Friedel-Crafts alkylation in supercritical fluids: continuous, selective and clean

Article abstract of DOI:10.1039/a707149c

Continuous Friedel-Crafts alkylation of mesitylene C6H3Me3, and anisole, C6H5OMe, with propene or propan-2-ol has been carried out in supercritical propene or CO2 using a heterogeneous polysiloxane-supported solid acid Deloxan catalyst in a small fixed bed reactor (10 ml volume); 100percent selectivity for mono-alkylated products with 50percent conversion could be obtained by adjusting the reaction parameters, e.g. temperature, pressure, flow rates, etc.

Full text of DOI:10.1039/a707149c

View Article Online / Journal Homepage / Table of Contents for this issue
Friedel–Crafts alkylation in supercritical fluids: continuous, selective and clean
a
a
b
a
Martin G. Hitzler, Fiona R. Smail, Stephen K. Ross and Martyn Poliakoff *†
a
Department of Chemistry, University of Nottingham, Nottingham, UK NG7 2RD
Thomas Swan & Co. Ltd, Crookhall, Consett, Co. Durham, UK DH8 7ND
b
Continuous Friedel–Crafts alkylation of mesitylene,
Me , and anisole, C OMe, with propene or propan-
acid catalyst based on the same Deloxan® support.‡ In our
process,§ the organic solvent is replaced by a supercritical fluid,
e.g. scCO . The separation of the product occurs simply by
expanding the fluid to atmospheric pressure, so the crude
product can be analysed without further work-up. Although
reaction temperatures are high, because of short residence
times, our process maintains high product selectivity.
C
6
H
3
3
6 5
H
2
-ol has been carried out in supercritical propene or CO
2
2
using a heterogeneous polysiloxane-supported solid acid
®
Deloxan catalyst in a small fixed bed reactor (10 ml
volume); 100% selectivity for mono-alkylated products with
5
0% conversion could be obtained by adjusting the reaction
parameters, e.g. temperature, pressure, flow rates, etc.
Our initial reactions involved mesitylene 1 (C
propene 2, which acts both as alkylating agent and as
supercritical solvent [scPropene, T = 91.9 °C, p = 46.0 bar
6 3 3
H Me ) with
Numerous chemical syntheses involve Friedel–Crafts or Frie-
del–Crafts-type alkylation reactions, both on laboratory and
industrial scales. Friedel–Crafts chemistry is therefore of major
importance and research in this field is of continuing chemical
interest.1 Frequently, the reaction is used to add only a single
alkyl group to the substrate, but addition of the first alkyl group
activates the substrate so that subsequent alkylation is made
easier. Thus, conditions have to be controlled precisely, often
with long reaction times and low temperatures, to minimise the
formation of poly-alkylated by-products.
c
c
(10 bar = 1 MPa)]. This reaction was chosen because the mono-
3, di- 4 and tri-alkylated product 5 each have only one possible
isomer (Scheme 1), which simplifies product analysis.¶
Table 1 shows that Friedel–Crafts chemistry can indeed be
carried out continuously under these conditions. The selectivity
is reasonable for 1-isopropyl-2,4,6-trimethylbenzene 3 but
significant amounts of 4 and dimers of propene 2 are formed.
Table 1 also shows that the yield of 3 cannot easily be improved
by varying the reaction conditions.
–5
Friedel–Crafts reactions are usually carried out batch-wise in
a large excess of the substrate or in an organic solvent.
Generally, the reaction is carried out in the presence of Lewis
By contrast, selectivity for the formation of 3 is much higher
in the alkylation of 1 with propan-2-ol, 6, in supercritical CO
(scCO , T = 31.1 °C, p = 73.8 bar) using the same solid acid
2
2
c
c
®
acid catalysts (e.g. AlCl
HF, H SO ). Relatively high concentrations of catalyst are
needed; often the amounts are near stoichiometric, especially
when H O is generated. These problems make most Friedel–
3
, BF
3
4
, TiCl ) or strong protic acids (e.g.
Deloxan catalyst. The reaction forms water but this is not a
serious problem; the organic and water layers are easily
2
4
separated after the scCO
2
has been vented. Initially, the reaction
2
was carried out with a three-fold excess of 6 over 1 (total flow
2
1
Crafts processes inherently dirty. Therefore, the development of
environmetally more benign Friedel–Crafts processes is a high
priority. Solid acid catalysts based on clays, zeolites or even
graphite6 could be the answer to cleaner Friedel–Crafts
chemistry, but until recently rapid catalyst deactivation has
rate of 1 + 6: 0.50 ml min ) at 220 bar and Tcat = 200 °C with
2
1
2
a gaseous CO flow rate of 0.43 l min . Analysis of the organic
layer showed that the conversion of 1 was 46% with 3 as the
major product (40% yield). As side-products, 4 was found in 5%
yield along with traces of diisopropyl ether (1%). Table 2
summarises the results obtained by varying the reaction
conditions. The yield of 3 can be improved by optimising the
temperature and raising the pressure. When 1 is in excess, 3 is
the only detectable organic product.
7
prevented the development of a commercial process. Not only
can solid catalysts be easily separated from liquid products, they
also offer the possibility of carrying out Friedel–Crafts reactions
in continuous rather than batch reactors.8 Continuous flow
reactors can be smaller than the batch reactors needed to
Previous attempts to run continuous Friedel–Crafts alkyla-
tion have frequently been thwarted by poor catalyst lifetimes
9
generate comparable amounts of product. A reactor of smaller
®
size usually increases safety and reduces capital cost.
but the lifetime of the Deloxan catalyst is relatively long. In
10
Very recently, we described how the supercritical hydro-
genation of a wide a range of organic compounds can be carried
out extremely efficiently in small-scale continuous flow
reactors (5 or 10 ml) using Pd and Pt heterogeneous catalysts
fact, when the reaction was started using a fresh sample of ASP
®
I/7 acid Deloxan catalyst, the activity actually increased after
an induction period of ca. 100 min and was maintained for at
least a further 15 h operation. We initially attributed this
induction to some activating effect of the H O formed in the
2
®
supported on polysiloxane (Deloxan , Degussa AG). Here we
show how the same reactor can be used to carry out continuous
reaction. However, the yields of 3 were dramatically lower
Friedel–Crafts alkylation using a commercially available solid
Table 1 Alkylation of mesitylene 1 with propene 2 at 200 bar pressurea
Yield (%)
Flow
Flow
+
b
rate of 1/ rate of
Dimers
Trimers
1
T
cat/°C ml min2 propene
1
3
4
5
of 2
of 2
1
2
acid–Deloxan®
scPropene
1
1
1
60
60
80
0.30
0.30
0.30
0.65
0.43
0.43
57 25
47 25 10
32 27 14
6
0
0
2
12
14
16
0
4
9
+
+
a
The reaction was carried out in a heated reactor of 10 ml volume
containing 9 ml of the solid acid Deloxan® catalyst (ASP I/7, particle size:
.1–0.4 mm). No reaction was observed when the catalyst was replaced by
0
3
4
5
b
21
Nafion as the solid acid. Flow rate of propene in 1 min measured at 1
atm and 20 °C, as determined by bubble flow meter.
Scheme 1
Chem. Commun., 1998
359

View Article Online
a
Table 2 Investigation of (I) temperature, (II) pressure and (III) molar ratio on yield of 3
Flow rate
Production
rate of
3/g min
Molar ratio of 1 + 6/
Foundb
3 (%)
Yieldc
of 3 (%)
21
21
of 1:6
g min
Tcat/°C
2
p CO /bar
I
2.0:1.0
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
1.51
0.60
0.52
200
230
250
270
300
250
250
250
250
250
250
200
200
200
200
200
200
150
100
150
150
150
15.0
21.0
21.0
15.0
7.5
21.0
18.5
16.0
19.0d
18.5
10.0
30.0
42.0
42.0
30.0
15.0
42.0
37.0
32.0
19.0d
37.0
50.0
0.09
0.13
0.13
0.09
0.05
0.13
0.11
0.10
0.29
0.11
0.05
2.0:1.0
2.0:1.0
2.0:1.0
2.0:1.0
II
2.0:1.0
2
2
.0:1.0
.0:1.0
III
1.0:5.0
2
5
.0:1.0
.0:1.0
a
(0.65 l min21) with 3 as the only detectable product. b Product analysis was by H NMR spectroscopy (CDCl
1
). c Based
All reactions carried out in scCO
2
3
d
on the amount of 6 in the initial reaction mixture. Yield based on 1. In addition, 3% of 4 and traces of side-products were found.
OMe
thank M. W. George, M. Guyler, S. M. Howdle, K.-H. Pickel,
K. Stanley, T. Tacke and S. Wieland for their help. M. P. thanks
the EPSRC/Royal Academy of Engineering for a Fellowship.
+
OH
7
6
Notes and References
acid–Deloxan®
scCO2, –H2O
†
‡
Web page: http://www.nottingham.ac.uk/supercritical/
Deloxan® ASP catalysts (ref. 11) are specially designed polysiloxane
OMe
OMe
OMe
based solid acids, formed by sol-gel condensation of alkyl sulfonic acid
functionalized organosilane monomers, e.g. HO SCH CH CH Si(OH)
which overcome the drawbacks of organic polymers by virtue of their inert
matrix material and excellent compatability with almost all organic
solvents. The sol-gel process yields products with a relatively narrow
particle size distribution, high porosity, large pore diameters ( > 20 nm) and
3
2
2
2
3
,
+
+
Pri
Pri2
Pri3
8
9
10
Scheme 2
2
21
high BET surface areas (300–600 m g ).
The substrate, reactant and supercritical fluid are brought together in a
when the catalyst was conditioned with water for 30 min prior
to use. When operated for maximum flow of 3 (see Table 2) the
catalyst can produce twice its volume of 3 per hour.
The supercritical reactor can also be applied sucessfully to
the continous alkylation of anisole 7 by scPropene 2 or by
§
heated mixer, passed through the hot reactor (10 ml volume; 9 mm i.d.,
length 152 mm) containing the catalyst, and then expanded to separate the
product from the fluid. The reactor is assembled from commercially
available units: scCO and scPropene pump PM 101 and Expansion Module
2
propan-2-ol 6 in scCO
2
(Scheme 2). As with the alkylation of 1,
PE 103 (all from NWA GmbH, L o¨ rrach, Germany), a high pressure mixer
(Medimix) and Gilson pumps 303, 305 (for substrate and reactant).
SAFETY NOTE: Reactions in supercritical fluids involve high pressures
and should only be carried out with appropriate equipment.
there is significant selectivity in favour of the mono-alkylated
product. Thus, 6 reacted with 7 at a molar ratio of 1.0:3.0 to
form isomers of 8 (30% yield) and 9 (only 6% yield).∑ No
further products could be detected by GC–MS. When 7 was
1
¶
Product analysis by GC–MS or H NMR spectroscopy.
∑
Pressure, 220 bar; catalyst temperature, 200 °C, flow rate of gaseous CO
2
2
1
alkylated at a flow rate of 0.2 ml min with scPropene (flow
2
1
21
of 0.2 l min ; total flow rate of 6 + 7, 1.0 ml min . The product analysis
is more complicated for the reactions of 7 than for those of 1 becasue the
products (8, 9 and 10) all have more than one possible isomer. Thus, GC–
MS gave three peaks for 8 (m/z 150) in the ratio of 13 :12:1, corresponding
to the ortho, para and meta isomers. Similarly, two peaks (m/z 192) were
found in the ratio 4:1 for the isomers of 9.
2
1
rate of gaseous propene: 0.43 l min ) under the same
conditions, the yield of 8 was 38% with the same isomer ratio as
was found with propan-2-ol.∑ However, the yield of 9 increased
to 18% (three isomers at a ratio of 5:2:1). In addition, 10 was
found in 5% yield (peak ratio 5:2 at a mass of 234) as well as
trace amounts (3%) of propene di- and tri-mers.
The supercritical fluid probably plays several roles in these
reactions. Compared to liquid phase reactions, the fluid reduces
mass transport restrictions at the surface of the catalyst.
Compared to a gas phase reaction, it increases the density of the
reaction medium and hence increases the residence time of the
substrate in a given size of reactor. This allows continuous
alkylation to be caried out on a reasonable scale in a small
reactor. Finally, it may reduce coking of the catalyst preventing
premature deactivation of the catalytic sites.
This investigation has shown that aromatic substrates can
undergo continuous and sustainable Friedel–Crafts alkylation in
supercritical fluid solution over solid acid heterogeneous
catalysts.Thereactionscanbecarriedoutinaflowreactor,which
differs only in the peripheral pipework from that used for the
supercritical hydrogenation of organic compounds.10 The
method of alkylation has features of potential importance for the
manufactureoffinechemicals.Itisselective,organicsolventsare
eliminated and a clean heterogeneous catalyst replaces a liquid–
based system. Thus, supercritical alkylation is a step closer to
ennvironmentally acceptable Friedel–Crafts chemistry.
1 G. A. Olah, Friedel–Crafts and Related Reactions, Wiley Interscience,
New York and London, 1963–1964, vol. I–IV; G. A. Olah, R. Krishna-
muri and G. K. Surya Prakash, in Comprehensive Organic Synthesis, ed.
B. M. Trost, Pergamon, Oxford, 1991, vol. 3, ch. 1.8.
2
J. H. Clark and D. J. Macquarrie, Org. Proc. Res. Develop., 1997, 1,
49.
1
3
4
R. Streekumar and R. Padmakumar, Synth. Commun., 1997, 27, 781.
M. Y. Croft, E. J. Murphy and R. J. Wells, Anal. Chem., 1994, 66,
4
459.
5
Y. Gao, Z. N. Zhu and W. K. Yuan, Prog. Nat. Sci., 1996, 6, 625.
6 M. Kodomari, Y. Suzuki and K. Yoshida, Chem. Commun., 1997,
1567.
7 L. Fan, I. Nakamura, S. Ishida and K. Fujimoto, Ind. Eng. Chem. Res.,
1
997, 36, 1458 and references cited therein.
8
9
J. S. Beck and W. O. Haag, Handbook of Heterogeneous Catalysis, ed.
G. Ertl, H. Kn o¨ zinger and J. Weitmap, Wiley-VCH, Weinheim, 1997,
vol. 5, p. 2123.
P. Tundo, Continuous Flow Methods in Organic Synthesis, Ellis
Horwood, 1991.
1
0 M. G. Hitzler and M. Poliakoff, Chem. Commun., 1997, 1667;
M. G. Hitzler, F. R. Smail, S. K. Ross and M. Poliakoff, Org. Proc. Res.
Develop., 1998, in the press.
11 S. Wieland, E. Auer, A. Freund, H. Lansink Rotgerink and P. Panster,
in Catalysis of Organic Reactions, ed. R. E. Malz, Jr., Marcel Dekker,
New York, 1996, pp. 277–286.
We thank Thomas Swan & Co Ltd for fully funding this work
and D. Campbell, S. K. Ross and J. C. Toler for their assistance.
We are grateful to Degussa AG for donating the catalysts. We
Received in Liverpool, UK, 1st October 1997; 7/07149C
360
Chem. Commun., 1998