Simple, versatile, and efficient catalysts for guanylation of amines

Article abstract of DOI:10.1021/om1003122

Commercially available products such as ZnEt₂ (1), MgBu ₂ (2), and n-BuLi (3) can act as excellent catalytic precursors for the catalytic addition of amines to carbodiimides. Aromatic primary amines bearing different substituents, se

Full text of DOI:10.1021/om1003122

Organometallics 2010, 29, 2789–2795 2789
DOI: 10.1021/om1003122
Simple, Versatile, and Efficient Catalysts for Guanylation of Amines
,†
Carlos Alonso-Moreno,* Fernando Carrillo-Hermosilla, Andres Garces,
†
‡
ꢀ
Antonio Otero,* Isabel Lopez-Solera, Ana M. Rodrıguez, and Antonio Antinolo
ꢀ
,†
†
†
†
ꢀ
~
´
†
´
ꢀ
ꢀ
´
´
Departamento de Quımica Inorganica, Organica y Bioquımica, Facultad de Ciencias Quımicas, Universidad
de Castilla-La Mancha, Campus Universitario de Ciudad Real, 13071-Ciudad Real, Spain, and
´ ꢀ ´ ꢀ
Departamento de Quımica Inorganica y Analıtica, Universidad Rey Juan Carlos, 28933-Mostoles, Spain
‡
Received April 19, 2010
Commercially available products such as ZnEt2 (1), MgBu2 (2), and n-BuLi (3) can act as excellent
catalytic precursors for the catalytic addition of amines to carbodiimides. Aromatic primary amines
bearing different substituents, secondary amines, and heterocyclic amines can undergo this reaction.
The synthesis and structural characterization of the zinc guanidinate complex [Zn(Et){(4-t-BuC6H4)-
NdC(N-i-Pr)(NH-i-Pr)}]2 (15) and the synthesis and characterization of the lithium guanidinate
complex [Li{(2,4,6-Me3C6H4)NdC(N-i-Pr)(NH-i-Pr)}(THF)] (16), both of which act as catalysts in
the guanylation reaction, allowed us to propose a mechanism involving the formation of amido inter-
mediates. These amido compounds subsequently react with the carbodiimide in an insertion process.
Scheme 1. Catalyzed Guanylation Reactions
Introduction
The C-N bond formation reactions are of considerable
interest in both synthetic organic and industrial chemistry
due to the importance of amines and their derivatives in
almost all areas of chemistry. In particular, guanidine deri-
vatives are very useful pharmacophores in medicinal chemis-
try due to their capacity to interact with functional groups
present in enzymes or receptors through hydrogen bonds and
electrostatic interactions.1 As a result, guanidine derivatives
have also been widely used as ancillary ligands for the stabi-
lization of different metal complexes. The unique properties
of metal complexes containing these versatile guanidinate
ligands have led to the active study of these systems and their
applications.2 The synthesis of guanidines has been inten-
sively investigated by a variety of methods.3 Primary alipha-
tic amines can undergo direct addition to carbodiimides
under rather forcing conditions.4 However, aromatic amines
or secondary amines do not react with carbodiimides under
the same (or even harsher) conditions. Among the direct
methods for the synthesis of guanidines, the catalytic hydro-
amination of carbodiimides is the most relevant, as it has an
atom-economy of 100% and is a waste-free process.5 In 2003,
Richeson et al. reported the first example of the transition
metal-catalyzed guanylation of aromatic amines with carbo-
diimides.6 Titanium complexes undergo this kind of reaction,
but these complexes are not able to promote the reaction
with secondary aromatic amines because the process requires
the regeneration of an MdN imido moiety. The guanylation
of aromatic amines catalyzed by lithium bis(trimethylsilyl)-
amide has been reported.7 The guanylation of both aromatic
and secondary amines was catalyzed, for the first time, by
using half-sandwich lanthanide.8a Titanacarboranyl amido
or vanadium imido complexes can catalyze the guanylation
*To whom correspondence should be addressed. Phone: þ34-26-295-
326. Fax: þ34-26-295-318. E-mail: Antonio.Otero@uclm.es; Carlos.
AMoreno@uclm.es.
(1) (a) Berlinck, R. G. S.; Burtoloso, A. C. B.; Kossuga, M. H. Nat.
Prod. Rep. 2008, 25, 919. (b) Durant, G. J. Chem. Soc. Rev. 1985, 14, 375.
(c) The Pharmacological Basis of Therapeutics, 7th ed.; Gilman, A.,
Gododman, L. S., Rall, T. W., Murad, T., Eds.; Pergamon Press: New York,
1990; p 899.
(2) Coles, M. P. Dalton Trans. 2006, 985.
(6) Ong, T.-G.; Yap, G. P. A.; Richeson, D. S. J. Am. Chem. Soc.
2003, 125, 8100.
(7) Ong, T.-G.; O’Brien, J. S.; Korobkov, I.; Richeson, D. S. Orga-
(3) See for example: (a) Evindar, G.; Batey, R. A. Org. Lett. 2003, 5,
133. (b) Wu, Y.-Q.; Hamilton, S. K.; Wilkinson, D. E.; Hamilton, G. S.
J. Org. Chem. 2002, 67, 7553. (c) Ghosh, A. K.; Hol, W. G. J.; Fan, E. J. Org.
Chem. 2001, 66, 2161. (d) Tamaki, M.; Han, G.; Hruby, V. J. J. Org. Chem.
2001, 66, 1038.
(4) Thomas, E. W.; Nishizawa, E. E.; Zimmermann, D. C.; Williams,
D. J. J. Med. Chem. 1989, 32, 228.
(5) (a) Zhang, W.-X.; Hou, Z. Org. Biomol. Chem. 2008, 6, 1720.
(b) Schow, S. In Encyclopedia of Reagents for Organic Synthesis; Paquete,
L. A., Ed.; Wiley: Sussex, U.K., 1995; pp 1408-1410.
nometallics 2006, 25, 4728.
(8) (a) Zhang, W.-X.; Nishiura, M.; Hou, Z. Synlett 2006, 8, 1213.
(b) Zhang, W.-X.; Nishiura, M.; Hou, Z. Chem.;Eur. J. 2007, 13, 4037.
(c) Zhou, S.; Wang, S.; Yang, G.; Li, Q.; Zhang, L.; Yao, Z.; Zhou, Z.; Song,
H. Organometallics 2007, 26, 3755. (d) Li, Q.; Wang, S; Zhou, S.; Yang, G.;
Zhu, X.; Liu, Y. J. Org. Chem. 2007, 72, 6763. (e) Wu, Y.; Wang, S.; Zhang,
L.; Yang, G.; Zhu, X.; Liu, C.; Yin, C.; Rong, J. Inorg. Chim. Acta 2009,
2814.
r
2010 American Chemical Society
Published on Web 05/27/2010
pubs.acs.org/Organometallics

2790 Organometallics, Vol. 29, No. 12, 2010
Table 1. Catalytic Addition of Amines to N,N0-Diisopropylcarbodiimide with ZnEt2
Alonso-Moreno et al.
a
a Conditions: amine (1 mmol); N,N’-diisopropylcarbodiimide (1 mmol); 1.5 mol % catalyst. b Determined by 1H NMR.
of aromatic and secondary amines.9 Similarly, other lantha-
nide amido complexes,8 commercial alkyl aluminums,10a and
heavier group 2 amides have shown high levels of catalytic
activity.11 Nevertheless, the construction of the CN-rich
functionalized guanidines plays an essential role in their bio-
logical activity, and the design of efficient catalysts capable
of promoting addition of amines to inactivated carbodi-
imides still remains to be developed. In the context of the
information outlined above, we present here a new approach
to C-N bond formation12 that involves the use of commer-
cially available products such as ZnEt2, MgBu2, and n-BuLi,
as cheap, simple, versatile, and efficient catalysts for guany-
lation reactions.
(9) (a) Shen, H.; Chan, H. S.; Xie, Z. Organometallics 2006, 25, 5515.
(b) Montilla, F.; Pastor, A.; Galindo, A. J. Organomet. Chem. 2004, 689, 993.
(10) (a) Zhang, W.-X.; Li, D.; Wang, Z.; Xi, Z. Organometallics 2009,
28, 882. (b) In the course of the preparation of this article, new results about
the use of a commercial zinc compound, Zn(OTf)2, as an effective catalyst for
guanylation reactions were reported: Li, D.; Guang, J.; Zhang, W.-X.; Wang,
Y.; Xi, Z. Org. Biomol. Chem. 2010, DOI: 10.1039/b923249b.
(11) Lachs, J. R.; Barret, A. G. M.; Crimmin, M. R.; Kociock-Kohn,
G.; Hill, M. S.; Mahon, M. F.; Procopiou, P. A. Eur. J. Inorg. Chem.
2008, 4173.
Results and Discussion
In the development of catalytic systems for the ring-opening
polymerization of cyclic esters13 the synthesis of the organo-
metallic entities often required the use of commercial compounds
(such as ZnEt2, MgBu2, and n-BuLi) as starting materials. It
was envisaged that these commercially available products could
be used as catalyst precursors for guanylation reactions.
The catalytic activity of ZnEt2 (1), MgBu2 (2), andn-BuLi (3)
in the guanylation of primary, secondary, and heterocyclic
aromatic amines was investigated (Scheme 1, Tables 1-3).
ꢀ
(12) Alonso-Moreno, C.; Carrillo-Hermosilla, F.; Romero-Fernandez,
~
´
guez, A. M.; Otero, A.; Antinolo, A. Adv. Synth. Catal. 2009, 351,
J.; Rodrı
881.
(13) (a) Otero, A.; Fernandez-Baeza, J.; Lara-Sanchez, A.; Alonso-
Moreno, C.; Marquez-Segovia, M. I.; Sanchez-Barba Merlo, L. F.;
Rodriguez, A. M. Angew. Chem. 2009, 48, 2176. (b) Otero, A.; Fernandez-
1
The conversion values were determined by H NMR spec-
troscopy. The use of N,N0-diisopropylcarbodiimide at 50 or
120 °C with 2,4,6-trimethylaniline in deuterated toluene did
not lead to a reaction (Table 1, entries 1 and 2). In contrast,
the addition of a small amount (1.5 mol %) of the readily
available ZnEt2 (1) under mild conditions (50 °C) led to rapid
addition of 2,4,6-trimethylaniline to N,N0-diisopropylcarbo-
diimide to give more than 99% conversion to the N,N0,N00-
trisubstituted guanidine in 30 min (Table 1, entry 3). The pola-
rity of the solvents did not seem to have a significant influence
~
Baeza, J.; Antinolo, A.; Lara-Sanchez, A.; Tejeda, J.; Martinez-Caballero, E.;
Marquez-Segovia, I.; Lopez-Solera, I.; Sanchez-Barba Merlo, L. F.; Alonso-
Moreno, C. Inorg. Chem. 2008, 47, 4996. (c) Otero, A.; Fernandez-Baeza, J.;
Lara-Sanchez, A.; Martinez-Caballero, E.; Tejeda, J.; Sanchez-Barba Merlo,
~
L. F.; Antinolo, A.; Alonso-Moreno, C.; Lopez-Solera, I. Organometallics
2008, 27, 976. (d) Alonso-Moreno, C.; Garces Osado, A.; Sanchez-Barba
~
Merlo, L. F.; Fernandez-Baeza, J.; Otero, A.; Lara-Sanchez, A.; Antinolo,
A.; Broomfield, L.; Lopez-Solera, I. Organometallics 2008, 27, 1310.
ꢀ
ꢀ
(e) Sanchez-Barba, L. F.; Garces, A.; Fajardo, M.; Alonso-Moreno, C.;
ꢀ
~
ꢀ
Otero, A.; Fernandez-Baeza, J.; Antinolo, A.; Tejeda, J.; Lara-Sanchez, A.;
ꢀ
Lopez-Solera, I. Organometallics 2007, 26, 6403.

Article
Organometallics, Vol. 29, No. 12, 2010 2791
a
Table 2. Catalytic Addition of Amines to N,N0-Diisopropylcarbodiimide with MgBu2
a Conditions: amine (1 mmol); N,N’-diisopropylcarbodiimide (1 mmol); 1.5 mol % catalyst. b Determined by 1H NMR.
Table 3. Catalytic Addition of Amines to N,N0-Diisopropylcarbodiimide with n-BuLia
a Conditions: amine (1 mmol); N,N’-diisopropylcarbodiimide (1 mmol); 1.5 mol % catalyst. b Determined by 1H NMR.
on the catalytic activity in the reaction, but the temperature
is crucial for the reaction to proceed efficiently (Table 1,
entries 3-5). Encouraged by the performance of ZnEt2 in the
guanylation of N,N0-diisopropylcarbodiimide with 2,4,6-tri-
methylaniline, we decided to expand the scope of the sub-
strates to include examples with different substitution pat-
terns in the aniline (Table 1, entries 6-8). Electron-donating
and electron-withdrawing susbtituents on the aniline ring
were well tolerated in this process. The conversion levels in
this reaction were found to be very high in all cases, and
complete conversion was achieved in 30 min. The versatility
of ZnEt2 in the guanylation reactions under mild conditions
was supported by the fact that the precatalyst was able to add
secondary aromatic and cyclic amines to inactivated carbo-
diimide in 30 min (Table 1, entries 9, 12, and 13; secondary
cyclic amines do no react in the absence of catalyst, after 20 h
at 50 °C) and that the reaction also occurs with heterocyclic
amines (Table 1, entries 10 and 11) in moderate to high yields.
The direct synthesis of all guanidines mentioned above by
reaction of the amines with carbodiimides catalyzed by ZnEt2
was also achieved on a Schlenk-tube scale, with the guani-
dine products isolated as white solids or colorless oily pro-
ducts (8) in high yields. The new guanidines were character-
ized by 1H NMR and 13C NMR spectroscopy and elemental
analysis. In the case of the previously described 1-(4-bromo-
phenyl)-2,3-diisopropylguanidine (7),8b the molecular structure

2792 Organometallics, Vol. 29, No. 12, 2010
Alonso-Moreno et al.
Scheme 2. Catalytic Addition of 2,4,6-Trimethylaniline to
Different Carbodiimides
Figure 1. ORTEP drawing of 1-(4-bromophenyl)-2,3-diisopro-
pylguanidine, 7. Hydrogen atoms, except those on the nitrogen
atoms, are omitted for clarity. Selected bond lengths [A] and
angles [deg]: N(1)-C(7) 1.287(8), N(2)-C(7) 1.374(9), N(3)-C(7)
1.357(9), N(1)-C(7)-N(2) 118.5(7), N(2)-C(7)-N(3) 113.2(6),
N(1)-C(7)-N(3) 128.3(7).
Table 4. Catalytic Addition of 2,4,6-Trimethylaniline to
Different Carbodiimidesa
entry
R1, R2
catalyst
product (conv %)b
1
2
3
4
5
6
7
8
9
i-Pr, i-Pr
i-Pr, i-Pr
i-Pr, i-Pr
Et, t-Bu
Et, t-Bu
Et, t-Bu
Cy, Cy
ZnEt2
4, >99
4, >99
4, >99
13, >99
13, >99
13, >99
14, 75
MgBu2
n-BuLi
ZnEt2
MgBu2
n-BuLi
ZnEt2
was established by X-ray diffraction (Figure 1). The crystal
data are summarized in the Experimental Section.
Commercially available MgBu2 (2) also behaves as a simple,
versatile, and efficient catalyst for the guanylation reaction
under mild conditions, as evidenced by the reaction of diffe-
rent amines with N,N0-diisopropylcarbodiimide (Table 2). It
can be seen from the results in Table 2 that MgBu2 behaves in
a similar way to ZnEt2 as a catalytic precursor for the guanyl-
ation reaction. In fact, MgBu2 can catalyze the reaction of
aniline as well as primary aromatic amines bearing electron-
donating or electron-withdrawing substituents in 30 min and
with high efficiency (Table 2, entries 1-4). It is noteworthy
that this species was also able to convert a secondary aro-
matic amine to the corresponding guanidine in high yield
(Table 2, entry 5). However, examination of the catalytic
activity in reactions with heterocyclic and secondary cyclic
amines showed MgBu2 to be a less efficient catalyst than
ZnEt2 (Table 2, entries 6-9). In the case of 4,6-dimethylpyridin-
2-amine, MgBu2 afforded only a trace of the corresponding
guanidine product. Comparison of the catalytic precursors
ZnEt2 and MgBu2 with other commercially available com-
pounds such as alkyl aluminums or the zinc triflate used by
Zhang et al.10 shows that ZnEt2 and MgBu2 have very similar
catalytic activities under milder conditions than the alumi-
num alkyl compounds at room temperature. This activity
includes guanylation reactions with secondary aromatic amines,
which demonstrates the high versatility of this approach.
Furthermore, these compounds were similar to or more effec-
tive than Zn(OTf)2. Finally, the lower toxicity and less hazar-
dous nature of these species in comparison with alkyl alumi-
nums lead us to propose them as more convenient catalytic
precursors. In 2006, Richeson et al. described LiN(SiMe3)2
as an effective precatalyst for guanylation reactions with
very high conversion yields.7 In that communication it was
suggested that other organolithium and lithium amides should
present similar rates of conversion based on an initiation step
in which the formation of lithium amide was expected. How-
ever, representative results for this hypothesis were not
presented. In the context of this work we became interested
in exploring n-BuLi as a catalytic precursor (Table 3). The
addition reaction between 2,4,6-trimethylaniline and N,N0-
diisopropylcarbodiimide catalyzed by commercially avail-
able n-BuLi was achieved at room temperature in less than
30 min (Table 3, entry 1). Although n-BuLi was highly efficient
as a catalyst for the guanylation reaction of primary (Table 3,
entries 1-4) and secondary (Table 3, entry 5) aromatic amines,
the conversion of heterocyclic and cyclic secondary amines
Cy, Cy
Cy, Cy
MgBu2
n-BuLi
14, 52
14, 75
a Conditions: 2,4,6-trimethylaniline (1 mmol); carbodiimide (1 mmol);
1.5 mol % catalyst; temperature 50 °C for 1 and 2, 25 °C for 3; toluene.
b Determined by 1H NMR at 30 min.
varied from low to moderate yields (Table 3, entries 6-9). As
for precatalysts 1 and 2, the presence of a p-bromo substi-
tuent (Table 3, entry 4) was well tolerated. Primary reactivity
of this organolithium compound toward acidic amine pro-
tons could explain this behavior. Similar results were obta-
ined with reactive lithium amide7 or AlMe3.10a
In order to extend the scope of the reaction to other carbo-
diimides, we explored the use of N,N0-tert-butylethylcarbo-
diimide and N,N0-cyclohexylcarbodiimide (Scheme 2, Table 4)
for the guanylation reaction with ZnEt2, MgBu2, and n-BuLi
as catalysts. In the case of N,N0-tert-butylethylcarbodiimide
the catalytic processes showed similar conversions to the diiso-
propyl system (Table 4, entries 1-6), but the use of the bulkier
N,N0-cyclohexylcarbodiimide gave lower conversions, pro-
bably due to steric hindrance (Table 4, entries 7-9).
Of the commercially available products tested as catalysts
in the guanylation reaction, ZnEt2 was chosen for kinetic
investigation, as it is cheap, harmless, and versatile and the
simplest catalyst used. In order to obtain quantitative kinetic
data, the reactions of N,N0-diisopropylcarbodiimide with an
equal number of equivalents of 2,4,6-trimethylaniline in the
presence of ZnEt2 as catalyst were carried out. The reaction
rates of the guanylation reaction were monitored over time
1
by H NMR spectroscopy. The kinetics were studied with
different catalyst precursor concentrations: [Zn] = 1.5 mol %,
5 mol %, and 8 mol %, at 50 °C, using toluene as solvent. The
semilogarithmic plot of ln([A]0/[A]t) versus reaction time is
shown in Figure 2, where [A]0 is the initial amine concentra-
tion and [A]t is the amine concentration after a given reaction
time. In all cases the linearity of the plot shows that the pro-
pagation was first order with respect to amine when the reac-
tion was carried out at 50 °C in toluene. An induction period
was not observed. The absence of an induction period indi-
cates that the catalyst was reactive from the beginning of the
process and that rearrangement of initiator aggregates was
not necessary to form the active species. The linearity of
the plot also illustrates that termination processes did not
occur during the reaction. The fastest reaction was observed

Article
Organometallics, Vol. 29, No. 12, 2010 2793
Figure 3. First-order kinetic plots for the guanylation reaction
of N,N0-diisopropylcarbodiimide and 2,4,6-trimethylaniline in
toluene with [ZnEt2] = 1.5 mol %: (a) 9 at 30 °C, kapp = 0.2 ꢀ
10-3 s-1 (linear fit, R=0.9987); (b) bat 40 °C, kapp=0.9ꢀ10-3 s-1
(linear fit, R=0.998); (c) 2 at 50 °C, kapp=1.6ꢀ10-3 s-1 (linear fit,
R=0.999);(d) (at 60 °C, kapp=1.8ꢀ10-3 s-1 (linearfit,R=0.998).
Figure 2. First-order kinetic plots for the guanylation reaction
of N,N0-diisopropylcarbodiimide and 2,4,6-trimethylaniline with
ZnEt2 as catalyst in toluene at 50 °C: (a) 9 [ZnEt2]=1.5 mol %,
k
app=1.6ꢀ10-3 s-1 (linear fit, R=0.999), (b) b[ZnEt2]=5mol%,
k
app=1.7ꢀ10-3 s-1 (linear fit, R=0.999),(c) 2[ZnEt2]=8mol%,
k
app = 0.9 ꢀ10-3 s-1 (linear fit, R= 0.998).
for [Zn]=1.5 mol %, with a pseudo-first-order rate constant
of 1.6ꢀ10-3 s-1. A very similar kapp was observed with [Zn]=
5 mol %. However, an increase in the charge of catalyst to
8 mol % gave a kapp = 0.9 ꢀ 10-3 s-1, which is much lower
than for [Zn] = 1.5 mol % under the same conditions. This
finding supports the formation of less reactive aggregates in
solution.
The influence of temperature on the guanylation reaction
using ZnEt2 was also investigated (Figure 3). It can be seen
from Figure 3 that the reaction rate increased with tempera-
ture, with a calculated activation energy (Ea) of 59.5 kJ mol-1
.
A possible mechanism for the guanylation reaction cata-
lyzed by ZnEt2 is proposed in Figure 4. A straightforward
acid-base reaction to yield the alkyl-amido species is postu-
lated as the first step. Participation of both ethyl groups in
this process cannot be ruled out. In a second step, nucleo-
philic addition of the amido species to the carbodiimide
would afford the guanidinate species, which would act as
the true catalyst. Protonolysis by another molecule of amine
could give the final product.
Two approaches were used in an effort to find evidence for
the intermediates in the proposed mechanism for the guany-
lation reactions. First, the addition of a hexane solution of
ZnEt2 to a toluene solution of {2-(4-(tert-butyl)phenyl)-1,3-
diisopropylguanidine}, previously obtained by catalytic
means, allowed us, after crystallization, to obtain white crys-
tals of compound 15 appropriate for X-ray diffraction. The
molecular structure is shown in Figure 5 along with the num-
bering system used in the crystallographic study. The crystal
data are summarized in the Experimental Section.
Complex 15 crystallized as a dimeric compound, [Zn(Et)-
{(4-t-BuC6H4)NdC(N-i-Pr)(NH-i-Pr)}]2, with a μ,η2:η1-co-
ordination mode of the guanidinate ligand, resulting in a
planar Zn2N2 ring metallacycle. Both the ethyl and guanidi-
nate groups show a relative anti arrangement around the
planar ring. This situation is in contrast with a similar com-
pound described by Coles and Hitchcock,14 where the ligands
adopt a syn distribution. The metal atoms show a distorted
Figure 4. Proposed mechanism for the guanylation reaction
catalyzed by ZnEt2.
tetrahedral geometry [angles in the range 62.3(4)-127.6(4)°].
Similarly, the bridging nitrogen atoms show an analogous
disposition [angles in the range 81.8(9)-123.5(7)°]. It is worth
noting that the metal-nitrogen distance between the ligand
of one [Zn(Et){(4-t-BuC6H4)NdC(N-i-Pr)(NH-i-Pr)}] unit
and the metal of the another [Zn-N1A = 2.04(1) A] is
shorter than that within the chelating guanidinate itself
[Zn-N1 = 2.34(1) A]. This finding suggests a strong inter-
action between the two monomeric units in the solid state.
Charge delocalization around the guanidinate atoms is postu-
lated on the basis of the C-N distances [1.30(1), 1.37(1), and
1.38(1) A for N(2)-C6, N(1)-C6, and N(3)-C6, respec-
tively], which are intermediate between the value expected
for C-N single and CdN double bonds.
Furthermore, the equimolar reaction of n-BuLi, 2,4,6-tri-
methylaniline, and N,N0-diisopropylcarbodiimide in hexanes/
THF at room temperature quantitatively gave the lithium
guanidinate complex [Li{C(NH-2,4,6-Me3C6H4)(N-i-Pr)2}-
(THF)] (16) as a white solid in high yield. Unfortunately,
crystals suitable for X-ray diffraction were not obtained.
Similar reactions with magnesium compound 2 gave rise to
unstable products that could not be isolated. Compounds 15
(14) Coles, M. P.; Hitchcock, P. B. Eur. J. Inorg. Chem. 2004, 2662.

2794 Organometallics, Vol. 29, No. 12, 2010
Alonso-Moreno et al.
deuterated solvent under nitrogen. Conversion of the starting
material to product was determined by integration of the pro-
duct resonances relative to the substrate peaks in the 1H NMR
spectrum.
Preparative Scale Synthesis of the Guanidines. In a glovebox,
a solution of amine (6.00 mmol) in toluene (20 mL) was added to
a solution of ZnEt2 in hexanes (0.09 mmol) in a Schlenk tube.
The carbodiimide (6.00 mmol) was then added to the above
reaction mixture. The Schlenk tube was taken outside the glove-
box, and the reaction was carried out at 50 °C for 2 h. The sol-
vent was removed under reduced pressure, and the residue was
extracted with diethyl ether and filtered to give a clear solution.
The solvent was removed under vacuum, and the residue was
recrystallized from ether to provide the solid guanidine products
or distilled under vacuum in the case of liquid guanidine 8.
Guanidines 4,9b 5,7 7,8b 11,9a 12,8c and 149b were previously des-
cribed in the literature.
Figure 5. ORTEP drawing of [Zn(Et){(4-t-BuC6H4)NdC(N-
i-Pr)(NH-i-Pr)}]2, 15. Hydrogen atoms, except those on the nitro-
gen atoms, are omitted for clarity. Selected bond lengths [A] and
angles [deg]: Zn(1)-C(1) 1.96(1), Zn(1)-N(1) 2.34(1), Zn(1)-
N(2) 2.04(1), Zn(1)-N(1A) 2.04(1), N(1)-C6 1.37(1), N(2)-C6
1.30(1), N(3)-C6 1.38(1), C(1)-Zn(1)-N(1) 121.2(4), C(1)-
Zn(1)-(N2) 127.6(4), N(1)-Zn(1)-N(2) 62.3(4), N(1)-Zn(1)-
N(1A) 93.9(4).
Characterization of {2-(4-(tert-Butyl)phenyl)-1,3-diisopropyl-
guanidine}, 6. Yield: 1.57 g, 95%. Anal. Calcd for C17H29N3: C,
74.13; H, 10.61; N, 15.26. Found: C, 74.28; H, 10.34; N, 15.31.
1H NMR (toluene-d8, 297 K): δ 7.27, 6.99 (2d, 2H each, 3JH-H
=
8.5 Hz, N-C6H4-C(CH3)3), 3.67 (m, 2H, N-CH(CH3)2), 3.61 (bs,
2H, NH), 1.26 (s, 9H, N-C6H4-C(CH3)3), 0.91 (d, 12H, 3JH-H
=
6.4 Hz, N-CH(CH3)2). 13C{1H} NMR (toluene-d8, 297 K): δ
149.5 (CdN), 143.4, 137.4, 126.3, 123.2 (N-C6H4-C(CH3)3),
43.3 (N-CH(CH3)2), 34.2 (N-C6H4-C(CH3)3), 31.7 (N-C6H4-
C(CH3)3), 23.2 (N-CH(CH3)2).
and 16 were tested as catalysts in the guanylation reaction of
2,4,6-trimethylaniline. Both compounds show the same acti-
vity as the precursor compounds 1 and 3, and this provides
plausible evidence that these or similar species play a role in
the mechanism proposed for the guanylation reaction.
Characterization of {2,3-Diisopropyl-1-methyl-1-phenylguanidine}
8. Yield: 1.10 g, 79%. Anal. Calcd for C14H23N3: C, 72.06; H,
9.93; N, 18.01. Found: C, 72.40; H, 10.01; N, 18.50. H NMR
1
(toluene-d8, 297 K): δ 7.19, 6.73 (m, 5H, N-C6H5), 3.54 (m, 2H,
N-CH(CH3)2), 3.02 (s, 3H, N-CH3), 1.02 (m, 12H, N-CH-
(CH3)2). 13C{1H} NMR (toluene-d8, 297 K): δ 149.7 (CdN),
146.8, 129.4, 118.4, 113.8 (N-C6H5), 45.4 (N-CH(CH3)2), 37.4
(N-CH3), 24.3 (N-CH(CH3)2).
Conclusions
In conclusion, we report the use of the commercially avail-
able compounds ZnEt2, MgBu2, and n-BuLi as cheap, simple,
versatile, and efficient catalysts for the guanylation reaction
of carbodiimides with a wide range of primary aromatic amines,
secondary aromatic amines, and heterocyclic amines. The
isolation and reactivity of zinc and lithium guanidinate inter-
mediates 15 and 16 confirms that the guanylation reaction
proceeds through nucleophilic addition of an amido species
to a carbodiimide, followed by amine protonolysis of the
resultant guanidinate species. Further studies are underway
aimed at examining thoroughly the use of other commer-
cially available zinc alkyls, Grignard reagents, and lithium
compounds, the effect of changes in the substituents on the
carbodiimide framework and the catalytic addition of term-
inal alkynes to carbodiimides, and to extend the scope of the
reaction to other primary aromatic amines, secondary amines,
and heterocyclic amines.
Characterization of {1,3-Diisopropyl-2-(pyridin-3-yl)guanidine},
9. Yield: 0.79 g, 60%. Anal. Calcd for C12H20N4: C, 65.42; H,
9.15; N, 25.43. Found: C, 65.29; H, 9.44; N, 25.21. H NMR
1
(toluene-d8, 297 K): δ 8.09, 7.95, 7.08, 7.07 (4 m, 1H each, N-C5N-
H4), 4.76 (bs, 2H, NH), 3.86 (m, 2H, N-CH(CH3)2), 1.10 (d,
12H, 3JH-H=6.6 Hz, N-CH(CH3)2). 13C{1H} NMR (toluene-d8,
297 K): δ 151.5 (CdN), 148.1, 145.3, 140.8, 129.2, 126.5 (N-C5-
NH4), 31.8 (N-CH(CH3)2), 22.8 (N-CH(CH3)2).
Characterization of {2-(4,6-Dimethylpyridin-2-yl)-1,3-diisopropyl-
guanidine}, 10. Yield: 0.83 g, 59%. Anal. Calcd for C14H24N4: C,
67.70; H, 9.74; N, 22.56. Found: C, 67.55; H, 9.94; N, 22.44. 1H
NMR (toluene-d8, 297 K): δ 6.47, 6.34 (2 m, 1H each, N-C5NH2-
Me2), 4.65 (bs, 2H, NH), 4.12 (m, 2H, N-CH(CH3)2), 2.35, 2.20
3
(2s, 3H each, N-C5NH2Me2), 1.26 (d, 12H, JH-H = 6.6 Hz,
N-CH(CH3)2). 13C{1H} NMR (toluene-d8, 297 K): δ 147.4
(CdN), 164.5, 154.7, 154.3, 120.7, 118.2 (N-C5NH2Me2), 49.4,
42.9 (N-C5NH2Me2), 32.5 (N-CH(CH3)2), 23.8 (N-CH(CH3)2).
Characterization of {1-(tert-Butyl)-3-ethyl-2-mesitylguanidine},
13. Yield: 1.44 g, 92%. Anal. Calcd for C16H27N: C, 73.51; H,
10.41; N, 16.07. Found: C, 73.85; H, 10.19; N, 15.99. 1H NMR
(toluene-d8, 297 K): δ 6.91 (s, 2H, N-C6H2Me3), 3.30 (bs, 2H,
NH), 2.61 (bs, 2H, N-CH2CH3), 2.31 (s, 3H, N-C6H2Me2-
(p-Me)), 2.26 (s, 6H, N-C6H2Me(o-Me)2), 1.50 (bs, 9H, N-C-
(CH3)2), 0.82 (bs, 3H, N-CH2CH3). 13C{1H} NMR (toluene-d8,
297 K): δ 145.2 (CdN), 137.3, 129.9, 129.7, 128.9 (N-C6H2-
Me3), 50.5 (N-CH2CH3), 37.1 (N-C(CH3)2), 31.9 (N-C(CH3)2),
20.9, 18.5 (N-C6H2Me3), 15.5 (N-CH2CH3).
Experimental Section
General Procedures. All reactions were performed using stan-
dard Schlenk and glovebox techniques under an atmosphere of
dry nitrogen. Solvents were distilled from appropriate drying
agents and degassed before use. Microanalyses were carried out
with a Perkin-Elmer 2400 CHN analyzer. H and 13C NMR
1
spectra were recorded on a Varian Unity FT-300 spectrometer
and are referenced to the residual deuterated solvent. The g-HSQC
spectra were recorded on a Varian Inova FT-500 spectrometer
using standard VARIAN-FT software. ZnEt2, MgBu2, n-BuLi,
amines, and carbodiimides were purchased from Aldrich. Li-
quid amines were distilled from CaH2.
General Procedure for Guanylation Reactions at NMR Tube
Scale. Catalytic reactions were performed on a small scale in an
NMR tube fitted with a concentric Teflon valve. Reactions were
performed in the NMR spectrometer with 1 mmol of amine, 1 mmol
of carbodiimide, and 1.5 mol % of catalyst in the appropriate
Synthesis and Characterization of [Zn(Et){(4-t-BuC6H4)Nd
C(N-i-Pr)(NH-i-Pr)}]2, 15. In the glovebox, a solution of ZnEt2
(3.00 mmol, 3 mL of a 1 M solution in hexanes) was added
dropwise to a solution of 2-(4-(tert-butyl)phenyl)-1,3-diisopro-
pylguanidine (3.00 mmol, 1.10 g) in toluene (30 mL), and the
mixture was stirred for 2 h. The mixture was concentrated and
stored at -25 °C to give compound 15 as white crystals. Yield:
0.72 g, 65%. Anal. Calcd for C38H68N6Zn2: C, 61.70; H, 9.27; N,

Article
Organometallics, Vol. 29, No. 12, 2010 2795
Table 5. Crystal Data and Structure Refinement for 7 and 15
1H NMR (THF-d8, 297 K): δ 6.68 (s, 2H, N-C6H2Me3), 3.78 (bs,
2H, CH(CH3)2), 3.58 (bs, 4H, THF), 2.12 (s, 3H, N-C6H2Me2(p-
Me)), 1.97 (s, 6H, N-C6H2(o-Me)2Me), 1.63 (bs, 4H, THF), 1.05
(bs, 12H, CH(CH3)2). 13C NMR (THF-d8, 297 K): δ 148.5,
(CdN), 146.0, 130.4, 129.7, 129.1 (C6H2Me3), 68.2 (THF), 43.8
(CH(CH3)2), 26.4 (THF), 23.7 (CH(CH3)2), 20.9 (N-C6H2(p-
Me)Me2), 18.6 (N-C6H2(o-Me)2Me).
General Procedures for the Kinetics Experiments. Carbodi-
imide (1 mmol) was added to a solution of catalyst and amine
(1 mmol) in deuterated toluene in a J-Young tube at room tem-
perature. The reaction rate was measured by monitoring the dis-
appearance of amine and formation of guanidine by 1H NMR
spectroscopy.
7
15
empirical formula
fw
C13H20BrN3
298.23
290(2)
0.71073
monoclinic
P21/n
8.721(1)
25.988(4)
13.289(2)
91.499(4)
3010.9(7)
8
C38H66N6Zn2
737.71
230(2)
0.71073
monoclinic
P21/c
12.392(4)
9.546(3)
17.778(5)
96.214(7)
2090.7(11)
2
temperature, K
wavelength, A
cryst syst
space group
a, A
b, A
c, A
β,°
volume, A3
Z
X-ray Crystallographic Structure Determination. Crystals of
compound 15 were of poor quality. We have attempted to repeat
the crystallization many times, and we have mounted several
crystals, but we were unable to obtain better data. However,
considering the importance of the structure, it was solved des-
pite the aforementioned problems. Single crystals of 7 and 15
were mounted on a glass fiber and transferred to a Bruker X8
APEX II CCD-based diffractometer equipped with a graphite-
monochromated Mo KR radiation source (λ=0.71073 A). Data
were integrated using SAINT,15 and absorption correction was
performed with the program SADABS.16 The software package
SHELXTL version 6.1217 was used for space group determina-
tion, structure solution, and refinement by full-matrix least-
squares methods based on F2. All non-hydrogen atoms were
refined with anisotropic thermal parameters. Hydrogen atoms
were placed using a “riding model” (except H3 in complex 15,
which was located in the difference map) and included in the
refinement at calculated positions. Crystallographic data are
presented in Table 5.
density (calcd),
g/cm3
1.316
1.172
absorp coeff, mm-1
F(000)
2.716
1232
1.178
792
cryst size, mm3
index ranges
0.20 ꢀ 0.15 ꢀ 0.12
-10 e h e 10,
-30 e k e 30,
-15 e l e15
13 829
5105 [R(int)=
0.0941]
5105/0/331
0.848
R1 =0.0658,
wR2 =0.1336
R1 =0. 2012,
wR2 =0.1732
1.038 and -0.533
0.22 ꢀ 0.19 ꢀ 0.12
-10 e h e 9,
-7 e k e 7,
-14 e l e14
5060
1172 [R(int)=
0.0997]
1172/0/219
1.000
R1 =0.0448,
wR2 =0.1063
R1 =0.0666,
wR2 =0.1195
0.372 and -0.256
reflns collected
indep reflns
data/restraints/params
goodness-of-fit on F2
final R indices
[I >2σ(I)]
R indices (all data)
largest diff peak
and hole, e A-3
11.36. Found: C, 61.88; H, 9.34; N, 11.31. 1H NMR (C6D6, 297
K): δ 7.33 (m, 8H, N-C6H4-t-Bu), 3.67 (bs, 4H, N-CH(CH3)2),
3.61 (bs, 2H, NH), 1.74 (t, 6H, JH-H = 8.1 Hz, ZnCH2CH3),
Acknowledgment. We gratefully acknowledge finan-
ꢀ
3
cial support from the Ministerio de Ciencia e Innovacion,
1.31 (s, 18H, N-C6H4-C(CH3)3), 1.16 (bs, 24H, N-CH(CH3)2),
0.78 (bs, 4H, ZnCH2CH3). 13C NMR (C6D6, 297 K): δ 159.1,
(CdN), 145.5, 138.5, 121.6, 119.5 (N-C6H4-t-Bu), 43.2 (N-CH-
(CH3)2), 41.4 (N-C6H4-C(CH3)3), 30.5 (ZnCH2CH3), 28.1 (N-
C6H4-C(CH3)3), 20.5, 19.6 (N-CH(CH3)2), 10.25 (ZnCH2CH3).
Synthesis and Characterization of [Li{(2,4,6-Me3C6H4)Nd
C(N-i-Pr)(NH-i-Pr)}(THF)], 16. In the glovebox, a solution of
2,4,6-trimethylaniline (6.40 mmol, 0.96 mL) in a mixture of
THF/hexane (9:1) (30 mL) was added to a solution of n-BuLi
(6.40 mmol, 4 mL of a 1.6 M solution in hexanes) in a Schlenk
tube. N,N0-Diisopropylcarbodiimide (6.40 mmol, 1 mL) was
added to the above reaction mixture. The mixture was concen-
trated and stored at -25 °C to give compound 16 as white
crystals. Yield: 1.84 g, 85%. Anal. Calcd for C20H34LiN3O: C,
70.77; H, 10.10; N, 12.38. Found: C, 71.01; H, 10.21; N, 12.44.
Spain (Grant Nos. Consolider-Ingenio 2010 ORFEOCS-
D2007-00006, CTQ2008-00318/BQU, CTQ2009-09214)
and the Junta de Comunidades de Castilla-La Mancha,
Spain (Grant No. PCI08-0010).
Supporting Information Available: Full crystallographic data
for 7 and 15 are available free of charge via the Internet at http://
pubs.acs.org.
(15) SAINTþ v7.12a, Area-Detector Integration Program; Bruker-
Nonius AXS: Madison, WI, 2004.
(16) Sheldrick, G. M. SADABSversion2004/1, A Programfor Empirical
€
€
Absorption Correction; University of Gottingen: Gottingen, Germany, 2004.
(17) SHELXTL-NT version 6.12, Structure Determination Package;
Bruker-Nonius AXS: Madison, WI, 2001.