Hydrogenolysis-hydrogenation of aryl ethers: Selectivity pattern

Article abstract of DOI:10.1039/c2cc34454h

The selectivity pattern of nickel-catalyzed hydrogenolysis-hydrogenation of aryl ethers has been studied in the micellar media. The micellar conditions selectively formed arenes and alcohols with enhanced yields.

Full text of DOI:10.1039/c2cc34454h

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 8658–8660
www.rsc.org/chemcomm
COMMUNICATION
Hydrogenolysis–hydrogenation of aryl ethers: selectivity patternw
Bhupesh S. Samant*a and George W. Kabalkabc
Received 21st June 2012, Accepted 4th July 2012
DOI: 10.1039/c2cc34454h
The selectivity pattern of nickel-catalyzed hydrogenolysis–
hydrogenation of aryl ethers has been studied in the micellar
media. The micellar conditions selectively formed arenes and
alcohols with enhanced yields.
carried out on Ni catalyzed hydrogenolysis in an effort to
achieve selectivity towards phenol and arene functionality.8
However, the need of the specific ligand, heavy catalyst loading
(Ni, 20 mol%), hydrogen gas, substrate specificity (substrates
with specific directing groups), and long reaction times limit the
use of this process at an industrial scale on the basis of scope,
safety and cost.2 Hence, the need for a selective, commercially
viable C–O scission process remains, especially in the fine
chemical industry and in the energy industries focused on
generating brown coal from lignocellulosic biomass with a
C–O bonds network.9 In the pharmaceutical industry for
example, the fragmentation study of drugs such as nitroscanate
(1-(4-isothiocyanatophenoxy)-4-nitrobenzene, an anthelmintic
drug) (Section S3 and Fig. S1, ESIw) is very important.
Here, we report the use of reverse micelles as a reaction
medium to improve the hydrogenolysis–hydrogenation (H–H)
of aromatic C–O bonds. We previously reported the use of
surfactants and various polymeric media in transition metal
catalyzed reactions,10 syntheses of bioactive compounds,11
and radiolabeling studies.12
Breaking aromatic C–O bonds is one of the most demanding
processes in synthetic organic chemistry especially because of
the stability of C–O linkages.1 However, cleavage of C–O
bonds is very important in the fields of deoxygenated fuels and
fine chemicals.2
Aliphatic C–O bonds are relatively easy to cleave using
either hydrolysis or dehydration reactions. However, aromatic
C–O bonds are resistant to these simple processes.3 In the case
of aryl ethers even harsh reaction conditions often fail to selectively
cleave C–O bonds. The breaking of aryl C–O bonds and hydro-
genation of arene rings generally lead to the mixture of phenols,
cycloalkanols, saturated cyclic hydrocarbons (cycloalkanes), and
arenes.4 Due to the low yields, lack of chemoselectivity (towards
an aryl C–O bond and towards a particular product), and
waste of hydrogen gas; hydrogenation cleavage reactions have
not been extensively utilized on an industrial scale. The few
methods currently available to achieve selective aryl C–O bond
cleavage generally require expensive electrocatalytic processes5
or the use of a stoichiometric quantity of alkali metal catalysts.6
Hence, the cost limits their use at an industrial scale.2
We initiated the H–H study by using diphenyl ether (DPE),
a tricyclohexylphosphine ligand (L), nickel(II) acetylacetonate
(Ni(acac)2), and sodium tert-butoxide (NaOtBu) as a base.
Lithium tri-tert-butoxyaluminum hydride (LiAl(OtBu)3H) was
used as the hydrogen source.w Under these reaction conditions the
yield of products (from hydrogenolysis and/or hydrogenation) was
negligible and DPE remained unreacted (Table 1, No. 1). We then
increased the reaction times along with the catalyst concentration
however; the selectivity of reaction was lost. The reaction
produced cyclohexane and cyclohexene as the main products
(Table 1, No. 2). The formation of these products presumably
was a result of destruction of the carbon–phosphine bond in
the L which ultimately destroyed the catalyst.
In the last few decades, the nickel (Ni) metal complexes have
been developed for C–O bond cleavage reactions because of
their ability to activate aromatic C–O bonds in the presence of
aliphatic C–O bonds.7 Efforts were made to improve reaction
conditions and catalytic efficiency.4 However, inhomogeneity of
the system due to reduction of catalysts and modest reactivity of
hydrogen restricts the use of these processes for hydrogenolysis
of aromatic C–O bonds. It is currently very difficult to achieve
selective cleavage of an aromatic C–O bond in the presence of
an aliphatic C–O bond. Recently, a very important study was
We then carried out the reaction in the presence of cetyl-
trimethylammonium bromide (CTAB) (0.60 mmol in toluene)
and a reduced quantity of Ni(acac)2 and of the L. Remarkable
enhancements in % conversion of DPE were observed.
Surprisingly, in the homogeneous catalysts system cyclohexane
and cyclohexanol were formed as the products in just 5 h
reaction time (Table 1, No. 3). On the other hand, when the
same reaction was performed in sodium dodecyl sulphate
(SDS) (0.60 mmol in toluene), conversion was very low
(Table 1, No. 4). Thus, the presence of cationic reverse micelles
in the reaction increases the conversion of diphenyl ether to
cyclohexane and cyclohexanol rather than a mixture of
phenols, cycloalkanols and saturated cyclic hydrocarbons.1–4
a Division of Pharmaceutical chemistry, Faculty of Pharmacy,
Rhodes University, Grahamstown-6140, South Africa.
E-mail: B.Samant@ru.ac.za; Fax: +27 466361205;
Tel: +27 466038395
b University of Tennessee, Departments of Chemistry and Radiology,
1420 Circle Drive, Knoxville, TN 37996-1600, USA
c UT Medical Hospital, Biomedical Imaging Center, Knoxville,
TN 37920-6999, USA
w Electronic supplementary information (ESI) available: Details of
materials, general procedure for hydrogenolysis–hydrogenation,
analytical information and additional optimization data. See DOI:
10.1039/c2cc34454h
c
8658 Chem. Commun., 2012, 48, 8658–8660
This journal is The Royal Society of Chemistry 2012

View Online
Table 1 Optimization of hydrogenolysis–hydrogenation of DPE
Reaction conditions
Total % conversiona
Yieldd (%)
1
2
3
4
Ni (10 mol%), L (20 mol%), LiAl(OtBu)3H (2.5 eq.), toluene (1.5 mL),
24 h, 70 1C, s. r.c 1.67 Hz
o5b
10b
—
Ni (20 mol%), L (40 mol%), LiAl(OtBu)3H (12 eq.), toluene (1.5 mL),
48 h, 120 1C, s. r. 1.67 Hz
—
Ni (5 mol%), L (10 mol%), LiAl(OtBu)3H (2.5 eq.), toluene (1.5 mL),
CTAB (0.60 mmol), 5 h, 70 1C, s. r. 1.67 Hz
60, 1a (50), 1b (50)
4, 1a (50), 1b (50)
1a (99), 1b (99)
Ni (5 mol%), L (10 mol%), LiAl(OtBu)3H (2.5 eq.), toluene (1.5 mL),
SDS (0.60 mmol), 5 h, 70 1C, s. r. 1.67 Hz
—
a
b
c
d
Determined by gas chromatography, (% selectivity). Mixture of benzene, phenol, cyclohexane and cyclohexene. Stirring rate. Isolated
yield, substrate recovered. Each experiment was repeated three times.
The yield and selectivity of H–H reaction were found to be
dependent on the head-group charges of the surfactant used in
the reaction (Section S5, ESIw). These results were unexpected
and distinct from the previous study done for palladium (Pd)
catalysed reactions, where there is no relationship between the
head-group charges and the yield of the reaction.10
LiAl(OtBu)3H was used as the hydrogen source instead of
hydrogen gas to test its utility as a solid hydrogen source and
also because of the safety issue. In preliminary runs, NaOtBu
and LiAl(OtBu)3H were not particularly soluble in toluene
even though fine powdered reagents were used. This problem
was overcome by the use of CTAB in toluene due to the
formation of reverse micelles in the reaction medium.
medium (0.60 mmol CTAB) the yield of H–H of DPE was not
influenced due to the presence of moisture. It also gave good
yield with reduced quantities of the Ni catalyst and L (5 mol%
Ni and 10 mol% L). The s. r. affected the yield of the reaction
in micellar medium. This optimization study proves the use of
micellar media for selective H–H of DPE to form cyclohexanol
and cyclohexane.
The exact mechanism for Ni catalyzed H–H in reverse micellar
media is still the unsolved obscurity (Section S6, ESIw). Never-
theless, in micellar medium the formation of a L–Ni(II) complex
seems to take place in hydrophobic bulk. Other reagents like
NaOtBu and LiAl(OtBu)3H may be present near the anisotropic
palisade layer. The separation of catalyst–L from other reagents
might protect the catalyst from decomposition.
Since the intermediate and/or final product may contain
benzene, which is a low boiling liquid (80 1C boiling point,
b.p.), we used 70 1C reaction temperature. At high reaction
temperatures (120 1C), percent conversions are detrimentally
affected (Table S2, No. 4 (ESIw)). To study the effect of
moisture on the product yield this reaction was carried out
in the reverse micellar media under atmospheric conditions
and no significant differences in the product yields were
observed (Table S2, No. 3 (ESIw)).
Due to the presence of a surfactant in the reaction medium,
there is always a problem with respect to the formation of
foam. In most reactions the formation of the foam negatively
affects the yield of the reaction. We know from our previous
studies10,11,13 that foam formation is directly related to the
stirring rate, and the position of the magnetic bar in the
reaction media. We performed extensive study on the effect
of stirring on the yield of the H–H by using a magnetic stirrer
to control stirring (Section S4, ESIw). Both slow stirring rate
(s. r.) and high s. r. in the reaction caused decreases in the yield
of the product without affecting selectivity (Table S1, ESIw).
This might be due to the lack of good contact between the
reactants in the case of slow s. r. and due to the formation of
foam in the case of high s. r. Also with slow or no stirring,
dispersion of the solid reagents became a problem. Hence, we
decided to maintain 1.67 Hz s. r. for the reaction.
The positive effect of strong base NaOtBu may be due to the
formation of an anionic Ni complex which is the main reactive
species for hydrogenolysis. The formation of an anionic Ni
complex is supported by the very low yield in the presence of
anionic surfactant SDS. On the other hand, increased yields
and shorter reaction times in the presence of CTAB (cationic
surfactant) can be explained by considering the attraction
between the anionic Ni complex (from bulk) and concentrated
positive charge (from reverse cationic micelles). This may
facilitate the contact between anionic Ni complex and the
ArOAr (near palisade layer). The spatial orientation of phenol
(intermediate product) in the palisade layer of micelles may be
responsible for restricting the further hydrogenolysis of the
C–O bond in phenol and conferring the selectivity towards
arene hydrogenation. In the past, we verified the spatial
orientation of aromatic compounds in palisade layers in the
nuclear magnetic resonance study.13 The formation of colloi-
dal Ni species was also considered as one of the reasons for
selective arene hydrogenation in the homogeneous catalysts
system (Section S7, ESIw). The inactivity of the Ni catalyst
system in the presence of mercury (Hg) proves the formation
of colloidal Ni species in the reaction medium.
To comprehend the scope of the H–H in the micellar media,
we studied various aryl ethers (Table 2). This process maintained
selectivity of H–H in the presence of other functional groups
(Table 2, No. 1: NO2 and NCS, No. 2 and 3: OMe). We found
that less reactive derivatives of DPEs produced good yields
and selectivity (Table 2, No. 2 and 3). In unsymmetrical DPEs,
The complete optimization study was done with the intention
to provide a process of value in large scale reactions of interest
to industry (Tables S1, S2, and Section S3, ESIw). All the
above reactions suggest that in the presence of reverse micellar
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 8658–8660 8659

View Online
Table 2 Hydrogenolysis–hydrogenation of representative compounds
of Ar–O–Ar bonds, cleavage was done at Me/EtO–Ar bonds
(Table 2, No. 5–8).
Substrate (total % conversion)a
Products (% yieldb)
An imperative pronouncement of this study is the obtained
selectivity pattern for H–H of aryl ethers because of the
anisotropic palisade layer of CTAB reverse micelles. The
stability of catalysts and the spatial orientation of the aromatic
alcohol (phenol) appear to show a positive effect on the yield
and selectivity of the particular products, respectively.
4-NCS–cyclohexanol (95),
NO2–cyclohexane (97)
1
2
3
4
(61)
3-MeO–cyclohexanol (88),
cyclohexanol (3), MeO–
cyclohexane (86), MeOH (nqc),
cyclohexane (2),
(60)
Notes and references
3-MeO–cyclohexanol (92),
(58) cyclohexanol (2), MeOH (nq),
cyclohexane (89)
1 E. Furimsky, Appl. Catal., A, 2000, 199, 147–190.
2 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098; J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and
B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
3 A. L. Marshall and P. J. Alaimo, Chem.–Eur. J., 2010, 16, 4970–4980.
4 E. M. Van Duzee and H. Adkins, J. Am. Chem. Soc., 1935, 57, 147;
G. Stork, J. Org. Chem., 1947, 69, 576; S. Utoh, T. Hirata, H. Oda
and C. Yokokawa, Fuel Process. Technol., 1986, 14, 221;
M. K. Huuska, Polyhedron, 1986, 5, 233–236.
Cyclohexanol (87),
(58)
cyclohexane (89)
Decahydronaphthalene (87),
MeOH (nq)
5
6
7
(60)d
5 A. Maercker, Angew. Chem., Int. Ed. Engl., 1987, 26, 972–989.
6 P. Dabo, A. Cyr, J. Lessard, L. Brossard and H. Me
Chem., 1999, 77, 1225–1229.
´
nard, Can. J.
Decahydronaphthalene (85),
EtOH (nq)
(57)
7 B. Guan, S. Xiang, T. Wu, Z. Sun, B. Wang, K. Zhao and Z. Shi,
Chem. Commun., 2008, 1437–1439; M. Tobisu, T. Shimasaki and
N. Chatani, Angew. Chem., Int. Ed., 2008, 47, 4866–4869; J. W.
Dankwardt, Angew. Chem., Int. Ed., 2004, 43, 2428–2432;
E. Wenkert, E. L. Michelotti, C. S. Swindell and M. Tingoli,
J. Org. Chem., 1984, 49, 4894–4898; E. Wenkert, E. L. Michelotti
and C. S. Swindell, J. Am. Chem. Soc., 1979, 101, 2246–2247.
8 A. G. Sergeev and J. F. Hartwig, Science, 2011, 332, 439–443.
9 D. G. Levine, R. H. Schlosenberg and B. G. Silbernagel, Proc.
Natl. Acad. Sci. U. S. A., 1982, 79, 3365.
Me–cyclohexane (85),
MeOH (nq)
(57)
Me–cyclohexane (82),
MeOH (nq)
8
(58)
a
b
Determined by gas chromatography. Isolated yield, substrate
c
recovered. Not quantified. Each experiment was repeated three
10 B. S. Samant and G. W. Kabalka, Chem. Commun., 2011, 47,
7236–7238; B. S. Samant and S. S. Bhagwat, Appl. Catal., A, 2011,
394, 191–194.
d
times. AlMe3 (1 equiv.).
11 B. S. Samant and M. G. Sukhthankar, Bioorg. Med. Chem. Lett.,
2011, 21, 1015–1018; B. S. Samant and S. S. Bhagwat, Chin. J.
Catal., 2011, 32, 231–234; B. S. Samant and M. G. Sukhthankar,
Med. Chem., 2009, 5, 293–300; B. S. Samant, Y. P. Saraf and
S. S. Bhagwat, J. Colloid Interface Sci., 2006, 302, 207–213.
12 G. W. Kabalka, J. F. Green, M. M. Goodman, J. T. Maddox and
S. J. Lambert, Radiopharmaceutical synthesis via boronated polymers, in
Current Topics in the Chemistry of Boron, ed. G. W. Kabalka, The Royal
Society of Chemistry, Cambridge, UK, 1994, pp. 199–203;
G. W. Kabalka, Acc. Chem. Res., 1984, 17, 215–221; G. W. Kabalka,
J. Labelled Compd. Radiopharm., 2007, 50, 888–894; H. M. Schuller and
G. W. Kabalka, US Pat. Appl. Pub., 2 006 067 879, 2006; G. W.
Kabalka and M.-L. Yao, J. Organomet. Chem., 2009, 694, 1638–1641;
L. Yong, M.-L. Yao, J. F. Green, H. Kelly and G. W. Kabalka, Chem.
Commun., 2010, 46, 2623–2625; Y. Li, M.-L. Yao, H. Kelly, J. F. Green
and G. W. Kabalka, J. Labelled Compd. Radiopharm., 2011, 54,
173–174.
cleavage of the C–O bond took place next to the aryl ring
containing electron withdrawing functionality. For 3,30-oxybis-
(methoxybenzene), selectivity in hydrogenolysis was maintained
towards the aromatic C–O bond (Table 2, No. 2). The formation
of decahydronaphthalene from 2-alkoxy-naphthalenes (Table 2,
No. 5 and 6) reveals selectivity in H–H of the aromatic C–O bond.
The interesting formation of methyl-cyclohexane from benzyl
ethers (Table 2, No. 7 and 8) indicates the preference for cleavage
towards more electron dense species. This series clearly
indicate the selectivity pattern including (i) selectivity towards
arene hydrogenation, (ii) selectivity towards aromatic C–O
bond cleavage, (iii) selectivity towards cleavage of Ar–O–Ar in
the presence of Ar–OMe (Table 2, No. 2 and 3). In the absence
13 B. S. Samant and S. S. Bhagwat, J. Dispersion Sci. Technol., 2011,
DOI: 10.1080/01932691.2011.596325; B. S. Samant and S. S. Bhagwat,
Monatsh. Chem., 2012, 143, 1039–1044.
c
8660 Chem. Commun., 2012, 48, 8658–8660
This journal is The Royal Society of Chemistry 2012