The continuous acid-catalysed etherification of aliphatic alcohols using stoichiometric quantities of dialkyl carbonates

Article abstract of DOI:10.1021/op1002243

A range of methyl and ethyl ethers of aliphatic alcohols have been synthesized cleanly in high yield by reacting the corresponding alcohol with dimethyl carbonate or diethyl carbonate over the solid acid catalyst, I-alumina. The reaction could be conducted at ambient pressure without the need for the large excess of dialkyl carbonate as previously reported in the literature. If the reaction was conducted at high pressure, the conversion of the starting alcohol was greatly reduced. However, high pressure CO₂ can be used as the solvent without significant reduction in yield. This has implications for tandem reactions.

Full text of DOI:10.1021/op1002243

Organic Process Research & Development 2010, 14, 1420–1426
The Continuous Acid-Catalysed Etherification of Aliphatic Alcohols Using
Stoichiometric Quantities of Dialkyl Carbonates
Andrew J. Parrott,† Richard A. Bourne,† Peter N. Gooden,† Han. S. Bevinakatti,‡ Martyn Poliakoff,*,† and Derek J. Irvine*,†,§
School of Chemistry, UniVersity of Nottingham, UniVersity Park, Nottingham NG7 2RD, U.K., Croda Enterprises Ltd, Wilton
Centre, Wilton, Redcar TS10 4RF, U.K., and Department of Chemical and EnVironmental Engineering, UniVersity of
Nottingham, UniVersity Park, Nottingham NG7 2RD, U.K.
Abstract:
synthesis of aliphatic ethers Via DMC in the literature to date.
For example, Selva et al.17 have successfully methylated several
benzyl-type alcohols with a similar method using Y- and
X-faujasites as catalysts at high temperature under autogenic
pressure in an autoclave. They suggest that the methylation and
carboxymethylation reactions occur simultaneously over this
catalyst and at elevated temperatures the methyl carbonate
undergoes decarboxylation, so that the methyl ether is the sole
observed product of the reaction. This method is batch mode
and requires a very large excess of DMC (>30 mol equiv) which
wastes materials and energy due to the heating and cooling of
the excess reagents which are not consumed by the reaction.
Tundo et al.18,19 have reported the synthesis of methyl ethers
from aliphatic alcohols using DMC but only Via adopting a
two-step process. They have proposed that due to their harder
nature, aliphatic hydroxyl groups will only react with the harder
carbonyl group of DMC to afford methyl carbonates; methyl
ethers can then be obtained by decarboxylation of this methyl
carbonate. This two-step procedure has been carried out in
multipot and one-pot batch reactions. Both methods required
an excess of DMC >4 and >18 mol equiv for multipot and one-
pot, respectively, to drive the formation of the methyl carbonate.
Tundo et al.19 have also reported a continuous flow method
for the synthesis of methyl ethers, where a mixture of the alcohol
and DMC (3 mol equiv) are continuously passed over a column
of basic alumina at 180 °C and atmospheric pressure to yield
∼90% of the methyl ether. Such continuous methods are
preferable to the batch methods, especially when scale up of
production needs to be considered.
A range of methyl and ethyl ethers of aliphatic alcohols have been
synthesized cleanly in high yield by reacting the corresponding
alcohol with dimethyl carbonate or diethyl carbonate over the solid
acid catalyst, γ-alumina. The reaction could be conducted at
ambient pressure without the need for the large excess of dialkyl
carbonate as previously reported in the literature. If the reaction
was conducted at high pressure, the conversion of the starting
alcohol was greatly reduced. However, high pressure CO2 can be
used as the solvent without significant reduction in yield. This has
implications for tandem reactions.
Introduction
Aliphatic methyl ethers are important compounds in synthetic
organic chemistry and have applications as fuel additives and
solvents.1-4 Typically, methyl ethers of aliphatic alcohols are
synthesized using the Williamson ether synthesis which uses
highly toxic reagents such as dimethyl sulfate and methyl halides
and generates stoichiometric quantities of inorganic salt.5-7
Dimethyl carbonate (DMC) has received much attention as
an alternative methylating agent as it has low toxicity, is
biodegradable, and has been used to perform methylation and
carboxymethylation reactions with a wide variety of nucleo-
philes.8-16 However, there are relatively few examples of the
* Authors for correspondence. E-mail: martyn.poliakoff@nottingham.ac.uk.
E-mail: derek.irvine@nottingham.ac.uk.
† School of Chemistry, University of Nottingham.
‡ Croda Enterprises, Ltd.
Recently, we reported the continuous methylation of 1-oc-
tanol with DMC with a variety of heterogeneous acid catalysts
at high pressure using supercritical CO2 (scCO2) as the solvent.20
This method allowed the use of lower amounts of DMC (2
mol equiv) whilst maintaining a high yield (86%) of the methyl
ether, with the best-performing catalyst being an acidic γ-alu-
mina. This method was applied successfully to a range of
aliphatic alcohols and diols, and it was also shown that DMC
could be used for clean esterification and N-methylation
reactions.20
§ Department of Chemical and Environmental Engineering, University of
Nottingham.
(1) Ancillotti, F.; Fattore, V. Fuel Process. Technol. 1998, 57, 163–194.
(2) Nel, R. J. J.; de Klerk, A. Ind. Eng. Chem. Res. 2009, 48, 5230–
5238.
(3) Snelling, J.; Curtis, C.; Park, Y. Fuel Process. Technol. 2003, 83, 219–
234.
(4) Watanabe, K.; Yamagiwa, N.; Torisawa, Y. Org. Process Res. DeV.
2007, 11, 251–258.
(5) Williamson, A. Justus Liebigs Ann. Chem. 1851, 77, 37–49.
(6) Reed, C.; Gaskell, B.; Banger, K.; Lock, E. Arch. Toxicol. 1995, 70,
51–56.
(7) Schwartz, M. D.; Obamwonyi, A. O.; Thomas, J. D.; Moorhead, J. F.;
Morgan, B. W. Am. J. Ind. Med. 2005, 47, 550–556.
(8) Ono, Y. Appl. Catal., A 1997, 155, 133–166.
(9) Selva, M. Pure Appl. Chem. 2007, 79, 1855–1867.
(10) Selva, M.; Perosa, A. Green Chem. 2008, 10, 457–464.
(11) Tundo, P.; Perosa, A. Chem. Rec. 2002, 2, 13–23.
(12) Selva, M.; Tundo, P. J. Org. Chem. 2006, 71, 1464–1470.
(13) Tundo, P.; Selva, M. Acc. Chem. Res. 2002, 35, 706–716.
(14) Fu, Y.; Baba, T.; Ono, Y. Appl. Catal., A 1998, 166, 419–424.
(15) Shaikh, A. A. G.; Sivaram, S. Chem. ReV. 1996, 96, 951–976.
(16) Parrish, J. P.; Salvatore, R. N.; Jung, K. W. Tetrahedron 2000, 56,
8207–8237.
Here we evaluate our original process20 at several different
reaction conditions and show that a range of primary and
(17) Selva, M.; Militello, E.; Fabris, M. Green Chem. 2008, 10, 73–79.
(18) Tundo, P.; Rossi, L.; Loris, A. J. Org. Chem. 2005, 70, 2219–2224.
(19) Tundo, P.; Memoli, S.; Herault, D.; Hill, K. Green Chem. 2004, 6,
609–612.
(20) Gooden, P. N.; Bourne, R. A.; Parrott, A. J.; Bevinakatti, H. S.; Irvine,
D. J.; Poliakoff, M. Org. Process Res. DeV. 2010, 14, 411–416.
1420
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development
10.1021/op1002243  2010 American Chemical Society
Published on Web 10/27/2010

Scheme 1. Reaction of 1-octanol (1) with DMC
secondary aliphatic alcohols can be successfully methylated in
a continuous process which operates at ambient pressure and
uses stoichiometric amounts of DMC, thus potentially saving
energy and material costs associated with our original process.
This concept is also extended to ethylation reactions by using
diethyl carbonate (DEC) as the alkylating agent. Additionally,
the effect of the pressure on the reaction outcome has been
investigated, and we show that an increase in pressure leads to
a significant decrease in yield; however, this can be overcome
by using scCO2 as the solvent.
Figure 1. Plot showing the effect of temperature on the reaction
of 1 with DMC (1:1 molar ratio). Liquid reagents pumped at
0.2 mL/min, 10 cm3 tubular reactor packed with γ-alumina
catalyst, ambient pressure. Traces are labeled as follows: 0 1,
b 1b, 3 1c, 4 1d, ] 1e, O 1f.
Scheme 2. Alkylation of aliphatic alcohols with DMC and
DEC at ambient pressure
Results and Discussion
Reactions at Ambient Pressure. In all our experiments,
reactions were carried out in an automated continuous reactor
equipped with online gas liquid chromatography (GLC) analysis
as described in the Experimental Section. To determine if DMC
could be used as a methylating agent in stoichiometric amounts
at ambient pressure, the methylation of 1-octanol (1), Scheme
1, was chosen as a model reaction, as it is a well-understood
reaction which formed the basis of our previous studies.20
A solution of DMC and 1 (1:1 molar ratio of DMC:1) was
pumped at 0.2 mL/min at ambient pressure over a fixed bed of
γ-alumina, which we have already shown to be an excellent
catalyst for this reaction in scCO2.20 The temperature of the
reactor was gradually increased from 100 to 250 °C, and GLC
samples were taken periodically so that the effect of the
increasing temperature on the product composition could be
determined, as shown by Figure 1.
At low temperatures (<120 °C) 1 underwent transesterifi-
cation with DMC to form octyl carbonate (1c) which, as the
temperature was increased (>150 °C), underwent further trans-
esterification with 1 to generate dioctyl carbonate (1d). Small
amounts of the desired methyl ether, 1-methoxyoctane (1b),
were also formed at these low temperatures. As the temperature
of the reaction was increased (>160 °C) both overall conversion
of 1 and selectivity towards 1b increased markedly so that the
yield of 1b increased rapidly with increasing temperature,
reaching a maximum yield of >99% at around 220 °C. Above
this temperature unimolecular dehydration of 1 to form 1-octene
and its isomers (1e) occurs, lowering the yield of 1b. Small
amounts (<10%) of dioctyl ether (1f), which is formed from
the bimolecular dehydration of 1, were also detected at high
temperatures (>180 °C). Interestingly this product composition
versus temperature profile is very similar to those observed
previously when the reaction was performed at 100 bar with
the same organic feed rate but with an additional 1 mL/min
CO2.20 This result confirms that high yields of aliphatic methyl
ethers can be obtained when using DMC as the methylating
agent in stoichiometric quantities, at ambient pressure and with
no solvent. The catalyst lifetime was tested in a separate
experiment with the temperature held at 220 °C, and a solution
of 1 and DMC was pumped at 0.2 mL/min over a fixed bed of
γ-alumina at ambient pressure; the yield of 1b remained
constant at around 90% for a period of greater than 110 h.
Consequently, the scope of this ambient pressure route was
further investigated by attempting the alkylation of several
aliphatic alcohols with both DMC and DEC, Scheme 2.
In all experiments the substrate alcohol and the dialkyl
carbonate (DAC) were mixed to form a 1:1 molar solution
which was pumped at 0.2 mL/min over a fixed bed of γ-alumina
catalyst at ambient pressure. The temperature of the preheater
and reactor were initially set to 100 °C and were gradually
increased; GLC samples were taken periodically so that the
effect of temperature on the outcome of the reaction could be
determined. The results are summarized in Table 1.
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1421

Table 1. Results of alkylation experiments using DMC and
DEC at ambient pressurea
max. yield
(%) of
desired
ether
temp.
(°C) at
max. yield
alkylating
agent
substrates
1
DMC
DEC
DMC
DMC
DEC
DMC
DEC
DMC
DEC
99
94
87
90
87
49
41
54
42
222
211
221
219
221
174
193
177
206
2
3
4
5
a Reaction conditions - substrate solution pumped at 0.2 mL/min (1:1 molar
ratio of DMC:substrate), ambient pressure, fixed bed of γ-alumina catalyst (tubular
reactor, internal volume 10 mL), temperature ramped linearly at 0.2 °C/min.
Figure 3. Plot showing the effect of mole fraction of DMC on
the composition of products from the reaction of 4 with DMC.
Total organic flow rate 0.2 mL/min (composition varied by using
two programmable HPLC pumps), 10 cm3 tubular reactor
packed with γ-alumina catalyst, ambient pressure, 170 °C.
Traces are labeled as follows: 0 4, b 4b, ] 4e.
to favor the methyl ether, 2-methoxyoctane (4b); however, the
dehydration reaction to form octenes (4e, same as 1e) occurs
at a much lower temperature (>160 °C) compared to that for 1.
The formation of octenes at lower temperatures limits the
maximum yield of the methyl ether to only ∼50%. 2-Pentanol
(5) reacts in a manner similar to that of 4 as shown in Table 1
and the temperature-composition plot in the Supporting
Information (ESI-2).
The effect of temperature on product composition for
reactions of 1 and 3-5 with DEC was very similar to that
observed for the reactions with DMC. The ethyl alkyl carbonate
was formed at low temperatures, with the target ethyl ether
formed as the temperature was increased, until a maximum yield
was reached. After which the formation of alkenes dominated
as the temperature was increased further, thus reducing the ether
yield. DEC generally afforded a slightly lower maximum yield
of the target ether than did DMC; the full temperature-compo-
sition plots for the reactions of 1 and 3-5 with DEC can be
found in the Supporting Information (ESI-3).
Effect of DMC Concentration. To determine if higher
yields of methyl ethers of secondary alcohols could be achieved
by using an excess of DMC, the reaction of 4 with DMC was
conducted at 170 °C and ambient pressure and the molar
fraction of DMC was changed gradually. This was achieved
by using two HPLC pumps, one feeding a solution of 4 and
DMC (1:1 molar ratio) and the other feeding pure DMC. The
pump flow rates were then programmed to change over time
so that the molar ratio of DMC to 4 could be gradually
decreased from 12.5:1 (0.93 mol fraction DMC) to 1:1 (0.5
mol fraction DMC) whilst maintaining the total flow of reagents
at a constant 0.2 mL/min. The results are shown in Figure 3,
which shows that there is a maximum in the yield of 4b of
87% at DMC mol fraction ca. 0.8 (4.2:1 DMC to 4 molar ratio).
Increasing the DMC composition beyond this point enhances
the dehydration reaction to form 4e, at the expense of the
methylation reaction. To investigate this effect further, a series
Figure 2. Plot showing the effect of temperature on the reaction
of 4 with DMC (1:1 molar ratio). Liquid reagents pumped at
0.2 mL/min, 10 cm3 tubular reactor packed with γ-alumina
catalyst, ambient pressure. Traces are labeled as follows: 0 4,
b 4b, 3 4c, ] 4e.
Table 1 shows that high yields of methyl ethers could be
obtained when the primary aliphatic alcohols, 1, 1-pentanol (2),
and 1-decanol (3) were used as substrates. The reaction of 1
with DMC has been discussed above and is shown in more
detail in Figure 1. The reaction of the other primary aliphatic
alcohols (2, 3) with DMC proceeds in a manner similar to that of
1; at low temperatures (<130 °C) the main product is the methyl
alkyl carbonate with small amounts of methyl ether being formed.
As the temperature increases, the selectivity of the reaction switches
to favor the methyl ether, reaching a maximum yield of ∼90% at
∼220 °C. At higher temperatures (>250 °C) dehydration of the
alcohol to form the corresponding alkene and its isomers becomes
the dominant process, and the yield of the methyl ether decreases.
The full temperature-composition plots can be found in the
Supporting Information (ESI-1).
Figure 2 shows the that the secondary alcohol, 2-octanol (4)
reacts with DMC in a manner similar to that of its primary
analogue, 1, and forms the methyl carbonate, methyl octan-2-
yl carbonate (4c), at low temperatures (<120 °C). As the
temperature is increased, the selectivity of the reaction switches
1422
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development

Table 2. Effect of DMC:4 molar ratios on the product
composition versus temperaturea
DMC:4 molar ratio
1:1
38
5:1
52
76
1
82
172
10:1
78
yield (%) of 4b at 150 °C
yield (%) of 4b at 175 °C
yield (%) of 4b at 200 °C
max. yield (%) of 4b
45
62
34
60
175
<1
94
157
temp. (°C) at max. yield
a Reaction conditions
- substrate solution pumped at 0.2 mL/min, ambient
pressure, fixed bed of alumina catalyst (tubular reactor, internal volume 10 mL),
temperature ramped linearly at 0.2 °C/min.
of temperature ramp experiments similar to those conducted
for Table 1 were undertaken but with various DMC:4 molar
ratios. The results are summarized in Table 2 and full
temperature-composition plots can be found in the Supporting
Information (ESI-4).
Table 2 shows that increasing the amount of DMC also
increased the overall yield of 4b and lowered the temperature
needed to achieve this yield. However increasing the amount
of DMC also lowers the temperature at which formation of 4e
occurs and hence reduces the temperature window over which
the reaction is selective towards methylation, reinforcing the
trend shown in Figure 3. A possible reason for this trend is
that the reaction temperature (170 °C) is lower than the boiling
point of 2-octanol (181 °C); thus, increasing the proportion of
DMC in the feed may also increase the amount of 4 which is
in the gas phase which might enhance the dehydration of 4 to
4e. Another possible explanation is that by increasing the DMC
loading (i.e., reducing the concentration of 4 in the feed) one
reduces the amount of time (and hence length of the catalyst
bed) required in the reactor to convert 4 fully to 4b. Thus, if
4b is not stable under the reaction conditions, it might undergo
cracking as it passes through the remainder of the catalytic bed
to produce 4e, as shown by the experiments described in the
next section. This might explain why there is a sharp increase
in the yield of 4e shortly after the DMC concentration exceeds
the loading required for full conversion of 4, as shown in Figure
3.
Stability of the Ether Products. Work previously carried
out at Nottingham has shown that ethers are not stable at high
temperature over solid acids.21 Gao et al.22-24 have also observed
that 1-alkenes can be produced by cracking 1-ethers over
γ-alumina, including the formation of 1-octene from 1b.
Therefore, to determine whether this reaction could be occurring
during the methylation of 1 and 4, temperature ramp experi-
ments were performed using 1, 4, 1b, and 4b as starting
materials in the absence of DMC, and the variation in yield of
1e/4e with temperature is shown in Figure 4.
Figure 4. Plot showing the effect of reactor temperature on
the total yield of octenes (1e/4e) from different starting materi-
als. Organic substrates pumped at 0.2 mL/min, 10 cm3 tubular
reactor packed with γ-alumina catalyst, and ambient pressure.
Starting materials are as follows: 0 1, 9 1b, O 4, b 4b.
Scheme 3. Possible routes to alkenes during the alkylation
of aliphatic alcohols with DACs
suggestion that the reason why 4e yield increases with DMC
composition as shown in Figure 3 is due to the cracking of the
ether product 4b.
The reason for the large offset in yield of 1e from 1 and 1b
at temperatures between 225 and 275 °C is the formation of 1f
Via bimolecular dehydration of 1. The yield of 1f during the
dehydration of 1 reaches a maximum of around 40% at 250
°C, whereas the yield of the branched dioctyl ether (4f) never
exceeded 1% during the dehydration of 4. This is presumably
due to the secondary centre in 4f being less stable than the
primary centre in 1f, and 4f itself undergoes cracking to 4e.
This formation of linear ethers also contributes to the upper
temperature limit for selective methylation with DMC. Thus,
the upper temperature for effective alkylation with DACs is
probably determined by three effects; the dehydration of the
starting materials, the cracking of the ether products, and the
decomposition of the DAC. In addition, alkenes may possibly
be formed from the decomposition of methyl (or ethyl) alkyl
carbonates and DACs Via pyrolysis,19,25,26 as illustrated in
Scheme 3.
Figure 4 shows that 1e is formed from both 1 and 1b at
around the same temperature (200 °C); this is also the case for
4 and 4b although the temperature is lower (160 °C). These
temperatures are similar to those observed for octene formation
during the methylation reactions of 1 and 4, supporting the
(21) Walsh, B.; Hyde, J. R.; Licence, P.; Poliakoff, M. Green Chem. 2005,
7, 456–463.
(22) Gao, X.; Johnston, S. D. R.; Krzywicki, A.; Krzywickl, A.; Krzywicki,
A. Z. CA2524940-A1, 30-April-2007.
(23) Gao, X.; Krzywicki, A.; Johnston, S. D. R.; Kalivoda, A.; Krzywicki,
A. Z. CA2538206-A1, 28-August-2007.
(24) Gao, X.; Johnston, S. D. R. CA2578494-A1, 14-April-2008.
(25) Gordon, A. S.; Norris, W. P. J. Org. Chem. 1965, 69, 3013–3017.
(26) Notario, R.; Quijano, J.; Sa´nchez, C.; Ve´lez, E. J. Phys. Org. Chem.
2005, 18, 134–141.
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1423

withdraws random samples from a multiphasic mixture.27 As
the pressure was increased above 125 bar, there was a gradual
decrease in the yield of 1b from 100 to 85%, and again it was
observed that the decrease in yield with increasing pressure was
due to a decrease in conversion of 1 rather than to a change in
selectivity of the reaction. Thus, the addition of CO2 clearly
has a marked effect on the outcome of the reaction when
performed at high pressure and increases the useful pressure
range over which effective methylation can be conducted by
∼120 bar.
To determine whether this effect was simply due to dilution
of the reactants by a solvent, the experiment was repeated with
n-heptane as the solvent instead of CO2. From Figure 5, it can
be seen that the yield versus pressure trend for n-heptane was
very similar to that of the experiment with no solvent. This
clearly shows that the role of CO2 is not simply that of a diluent.
To determine whether this large decrease in yield with pressure
was due to the phase change from vapour to liquid, which
Figure 5. Plot showing the effect of pressure on the yield of
1b. Solutions of DMC and 1 (2:1 molar ratio) pumped at 0.2
mL/min, 10 cm3 tubular reactor packed with γ-alumina catalyst,
temperature set to 220 °C. Additional solvent key: OCO2
pumped at 1.0 mL/min (pump head at -10 °C, 56 bar), 0
n-heptane pumped at 1.0 mL/min, 9 n-pentane pumped at 1.0
mL/min, b no additional solvent.
occurs at around 13.6 bar at 220 °C for pure n-heptane,28
a
similar experiment was conducted but using n-pentane. At 220
°C n-pentane is above its critical temperature,28 and so it should
not cross through a vapor-liquid phase transition as the pressure
is increased in this experiment. Figure 5 shows that the results
were very similar to those observed for n-heptane. Thus, the
effect of CO2 is probably unconnected with phase changes.
As these experiments are designed to have only a small
excess of DMC, if some of the DMC decomposes to form DME
instead of reacting with the alcohol, it then becomes impossible
for the alcohol to form the methyl ether as there is insufficient
methylating agent remaining for 100% conversion. Therefore,
any process that accelerates the rate at which DMC is
decomposed to DME will reduce the amount of DMC available
to react with the alcohol, and hence, the conversion of the
alcohol and the yield of the methyl ether will also be reduced.
Thus, to explain why the yield of 1b is reduced at high pressure
without the use of CO2, we propose two possible scenarios either
of which could explain why the rate of decomposition of DMC
is affected compared to the rate of reaction between substrate
and DMC. First, it might be the case that, CO2 could be reducing
the rate at which DMC is decomposing to DME, possibly by
blocking the catalytic sites which are responsible for this
reaction. This would result in a higher proportion of DMC
remaining to react with the alcohol to form the methyl ether in
higher yield when CO2 was the solvent. As the other solvents
used may not be able to block or hinder access of DMC to the
catalyst surface, they would have little effect on the rate at which
DMC decomposes to DME; hence, this could explain why,
when the other solvents are used at high pressure, the yield of
the ether is lower.
Effect of Pressure and Solvent Medium. Our previous
process20 used high pressure (100 bar) with scCO2 as the
solvent, and we obtained similar yields of methyl ethers at very
similar temperatures, suggesting that pressure has little effect
on the outcome of the reaction. To determine whether this was
the case in the absence of CO2, a solution of 1 and DMC (1:2
molar ratio) was pumped at 0.2 mL/min over a fixed bed of
γ-alumina catalyst at 220 °C, and the pressure was gradually
increased from 5 to 250 bar.
From Figure 5 it can be seen that when no additional solvent
was used, pressure did in fact significantly influence the outcome
of the reaction. At 5 bar the yield of 1b is close to 100% but it
rapidly decreased with increasing pressure to be only 50% at
50 bar; and only 40% at 240 bar. Additionally this decrease in
yield was found to be due to a decrease in the conversion of 1
and not due to the formation of byproducts. Interestingly the
conversion of DMC is unaffected by changing the pressure,
but instead of reacting with 1, it undergoes a thermal catalytic
reaction to form DME. The decrease in conversion of 1 with
increasing pressure is unexpected because increasing the pres-
sure also increases the overall density of the mixture in the
reactor, increasing concentration and residence time which could
be expected to increase conversion.
Figure 5 shows that the use of scCO2 in our earlier work20
enhanced the yield of the methyl ether with ∼86% yield at 100
bar with scCO2 compared with only 46% at 100 bar with no
solvent. In order to investigate this effect further the pressure
ramp experiment was repeated but with an additional stream
of CO2 pumped into the reactor at 1 mL/min, all other conditions
were the same as those for the experiment without solvent, and
these results are also included in Figure 5. As the pressure was
increased from 5 to 125 bar, the yield of 1b remained around
100% apart from a “noisy” region between 55 and 100 bar
which is probably due to the way in which the sample loop
Alternatively if, at high pressure, the rate at which the alcohol
or methyl ether adsorbs or desorbs from the catalyst surface is
slow, then there will be a lower proportion of the alcohol at
the surface for DMC to react with and hence DMC will instead
decompose to DME. Therefore it is also possible that CO2 may
(27) Licence, P.; Gray, W. K.; Sokolova, M.; Poliakoff, M. J. Am. Chem.
Soc. 2005, 127, 293–298.
(28) Lemmon, E.; McLinden, M.; Friend, D. NIST Chemistry WebBook,
NIST Standard Reference Database Number 69; National Institute of
Standards and Technology: Gaithersburg, MD, 2010.
1424
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development

also affect the rate at which the alcohol/methyl ether adsorbs/
desorbs from the catalyst surface and hence increase the rate at
which DMC reacts with the alcohol and thus increases the yield
of methyl ether.
It should be stressed that neither of these scenarios has yet
been tested experimentally and there is no clear evidence to
support or refute either. Thus, it is possible that there is an
entirely different reason for the trends observed in Figure 5 and
we are currently conducting further work to determine the exact
role of CO2. However, the fact remains that our results are
consistent with any process that accelerates the rate of decom-
position of DMC relative to its rate of reaction with the alcohol.
Conclusion
We have reported a high-yielding, greener route to the
continuous synthesis of ethers Via the reaction of aliphatic
alcohols with DACs over a solid acid catalyst. The reaction
can be conducted at ambient pressure with no solvent, giving
high yields of several primary methyl and ethyl aliphatic ethers
whilst requiring only stoichiometric amounts of DMC or DEC.
This methodology improves upon previous routes to alkylation
of aliphatic alcohols with DACs, which have often been
conducted in batch mode with large excesses of the alkylating
agent.
Good yields of secondary ethers were also formed using
stoichiometric amounts of DACs and were increased further
by using a slight excess of DAC. The upper temperature for
effective alkylation was restricted by the formation of dehy-
drated products; these are produced by both the dehydration of
the starting alcohols and by the cracking of the ether products.
The use of pressures above 5 bar has negative impact upon
conversion of aliphatic alcohols when no solvent was used or
when n-heptane or n-pentane was used as solvent. However,
when CO2 was used as the solvent, high yields of the methyl
ether could be obtained at pressures up to 125 bar and
reasonable yields even at 250 bar. This means that, by using
CO2, this reaction procedure has the flexibility to be integrated
more easily into an existing process which requires the use of
high pressure.
Figure 6. Schematic of the automated supercritical flow reactor
developed at the University of Nottingham. The reactant
solutions are pumped by two HPLC pumps (OP and OP2), and
CO2 is pumped by a chilled CO2 pump (CP), which is connected
to a back pressure regulator (BPR) which allows pressures
below that of the cylinder to be investigated. The N2 flow is
regulated by a mass flow controller (M), the streams are then
mixed in a heated mixing unit (PH) and passed over a heated
fixed catalyst bed (R). The system pressure is controlled by
another back pressure regulator (BPR-2), and the products are
collected once the pressure is released in a collection vessel (W).
Whilst the product stream is flowing from the exit of R to the
BPR, the high pressure sample loop (HPSL) can remove
aliquots of the product mixture and introduce them directly
into the carrier gas stream of the gas liquid chromatograph
(GC) for analysis.
2216). There were two programmable BPRs (Jasco BP-
1580-81) present in the system the first was placed in between
the outlet of the CO2 pump and the inlet of the preheater and
was permanently set to 70 bar, allowing pressures below that
of the CO2 cylinder to be investigated. The second BPR was
placed at the outlet of the system to control the system pressure.
The setup uses an online sample loop (VICI EPCI4W.06, 0.06
mL, micro-electronically actuated, 250 ms switching time)
which is positioned upstream of the expansion system to allow
direct sampling of the product stream Via GLC analysis using
a Shimadzu GC-14B equipped with a CP-Sil 8 CB column
(Varian, 30 m, 0.25 mm I.D., 0.25 µm F.T.) using H2 carrier
gas and a flame ionisation detector. This sampling provides a
method to obtain an instantaneous ‘snapshot’ of the reaction
products at the current reactor conditions. The identity of
compounds was confirmed by injection of authentic samples
and by GLC-MS using a Thermo-Finnigan Polaris-Q instrument
fitted with a RTX-1MS column (30 m, 0.25 mm ID, 0.25 µm
FT). Reaction conditions are monitored using a PicoLog TC-
08 connected to thermocouples positioned internally and
Experimental Section
CAUTION! The experiments described in this paper inVolVe
the use of relatiVely high pressures and require equipment with
the appropriate pressure rating.
Figure 6 shows a simplified schematic of the automated
supercritical flow reactor. This system is a modified version of
the one described previously,20,21 and only a brief overview of
the equipment will be provided here. A programmable HPLC
CO2 pump (Jasco PU-1580-CO2), two programmable liquid
HPLC pumps (Jasco PU-980) and a nitrogen mass flow
controller (Brooks Instrument 5850TR) were connected to a
10 mL 316 stainless steel tube (156 mm ×12 mm OD) packed
with sand. This first tube acts as both mixer and preheater and
directly feeds the fixed bed reactor, which consists of a 10 mL
316 stainless steel tube (156 mm ×12 mm OD) packed with
catalyst. The preheater and reactor tubes are heated by cartridge
heaters within an aluminium heating block and the temperature
controlled Via a programmable heating controller (Eurotherm
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1425

externally on the reactor and to the pressure transducers located
within the HPLC pumps and the BPRs.
ics), 1-decanol 99% (Acros Organics), N2 99.998% (BOC), and
CO2 99.9% (Cryoservice) were used without further purification.
1-Methoxyoctane 96% and 2-methoxyoctane 92% were ob-
tained by fractional distillation at reduced pressure of the
products collected at the end of methylation experiments of
1-octanol and 2-octanol, respectively; the ethers were obtained
as a colourless liquid, and the purity was analysed by GLC.
In a typical experiment, the catalyst was loaded in the reactor
and sealed into the apparatus. The initial conditions for the
experiment were set at the pumps, BPR, and temperature
controllers, and the system was allowed to equilibrate over at
least 30 min. The experimental parameters were then pro-
grammed into pumps, BPR, temperature controllers, and GLC
such that samples are injected into the GLC periodically as one
or more reaction parameters (e.g., temperature and pressure)
were varied.
For all the reactions conducted at ambient pressure, BPR-2
was removed from the system and the outlet from HPSL directly
fed into W. The back pressure required to pump the reagents
through the pipe work was measured at the pumps and never
exceeded 2 bar.
Acknowledgment
We thank Croda Europe Ltd., Thomas Swan & Co. Ltd.,
the EPSRC and the DICE initiative for support. We also thank
Dr. S. K. Ross for helpful discussion and Mr. M. Dellar, Mr.
M. Guyler, Mr. R. Wilson, and Mr. P. Fields for their technical
support at the University of Nottingham.
A commercially available acidic PURALOX NWa-155
γ-alumina (Supplied by Sasol; powder, 150 m2/g surface area
and containing trace impurities of SiO2, Fe2O3, and Na2O) was
used as a catalyst in all experiments. Dimethyl carbonate 99%
(Acros Organics), diethyl carbonate 99% (Acros Organics),
1-pentanol 99% (Acros Organics), 2-pentanol 99% (Alfa Aesar),
1-octanol 99% (Sigma-Aldrich), 2-octanol 97% (Acros Organ-
Supporting Information Available
This material is available free of charge via the Internet at
http://pubs.acs.org.
Received for review August 13, 2010.
OP1002243
1426
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development