Highly chemoselective hydrogenation of active benzaldehydes to benzyl alcohols catalyzed by bimetallic nanoparticles

Article abstract of DOI:10.1016/j.tetlet.2015.09.154

By using novel Pd/Ni bimetallic nanoparticles as a catalyst, the active benzaldehydes were hydrogenated to the corresponding benzyl alcohols as unique products in practical quantitative yields. The undesired catalytic hydrogenolysis of the benzyl alcohol was inhibited completely. By using this hydrogenation as a key step, the total synthesis of the natural product gastrodin was achieved with less total steps and a higher total yield.

Full text of DOI:10.1016/j.tetlet.2015.09.154

Tetrahedron Letters 56 (2015) 6460–6462
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Highly chemoselective hydrogenation of active benzaldehydes
to benzyl alcohols catalyzed by bimetallic nanoparticles
⇑
⇑
Chulong Liu, Hailin Bao, Dingsheng Wang, Xinyan Wang , Yadong Li, Yuefei Hu
Department of Chemistry, Tsinghua University, Beijing 100084, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
By using novel Pd/Ni bimetallic nanoparticles as a catalyst, the active benzaldehydes were hydrogenated
to the corresponding benzyl alcohols as unique products in practical quantitative yields. The undesired
catalytic hydrogenolysis of the benzyl alcohol was inhibited completely. By using this hydrogenation
as a key step, the total synthesis of the natural product gastrodin was achieved with less total steps
and a higher total yield.
Received 28 August 2015
Revised 28 September 2015
Accepted 30 September 2015
Available online 18 October 2015
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
Chemoselectivity
Catalytic hydrogenation
Active benzaldehydes
Benzyl alcohols
Bimetallic nanoparticles
EDG (electron donating group)-substituted benzaldehydes 1 are
also called active benzaldehydes, which can be prepared easily by
Vilsmeier reaction.1 Reduction of 1 to the corresponding benzyl
alcohols 2 is a fundamental transformation in organic synthesis2
and numerous important compounds have been synthesized by
this transformation. As shown in Figure 1, the natural product gas-
trodin is an OTC medicine that can be utilized for sedation, anti-
convulsion and anti-epilepsy.3 4-Methoxy-3-(3-methoxypropoxy)
benzyl alcohol is a key precursor for the synthesis of aliskiren, a
clinical medicine used as an rennin inhibitor that can be taken
orally.4
OH
O
HO
HO
O
OH
OH
MeO
O
MeO
OH
Gastrodin
4-Methoxy-3-(3-methoxypropoxy)
benzenemethanol
Figure 1. Some important EDG-substituted benzyl alcohols.
Investigation showed that the transformation of
1 to 2
CHO
CH2OH
CH3
is achieved normally by using hydride reagent reduction5 rather
than metal-catalyzed hydrogenation in the literature. For example,
Pd-catalyzed hydrogenation is characterized by its extremely clean
hydrogen resource and convenient work-up procedure, but it is
rarely used for this purpose. As shown in Scheme 1, this must have
resulted from the fact that the catalytic hydrogenolysis of benzyl
alcohol 2 also proceeds smoothly under the catalytic hydrogena-
tion conditions to give the byproduct toluene 3.6 In many cases,
the byproduct 3 was even a major or a unique product.7
Pd-Catalyed
Hydrogention
Pd-Catalyed
Hydrogenolysis
EDG
EDG
EDG
1
2
3
Scheme 1. Pd-catalyzed hydrogenation and hydrogenolysis.
catalyzed hydrogenation, by which highly efficient and chemoselec-
tive transformation of 1 to 2 was achieved in practical quantitative
yields. The method meets the needs of the laboratory scale trans-
formation of 1 to 2 with three advantages: (a) easy preparation
of the Pd/Ni bimetallic nanoparticles; (b) no need for any additives;
(c) convenient work-up by magnetic separation.
In recent years, several novel catalytic systems8 were reported
for highly chemoselective hydrogenation of 1 to 2. The fact that
they were not used in practice may be caused by the difficult
preparation of the catalyst and the use of additives. Herein, we
would like to report a novel Pd/Ni bimetallic nanoparticles-
In our recent work,9 novel Pd/Ni bimetallic nanoparticles were
prepared by simply boiling the mixture of Raney-Ni and Na2PdCl4
in H2O for 6 h. When they were used as a catalyst, nitrobenzylamines
⇑
Corresponding authors.
http://dx.doi.org/10.1016/j.tetlet.2015.09.154
0040-4039/Ó 2015 Elsevier Ltd. All rights reserved.

C. Liu et al. / Tetrahedron Letters 56 (2015) 6460–6462
6461
Table 1
were chemoselectively hydrogenated into aminobenzylamines.
The hydrogenations of 1a catalyzed by the Pd/Ni bimetallic catalysta
Since the Pd/Ni ratio was about 1:99 (by wt %), we hypothesized
that most or all of the Pd-atoms on the surface of this catalyst were
surround by Ni-atoms to form the heterometallic Pd–Ni bonds.
Unlike the Pd–Pd and Ni–Ni bonds, Pd–Ni bond is a polarized bond
that may be described as dÀPd–Nid+ (electronic effect or ligand
effect), by which the Pd/Ni bimetallic catalyst demonstrated novel
catalytic ability and chemoselectivity that differ from their parent
metals.10 As shown in Scheme 2, the polarized Pd/Ni bimetallic cat-
alyst may have stronger adsorptivity for the heteroatom double
bond than for the heteroatom single bond. This hypothesis also
strongly implied that the EDG-substituted benzaldehydes 1 (with
heteroatom double bond) may be chemoselectively hydrogenated
into the corresponding benzyl alcohols 2 (with heteroatom single
bond) by using the same catalyst.
1% Pd/Ni, H2
MeOH, rt, time
71-98% yield
100% chemoselectivity
MeO
MeO
MeO
MeO
OH
O
OMe
1a
OMe
2a
Entry
1% Pd/Ni (wt %)
Timeb (h)
Pressure (psi)
Yield of 2ac (%)
1
2
3
4
100
50
30
20
20
10
10
1.5
2.5
3.5
6.5
6.0
6.0
6.0
—
—
—
—
30
30
60
94
97
98
91d
98
5e
6e
7e
71d
96
To prove our hypothesis, 3,4,5-trimethoxybenzylaldehyde (1a)
was used as a model substrate because its hydrogenation product
3,4,5-trimethoxybenzyl alcohol (2a) is highly sensitive to further
catalytic hydrogenolysis. As shown in Scheme 3, five commercial
Pd-catalysts A–E were initially tested on the hydrogenated of 1a
and three products 2a, 3a and 4a were detected with variable
ratios. But, none of them gave an acceptable hydrogenation
chemoselectivity.
a
A mixture of 1a (196 mg, 1 mmol) and the Pd/Ni bimetallic catalyst in MeOH
(10 mL) was stirred under H2 at room temperature (on an atmospheric pressure
hydrogenation apparatus).
b
The time was when the absorption of hydrogen ceased.
Separated yield was obtained.
Only the unreacted substrate 1a was recovered.
The experiments proceeded on a Parr-hydrogenator.
c
d
e
As was expected, the hydrogenation of 1a gave 2a as a unique
product over the Pd/Ni bimetallic catalyst. As shown in
Table 1, 2a was obtained in 94% yield and 100% chemoselectivity
under the conditions listed in entry 1. Since the net weight of pal-
ladium metal used in this entry was much more than that used in
Scheme 3, the chemoselectivity must arise from the synergistic
effects of the Pd/Ni bimetallic catalyst. The quantitative yield and
chemoselectivity of 2a were obtained when 30 wt % Pd/Ni bimetal-
lic catalysts were used (entry 3). Although 2a was obtained in
lower yield (91%) by use of 20 wt % Pd/Ni bimetallic catalysts, its
chemoselectivity was not affected (entry 4). When the hydrogena-
tion was operated on a Parr-hydrogenator by slightly increasing
the hydrogen pressure, 10 wt % Pd/Ni bimetallic catalysts were
good enough to produce excellent results (entries 5–7).
The conditional experiments for reaction solvents and recycling
studies of the Pd/Ni bimetallic catalyst gave very similar results to
those reported in our previous work.9 For example, MeOH was the
best solvent for this hydrogenation and the catalytic activity of the
catalyst decreased sharply after four recycles. Thus, the standard
1% Pd/Ni (30 wt%), H2 (atm)
previous work
NR2
NR2
OH
MeOH, rt, 3.5-5.3 h
96-98% yields
100% chemoselectivity
OH
OH
O
EDG
EDG
O2N
H2N
Pd/Ni, H2 (atm), MeOH, rt
1a-1t
2a-2t
?
O
EDG
EDG
MeO
MeO
proposed work
OH
OH
OH
1
2
MeO
EtO
EtO
OMe
OMe
OMe
OEt
Scheme 2. Proposed chemoselective hydrogenation.
2a (3.5 h, 98%)
2b (4.1 h, 98%)
2c (4.0 h, 98%)
2d (4.0 h, 97%)
HO
O
OH
Me2N
OH
HO
2g (5.5 h, 98%)
OH
OH
MeO
catalysts A-E (5 wt%), H2 (atm), MeOH, rt, 3.5 h
the ratios of 2a:3a:4a
O
O
MeO
were determined by 1H NMR spectra
2e (4.0 h, 97%)
2f (4.0 h, 97%)
2h (5.3 h, 98%)
1a
OMe
MeO
OH
OH
OH
OH
MeO
MeO
MeO
MeO
Me
MeO
MeO
OH
OMe
+
+
MeO
2j (4.5 h, 98%)
OH
OMe
2l (3.5 h, 98%)
2i (5.3 h, 97%)
2k (4.7 h, 98%)
PhO
OMe
2a
71%
0%
33%
13%
0%
OMe
3a
OMe
4a
0%
58%
0%
0%
0%
F3CO
OH
OH
OH
OH
OH
Cat-A
Cat-B
Cat-C
Cat-D
Cat-E
29%
42%
PrnO
BnO
67%
2m (4.1 h, 97%)
2n (4.0 h, 97%)
2o (4.3 h, 99%)
2p (5.0 h, 98%)
87%
100%
Me
Scheme 3. Hydrogenations over five commercial Pd-catalysts. Cat-A: Acros-
199620100, palladium hydroxide on carbon powder, 20% Pd. Cat-B: Aldrich-
330108, palladium, 10 wt % on activated carbon. Cat-C: Alfa Acesar-A26D10,
palladium, 5 wt % on activated carbon powder, unreduced. Cat-D: Alfa Acesar-
K08P30, palladium, 5 wt % on activated carbon powder, reduced, acidic catalyst.
Cat-E: Shanxi Kaida product, palladium, 10 wt % on activated carbon powder.
OH
OH
OH
Me
2s (5.0 h, 98%)
Me
2q (4.5 h, 96%)
2r (4.7 h, 97%)
2t (5.0 h, 96%)
Scheme 4. The scope of the hydrogenation.

6462
C. Liu et al. / Tetrahedron Letters 56 (2015) 6460–6462
OAc
O
the natural product gastrodin, less total steps and a higher total
yield were achieved.
OAc
AcO
AcO
AcO
AcO
i
ii or v
O
O
O
Br
Acknowledgment
OAc
5
OAc
This work was supported by National Natural Science Founda-
tion of China (Nos. 21221062, 21372142 and 21472107).
OAc
OH
AcO
AcO
HO
HO
O
O
OH
O
OH
iv
O
Supplementary data
OAc
OH
6
Gastrodin
OAc
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2015.09.
154.
AcO
AcO
O
O
OAc
iii
iv
OAc
7
References and notes
Scheme 5. Reported method: (i) 4-hydroxy-benzyldehyde, aq NaOH, acetone, rt,
12 h, 35%; (ii) KBH4, i-PrOH, rt, 0.5 h, 87%; (iii) Ac2O, pyridine, 100 °C, 6 h, 100%; (iv)
MeONa, MeOH, rt, 12 h, 90%. Total 27.4% yield for four steps. Our method: (v) 1% Pd/
Ni bimetallic catalyst (30 wt %), H2 (atm), MeOH, rt, 2 h, 100%. Total 31.5% yield for
three steps.
1. For a review, see: Jones, G.; Stanforth, S. P. Org. React. 1997, 49, 1–330.
2. Hudlicky, M. Reductions in Organic Chemistry, 2nd ed.; ACS: Washington, DC,
1996.
3. (a) Cai, Z.; Lei, X.; Lin, Z.; Zhao, J.; Wu, F.; Yang, Z.; Pu, J.; Liu, Z. Acta Pharm.
Sinica B 2014, 4, 86–93; (b) Song, C.; Fang, S.; Lv, G.; Mei, X. Neural Regen. Res.
2013, 8, 1383–1389; (c) Lin, L.-C.; Chen, Y.-F.; Lee, W.-C.; Wu, Y.-T.; Tsai, T.-H. J.
Pharm. Biomed. Anal. 2008, 48, 909–917.
4. Wang, F.; Xu, X.-Y.; Wang, F.-Y.; Peng, L.; Zhang, Y.; Tian, F.; Wang, L.-X. Org.
Process Res. Dev. 2013, 17, 1458–1462.
conditions12 for the chemoselective hydrogenation of 1 to 2 over
the Pd/Ni bimetallic catalyst were assigned as shown in Scheme 4
and its scope was studied. As was expected, the quantitative
chemoselectivity was obtained for all tested substrates. Since the
products 2f–2i contain strong coordination groups, they are usu-
ally difficult to be synthesized in excellent yields by using hydride
reagent reduction. However, they were obtained in excellent yields
under our standard conditions. In the preparation of 2n, the alde-
hyde group was hydrogenated into the alcohol group smoothly,
while the benzyl ether group stayed intact throughout. This result
gave new evidence to support our hypothesis that the polarized
Pd/Ni bimetallic catalyst may have much weaker adsorptivity for
the heteroatom single bond.
In literature, the most practical route for the total synthesis of
gastrodin11 is shown in Scheme 5. Although the reduction of 5 by
KBH4 (step ii) proceeded smoothly, the desired product 6 usually
was obtained as a mixture with some partial deacylated products.
When the mixture of 6 was saponified directly, the final product
gastrodin must be purified by an ion-exchange resin chromatogra-
phy since the deacylated products strongly coordinated with boron
and sodium ions. Therefore, a reacylation for the mixture of 6 (step
iii) was an essential step because the reacylated product 7 can be
purified conveniently by recrystallization or by a normal flash
chromatography. However, when 5 was hydrogenated over the
Pd/Ni bimetallic catalyst (step v), the desired 6 was obtained in
almost quantitative yield. Since the Pd/Ni bimetallic catalyst was
used barely without any additives and was recovered by a mag-
netic separation, the product 6 was pure enough for the direct
saponification (step iv) without further purification. Thus, our
method has less total steps and a higher total yield for the total
synthesis of gastrodin.
5. For selected recent references: (a) Suman, P.; Murthy, T. R.; Rajkumar, K.;
Srikanth, D.; Dayakar, Ch.; Kishor, C.; Addlagatta, A.; Kalivendi, S. V.; Raju, B. C.
Eur. J. Med. Chem. 2015, 90, 603–619; (b) Zheng, Y.; Liu, Y.; Wang, Q. J. Org.
Chem. 2014, 79, 3348–3357; (c) Worayuthakarn, R.; Boonya-udtayan, S.;
Ruchirawat, S.; Thasana, N. Eur. J. Org. Chem. 2014, 2496–2507; (d) Yoshida,
Y.; Terauchi, T.; Naoe, Y.; Kazuta, Y.; Ozaki, F.; Beuckmann, C. T.; Nakagawa, M.;
Suzuki, M.; Kushida, I.; Takenaka, O.; Ueno, T.; Yonaga, M. Bioorg. Med. Chem.
2014, 22, 6071–6088; (e) Azizi, N.; Batebi, E.; Bagherpour, S.; Ghafuri, H. RSC
Adv. 2012, 2, 2289–2293.
6. (a) Nishimura, S. Handbook of Heterogeneous Catalytic Hydrogenation for Organic
Synthesis; John Wiley & Sons Inc: New York, 2001. Chapter 13; (b) Rylander, P.
N. Hydrogenation Methods; Academic Press: London, 1985. Chapter 13.
7. (a) Bielitza, M.; Pietruszka, J. Chem. Eur. J. 2013, 19, 8300–8308; (b) Varadaraju,
T. G.; Hwu, J. R. Org. Biomol. Chem. 2012, 10, 5456–5465; (c) Dinh, T. N.; Chen,
A.; Chai, C. L. L. Tetrahedron 2011, 67, 3363–3368; (d) Tan, N. P. H.; Donner, C. D.
Tetrahedron Lett. 2008, 49, 4160–4162; (e) Xing, L.; Wang, X.; Cheng, C.; Zhu, R.;
Liu, B.; Hu, Y. Tetrahedron 2007, 63, 9382–9386; (f) Connolly, T. J.; Matchett, M.;
McGarry, P.; Sukhtankar, S.; Zhu, J. Org. Process Res. Dev. 2004, 8, 624–627; (g)
Lebrun, S.; Couture, A.; Deniau, E.; Grandclaudon, P. Org. Biomol. Chem. 2003, 1,
1701–1706; (h) Yamaguchi, S.; Muro, S.; Kobayashi, M.; Miyazawa, M.; Hirai, Y.
J. Org. Chem. 2003, 68, 6274–6278; (i) Lai, C.; Shen, Y.; Wang, M.; Kameswara
Rao, N. S.; Liao, C. J. Org. Chem. 2002, 67, 6493–6502; (j) Burk, M. J.; Gerlach, A.;
Semmeril, D. J. Org. Chem. 2000, 65, 8933–8939.
8. (a) Ngo, A. H.; Adams, M. J.; Do, L. H. Organometallics 2014, 33, 6742–6745; (b)
Manas, M. G.; Graeupner, J.; Allen, L. J.; Dobereiner, G. E.; Rippy, K. C.; Hazari,
N.; Crabtree, R. H. Organometallics 2013, 32, 4501–4506; (c) Jia, Z.; Zhou, F.; Liu,
M.; Li, X.; Chan, A. S. C.; Li, C.-J. Angew. Chem., Int. Ed. 2013, 52, 11871–11874;
(d) Fleischer, S.; Zhou, S.; Junge, K.; Beller, M. Angew. Chem., Int. Ed. 2013, 52,
5120–5124; (e) Tlili, A.; Schranck, J.; Neumann, H.; Beller, M. Chem. Eur. J. 2012,
18, 15935–15939; (f) Zhang, G.; Scott, B. L.; Hanson, S. Angew. Chem., Int. Ed.
2012, 51, 12102–12106.
9. Bao, H.; Wang, D.; Wang, X.; Cheng, C.; Li, Y.; Hu, Y. RSC Adv. 2015, 5, 47125–
47130.
10. (a) Cai, S.; Wang, D.; Niu, Z.; Li, Y. Chin. J. Catal. 2013, 34, 1964–1974; (b) Yu,
W.; Porosoff, M. D.; Chen, J. G. Chem. Rev. 2012, 112, 5780–5817; (c) Liu, X.;
Wang, D.; Li, Y. Nano Today 2012, 7, 448–466; (d) Coqa, B.; Figueras, F. J. Mol.
Catal. A 2001, 173, 117–134.
11. Zhou, J.; Yang, Y.-B.; Yang, C.-R. Acta Chim. Sinica 1980, 38, 162–166.
12. Typical procedure for hydrogenation of 3,4,5-trimethoxybenzyl-aldehyde (1a) to
3,4,5-trimethoxybenzyl alcohol (2a). A mixture of 1a (196 mg, 1 mmol) and 1%
Pd/Ni bimetallic catalyst9 (60 mg, 30 wt %) in MeOH (10 mL) was stirred under
H2 at room temperature and atmospheric pressure (on an atmospheric
pressure hydrogenation apparatus) until the absorption of hydrogen ceased
(3.5 h). After the catalyst was removed off by a magnetic stirring bar, the
solution was evaporated in a vaporator to give the product 2a as a colorless oil
(195 mg, 98%), which is pure enough for 1H and 13C NMR determinations. 1H
NMR (300 MHz, CDCl3) d 6.59 (s, 2H), 4.61 (d, 2H, J = 5.49 Hz), 3.86 (s, 6H), 3.83
(s, 3H), 1.92 (t, 3H, J = 5.49 Hz); 13C NMR (75 MHz, CDCl3) d 153.3, 137.2 (2C),
136.6 (2C), 103.7, 65.4 (2C), 60.8, 56.0.
In conclusion, the catalytic hydrogenation of the active ben-
zaldehydes to the corresponding benzyl alcohols is a challenging
task in organic synthesis because the further catalytic hydrogenol-
ysis of the benzyl alcohols into the corresponding methylbenzenes
proceeds under the same conditions. However, this problem can be
solved completely by simply using Pd/Ni bimetallic nanoparticles
as a catalyst. The highly chemoselective hydrogenation may have
resulted from the fact that the polarized Pd/Ni bimetallic catalyst
has weak adsorptivity for the heteroatom single bonds. When this
hydrogenation was used as the key step in the total synthesis of
The same procedure was used to chemoselectively hydrogenate the substrates
1b–1t to 2b–2t. In some cases, the flash chromatography was required for the
purification of the products.