An efficient cleavage of the aryl ether C-O bond in supercritical carbon dioxide-water

Article abstract of DOI:10.1039/c3cc41522h

A simple and highly efficient Rh/C catalyzed route for the cleavage of the C-O bond of aromatic ether at 80 °C in the presence of 0.5 MPa of H ₂ in the scCO₂-water medium is reported; CO₂ pressure and water play a key role under the tested conditions.

Full text of DOI:10.1039/c3cc41522h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
An efficient cleavage of the aryl ether C–O bond in
supercritical carbon dioxide–water†
Cite this: Chem. Commun., 2013,
49, 4567
Maya Chatterjee,* Takayuki Ishizaka, Akira Suzuki and Hajime Kawanami*
Received 28th February 2013,
Accepted 28th March 2013
DOI: 10.1039/c3cc41522h
www.rsc.org/chemcomm
A simple and highly efficient Rh/C catalyzed route for the cleavage efficiency for hydrogenolysis of the C–O bond. However, homogenous
of the C–O bond of aromatic ether at 80 8C in the presence of catalysts suffer from difficulties in product separation, reusability and
0.5 MPa of H2 in the scCO2–water medium is reported; CO2 are sometimes difficult to handle because of their air sensitivity.6
pressure and water play a key role under the tested conditions.
Thus, it was necessary to develop a heterogeneous catalyst for the
hydrogenolysis of the aryl C–O linkage, which is easy to handle, and
Lignocellulosic biomass comprises of cellulose, hemicellulose and advantageous for large-scale application. In a recent approach,
lignin and has a huge appeal for production of second generation 57 wt% of the Ni/SiO2 catalyst was used for the cleavage of the aryl
bio-fuels. Lignin, which takes up B40% of the lignocellulosic ether bond in aqueous media.7 The heterogeneous catalytic system
biomass, is a complex structure with strength and stability mainly required an excessive amount of catalyst or a longer reaction time.
due to the presence of various kinds of aryl ether bonds. Breaking of
In this work, we attempted to develop a simple and convenient
the aryl ether C–O bond is a challenging task as the C–O bond is method for the cleavage of the aryl C–O bond. Here, diphenyl ether
highly stable and typically difficult to activate.1 In the past few years, (DPE) was used as a model compound for hydrogenolysis over 5%
several attempts have been made to develop suitable catalysts or Rh/C as a heterogeneous catalyst in a supercritical carbon dioxide
processes for the cleavage of the aryl ether bond, and this might be (scCO2)–water system (Scheme 1). The use of a diaryl ether model
the best strategy for removing oxygenated groups from the aryl ring.2 compound represents a firm test of the catalyst’s effectiveness in the
However, it was difficult to achieve desired cleavage of the C–O cleavage of ether linkages. The reason behind using scCO2 as the
bond under the typical heterogeneous conditions even after the reaction medium was its tunable properties related to high product
application of very high pressure and temperature, which resulted selectivity, clean product separation as well as non-toxicity and non-
in a mixture of several compounds.3 Recently, a remarkable work flammability. Instead of expensive metal hydrides, molecular H2 was
of Sergeev et al. revealed an interesting strategy to the selective used as the source of H2. In the studied process, activated H2 breaks
cleavage of the aryl C–O bond using Ni(COD)2 as a homogeneous the aryl C–O bond at 80 1C within a reaction time of 5 h.
catalyst in the presence of a specific carbene ligand and NaOBu as
As we targeted the development of simple and practical hetero-
a base.4 Furthermore, in a new finding by the same group it was geneous catalysts, the reaction was conducted over different solid
suggested that without any additional dative ligand, selective catalysts such as Ni, supported on MCM-41, and noble metals like
hydrogenolysis of aryl ethers was also possible using a homo- Pt, Pd and Rh on MCM-41 as well as on activated carbon (C). Among
geneous catalyst formed in situ from the well-defined soluble the solid catalysts, activated carbon supported catalysts were pre-
nickel precursor Ni(COD)2 or Ni(CH2TMS)2(TMEDA) and a base viously used for hydrogenolysis of organosolv lignin under harsh
additive (tBuONa) in m-xylene at 120 1C.4 Following the first reaction conditions.8 Table 1 shows the results of catalyst screening
development by Sergeev et al., various efforts were made to for the hydrogenolysis of DPE. Among the MCM-41 supported
improve the reaction conditions and efficiency of the Ni catalysts
using different reducing agents or reaction media.5 From earlier
and recent literature, it is evident that Ni-based catalysts (mainly
homogeneous) are among the potential candidates with high
Research Center for Compact Chemical System, AIST, Tohoku, 4-2-1 Nigatake,
Miyagino-ku Sendai, Japan. E-mail: c-maya@aist.go.jp, h-kawanami@aist.go.jp;
Fax: +81 22 237 5388; Tel: +81 22 237 5213
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
Scheme 1 Possible reaction path of DPE hydrogenolysis–hydrogenation.
Chem. Commun., 2013, 49, 4567--4569 4567
c3cc41522h
c
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
a
Table 1 Catalyst screening for DPE hydrogenolysis–hydrogenation in scCO2
the path at higher pressure. In the lower pressure region hydro-
genolysis of DPE was favoured, which results mainly in cyclohexanol
and the possible source of cyclohexanol was phenol. To confirm this
process, a separate experiment was conducted using phenol under
similar working conditions, in which phenol was successfully con-
verted to cyclohexanol (Table 1; entry 9). No direct hydrogenolysis
from phenol or benzene or cyclohexanol to cyclohexane was
detected. In the higher pressure region the formation of dicyclohexyl
ether might be expected from the dehydrogenation or condensation
(dehydration) of cyclohexanol or hydrogenation of DPE. Under the
studied reaction conditions, we can neglect the first two options,
thus, hydrogenation of DPE might be the possible route for genera-
tion of dicyclohexyl ether. At higher pressure, DPE was highly soluble
in scCO2 (Fig. S1, ESI†). Hence, a large amount of H2 would be
available in the medium, which easily hydrogenated DPE to dicyclo-
hexyl ether. The hydrogenation of DPE to dicyclohexyl ether required
Product (selectivity (%))
Entry Catalyst
Conv. (%) DCHE CHPE CHOH CH/BZ
1
2
3
4
Ni/MCM-41
50.1
—
82.3
65.6
—
0.0
—
0.0
0.0
—
78.9
—
90.2
68.4
—
11.2
5.2/4.7
Pt/MCM-41
Pd/MCM-41
Rh/MCM-41
Pt/C
—
—
9.8
0.0/0.0
1.6/0.0
30.0
—
5
—
6
Pd/C
Rh/C
Rh/C
Rh/C
52.1
100.0
100.0
100.0
0.0
0.0
98.2
0.0
62.7
0.0
0.0
17.3
96.0
1.8
11.2/8.8
2.2/1.8
0.0/0.0
0.0/0.0
7
8b
9c
0.0
100.0
a
Reaction conditions: catalyst : substrate = 1 : 5, PCO = 10 Mpa, PH
=
2
0.5 Mpa, temp. = 80 1C, time = 5 h, water = 4 ml,2 entries 1–4 were
b
prepared in our laboratory; entries 5–7 are from Aldrich. Entry 8 = no
water. Entry 9 = phenol as the substrate; DCHE = dicyclohexyl ether;
c
CHPE = cyclohexyl phenyl ether; CH = cyclohexane, BZ = benzene.
catalysts, Pt was inactive, under the condition used (Table 1; higher H2 pressure over the Rh/C catalyst.10 To explore the specific
entry 2), whereas Ni, Pd and Rh exhibited high conversion from 50 contribution of the individual medium, the reaction was carried out
to 80%. However, the main product was partially hydrogenated separately in CO2 and in water. In CO2, DPE was completely
cyclohexyl phenyl ether (Table 1; entries 1, 3 and 4). As the catalyst converted to dicyclohexyl ether (98%) and cyclohexanol (2%)
support for Pt, Pd and Rh was changed to activated carbon, Pt (Table 1; entry 8). Similarly, in water the conversion of DPE dropped
remained inactive (Table 1; entry 5). Interestingly, Pd and Rh showed sharply to B15% with the formation of dicyclohexyl ether (68%),
impressive results in the formation of the C–O bond cleavage cyclohexanol (22.5%) and cyclohexane (9.5%) (see Table S1, entry 1
product. In the presence of Pd, DPE was converted to cyclohexyl (ESI†)). Hence, hydrogenation of the aromatic ring without any
phenyl ether (62.7%), cyclohexanol (17.3%), cyclohexane (11.2%) cleavage of the C–O bond was the major process, when CO2 and
and benzene (8.8%) (Table 1; entry 6). On the other hand, DPE was water were used independently. Therefore, a combined effect of
completely converted to mainly cyclohexanol under similar reaction water and scCO2 as the biphasic medium was necessary for hydro-
conditions over the Rh/C catalyst (Table 1; entry 7). Thus, depending genolysis of DPE because CO2 helps to accelerate the process and
on the performances, Rh/C was selected as an effective catalyst for water dominates the hydrogenolysis over hydrogenation at lower
hydrogenolysis of the C–O bond of DPE and further used to study pressure. It is too early to predict the exact route for this reaction as
the different reaction parameters. The uniqueness in the product study of the reaction mechanism is still underway.
selectivity over the Rh catalyst was also observed previously.9
From the accumulated results, one can understand that water
Fig. 1 exhibits the product distribution profile of the reaction played an important role. Keeping the other reaction parameters
with the variation in CO2 pressure at a fixed temperature of 80 1C. fixed, the amount of water was varied from 0 to 4 ml. We have found
No change in the conversion within the studied range of 7–18 MPa that the conversion of DPE remained unaffected in the presence or
was observed. The result shows a significant effect of CO2 pressure absence of water. In any case 100% conversion was observed.
on product distribution. In the low pressure region (7–12 MPa), However, there was a strong correlation between the product
cyclohexanol was detected as the major product; however when the distribution and the amount of water used (see Fig. S2, ESI†). As
pressure was enhanced to >12 MPa, dicyclohexyl ether was formed mentioned before, in the absence of water, mainly dicyclohexyl ether
and reached a selectivity of B50%. These results implied that the was formed. Surprisingly, when the same reaction was performed
catalytic path for this reaction at lower pressure was distinct from with the addition of 0.5 ml of water along with CO2, the selectivity to
cyclohexanol was increased to 32% and finally reached the highest
selectivity of 96% when the amount of water was 4 ml within the
reaction time of just 5 h. Thus, an optimum amount of 4 ml water
was used throughout the reaction. The increasing selectivity of
cyclohexanol did not result from dicyclohexyl ether because it was
inactive under the tested conditions.
Regarding the effect of H2 pressure, no reaction occurred in
the absence of H2. Thus, H2 was introduced into the system and
the pressure was changed from 0.1 to 1.5 MPa. As a result, the
conversion of DPE was enhanced from 38.2% to 100% when the
pressure increased from 0.1 to 0.5 MPa, and the detected
product was mainly cyclohexanol. However, higher H2 pressure
(>0.5 MPa) accelerated the reaction with dicyclohexyl ether as a
major product and confirmed the origin of dicyclohexyl ether at
Fig. 1 Effect of CO2 pressure on the hydrogenolysis–hydrogenation of
higher pressure of CO2. The most significant point for this
system was the potential to control the formation of a C–O bond
DPE; reaction conditions: PH = 0.5 Mpa; catalyst: substrate = 1 : 5; temp. = 80
2
1C; time = 5 h; water = 4 ml.
c
4568 Chem. Commun., 2013, 49, 4567--4569
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
Table 2 Hydrogenolysis–hydrogenation of different aryls and aryl alkyl ethersa
correlation was observed depending on the presence of electron
withdrawing and electron donating groups, which dictated the
formation of saturated or unsaturated alcohol or arene compounds.
In addition, the cleavage of the C–O bond of alkyl aryl ether
compounds was also achieved successfully (Table 2; entries 6–8).
For this system, instead of scCO2, different organic solvents
such as methanol, ethanol, isopropanol, tetrahydrofuran and
hexane were used along with water (see Table S1, ESI†). According to
the results, highest and lowest conversion of DPE was detected in
hexane (100%) and methanol (47%), respectively. In addition,
depending on the solvent used, product distribution was changed.
In isopropanol, the highest selectivity of cyclohexanol (92%) was
observed but the selectivity of dicyclohexyl ether was very high about
70% when hexane was used. The selectivity of cyclohexanol follows
the order of isopropanol > ethanol > methanol > tetrahydrofuran >
hexane. No definite trend of solvent polarity was detected in the
catalytic performances, but it could be stated that the reaction was
faster in non-polar solvents, which preferred hydrogenation rather
than hydrogenolysis. A strong role of the solvent was reported before
as they affected the activity and selectivity of the reaction using the
RANEYs Ni catalyst.11 It should be mentioned that although scCO2 is
considered as a non-polar solvent, it is superior to similar organic
solvents as it possesses substantially different characteristics, which
can direct the activity and selectivity of a reaction. Recycling of the
catalyst confirmed its stability (see Table 1, entries 7 and 8; ESI†).
In conclusion, we have developed a simple and highly
reactive catalytic system for the cleavage of the aryl ether bond
of DPE. Under the mild reaction conditions, it was possible to
achieve very high catalytic performance. The studied method was
also extended to different types of symmetrical and unsymmetrical
aryl and alkyl aryl ether compounds. Future work will be
directed to the mechanistic aspects to understand the actual
process for large-scale applications.
as represented in Scheme 2
Selectivity (%)
Entry Substrate
1
Conv. (%) 2(6)
3(7)
4(8)
5(9)
100
42.1 (57.9)
0.0
0.0
2
85.2
48.2 (38.6) (8.9)
46.2 (50.1) (1.2)
(7.3)
(2.5)
3
4
87.3
22.1
1.1
1.3 (55.9) (41.7)
5b
6
100
(49.9) (48.6)
0.0
0.0
60.2
(100)
—
—
—
7
52.1
55.5
(100)
(100)
—
—
—
—
—
—
8
a
Reaction conditions: catalyst : substrate = 1 : 5, PCO = 10 Mpa, PH = 0.5 Mpa,
temp. = 80 1C, water = 4 ml; reaction time: entry21 = 5 h, entry 2 = 12 h;
2
entry 3–8 = 18 h.; ‘‘—’’ = not determined. b 6 = cyclohexylmethanol.
Notes and references
1 E. Furimsky, Appl. Catal., 2000, 199, 147.
´
2 P. A. Bercedo and R. Martin, J. Am. Chem. Soc., 2010, 132, 17352;
A. Maercker, Angew. Chem., Int. Ed. Engl., 1987, 26, 972; P. Dabo,
Scheme 2
´
A. Cyr, J. Lessard, L. Brossard and H. Menard, Can. J. Chem., 1989,
77, 1225; M. Tobisu, K. Yamakawa, T. Shimasaki and N. Chatani,
Chem. Commun., 2011, 47, 2946.
cleavage product or an aromatic ring hydrogenation product
without any cleavage by simply tuning the reaction parameters.
After finding primary reaction conditions, we applied this
method to other aryl ether compounds with different substitution
patterns and the results are shown in Table 2. The results show that
the developed method was applicable for C–O bond cleavage of
unsymmetrical aryl ethers and alkyl aryl ethers. Depending on the
nature of the substrate, conversion varied from 20–100%. The
reaction of symmetrical aryl ether was faster and produced corres-
ponding hydrogenated arene and phenol compounds (Table 2;
entry 1). On the other hand, unsymmetrical aryl ether bearing
either electron withdrawing or electron donating groups or
both required a longer reaction time for the cleavage (Table 2;
3 J. Shabtai, N. K. Nag and F. E. Massoth, J. Catal., 1987, 104, 413.
4 A. G. Sergeev and J. F. Hartwig, Science, 2011, 332, 439; A. G. Sergeev,
J. D. Webb and J. F. Hartwig, J. Am. Chem. Soc., 2012, 134, 20226.
5 B. S. Samant and G. W. Kabalka, Chem. Commun., 2012, 48, 8658;
M. Tobisu and N. Chatani, ChemCatChem, 2011, 3, 1410; P. Kelley,
S. Lin, G. Edouard, M. W. Day and T. Agapie, J. Am. Chem. Soc., 2012,
134, 5480.
6 J. S. Miller and K. L. Pokhodnya, J. Mater. Chem., 2007, 17, 3585.
7 J. He, C. Zhao and J. A. Lercher, J. Am. Chem. Soc., 2012, 134, 20768.
8 M. Nagy, K. David, G. J. P. Britovsek and A. J. Ragauskas, Holz-
forschung, 2009, 63, 513; N. Yan, C. Zhao, P. J. Dyson, C. Wang, L. Liu
and Y. Kou, ChemSusChem, 2008, 1, 626; N. Yan, Y. A. Yuan,
R. Dykeman, Y. A. Kou and P. J. Dyson, Angew. Chem., Int. Ed.,
2010, 49, 5549.
9 M. Chatterjee, H. Kawanami, T. Ishizaka, M. Sato, T. Suzuki and
A. Suzuki, Catal. Sci. Technol., 2011, 1, 1466.
10 P. N. Rylander and M. Kilroy, Engelhard Ind. Tech. Bull., 1968, 9, 14.
entries 2–4). In terms of product selectivity, an apparent 11 X. Wang and R. Rinaldi, ChemSusChem, 2012, 5, 1455.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4567--4569 4569