1116
Chemistry Letters 2002
Selective Hydrogenation of Aromatic Compounds Containing Epoxy Group over Rh/Graphite
Yoshinori Haraà and Hiroko Inagaki
Mitsubishi Chemical Corporation, MCC-Group Science & Technology Research Center,
1000, Kamoshida-cho, Aobaku, Yokohama 227-8562
(Received August 22, 2002; CL-020712)
Catalytic hydrogenation of aromatic compounds containing
epoxy group to alicylic compounds has been investigated. Rh
supported on graphite with high surface area exhibited superior
performance to other supported catalysts for the selective
reduction of the aromatic group while retaining epoxy group.
to find out the selective hydrogenation catalysts which can
exclusively hydrogenate the aromatic ring while retaining epoxy
group. There have been scare literatures referring to the reaction
of kinds except a few patents disclosed in which RuO2 and Rh/
active carbon were employed as catalysts.5{7 Table 1 shows the
effect of various kinds of hydrogenation catalyst.
Controlling selectivity in the reduction of the organic
compounds which have both reduciable groups is one of the
most important key technologies in academia as well as in
industry. For instance, many efforts have been made for the
hydrogenation of ꢀ; ꢁ-unsaturated aldehydes or ketones to give
ꢀ; ꢁ-unsaturated alcohols.1{3 We describe herein the effect of
catalyst metal and supports on the catalytic property for the
hydrogenation of a bisphenol A diglycidyl ether which has widely
used for the epoxy cured resins. Catalytic hydrogenation of
aromatic compounds to alicyclic ones is now easily accomplished
over many platinum group catalysts under relatively mild
condition.4 However, highly selective catalytic hydrogenation
of an aromatic ring in the vicinity of an epoxy group is a unique
problem because the epoxy groups are vulnerable to acid and base
and can easily undergo hydrogenolysis. It is of great importance
to establish the technology for the transformation taking into
account the industrialapplications. Epoxy resins prepared from a
diglycidyl ether of hydrogenated bisphenol A are characterized to
be superior in weatherability and electric property to those from
the aromatic counterparts. There have been known two ways to
produce diglycidyl ether of hydrogenated bisphenol A as depicted
in Scheme 1. The first is reacting epichlorohydrin with
hydrogenated bisphenolA, whereas another is directyl hydro-
genating bisphenolA deglycidylethers.
Table 1. Various catalysts for hydrogenation of bisphenol A
diglycidyl ether
Conversion of
aromatic group/%
Residualepoxy group
/%
Catalyst
RuO2
Ru/active carbon
Ru/SiO2
Pd/active carbon
Rh/Al2O3
Rh/SiO2
Rh/active carbon
Rh/graphitea
Rh/graphiteb
Rh/graphitec
100
30
0
0
35
0
100
100
70
9
92
99
99
44
50
98
77
92
93
100
Reaction conditions: bisphenolA digyl cidylether 5 g, THF
5 g, catalyst 0.25 g, H2 15 MPa (at room temp.), 50 ꢁC, 3 h.
Metal loading of supported catalysts were 5 wt% and all the
catalysts used except Rh/graphite were commercially avail-
able. Rh/graphite was prepared by wet impregnation of
graphite with a aqueous solution of RhCl3 followed by H2
a
gas reduction at 300 ꢁC for 2 h. Lonza, HSAG100, surface
area 130 m2/g, bLonza, HSAG300, surface area 280 m2/g,
cKishida F08126D, surface area 3 m2/g.
Catalytic property significantly changed by the catalyst
metals as well as their supports. Ru and Rh have tendency to
hydrogenate aromatic ring, whereas Pd predominantly attacks the
epoxy group and Rh is much more active than Ru. Support also
exerts an influence on the reaction feature. SiO2 was quite inert in
both Ru and Rh catalyst and Al2O3 is liable to decompose epoxy
group possibly due to its acidic character. Active carbon support
gave relatively high catalytic activity and selectivity, which is
probably due to high surface area and relatively neutral property
(acidity-basicity) on the surface. Therefore, we have focused on a
series of carbonaceous supports as suitable for the reaction.
Graphite with a certain surface area (SA) of 100 m2/g turned
out to exhibit extraordinary catalytic behavior. No significant
H2
HO
OH
OH
HO
O
Cl
O
O
O
Cl
O
O
H2
O
O
O
O
ꢀ
difference in Rh particle size (<50 A) between HSAG100 and
HSAG 300 was observed from XRD though Rh particle supported
2
Scheme 1.
The product obtained from the former process contains a
large amount of chlorines and it is unsuitable for electric uses. On
the other hand, the problems in the latter process to be solved are
ꢀ
on graphite with SA of 3 m /g was surely large (ꢀ200 A). In order
to accurately compare the catalytic performance between Rh/
active carbon and Rh/graphite, the hydrogenation were carried
out monitoring H2 uptake at the constant pressure as listed in
Copyright Ó 2002 The ChemicalSociety of Japan