Mendeleev Commun., 2019, 29, 438–440
ammonium salt that does not participate in further processes.
and can proceed in the absence of CO2 without loss of selectivity.
Attempts to accelerate the reaction by heating (45–60°C) led
to the formation of a completely insoluble (even in DMSO)
precipitate. Compared with DMF, the reaction in DMA is slightly
faster. Note that raising the pressure of CO2 decelerates the
reaction (the maximum rate was observed at 10 bar). Apparently,
this fact is associated with the dependence of the carbamic acid
decomposition rate on the concentration of CO2 (see Scheme 2,
step 3). However, CO2 pressure of 1 bar is insufficient to provide
good selectivity.
Methylation of diamine 2 with an excess of MeI in DMA at a
pressure of CO2 of 60 bar led to unexpected results. Instead of
a mixture of four ammonium salts ArN+MenH3–nI– (n = 0–3),
the major product (92%) was a quaternary ammonium salt 5
(Scheme 3). Thus, more than 2 equiv. of MeI react without any
additional base. This fact can be explained by the increase in the
acidity of the arylammonium salts and simultaneous amide solvent-
promoted decrease in the liberating HHal acidity. As a result,
an aromatic amine capable of undergoing further alkylation is
constantly present in the solution in a small steady-state con-
centration. For DMSO solutions, so it is (see, e.g., the pKa values
of HBr and PhNMe2H+ presented in Table S2, Online Supple-
mentary Materials). It is also known that the pKa values of weak
acids in DMF are 0.5–1.5 units higher than those in DMSO.5 DMA
in this respect, apparently, occupies an intermediate position
between DMSO and DMF. Similar results were obtained earlier
for the exhaustive methylation of primary aliphatic and aromatic
amines with MeI in DMF in the presence of 2 equiv. 4-hydroxy-
2,2,6,6-tetramethylpiperidine and in its absence. Unlike aliphatic
amines, 4-aminobenzoic acid and 3-nitroaniline give the corre-
sponding quaternary ammonium salts in the absence of a base
in significant yields.6
For better rationalization, we carried out model experiments
on the methylation of o-, m- and p-toluidines with an excess of
MeI in MeCN, DMF and DMA. In contrast to the reactions in
MeCN, the final methylation products in the amide solvents
were 2-MeC6H4N+HMe2I– and 3-, 4-MeC6H4N+Me3I–, respec-
tively. The reaction rate in DMA was noticeably higher than that
in DMF (Table S1, Schemes S1–S3, Figures S2–S13, see Online
Supplementary Materials). Thus, the exhaustive alkylation of
aromatic amines in amide solvents does not require application
of additional base and can be a real alternative to the known
methods for the synthesis of quaternary arylammonium salts.7
Similar alkylation of diamine 2 with less reactive EtBr led to
the expected diethylammonium derivative 6 (Scheme 3) with
~90% selectivity, the ArN+Et3Br– quaternary salt having not
formed. The main by-product was the monoethyl derivative of
the aromatic amino group, and the reaction time was much longer
as compared to methylation.
Thus, the by-products of methylation of the aliphatic amino group
(for strictly equimolar mixtures of amines) can be formed only in
step 2, i.e. before the complete consumption of the first equivalent
of MeI. The proposed step sequence was proved by reacting
two crude materials obtained independently, namely, methylated
2-MeC6H4NH2 and carbamine derivative of PhCH2NH2. The result
of this reaction was the same as in the case of ‘one-pot’ process.
The main experiments were carried out for model diamines,
namely, 2-H2NC6H4CH2NH2 1, 3-H2NC6H4CH(Me)NH2 2, and
4-H2NC6H4CH2CH2NH2 3. In the absence of carbon dioxide, their
reactions with MeI in MeCN proceeded unselectively and led
to mixtures of products (up to 16 compounds). In particular,
alkylation of diamine 1 gave a mixture of 12 ammonium salts
since the steric hindrance in 1 prevents exhaustive methylation of
arylamino group. Compared with model mixtures of alkyl- and
arylamines, alkylation of diamines 1–3 in MeCN in the presence
of CO2 has a very important difference. In the former case, alkyl-
ammonium carbamate precipitates, while the soluble aromatic amine
is alkylated, i.e., the slow stage of this reaction is homogeneous.
In the case of diamines, this stage becomes heterogeneous due to
the complete insolubility of ammonium carbamate, resulting in the
unpractically prolonged reaction time.
It is known that the products of the reaction of aliphatic amines
with carbon dioxide, depending on the solvent, can be either
carbamic acids or ammonium carbamates (or a mixture thereof),3
e.g., a- and w-(1-naphthyl)alkylamines in DMSO, DMF and
pyridine are quantitatively converted into soluble carbamic acids.4
In THF or dioxane, mixtures of carbamic acid and ammonium
carbamate would form, while in C6H6, CHCl3, MeCN, MeOH or
PriOH, the only products are poorly soluble ammonium carbamates.4(b)
Unlike DMSO and pyridine, amide solvents do not react with
alkylating agents; therefore, DMF and DMA were the solvents of
choice for the alkylation of diamines 1–3 in the presence of CO2.‡
Methylation of diamine 1 with an excess of MeI in DMF
or DMA at 10–60 bar CO2 resulted in the expected dimethyl-
ammonium derivative 4 (Scheme 2) with high selectivity (~90%).
The main by-product, as in the case of model mixture of amines,
was the monomethylation product of the aliphatic amino group.
The methylation in DMF for 72 h at ambient temperature did not
occur completely, anyway, more than 1 equiv. of MeI was consumed
within this time. Reaction completion requires additional 72 h
NH2
NHCOOH
NH2
step 2
step 1
CO2
NH2
MeI, 24–40 °C
(slow)
1
Unlike substrates 1 and 2, alkylation of diamine 3 may be
carried out at elevated temperatures (55–60°C), which signi-
ficantly reduces the reaction time (Scheme 4). This is probably
NH3 I–
NHCOOH
NMenH3–nI–
step 3
NMenH2–n
– CO2, proton
transfer (fast)
H2N
n = 0–2
H2N
NHCOOH
NH2
Me
NH3 I–
i
Me
NMe2H I–
+
step 4
(slow)
2
minor products (~10%)
MeI
EtBr
4
Br–
Me3N I–
Et2HN
NH3 I–
Me
NH3 Br–
Me
Scheme 2 Reagents and conditions: CO2 (10–60 bar), DMF or DMA,
24–40°C.
5
6
‡
Our data on the interaction of diamines with CO2 in DMSO and DMF
are in good agreement with the known data. These results will be published
elsewhere.
Scheme 3 Reagents and conditions: i, AlkHal (6 equiv.), DMA, CO2
(60 bar), 37°C, 48 h (for MeI) or 144 h (for EtBr).
– 439 –