Trimethyl orthoformate as a highly selective mono-C-methylating agent for arylacetonitriles

Full text of DOI:10.1021/jo980914s

9540
J . Org. Chem. 1998, 63, 9540-9544
Ch a r t 1. Non ster oid a l An a lgesics
Tr im eth yl Or th ofor m a te a s a High ly
Selective Mon o-C-Meth yla tin g Agen t for
Ar yla ceton itr iles
Maurizio Selva*,† and Pietro Tundo
Dipartimento di Scienze Ambientali dell’Universita` Ca`
Foscari, Calle Larga S. Marta, 2137-31023 Venezia, Italy
Received May 12, 1998
The mono-C-methylation of arylacetonitriles (ArCH2-
CN, 1) to produce 2-arylproprionitriles [ArCH(CH3)CN,
2] represents a valuable reaction especially from a
pharmaceutical standpoint. In fact, a number of com-
pounds 2 are key intermediates for the synthesis of
nonsteroidal analgesics of the hydratropic acid (2-aryl-
propanoic acid) class.1 Common well-known examples
are Ibuprofen, Ketoprofen, and Naproxen (Chart 1).
However, synthetic procedures for the direct mono-
methylation of 1 fail with classical alkylating agents
(methyl halides and dimethyl sulfate) because mixtures
of mono- and dimethylated products are always obtained
(Scheme 1).2 For instance, the alkylation of phenyl-
acetonitrile with CH3I is reported with a mono- to
dimethyl selectivity of 84%, at a conversion of 86%.3
Although a number of multistep alkylation methods
have been developed for the preparation of 2-aryl-
propanoic acids,1 the achievement of an effective one-pot
procedure still represents a challenging task and may
deserve attention from both the economical standpoint
and the synthetic feasibility.
Concerning this, a very efficient procedure is the
ruthenium-catalyzed reductive methylation of active
methylene compounds carried out at 135-230 °C with
paraformaldehyde.4 However, we extensively reported
that direct highly selective mono-C-methylations of CH2-
acidic compounds (YCH2X) can also be performed by the
use of dimethyl carbonate (DMC) as a methylating agent,
without any metal catalyst.5-11 Thus, at 180-210 °C in
the presence of weak bases (K2CO3), aryl- and aroxy-
acetonitriles, methyl aryl- and aroxyacetates (Y ) Ar,
ArO; X ) CN, CO2CH3), and R-methylene sulfones (Y )
Ar, X ) SO2Ar, SO2R) yield the corresponding mono-C-
methyl derivatives with selectivities >99% at a complete
substrate conversion. In addition, the procedure is a true
environmentally benign one: DMC is a nontoxic reagent,
Sch em e 1
the base can be used catalytically, and neither organic
nor inorganic byproducts are formed and need to be
disposed of.12,13
In a further effort to conceive new methods for the
selective monoalkylation of arylacetic acid derivatives,
we explored the applicability of ortho esters as alkylating
agents; the attention was focused on trimethyl orthofor-
mate (TMOF). Although ortho esters are most commonly
used for the preparation of ketals and acetals through
transacetalation, transetherification, and reduction reac-
tions, 14a,15-18 some successful TMOF-mediated N-methy-
lations of aromatic amines and imidazole-like compounds
have also been claimed.19-21 More generally, ortho esters
have been reported as highly selective O-alkylating
agents of primary alcohols in the presence of a mont-
morillonite catalyst.22 Some years ago, we also reported
that, at 195 °C and under basic conditions, TMOF could
react with phenol, thiophenol, and phenylacetonitrile to
yield the corresponding O-, S-, and C-methylated deriva-
tives;23 however, while anisole and thioanisole were
obtained by using K2CO3 as a base, the reaction of
phenylacetonitrile proceeded only with t-BuOK and we
noticed that a selective mono-C-methylation was elusive.
(12) Selva, M.; Tundo, P. In Green Chemistry: Designing Chemistry
for the Environment; Anastas, P., Williamson, T., Eds.; ACS Sympo-
sium Series No. 626; American Chemical Society: Washington, DC,
1996; Chapter 7, pp 81-91.
† Phone: +39 41 2578687. Fax: +39 41 2578620. E-mail: selva@
unive.it.
(1) Rieu, J . P.; Boucherle, A.; Cousse, H.; Mouzin, G. Tetrahedron
1986, 42, 4095-4131.
(2) Carruthers, W. In Some Modern Methods of Organic Synthesis,
3rd ed.; Cambridge University Press: Cambridge, U.K., 1989.
(3) Brandstrom, A.; J unggren, U. Tetrahedron Lett. 1972, 472.
(4) Abe, F.; Hayashi, T.; Tanaka, M. Chem. Lett. 1990, 5, 765-768.
(5) Tundo, P.; Moraglio, G.; Trotta, F. Ind. Eng. Chem. Res. 1989,
28, 881.
(13) Ono, Y. Pure Appl. Chem. 1996, 68, 367-75.
(14) March, J . In Advanced Organic Chemistry, 4th ed.; Wiley &
Sons: New York, 1992: (a) pp 390, 443, 890; (b) pp 378-386.
(15) Branalt, J .; Kvarnstrom, I.; Classon, B.; Samuelsson, B. J . Org.
Chem. 1996, 61, 3611-3615.
(16) J ellen, W.; Mittelbach, M.; J unek, H. Monatsh. Chem. 1996,
127, 167-172.
(17) Faja, M.; Reese, C. B.; Song, Q. L.; Zhang, P. Z. J . Chem. Soc.,
Perkin Trans. 1 1997, 191-94.
(6) Tundo, P.; Trotta, F.; Moraglio, G. J . Chem. Soc., Perkin Trans.
1 1989, 1070.
(18) Inoue, S.; Asami, M.; Honda, K.; Miyazaki, H. Chem. Lett. 1996,
10, 889-90.
(7) Selva, M.; Marques, C. A.; Tundo, P. J . Chem. Soc., Perkin Trans.
1 1994, 1323-1328.
(19) Todter, C.; Lackner, H. Synthesis 1997, 5, 567.
(20) Padmanabban, S.; Reddy, N. L.; Durant, G. J . Synth. Commun.
1997, 27, 691-699.
(8) Bomben, A.; Marques, C. A.; Selva, M.; Tundo, P. Tetrahedron
1995, 51, 11573-11580.
(21) Katritzky, A. R.; Musgrave, R. P.; Rachwal, B.; Zaklika, C.
Heterocycles 1995, 41, 345-52.
(9) Tundo, P.; Selva, M. Chemtech. 1995, 25 (5), 31-35.
(10) Bomben, A.; Marques, C. A.; Selva, M.; Tundo, P. Recl. Trav.
Chim. Pays-Bas 1995, 51, 11573-11580.
(22) Sampath Kumar, H. M.; Subba Reddy, B. V.; Mohanty, P. K.;
Yadav, J . S. Tetrahedron Lett. 1997, 38, 3619-22.
(23) Selva, M.; Trotta, F.; Tundo, P. J . Chem. Soc., Perkin Trans. 2
1992, 519-22.
(11) Bomben, A.; Selva, M.; Tundo, P. J . Chem. Res., (Miniprint)
1997, 2688-2696; J . Chem. Res., Synop. 1997, 448-449.
10.1021/jo980914s CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/17/1998

Notes
J . Org. Chem., Vol. 63, No. 25, 1998 9541
Ta ble 1. Rea ction of P h en yla ceton itr ile w ith Tr im eth yl Or th ofor m a te Ca r r ied Ou t in th e P r esen ce of t-Bu OK a n d
Differ en t Cosolven tsa
products (%)d
entry
T (°C)
cosolvent, (A/Cs)b
time (min)
convn,c (%)
2a
3a
4a
others
1
2
3
4
5
190
140
190
190
160
60
90
180
200
180
96
85
72
98
42
28
19
36
70
4
11
4
5
25
32
60
31
DMF (4)
MeOH (4)
MeOH (4)
trace (<1)
28
38
a
All reactions were carried out in an autoclave loaded with a mixture of PhCH2CN (0.5 g, 4.3 mmol), TMOF (20 mL), and t-BuOK (0.96
b
g, 8.6 mmol) in a 1:43:2 molar ratio, respectively. A/Cs is the volumetric ratio (mL/mL) between TMOF (A) as the alkylating agent and
the cosolvent (Cs) (entries 3-5). c % determined by GC. 2a , 2-phenylpropionitrile [PhCH(CH3)CN]; 3a , 2-phenylisobutyronitrile
d
[PhC(CH3)2CN]; 4a , phenylacetic acid (PhCH2COOH); others, unidentified high-boiling products. % determined by GC.
We wish to report here that, in the presence of t-BuOK,
TMOF may allow a one-pot transformation of arylaceto-
nitriles into the corresponding 2-aryl proprionitriles with
excellent monomethyl selectivities (up to 98-99% at
conversions of 96-98%) providing that reactions be
performed in the presence of suitable amounts of metha-
nol as a cosolvent.
Resu lts a n d Discu ssion
A first set of experiments was planned by using
phenylacetonitrile (1a ) as a model compound. As before
mentioned, under basic conditions, a high temperature
(g190 °C) was necessary for TMOF to act as a C-
F igu r e 1. TMOF-mediated methylation of phenylacetonitrile
methylating agent, plausibly through a BAl2 mechanism.14b
carried out at 190 °C and in the presence of different amounts
of MeOH as a cosolvent.
Thus, all the reactions were carried out at 190-210 °C
by loading an autoclave with a mixture of 1a (0.5 g; 4.3
mmol), TMOF (20 mL), and t-BuOK (amount: see Table
1). Experiments were performed in the presence of
different cosolvents in order to explore whether the
medium polarity could have an effect in tuning the
selectivity toward the monomethylated product: to this
aim, DMF and MeOH were used.7 Each was added
separately to the mixture of the reagents in different
volumetric ratios with respect to TMOF (see Table 1).
This latter reagent was used in a large excess acting both
as the methylating agent and the solvent. Table 1
reports the results.
Experiments 1-4 refer to the use of the base in a 2
molar excess with respect to the substrate. When no
cosolvents are used, the reaction of 1a with TMOF is
rapid though nonselective: at nearly quantitative con-
versions (85-96%), mixtures of mono- and dimethylated
products [PhCH(CH3)CN (2a ) and PhC(CH3)2CN (3a )] are
always observed along with PhCH2COOH (4a ) and other
unidentified high-boiling compounds (entries 1 and 2).24
Instead, the use of cosolvents dramatically influences
the reaction outcome. At 190 °C, in the presence of DMF
(TMOF/DMF ) 4 volume ratio), although the reaction
becomes slower (72% conversion after 180 min), the
extent of monomethylated products increases: 2a and 3a
are observed in a 36 and 5% amounts, respectively, while
the sum of other products is 31% (entry 3). A further
and marked improvement of the selectivity toward the
monomethyl derivative 2a is achieved by the addition of
MeOH as a cosolvent. Under the same conditions used
for DMF (190 °C; TMOF/MeOH ) 4 volume ratio), the
presence of MeOH allows the methylation of 1a to
Sch em e 2
proceed quantitatively (98% conversion after 200 min)
yielding 2a in a 70% amount, only traces of 3a (<1%),
and 4a as the sole detectable byproduct (25%) (entry 3
and Scheme 2). At a lower temperature (160 °C), the
methylation is markedly slower and PhCH2COOH be-
comes the major product (entry 5).
Under the conditions of entry 4, both the rate and the
obtainable monomethyl selectivity appear to be rather
independent from the TMOF/substrate ratio. In fact,
when this is decreased from 43 (the value of Table 1) to
11 (or even to 4.3) by increasing by 4 (or 10) the substrate
quantity, no appreciable changes in the reaction time or
in the product distribution are observed: after 200 min,
conversion is 98 (99) and the 2a , 3a , and 4a amounts
are 69 (70), 2 (2), and 25 (25)%, respectively.
Encouraged by the promising monomethyl selectivity
observed, we began to investigate whether the TMOF-
mediated methylations could be affected by different
amounts of both MeOH and t-BuOK.
At first, some tests were set up by varying the volume
of the added MeOH. Experiments were run at 190 °C
by reacting PhCH2CN in the presence of a 2 molar excess
of t-BuOK. Figure 1 shows the results for reactions
stopped after 180 min. The conversion of the substrate
(1a ) and the related product distributions are reported
versus the TMOF/MeOH (A/M) volumetric ratio; two
considerations emerge: (i) Under the explored reaction
conditions, a very high mono- to di-methyl selectivity
(24) Unless otherwise indicated, the product % indicated in Table
1 as well as those of Tables 2 and 3 represents the % areas of the
corresponding gas-chromatographic peaks. However, when authentic
samples of compounds 1a -4a were compared to tetradecane as a
standard, very similar GC-response factors were observed for them.

9542 J . Org. Chem., Vol. 63, No. 25, 1998
Notes
Ta ble 2. Mon om eth yla tion of P h en yla ceton itr ile w ith
Tr im eth yl Or th ofor m a te in th e P r esen ce of Differ en t
Am ou n ts of Meth a n ol a n d t-Bu OKa
in fact, such a reaction becomes important only when the
base is in a 2-fold excess with respect to 1a (B/S ) 2,
Table 1 and Figure 1).
products (%)e
T
B/Sb
A/Mc
time convnd
The data of Figure 1 and of entries 1-4 of Table 2 allow
one to get a measure of the importance of both the
cosolvent MeOH and the base. While the former (MeOH)
deeply influences the methylation selectivity, the latter
(t-BuOK) mainly affects the reaction rate and the extent
of the nitrile hydrolysis. As a further support to this,
Table 2 reports the outcomes of the reaction of 1a with
TMOF carried out by using A/M ratios of 8 and of 20 and
B/S of 0.5, 1.2, and 1.5 (entries 5-9). These results shows
the following: (i) At every given B/S ratio, the decrease
of the added methanol produces a drop in the monom-
ethyl selectivity (compare entries 4, 7, and 8) and,
concurrently, an increased methylation rate (compare
entries 3, 5, and 6); accordingly, the reduction of the
cosolvent also allows the methylation to occur at a lower
temperature (190 vs 210 °C; compare entries 4 and 8-9).
(ii) At every given A/M ratio, the increase of the base
amount results in a marked increase of the reaction rate
as well, while selectivity is scarcely, if at all, affected
(compare entries 3 and 4, 5 and 7, and 6 and 8). Finally,
at a S/B ratio of 0.5, the reaction is extremely slow even
by using small volumes of MeOH (entry 9).
entry (°C) (mol ratio) (vol ratio) (min)
(%)
2a 3a 4a
1
190
1
4
180
420
465
435
840
300
570
470
300
300
950
43
55
67
62
84
96
85
91
95
93
40
41
53 <0.5
2
3
210
210
1
1.2
4
4
66
60
80
92
82
81
88
82
0.6
0.6
1
1
2
8
4
9
4
5
6
7
8
9
200
210
210
190
190
210
1.5
1.5
1.2
1.2
1.2
0.5
4
8
20
8
20
8
1
1
2
37 <0.5
a
All reactions were carried out in an autoclave loaded with
PhCH2CN (0.5 g; 4.2 mmol) and TMOF (20 mL); MeOH as a
cosolvent and t-BuOK were added as reported in the footnotes b
b
and c. B/S is the molar ratio between the base (B) and the
substrate (S). c A/M is the volumetric ratio between TMOF (A) as
d
the alkylating agent and methanol (M) as the cosolvent.
%
determined by GC. e 2a , 3a , and 4a are defined as in Table 1; %
determined by GC.
(SM/D ) 96-99%)25 is always attained. However, an
optimal A/M of 4 may be identified whereby SM/D reaches
a maximum of 99%. (ii) Although experiments are
performed under a N2 atmosphere, the formation of
PhCH2COOH appears unavoidable and quite constant
throughout all the examined reactions: 4a is observed
in a 25-30% amount regardless of the added methanol.
Under the conditions we found for the highest SM/D
(A/M ) 4), we then explored whether any base effects
could be observable; PhCH2CN was reacted with TMOF
by varying the t-BuOK amount over the range of 0.5-
1.5 molar equiv with respect to 1a . Table 2 reports the
results.
The decrease of the quantity of t-BuOK drastically
depresses the reaction rate. At 190 °C and B/S (base/
substrate molar ratio) of 1, low conversions (43-55%) are
observed even for a prolonged reaction time [compare
entry 4 of Table 1 (B/S ) 2) to entry 1 of Table 2]. More
generally, when B/S e 1.5, a higher reaction temperature
becomes necessary to push the methylation at an ap-
preciable rate (entries 2-4). Thus, at a B/S of 1.5, a
distinct improvement is observed at 200 °C: after 300
min, the reaction goes to a substantial completion (96%
conversion: entry 4).
Despite the higher temperature (210 vs 190 °C) and
longer reaction times (300-850 min vs 180 min), all the
tested reactions proceed with a very high monomethyl
selectivity (SM/D g 99%). In addition, the formation of
PhCH2COOH is observed in only trace amounts (e2%).
A B/S of 1.5 appears to be the best compromise between
the monomethylation rate and the byproducts minimiza-
tion: after 300 min, a conversion of 96% is reached with
2a , 3a , and 4a formed in 92, 1, and 1% amounts,
respectively (entry 4).
As far as the formation of 4a is concerned, this has to
be ascribed to a side reaction of hydrolysis of PhCH2CN
taking place concurrently with respect to the methylation
process. This behavior is likely to be due to some water
(coming from the reagents) whose availability for the
hydrolysis is very sensitive to the quantity of t-BuOK;
Sodium methoxide was also used as a base. However,
under the conditions of entry 4, Table 2 (200 °C; B/S )
1.5; A/M ) 4), the reaction of phenylacetonitrile with
TMOF was not as satisfactory as in the case of t-BuOK:
after 360 min, the conversion was 75% and 2a and 4a
were observed in 65 and 2% amounts, respectively, the
remainder (8%) being unidentified byproducts.
To investigate the synthetic applicability of the ex-
plored methylation procedure, both phenylacetonitrile
and different arylacetonitriles [Ar: 4-CH3OC6H4 (1b),
2-CH3OC6H4 (1c), 4-CH3C6H4 (1d ), 4-ClC6H4 (1e), and
naphthyl (1f)] were reacted with TMOF in the presence
of MeOH and t-BuOK. Table 3 reports the results.
Data for 1a refer to a reaction carried out under the
conditions of entry 4 in Table 2 (A/M ) 4, 200 °C, B/S )
1.5) except for the substrate amount which is 10 times
larger (5 g instead of 0.5 g); the quantity of the base is
also proportionally increased.
As far as the other nitriles are concerned, Table 3
shows that the reaction conditions need to be tuned
according to the reactants’ structure. Electron-donating
substituents of weak and medium strength (4-CH3-,
4-CH3O-, and 2-CH3O-) produce a decrease of the
reaction rate with respect to phenylacetonitrile (compare
entries 1, 2-3, 7, and 9). The effect is much more evident
for 1c (2CH3OC6H4CH2CN) because also a relevant steric
hindrance operates at the ortho position (entries 7 and
8). Therefore, reactions have to be run at 210 °C by
increasing the base amount at B/S of 3 (compounds 1b,c)
and of 2 (1d ). Despite that, no hydrolysis of the substrate
to the respective arylacetic acid is observed. However,
although no dimethylation occurs, unidentified byprod-
ucts form (2-25%; entries 5-10).
The methylation of compounds 1e (4-ClC6H4CH2CN)
and 1f (C10H8CH2CN) with TMOF may proceed under the
same conditions used for 1a (200 °C, B/S ) 1.5, and A/M
) 4) with a SM/D of 97% in both cases (entries 12 and
13), though byproducts are observed for 1e (17-18%;
entries 11-12). Some dimethylation (19%) takes place
for 1f only at a very high conversion (96%; entry 14).
(25) Mono- to dimethyl selectivity (SM/D) is calculated as: {% of
PhCH(CH3)CN/[% of PhCH(CH3)CN + % of PhC(CH3)2CN]} × 100,
where % is defined in ref 24.

Notes
J . Org. Chem., Vol. 63, No. 25, 1998 9543
Ta ble 3. Mon o-C-Meth yla tion of Differ en t Ar yla ceton itr iles w ith Tr im eth yl Or th ofor m a te in th e P r esen ce of MeOH
a n d t-Bu OKa
products (%)d
T
(°C)
B/Sb
(molar ratio)
A/M,b
(vol ratio)
time
(min)
convnc
(%)
yielde
(%)
entry
ArCH2CN (g)
M
D
others
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1a , Ar ) Ph (5)
200
190
200
210
210
210
210
210
200
200
200
200
200
200
1.5
1.5
1.5
1.8
2.5
3
1.5
3
1.5
2
4
4
4
8
8
8
4
4
4
4
4
4
4
2.5
300
300
840
810
570
850
420
960
360
270
200
300
300
380
96
31
48
50
72
94
24
72
79
96
78
93
84
96
93
31
48
50
70
87
19
47
76
82
61
73
82
77
1
49
1b, Ar ) 4-CH3OC6H4 (0.5)
2
2
5
25
3
14
17
18
5
37
60
1c, Ar ) 2-CH3OC6H4 (0.5)
1d , Ar ) 4-CH3C6H4 (0.5)
1e, Ar ) 4-ClC6H4 (0.5)
1f, Ar ) C10H8 (0.5)
1.5
1.5
1.5
1.5
2
2
19
39
47
a
All reactions were carried out in an autoclave loaded with the substrate, TMOF (20 mL), and t-BuOK in the reported molar ratio.
B/S and A/M are the molar and volumetric ratio as defined in footnotes b and c of Table 2. c % determined by GC. % determined by
b
d
GC; M and D, monomethylated [ArCH(CH3)CN] and dimethylated [ArC(CH3)2CN] derivatives, respectively; others, unidentified high
boiling products. e Isolated yields.
mL), equipped with a purging valve, through which, at room
temperature, air was removed before each reaction by purging
with N2 stream. A magnetically stirred mixture of the alkylating
agent, the arylacetonitrile, the base (t-BuOK), and methanol
(where indicated) in the reported molar and volumetric ratios
(see Tables 1-3) was heated in the autoclave, itself heated in
an electrical oven, at the desired temperature (190-210 °C). The
corresponding internal pressure was of 8-12 bar. A thermo-
couple (T) and a needle valve (V) were fixed onto the autoclave
head: while the former (T) (dipping into the reaction mixture)
These results suggest that the reaction conditions for
the methylation of different arylacetonitriles with TMOF
need to be optimized case-by-case to avoid (or minimize)
the byproduct formation.
The isolated yields of products 2 appear to be moderate
(37-60%): these values correspond to the 50-70% of the
gas-chromatographic percent of the monomethyl deriva-
tives (M) reported in Table 3.24 Although yields have not
been optimized, this result can be also partly ascribed to
some decomposition of the starting reagents. This has
been observed, for instance, after the distillation of 2a :
a residual tar is recovered as a nondistillable and
nonanalyzable (by GC) material.
allowed a constant check of the reaction temperature, the latter
1
(V) was connected to a
/ in. stainless steel sampling pipe
8
immersed into the reaction mixture. In this way, the internal
pressure allowed samples to be withdrawn through V at
intervals, during the course of the reaction. Before GC analyses,
each sample (0.2-0.3 mL) was cooled to room temperature,
added to diethyl ether (2 mL), water (2 mL), and diluted HCl (3
drops), and finally shaken.26 The organic layer was then
analyzed by GC.
Con clu sion s
The here described procedure proposes a new one-pot
transformation of arylacetonitriles into 2-arylpropio-
nitriles (2a -f) by using trimethyl orthoformate as the
alkylating agent. The reaction occurs with a high
monomethyl selectivity (up to 99%) at complete substrate
conversions. Although this preliminary investigation is
far from explaining the mechanism responsible for such
an intriguing result, it has revealed that the reaction
outcome is mostly dependent upon the presence of
methanol as a cosolvent. In fact, it is this alcohol that
tunes the reaction toward a very selective monomethy-
lation process.
On the other hand, the base used (t-BuOK) has major
effects on the reaction rate.
Finally, the procedure may also have an environmental
significance; in fact, TMOF is by far a less toxic alterna-
tive to current methylating agents (e.g. methyl halides
or dimethyl sulfate).
Typ ica l Exp er im en ta l P r oced u r e. Mon om eth yla tion of
P h en yla ceton itr ile (En tr y 4, Ta ble 2). The above-described
autoclave was loaded with a solution of phenylacetonitrile (0.5
g, 4.3 mmol), trimethyl orthoformate (20 mL, 0.18 mol), and
methanol (5 mL, 0.12 mol). To this solution, t-BuOK (0.72 g,
6.4 mmol) was added. The autoclave was then closed, purged
with a N2 stream, and finally heated in an electrical oven at
200 °C, while the reaction mixture was kept under a magnetic
stirring. At intervals (30 min), samples were withdrawn and
analyzed by GC: a substantially quantitative conversion of the
substrate was observed after 300 min.
P u r ifica tion of P r od u cts. After the reaction was com-
pleted, the autoclave was rapidly cooled to room temperature
in a water bath. Then, the reaction mixture was transferred
into a separatory funnel, added to water (50 mL), and carefully
acidified with diluted HCl (10%) until a pH of 4-5 was reached
(checked by a pH paper). The organic phase was then extracted
with diethyl ether (3 × 50 mL) and the combined layers dried
over sodium sulfate and filtered. The light solvents (TMOF and
diethyl ether) were removed by rotary evaporation, and the
residue was distilled under vacuum (in the case of compound
2a ) or purified by gravity column chromatography for the
monomethylated derivatives 2b,d -f (silica gel, Merck F60;
eluting solvent, diethyl ether/petroleum ether in a 30:70 v/v
ratio). The vacuum distillation was performed in a micro-
Claisen distillation apparatus with a fused-on Liebig condenser.
Exp er im en ta l Section
All the compounds used were ACS grade and were employed
without further purification. 1H NMR spectra were recorded
at 400 MHz using CDCl3 as the solvent. GC analyses were
performed using a 30 m, DB5 capillary column. GC/MS analyses
were performed by a mass detector at 70 eV coupled to a gas
chromatograph fitted with a 30 m, DB5 capillary column.
Melting points are uncorrected.
Rea ction s Ca r r ied Ou t in Au tocla ve. Gen er a l P r oce-
d u r e. All methylation reactions by TMOF were carried out in
a stainless steel (AISI 316) autoclave (internal volume of 250
(26) The addition of HCl transforms anions such as ArCH-CN,
ArC-(CH3)CN, etc., into the corresponding conjugated acids (PhCH2-
CN, ...) that can be so analyzed by GC. This hydrolytic workup does
not certainly hydrolyze the reacting nitrile 1. If so, also ArCH(CH3)-
COOH (coming from the hydrolysis of 2) should be observed, but we
never detected it.

9544 J . Org. Chem., Vol. 63, No. 25, 1998
Notes
Compounds 2a ,b,d ,e were compared to authentic samples
whose full analytical data were previously reported by us;4 data
for 2f agreed with the reported ones.27 2c was not isolated; its
characterization was through GC/MS analysis by comparison
to an authentic sample.4
Ack n ow led gm en t. This work was supported by
Inca (Interuniversity Consortium “The Chemistry for
the Environment”) and MURST (Ministero Universita`
e Ricerca Scientifica Tecnologica).
(27) Freerksen, R. W.; Selikson, S. J .; Wroble, R. R. J . Org. Chem.
1983, 48, 4087-96.
J O980914S