamides
under relatively mild conditions.8 Lewis acids, such as Sc(OTf)3,
promote thermodynamically favorable transamidation reactions
between alkylamines and N-aryl carboxamides, but they are
ineffective in the more difficult (and potentially more interesting)
thermoneutral reactions between alkylamines and N-alkyl car-
boxamides. Alkali metal-amido reagents9 (e.g., LiN(SiMe3)2)
proved to be ineffective for both thermodynamically favorable
and thermoneutral reactions because these basic reagents
preferentially deprotonate secondary carboxamides without
effecting transamidation.
mides by identifying a substrate pair that could be assayed
readily by NMR spectroscopy and gas chromatography. The
substrates N-benzylheptanamide (2) and 2-ethylhexylamine (3)
(eq 4) proved suitable for this purpose. The combination of these
substrates in a 1:1 ratio (0.17 M each) with 2.5 mol % Al2-
(NMe2)6 in dry toluene at 90 °C results in transamidation,
producing an equilibrium mixture of products 2, 3, 4, and 5 in
approximately 20 h (Figure 1A). Under these conditions, the
final ratio of carboxamides, 2:4 ) 1.2:1.0 based on GC analysis,
confirms the near-thermoneutrality of the transamidation reac-
tion. An identical product ratio is obtained if the reaction is
performed in reverse, namely by initiating the reaction with the
alternate pair of substrates, 4 and 5. Formation of N,N-
dimethylheptanamide from incorporation of the Me2N- fragment
1
of Al2(NMe2)6 into the carboxamide was not detected by H
NMR spectroscopy or GC (see further analysis below).
In our initial report, we speculated that the ability of the
amidoaluminum complex to promote thermoneutral transami-
dation might reflect a bifunctional mechanism. The aluminum
center could serve both as a Lewis acid to enhance the
electrophilicity of the carboxamide group and as a site for
activation of the amine nucleophile via formation of an
amidoaluminum fragment. To test this hypothesis and support
future efforts to develop improved catalysts, we undertook a
mechanistic investigation of the Al2(NMe2)6-catalyzed transa-
midation involving primary alkylamines and secondary N-alkyl
carboxamides. The results of this study, presented below,
provide valuable insights into this unique equilibrium-controlled
transformation and provide the basis for further development
of this important class of reactions.
The reaction time-course of eq 4, monitored by GC, exhibits
well-behaved kinetics, and the data fit well to a calculated curve
for a bimolecular approach to equilibrium (Figure 1A). A
bimolecular rate law is supported further by the rate data
obtained at different “reaction concentrations” (Figure 1B). (The
“reaction concentration” was varied by changing the solvent
volume while maintaining a constant absolute quantity of all
reagents.) The initial rates exhibit a linear correlation with the
square of the reaction concentration.
The origin of the bimolecular rate law was examined by
evaluating the rate-dependence on each component of the
reaction: Al2(NMe2)6, amine, and carboxamide. The initial rate
increases linearly with respect to [Al2(NMe2)6] over a range of
1-10 mol % of the dimer (i.e., 2-20 mol % [monomeric AlIII])
(Figure 2). Since the dimeric complex dissociates into mono-
meric AlIII species under the reaction conditions, as we elaborate
below, the linear plot reflects a first-order dependence of the
rate on the concentration of monomeric AlIII species.
Results and Discussion
Kinetic Studies of AlIII-Catalyzed Transamidation. We
initiated the study of Al2(NMe2)6-promoted transamidation
between primary alkylamines and secondary N-alkyl carboxa-
(7) (a) Beste, L. F.; Houtz, R. C. J. Polym. Sci. 1952, 8, 395. (b) Ogata, N.,
Makromol. Chem. 1959, 30, 212. (c) Miller, I. K., J. Polym. Sci., Part A:
Polym. Chem. 1976, 14, 1403. (d) McKinney, R. J. U.S. Patent 5,302,756,
1994. (e) McKinney, R. J. U.S. Patent 5,395,974, 1995.
(8) Eldred, S. E.; Stone, D. A.; Gellman, S. H.; Stahl, S. S. J. Am. Chem. Soc.
2003, 125, 3422.
(9) Hereafter, we avoid using the term “amide” in order to minimize confusion
between organic “carboxamides” and metal-stabilized anionic “amido”
groups (R2N-).
9
5178 J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006

AlIII-Catalyzed Transamidation Mechanism
A R T I C L E S
1
dimethylamine, as indicated by H NMR spectroscopy (0.15
and 2.19 ppm for the N-H and N-CH3 resonances, respectively)10
(eq 6). The reaction of benzylamine (5) with Al2(NMe2)3
generates a precipitate, presumably oligomeric amido- or imido-
bridged AlIII aggregates, that remains insoluble even at 90 °C.
The reaction solutions for each of the other substrates remain
homogeneous. These observations suggest that AlIII prefers
primary amido and amidate ligands over secondary amido
ligands. Furthermore, the lack of Me2N-AlIII fragments in the
presence of the transamidation substrates provides an explana-
tion for the absence of N,N-dimethylcarboxamide products under
the catalytic reaction conditions.
Figure 2. Dependence of the initial rate on catalyst concentration for the
transamidation of carboxamide 2 with amine 3. Initial rates, derived from
monitoring the first 5% conversion, were determined by sampling [2] every
2 min. Conditions: [2] ) [3] ) 0.17 M, 5 mL toluene, 90 °C.
The reaction of N-benzylcarboxamide 2 with Al2(NMe2)6 was
studied by FTIR spectroscopy using an in situ ZnSe ATR probe.
Solution-IR spectra were acquired as the carboxamide was added
via syringe pump to a solution of Al2(NMe2)6 in toluene.
Significant changes were observed in the spectra between 1300
and 1700 cm-1 (Figure S1). The presence of numerous
overlapping bands hindered spectral deconvolution and detailed
spectral assignment. Nevertheless, diagnostic IR bands associ-
ated with the formation and disappearance of intermediates were
evident during the titration (Figures S1 and S2). Only after >3
equiv of 2 were added relative to [AlIII] did the carbonyl IR
band of the free carboxamide (1674 cm-1) become apparent in
the spectrum. Similar insights were gained from 13C NMR
spectroscopic studies in which N-benzylcarboxamide 2 was
titrated into a solution of Al2(NMe2)6 in toluene-d8. The chemical
shifts of the carbonyl carbon atom at low concentrations of 2
are distinct from that of free carboxamide and consistent with
formation of AlIII-amidate complexes (Figure 4). At a 2:AlIII
ratio of 1:1, two peaks are evident at 179.4 and 183.5 ppm.
Although the precise identity of the complexes corresponding
to these peaks is not known, neither of the peaks is associated
with free carboxamide (171.8 ppm). At ratios of 2:AlIII ) 2:1
and 3:1, single (distinct) peaks are observed at 179.8 and 183.6
ppm, respectively. Finally, at ratios of 2:AlIII > 3:1, two peaks
are present, one of which is identical to that observed when
2:AlIII ) 3:1 (183.6 ppm), and the other corresponds to the free
carboxamide.
Figure 3. Initial-rate dependence on [2], [3], or [3-d2] for the catalytic
transamidation of 2 and 3 or 2-d1 and 3-d2. Initial rates, derived from
monitoring the first 5% conversion, were determined by sampling [2] or
[2-d1] every 2 min. Conditions: [Al2(NMe2)6] ) 4.2 mM; [2], [2-d1], [3]
and [3-d2] ) 0.17 M.
Initial rates were measured also at different concentrations
of the substrates, carboxamide 2 and amine 3, at 5 mol % [AlIII].
The rate exhibits a linear, first-order dependence on [amine],
but no dependence on [carboxamide] (Figure 3, red circles and
blue squares, respectively). The rate slows significantly if
N-deuterated substrates, 2-d1 and 3-d2, are used in the reaction.
The linear increase in the rate with respect to [3-d2] (Figure 3,
black diamonds) reveals a deuterium kinetic isotope effect of
kH/kD ) 2.9. (Proton exchange between the amine and carboxa-
mide substrates under the reaction conditions prevents inde-
pendent determination of isotope effects associated with the two
different substrates. The lack of rate-dependence on [carboxa-
mide], however, implies that the isotope effect arises solely from
the deuterated amine.)
The kinetic data in Figures 2 and 3 support a simple
bimolecular rate law (eq 5), consistent with the preliminary
analysis of the reaction time-course data in Figure 1. The first-
order dependence of the rate on [AlIII] and [amine] indicates
these species participate in the turnover-limiting step. The
substantial deuterium kinetic isotope effect indicates that proton
transfer occurs in or before the turnover-limiting step.
In contrast to the 13C NMR data, 1H NMR spectra of mixtures
of carboxamide and Al2(NMe2)6 are rather complex because of
the presence of numerous resonances arising from heptanoyl
methylene groups in the carboxamide substrate and metal-
amidate complexes (Figure S3). Nevertheless, distinct ArCH2N-
resonances can be identified for free carboxamide and aluminum-
coordinated amidates in a relatively open region of the spectrum,
rate ) k[AlIII][RNH2]
(5)
Studies of Substrate Reactions with Al2(NMe2)6. To gain
additional insights into the catalytic mechanism, we spectro-
scopically probed fundamental reactions between the organic
substrates and Al2(NMe2)3. Further, we sought to characterize
the catalyst resting state under the reaction conditions.
The independent addition of g6 equiv of primary amine 3
or 5 or carboxamide 2 or 4 to a toluene solution of Al2(NMe2)3
at room temperature results in nearly quantitative formation of
1
at 4.25 and 3.90 ppm, respectively. The H and 13C NMR
spectral assignments of the aluminum amidate complex formed
when >3 equiv of 2 are present relative to [AlIII] are supported
by gHMBC NMR spectra, which reveal a three-bond correlation
between the CdO carbon atom and methylene protons of the
benzyl group of the coordinated amidate (Figure S4).
(10) See Supporting Information for additional details.
9
J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006 5179

A R T I C L E S
Hoerter et al.
Figure 4. 13C NMR spectra monitoring the changes that occur in the carbonyl region upon addition of N-benzylcarboxamide 2 to a solution of Al2(NMe2)6
in toluene-d8. The AlIII:carboxamide ratio present in each solution is indicated on the right-hand side of the spectra. [Note: 1H NMR data (Figure S3) reveal
that the peak labeled with an asterisk in spectrum for 2:Al ) 1:1 does not arise from the same complex as that present at 2:Al ratios g 3:1.]
Collectively, the FTIR and NMR spectroscopic data indicate
that the secondary N-benzylcarboxamide, 2, undergoes rapid
and quantitative proton-coupled exchange with the dimethyla-
mido ligands of Al2(NMe2)6 to generate aluminum amidate
complexes and free dimethylamine. Previous examples of AlIII-
amidate complexes include both mononuclear and bridged-
multinuclear structures, that possess alkyl groups or sterically
encumbered phenoxides as ancillary ligands.11,12 The amidate
ligand generally adopts a κ2 coordination mode in which both
nitrogen and oxygen atoms bind to the metal center. The amidate
may chelate a single AlIII center or bridge two different AlIII
centers. On the basis of these literature precedents, together with
the spectroscopic data described above, we propose that
secondary carboxamides react with Al2(NMe2)6 to produce a
monomeric, pseudo-octahedral tris-amidate aluminum complex,
AlIII(κ2-amidate)3, 6. Several stereoisomers are possible for such
structures; however, the spectroscopic data do not distinguish
between them, if they are present. The monomeric nature of
this complex is consistent with the presence of only one benzylic
of the tris-N-benzylamidate 6a (which consumes 0.6 equiv of
2), 0.4 equiv of free carboxamide 2, and 1.0 equiv of free
benzylamine (eq 7). Nearly quantitative formation of dimethy-
lamine is detected (0.6 equiv), although integration is compli-
cated by the presence of overlapping resonances in the 1H NMR
spectrum. Upon heating the solution to 90 °C, no change is
observed in this product ratio. The 13C NMR spectrum of this
solution confirms that 6a and 2 are the sole carbonyl-containing
products in solution. These results indicate that, in the presence
of a mixture of excess primary amine and secondary carboxa-
mide, AlIII exists primarily (>95%) as a monomer with three
κ2-amidate ligands.
1
NMR Spectroscopic Studies under Catalytic Conditions.
To study the transamidation reaction mixture (eq 4) under
catalytic conditions, 1H NMR spectra were acquired for a
reaction mixture consisting of 10 mol % Al2(NMe2)6 (20 mol
% AlIII) relative to N-benzylcarboxamide (2) and 2-ethylhexy-
lamine (3) in toluene-d8. A room-temperature spectrum reveals
the presence of tris-N-benzylamidate complex 6a, which
consumes 60% of available 2 (cf. eq 7). Unreacted 2 (0.4 equiv)
and amine 3 (1 equiv) are also present. Upon heating this
solution to 90 °C, transamidation occurs, producing carboxamide
4 and benzylamine 5. 1H NMR spectra were acquired at 5-min
intervals until the reaction reached equilibrium (Figure 5).
Initially, 2 is the only carboxamide present in solution, and all
of the AlIII is present as the tris-(N-benzylamidate)AlIII complex
6a. As the reaction proceeds, the resonance associated with 6a
decays with concomitant growth of new resonances, attributed
to new AlIII-amidate complexes 7a, 7b, and 6b.
(ArCH2N-) and one carbonyl (CdO) resonance in the H and
13C NMR spectra, respectively, for the aluminum-amidate
complex present at 2:AlIII ratios g3:1; multinuclear aggregates
with both bridging and nonbridging amidates might be expected
to exhibit multiple resonances.
To probe the relevance of the tris-amidate complex 6 to the
transamidation reactions, we investigated the fate of Al2(NMe2)6
with both amine and carboxamide substrates present. Ten mol
% Al2(NMe2)6 (i.e., 20 mol % AlIII) was added to a solution
containing a 1:1 ratio of benzylamine (5) and N-benzylhep-
tanamide (2) in toluene-d8, and the resulting solution was
1
1
examined by H and 13C NMR spectroscopy. The H NMR
1
The individual resonances in each H NMR spectrum were
spectrum at room temperature reveals the presence of 0.2 equiv
integrated in order to track the growth and/or decay of each
species present in solution (Figure 6). The concentrations of
amines 3 and 5 (open circles, Figure 6A) provide a direct
indication of the reaction progress. The concentrations of free
carboxamides 2 and 4 (filled triangles, Figures 6A) are
(11) (a) Kai, Y.; Yasuoka, N.; Kasai, N.; Kakudo, M. J. Organomet. Chem.
1971, 32, 165. (b) Huang, Y.-L.; Huang, B.-H.; Ko, B.-T.; Lin, C.-C. J.
Chem. Soc., Dalton Trans. 2001, 1359. (c) Huang, B.-H.; Yu, T.-L.; Huang,
Y.-L.; Ko, B.-T.; Lin, C.-C. Inorg. Chem. 2002, 41, 2987.
(12) Power, M. B.; Bott, S. G.; Clark, D. L.; Atwood, J. L.; Barron, A. R.
Organometallics 1990, 9, 3086.
9
5180 J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006

AlIII-Catalyzed Transamidation Mechanism
A R T I C L E S
Figure 5. 1H NMR spectral time-course for AlIII-catalyzed transamidation reaction between 2 and 3 (90 °C in toluene-d8). Data were collected until the
reaction reached equilibrium. The first spectrum was acquired 10 min after mixing, and additional spectra were acquired at 5-min intervals. The spectra
display the diagnostic resonances of the CH2 group adjacent to nitrogen in the amine and carboxamide reagents and products. The internal standard peak at
3.41 ppm (1,3,5-trimethoxybenzene) has been removed for clarity. Initial conditions: [Al2(NMe2)6] ) 0.083 M, [2] ) [3] ) 0.17 M.
Figure 6. (A) Reaction time-course of the AlIII-catalyzed transamidation between 2 and 3 in toluene-d8 at 90 °C. The concentrations of the species were
determined by 1H NMR spectroscopy, sampling every 5 min, with the initial time point at 10 min. Concentrations are plotted for amines 3 and 5, carboxamides
2 and 4, and the aluminum amidates 6a, 6b, 7a, and 7b. The concentrations of mixed AlIII-amidates 7a and 7b were determined from the weighted average
of the two resonances associated with each amidate. Initial conditions: [Al2(NMe2)6] ) 0.083 M, [2] ) [3] ) 0.17 M. (B) Total carboxamide concentration
present in the reaction, accounting for both free carboxamides and AlIII-coordinated amidates. (C) Expansion of the time-course data revealing the AlIII-
amidate speciation during the reaction.
9
J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006 5181

A R T I C L E S
Hoerter et al.
Scheme 1. Proposed Mechanism for AlIII-Catalyzed
Transamidation
complex A enters the catalytic cycle via bimolecular reaction
with a primary amine to form the amine adduct B (step I).
Coordination of the amine to AlIII forces one of the amidate
ligands to adopt a κ1 binding mode. This preequilibrium step
strongly favors the tris-amidate complex A. Proton transfer from
the coordinated amine to the κ1-amidate (step II) can account
for the deuterium kinetic isotope effect and serves to activate
both substrates. The anionic amido ligand is a better nucleophile
than is a primary amine, and the electrophilicity of the neutral
carboxamide is enhanced by virtue of its coordination to the
Lewis acidic AlIII center. The tetrahedral intermediate D formed
upon nucleophilic attack of the amido ligand on the coordinated
carboxamide (step III) can either revert back to complex C or
isomerize via coordination of the other nitrogen atom to AlIII
to yield intermediate F. Subsequent breakdown of the new
tetrahedral intermediate F (step V) and exchange of the neutral
carboxamide ligand (step VI) completes the transamidation
reaction. The inability to observe intermediates B-G spectro-
scopically and the unimolecular nature of steps II-V complicate
efforts to probe this mechanism further. The precise identity of
the turnover-limiting step is not known; however, the sizable
kinetic isotope effect indicates that this step can occur no earlier
than step II but may be as late as step IV (note that steps V and
VI are approximately isoenergetic to steps III and II, respec-
tively).
Implications of the Catalytic Mechanism. In our original
catalyst screening study,8 we tested three classes of metal
complexes as potential transamidation catalysts: (1) Lewis acids,
(2) alkali-metal-amido reagents, and (3) transition metal- and
main group metal-amido complexes. Among the complexes
tested, Al2(NMe2)6 proved to be the best at promoting thermo-
neutral transamidation between primary alkylamines and sec-
ondary N-alkyl carboxamides. In the mechanism proposed above
(Scheme 1), intermediate C plays a central role and is
noteworthy because it embodies the bifunctional role of AlIII
via activation of both the amine and carboxamide substrates.
Lewis acids, such as Sc(OTf)3, should be capable of electrophilic
activation of the carboxamide, but the triflate anion is insuf-
ficiently basic to form a Sc-amido species. In contrast, alkali-
metal amido species are quite basic. They will deprotonate the
carboxamide N-H to yield a metal amidate complex analogous
to the amidoaluminum complex 1 (cf. intermediate A, Scheme
1). However, alkali metal cations possess only a single positive
charge and are relatively weak Lewis acids. Therefore, they are
not effective at organizing multiple substrates in their coordina-
tion sphere and do not effect the bifunctional substrate activation
mechanism proposed for AlIII.
significantly lower than those of the amines because a large
fraction (60%) coordinates to aluminum. Therefore, the con-
centrations of 2 and 4 must be combined with the concentrations
of AlIII-coordinated carboxamides in 6a, 6b, 7a, and 7b to
achieve an accurate depiction of the reaction progress (Figure
6B). The final equilibrium ratio of carboxamides detected in
1
this H NMR experiment (2:4 ) 1:1.2) is identical to that
observed in the GC study (Figure 1A).
The equilibrium evolution of AlIII-amidate species follows
the sequence 6af7af7bf6b (Figure 6C). Complexes 6a and
6b have been prepared independently in solution and character-
ized by NMR spectroscopy (Figure S5), whereas the identities
of 7a and 7b are inferred from the order in which they appear
during the reaction course. The distribution of these four AlIII-
amidate complexes at equilibrium deviates somewhat from a
statistical ratio of 6a:7a:7b:6b ) 1:3:3:1. The observed ratio
of 1.1:3.6:2.8:1.0 suggests that the N-benzylamidate anion has
a somewhat higher affinity for AlIII relative to a proton than
does the N-(2-ethylhexyl)amidate anion.
Proposed Catalytic Mechanism. The NMR spectroscopic
studies described above provide strong evidence for a catalyst
resting state consisting of an aluminum tris-amidate complex,
such as 6. To our knowledge, tris-amidate complexes of
aluminum have not been characterized previously. ꢀ-Caprolac-
tam has been proposed to form a homoleptic tris-lactamate
complex with AlIII upon reaction with AlEt3; however, the
complex was not isolated or characterized.13 Well-defined
examples of Al(κ2-amidate) complexes bearing ancillary phe-
noxide and alkyl ligands have been reported previously by
Barron12 and Lin.11 An Al(κ2-amidate)3 resting state accounts
for the fact that variation of concentration of free carboxamide
in solution has no effect on the rate (Figure 3) and that no
[carboxamide] term appears in the rate law (eq 5).
This AlIII-catalyzed transamidation mechanism resembles
mechanisms proposed for metalloenzyme-catalyzed hydrolysis
reactions.14 The metal cofactor in these enzymes often serves a
bifunctional role, activating water via formation of a reactive
M-OH fragment and activating the electrophilic substrate via
coordination at the Lewis acidic metal center.
A critical challenge that must be overcome in these AlIII-
catalyzed transamidation reactions is the high stability, and
hence low reactivity, of the AlIII-amidate complex A. Attempts
to destablize this AlIII-amidate interaction and achieve higher
On the basis of the mechanistic insights described above,
including the rate law, deuterium kinetic isotope effect, and
identity of the catalyst resting state, we propose the following
catalytic mechanism (Scheme 1). The tris-amidate aluminum
(14) See, for example: (a) Bertini, I.; Luchinat, C. In Bioinorganic Chemistry;
Bertini, I., Gray, H. B., Lippard, S. J., Valentine, J. S., Eds.; University
Science Books: Mill Valley, CA, 1994; p 37. (b) Parkin, G. Chem. ReV.
2004, 104, 699.
(13) Konomi, T.; Tani, H. J. Polym. Sci., Part A: Polym. Chem 1971, 9, 325.
9
5182 J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006

AlIII-Catalyzed Transamidation Mechanism
A R T I C L E S
catalytic activity are currently being explored. The mechanistic
insights gained from this study have prompted us to evaluate
transamidation reactions between tertiary carboxamides and
secondary amines. Tertiary carboxamides do not possess a N-H
bond and, therefore, will not generate AlIII-amidate complexes.
Preliminary studies suggest amidoaluminum complexes, such
as 1, are effective catalysts for these transmidation reactions.
transfer from amine to a coordinated amidate ligand are
proposed to initiate the stepwise transamidation mechanism. The
results of this study highlight the ability of an AlIII center to
play a bifunctional role in the transamidation reaction, and the
mechanistic insights are now being used to develop improved
catalysts and target novel transformations of carboxamide-based
molecules.
Acknowledgment. We are grateful to the NSF Collaborative
Research in Chemistry program (Grant No. CHE-0404704) for
financial support of this work.
Conclusion
The combined use of kinetic and spectroscopic studies has
provided valuable insights into the mechanism of AlIII-catalyzed
transamidations. The amidoaluminum complex Al2(NMe2)6 was
selected from catalyst-screening studies as an effective precata-
lyst.8 Under catalytic conditions, Al2(NMe2)6 reacts rapidly with
secondary carboxamides to yield a tris-amidate-aluminum
complex (6) that has been identified as the catalyst resting state.
The reaction of 6 with a primary amine and subsequent proton
Supporting Information Available: Experimental procedures
and additional spectroscopic data, including FT-IR and NMR
spectra. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA060331X
9
J. AM. CHEM. SOC. VOL. 128, NO. 15, 2006 5183