CHEMSUSCHEM
COMMUNICATIONS
M3P was only 2%. Similar results
were also obtained in the pres-
ence of p-toluene sulfonic acid
functionalized ionic liquids 1c
and 2c (Table 1, entries 8 and
11). By using 1a (Table 1,
entry 6), the conversion of GVL
increased slightly to 53%; 14%
M3P and 5% M4P were ob-
tained. 2a (Table 1, entry 9) also
showed comparable catalytic
performance. These results sug-
Table 2. Catalytic distillation of GVL in the presence of 3 under different conditions.[a]
Entry
T
[8C]
Methanol feed
GVL conversion
[mol%]
Selectivity [mol%]
rate [mLhÀ1
]
M2P
M3P
M4P
MMP
MPA
1
2
3
4
5
6
7[b]
150
190
210
170
170
170
170
10
10
10
2
5
15
10
52
96
85
30
88
97
96
0
2
4
3
0
0
0
20
75
62
40
74
62
78
8
16
13
13
10
18
13
13
0
0
0
3
1
0
0
0
0
1
0
17
5
[a] GVL (10 g, 0.1 mol) with 3 (1.8 mmol) for 3.5 h. [b] 10 g GVL added into reactor every 4 h during a 24 h run.
À
gested that CF3SO3 was the
preferable anion for the catalytic distillation of GVL. By combin-
ing CF3SO3H and three germinal zwitterionic organic salts, 3, 4,
and 5 were synthesized. The conversion of GVL was clearly im-
proved to 94, 98, and 97% by 3, 4 and 5 (Table 1, entries 12–
14), respectively. Moreover, they also increased the selectivity
for M3P to 88, 71, and 73%, respectively. Note that in the pres-
ence of 3, only 7% M4P was generated and the isomer ratio of
M3P/M4P was over 12. The good discrimination between M3P
and M4P may facilitate further highly selective syntheses of
nylon monomers through the metathesis method we pro-
posed above. With regard to the high activity of H3PW12O40,
ionic liquid 6 was also used, which resulted in a medium yield
of M3P (Table 1, entry 15). A control experiment showed that
the zwitterionic salt (MIMPS; Table 1, entry 16) was incapable
of catalyzing this reaction. Additionally, for the tests involving
Figure 1. The structures of the by-products MMP and MPA formed from the
catalytic distillation of GVL.
low conversion of GVL at 2108C could be attributed to more
GVL vaporizing and entering into the collector with methanol.
By lowering the methanol feed rate to 2 mLhÀ1, (Table 2,
entry 4) only 30% GVL was converted and 3% M2P was
formed. At feed rates of 5 and 15 mLhÀ1, (Table 2, entries 5
and 6) the conversion of GVL was improved and the selectivity
for M3P was slightly lower than that at the rate of 10 mLhÀ1
(Table 1, entry 12). Moreover, higher methanol feed rates in-
creased MMP formation. These results suggested that the
methanol feed rate played a key role in the catalytic distillation
of GVL.
À
acidic ionic liquids containing CF3SO3 and PW12O403À, no M2P
was isomerized from M3P. These results revealed that acidic
ionic liquids synthesized through the combination of strong
acids and zwitterionic salts did not only maintain high catalytic
activity, but also exhibited good selectivity.
Compared with H2SO4, CF3SO3H, and H3PW12O40, the acid-
functionalized ionic liquids provided much less charcoal prod-
ucts and remained homogeneous and semitransparent; how-
ever, the color of these ionic liquids turned dark after 3.5 h run
time. For all tests conducted under 1708C, the weight loss of
methanol was 1–2 wt%. To verify the formation of dimethyl
ether, a liquid nitrogen trap was used to collect samples for
gas chromatography (GC) and NMR analysis; however, no di-
methyl ether was detected.
In an extended 24 h catalytic run (Table 2, entry 7), we re-
loaded fresh GVL into the reactor every 4 h and performed the
reaction at 1708C with a methanol feed rate of 10 mLhÀ1
.
A turnover number greater than 250 (based on M3P) was
achieved by using 3 as the catalyst. The overall yield of M3P
was 75% (yield=conversionꢁselectivity) and the selectivity for
M3P reached 78%. Because of the low volatility and high ther-
mal stability, most of 3 remained in the reactor and could be
easily recovered.
Next we examined the effect of temperature and methanol
feed rate on the reaction using 3 as the catalyst. As shown in
Table 2, the reaction did not proceed adequately at 1508C
(Table 2, entry 1), leaving a large amount of GVL in the reactor
after a 3.5 h run. The selectivities for M3P and M4P were only
20 and 8 %, respectively. Additionally, two unexpected by-
products, methyl 4-methoxypentanoate (MMP) and 4-methoxy-
pentanoic acid (MPA), were detected by GC–MS (Figure 1). The
mass spectra of them and plausible fragmentation pathways
are shown in Figure S4. The formation of a considerable
amount of MMP indicated that the elimination of water or
methanol to form a double bond could not proceed easily at
1508C. When the reaction was performed at 1908C and 2108C
(Table 2, entries 2 and 3) MMP and MPA were not detected;
however, isomerization of M3P to M2P occurred. The relatively
Because of the formation of MMP and MPA, we examined
the conversion of GVL with methanol in a Teflon-lined auto-
clave. As a result, only 2% methyl pentenoates were obtained
from GVL (Table 3, entry 1). However, GVL was converted
Table 3. Different substrates with methanol were tested in an autocla-
ve.[a]
Entry
Substrate
Product distribution [mol%]
GVL
pentenoates
MMP+MPA
1
2
3
GVL
MMP
M3P
53
51
0
2
8
58
43
26
1
[a] After reaction of substrate (1 g), methanol (3 g), and 3 (1.8 mmol)
loaded into a Teflon-lined autoclave and heated to 1708C for 3 h.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 600 – 603 602