Highly selective zeolite-catalysed mono-N-alkylation of arylenediamines by dialkyl carbonates

Article abstract of DOI:10.1016/j.tetlet.2006.12.131

Arylenediamines are mono-N-alkylated by dialkyl carbonates in the presence of NaY zeolite catalyst in a regioselective and nontoxic process.

Full text of DOI:10.1016/j.tetlet.2006.12.131

Tetrahedron Letters 48 (2007) 1641–1643
Highly selective zeolite-catalysed mono-N-alkylation of
arylenediamines by dialkyl carbonates
Warren J. Ebenezer,a Michael G. Hutchings,a,* Ken Jones,a
David A. Lamberta and Ian Wattb
aDyStar UK Ltd, University of Manchester, School of Chemistry, Oxford Road, Manchester M13 9PL, UK
bUniversity of Manchester, School of Chemistry, Oxford Road, Manchester M13 9PL, UK
Received 13 December 2006; revised 20 December 2006; accepted 21 December 2006
Available online 4 January 2007
Abstract—Arylenediamines are mono-N-alkylated by dialkyl carbonates in the presence of NaY zeolite catalyst in a regioselective
and nontoxic process.
Ó 2007 Elsevier Ltd. All rights reserved.
A recent analysis of the reaction classes used in the syn-
thesis of a range of drug candidate development samples
has shown that the most common is heteroatom alkyl-
ation (19% of the total), with alkylation of nitrogen
comprising more than half of these.1 Additionally, the
increasing problems resulting from the toxicity of com-
mon alkylating agents are highlighted, which leads the
authors to sound a call for an improved alkylation
methodology. A further well known and common diffi-
culty in synthesis is the prevention of over-alkylation
during the conversion of primary to secondary amines.
In a recent report Hudson and co-workers address this
problem by using alkylnitriles as the N-alkylating agents
for anilines under particularly mild conditions.2 An ear-
lier work by Selva and co-workers also goes part way to
addressing these concerns.3 They have demonstrated
that dialkyl carbonates efficiently mono-N-alkylate
anilines on heating in the presence of the zeolite NaY
(sodium-exchanged faujasite). Both the zeolite and the
carbonate esters are readily available and cheap,4 and
may be recovered and recycled. The carbonate esters
are much less toxic than traditional alkylating agents
such as alkyl halides, sulfates or sulfonates. The
by-products of reaction, the alcohol itself and carbon
dioxide, are also relatively innocuous. Selva established
selectivity for primary over secondary in amine alkyl-
ation and for N- over O-alkylation in the reaction of
aminophenols and aminobenzoic acids.3c
We now describe an additional useful and more surpris-
ing dimension to the selectivity displayed by this reagent
combination in the alkylation of arylenediamines. When
we initially exposed a phenylenediamine to the NaY-
catalysed dialkyl carbonate alkylation reaction condi-
tions, we had expected to observe N,N0-dialkylation.
In fact these conditions lead to highly selective—some-
times specific—mono-N-alkylation of just one of the
amino groups present (Scheme 1).
For a series of arylenediamines, application of the
published conditions3 [dimethyl carbonate (DMC) or
diethyl carbonate (DEC) at reflux, or ethylene carbonate
(EC) at 130 °C] yielded the results presented in the table.
These reactions were monitored by TLC or HPLC and
stopped when the analysis indicated maximal formation
of a single major product in cases 1–7 and two major
products in cases 8 and 9. A typical protocol is given
below.5a
All combinations of reactants show usefully high selec-
tivity for mono-N-alkylation. For the simplest diamines,
selectivities are higher for diethyl carbonate than for
dimethyl carbonate, and, surprisingly, o-phenylene-
R
H
NH2
N
(RO)2CO
+ ROH + CO2
NH2
NH2
NaY zeolite
X
X
*
Corresponding author. Tel.: +44 161 275 1880; fax: +44 161 275
1888; e-mail: mike.hutchings@man.ac.uk
Scheme 1. Mono-N-alkylation of arylenediamines.
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.12.131

1642
W. J. Ebenezer et al. / Tetrahedron Letters 48 (2007) 1641–1643
diamine (OPD) and m-phenylenediamine (MPD) show
comparable yields and high selectivities. In the best cases
(Table 1, entries 2, 4 and 5), HPLC analysis after filtra-
tion and evaporation of excess alkylating agent failed to
detect additional products of any kind; monoalkylations
by DEC and EC are essentially specific.
If the reaction is viewed as a sequence of first order con-
versions, the observed selectivity requires that the rate
constant for the first alkylation of a primary amine site
must be at least 103-fold greater than that for any sec-
ond alkylation, whether at the new secondary site or
at the remaining primary site. The mechanisms of these
alkylations and the origins of the selectivities are of con-
siderable interest for the optimisation of the processes,
but experiments to date do not distinguish whether the
major contribution to the monoalkylation/dialkylation
ratio is differential adsorption on the catalyst of mono-
and dialkylated material, or of differential reactivity of
these substrates once adsorbed. Nevertheless, as noted
earlier, selectivity between distinct primary amine sites
in the alkylations of unsymmetrically ring-methylated
diamines (entries 8 and 9) is low, and it seems unlikely
to us that any mechanism of alkylation within the zeolite
cage would favour reaction at a primary amine group in
a diamine over the primary amine group in a monoalky-
lated diamine. This consideration applies especially in
the case of MPD where the substitution pattern mini-
mises steric and electronic interactions between the
amine residues. In contrast to the OPD case, there can
be no intramolecular hydrogen bonding here to be
modified by an initial alkylation. (Such modification is
believed to be the reason for the selective monomethyl-
ation of OPD by excess methanolic methyl iodide
described by Brown and Nelson,7 with preferential pro-
tonation and deactivation of the mono-N-methylated
OPD by the HI produced in the reaction.) In the reac-
tions with carbonates, comparable selective protonation
of a more basic monomethylated product seems unlikely
since the alkylation releases a non-acidic alcohol. There
will, of course, be a statistical contribution from the
higher availability of primary amine groups in the
diamine, but we speculate that the major contributor
to the selectivity is the preferential adsorption of
diamine over monoalkylated diamine.
Of the more complex diamines, the reaction of 2,4,6-tri-
methyl-MPD (entry 6) is significantly slower, presum-
ably reflecting the steric hindrance at the amine
functionality. Introduction of an additional polar func-
tionality in the form of a carboxyl group, as in 5-car-
boxy-MPD (entry 7), also reduces the reactivity.
Nevertheless, a useful selectivity for mono-N-methyl-
ation persists in both cases. Reaction times for these
could be reduced by conducting them under pressure
in a closed vessel at temperatures above the normal boil-
ing point of the dialkyl carbonate solvent, with no
reduction of yield or selectivity. For the unsymmetri-
cally ring-methylated diamines (entries 8 and 9) both
possible mono-N-methyl isomers were produced and
no attempt has been made to separate them. Isomer ra-
tios (50:50 and 60:40, respectively, by NMR analysis)
show that although total reactivity is reduced, the pres-
ence of a m- or o-methyl group surprisingly does not sig-
nificantly affect the relative reactivity of the two primary
amine groupings in these compounds in this reaction. In
contrast to 2-methyl-p-phenylenediamine (entry 9),
application of the conditions to p-phenylenediamine
itself yielded only intractable tars, presumably the result
of air oxidation.
That the zeolite is essential to both the reactivity and
selectivity may be judged against the observation6 that
reaction of OPD in DMC at 160 °C in the absence of
the zeolite gave only 30% conversion to a 75:20:5 mix-
ture of N-methyl-OPD, N,N-dimethyl-OPD and 2-benz-
imidazolone. Incorporation of various metal salts
enhanced conversions, but also increased the proportion
of benzimidazolone. Furthermore, we have demon-
strated that the zeolite is recyclable by its isolation and
re-use in three successive reaction cycles between OPD
and DMC.5b There was no observable difference in reac-
tivity or selectivity between the three runs.
For the present, regardless of the underlying mechanistic
rationale for its selectivity, this reaction of readily
available arylenediamines offers a uniquely simple, non-
toxic and economical route to their mono-N-alkylated
derivatives, with potential additional environmental
Table 1. Zeolite catalysed alkylations of arylenediamines
Entry
Substratea
Agentb (time/h)
Major productc
Isolated yield (%)
Impuritiesd
1
2
3
4
5
6
7
8
9
OPD
OPD
MPD
MPD
DMC (2)
DEC (12)
DMC (3)
DEC (2)
N-MethylOPD
N-EthylOPD
N-MethylMPD
N-EthylMPD
N-2-HydroxyethylMPD
N,2,4,6-TetramethylMPD
N-Methyl-5-carboxyMPD
N,4- and N,5-DimethylOPD
N,2- and N,3-DimethylPPD
75
78
90
99
80
44e
64
60e
79
6% SM
—
10% Dimethylamines
—
—
20% Two unknown products
15% SM, 15% Dimethylated
1:1 Product ratio, 30% SM
3:2 Mixture of isomers; 10% SM,
10% dimethylated
MPD
EC (12)
2,4,6-TrimethylMPD
5-CarboxyMPD
4-MethylOPD
2-MethylPPD
DMC (48)
DMC (120)
DMC (12)
DMC (6)
a OPD = o-phenylenediamine; MPD = m-phenylenediamine; PPD = p-phenylenediamine.
b DMC = dimethyl carbonate; DEC = diethyl carbonate; EC = ethylene carbonate (1,3-dioxolan-2-one).
c All products were characterised either by spectroscopic and chromatographic comparison with authentic samples, or by NMR and mass spec-
troscopy and microanalysis.
d SM = starting material.
e After column chromatography.

W. J. Ebenezer et al. / Tetrahedron Letters 48 (2007) 1641–1643
1643
benefits.8 In contrast, a typical current process for
mono-N-alkyl MPD, for example, is likely to be longer
and could well involve isomer separation and possible
discard of unwanted isomer.8 The conventional process
is thus less economical and on a manufacturing scale
potentially less environmentally attractive. Many
N-alkylated arylenediamines are commercially impor-
tant building blocks in dyestuff applications,9 and are
components of pharmaceuticals.10 The new reaction
may therefore have immediate industrial relevance.11
material was detectable by HPLC (HP1100 chromato-
graph with diode array detector, on a LiChroCart 55-4
Purospher STAR RP-18 endcapped column, eluting with
an acetonitrile–water gradient containing 0.25% di-
cyclohexylamine phosphate). The relatively high molar
proportion of DEC follows the published procedure3a,e
and is convenient for small scale laboratory work. This
may be unrealistic for larger scale synthesis, especially
bulk manufacture, but satisfactory use of much lower
dialkyl carbonate/amine ratios has been reported3c and we
have also observed that the carbonate can be decreased
substantially on scale-up; (b) Demonstration of catalyst
reuse: OPD was methylated with DMC using the general
procedure described above. On completion of the reaction,
the zeolite catalyst was recovered by suction filtration and
washed with methanol. The recovered zeolite was reused,
following the same protocol with fresh OPD and DMC,
with identical results. In three successive cycles of catalyst
recovery and reuse in methylations of OPD there was no
measurable loss of activity or selectivity.
References and notes
1. Carey, J. S.; Laffan, D.; Thomson, C.; Williams, M. T.
Org. Biomol. Chem. 2006, 4, 2337; See also a statistical
analysis of scaled reactions used in a Pfizer facility:
Dugger, R. W.; Ragan, J. A.; Ripin, D. H. B. Org. Proc.
Res. Dev. 2005, 9, 253.
6. Fu, Y.; Baba, T.; Ono, Y. J. Catal. 2001, 197, 91.
7. Brown, H. C.; Nelson, K. L. J. Am. Chem. Soc. 1953, 75,
24.
2. Nacario, R.; Kotakonda, S.; Fouchard, D. M. D.;
Tillekeratne, L. M. V.; Hudson, R. A. Org. Lett. 2005,
7, 471, and references cited therein. These authors provide
a recent overview of approaches to the N-alkylation
question, and compare their new N-alkylation procedure
based on alkylnitriles with other methods generally
requiring harsher conditions; For a related approach see:
Sajiki, H.; Ikawa, T.; Hirota, K. Org. Lett. 2004, 6, 4977.
3. (a) Selva, M.; Bomben, A.; Tundo, P. J. Chem. Soc.,
Perkin Trans. 1 1997, 1041; (b) Perosa, A.; Selva, M.;
Tundo, P. Synlett 2000, 272; (c) Selva, M.; Tundo, P.;
Perosa, A. J. Org. Chem. 2001, 66, 677; (d) Selva, M.;
Tundo, P.; Perosa, A. J. Org. Chem. 2002, 67, 9238; (e)
Selva, M.; Tundo, P.; Perosa, A. J. Org. Chem. 2003, 68,
7374.
8. A formal process to mono-N-alkylated MPD based on the
new procedure would likely comprise benzene dinitration,
reduction and N-alkylation. This may be compared with
the conventional alternative starting from N-alkylaniline
(derived formally from benzene by halogenation and
displacement by primary alkylamine, or by nitration,
reduction, alkylation), involving nitration and isomer
separation (implying a possible discard of the p-isomer)
and finally reduction. It is not our intention to provide
environmental audits for these various processes, but
nevertheless
a simple unit operation count suggests
potential environmental and economic benefits would
derive from the new, more direct reaction.
4. The zeolite is currently available for $4/kg and the
carbonates for less than $1.9/kg.
9. Gordon, P. F.; Gregory, P. Organic Chemistry in Colour;
Springer: Berlin, 1987; Zollinger, H. Color Chemistry, 3rd
5. (a) General procedure for alkylations: MPD (3.98 g),
diethyl carbonate (146 mL) and zeolite NaY (3.98 g, pre-
dried at 70 °C in vacuo) were stirred under reflux for 2 h.
HPLC analysis then showed the absence of starting
material and the appearance of only one new component.
The mixture was cooled and filtered. The zeolite was
washed with methanol, and the combined filtrates evap-
orated under reduced pressure to give mono-N-ethyl MPD
(5.0 g; 99%), identified by its 1H NMR spectrum which
indicated also that the material was >99% pure. No other
ed.; VHCA: Zurich and Wiley-VCH: Weinheim, 2003.
¨
10. For example in TF476, a potent gastrin/CCK-B antago-
nist Semple, G.; Ryder, H.; Rooker, D. P.; Batt, A. R.;
Kendrick, D. A.; Szelke, M.; Ohta, M.; Satoh, M.;
Nishida, A.; Akuzawa, S.; Miyata, K. J. Med. Chem.
1997, 40, 331.
11. The material described in this Letter is the subject of
DyStar Patent Application WO2006/013164, priority date
July 29th, 2004.