Catalytic hydrogenation of 2-nitro-2′-hydroxy-5′-methylazobenzene over solid base-hydrogenation bifunctional catalysts: Effect of alkali metals on Pd/γ-Al2O3

Article abstract of DOI:10.1016/j.catcom.2016.11.015

Alkali metals doped Pd solid base-hydrogenation bifunctional catalysts were prepared, characterized, and employed in the hydrogenation of 2-nitro-2′-hydroxy-5′-methylazobenzene. The results indicated that the basicity of catalyst endowed by the alkali met

Full text of DOI:10.1016/j.catcom.2016.11.015

Catalysis Communications 90 (2017) 35–38
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short communication
Catalytic hydrogenation of 2-nitro-2′-hydroxy-5′-methylazobenzene
over solid base-hydrogenation bifunctional catalysts: Effect of alkali
metals on Pd/γ-Al2O3
⁎
Leilei Si, Bowei Wang, Shipeng Chen, Jingru Hou, Xilong Yan, Yang Li , Ligong Chen
School of Chemical Engineering and Technology, Tianjin University, Tianjin 300350, PR China
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 August 2016
Received in revised form 8 November 2016
Accepted 10 November 2016
Available online 11 November 2016
Alkali metals doped Pd solid base-hydrogenation bifunctional catalysts were prepared, characterized, and
employed in the hydrogenation of 2-nitro-2′-hydroxy-5′-methylazobenzene. The results indicated that the ba-
sicity of catalyst endowed by the alkali metals indeed hindered the formation of amino by-products. Among
them, K doped catalyst exhibited the best catalytic performance and 84.79% benzotriazole selectivity was obtain-
ed. Furthermore, besides the good distribution of palladium particles, it is important for basic sites to distribute
around the palladium particles strictly, which is the guarantee of high benzotriazole selectivity.
© 2016 Elsevier B.V. All rights reserved.
Keywords:
Benzotriazole
Catalytic hydrogenation
Palladium catalyst
Alkali metal
1. Introduction
products, the understanding of structure-activity relationship is still in-
sufficient. So we attempted to find the key factor of these solid base-hy-
2-(2′-Hydroxy-5′-methylphenyl) benzotriazole (BTA) is one of the
most important ultraviolet absorbers in industry, which is widely used
for the protection of plastics against sunlight [1,2]. Currently, the
major preparation methods of BTA are the catalytic reduction of 2-
nitro-2′-hydroxy-5′-methylazobenzene (NAB) with reducing agents
[3–9]. Although good yield of BTA is achieved, the serious pollution is
still a big challenge [10]. In our previous work, we have successfully
established a continuous synthetic technology of BTA, which is not
only a green method but also an efficient process [11]. In addition,
solid base-hydrogenation bifunctional catalysts have been also success-
fully employed in this reaction to avoid using additional base and then
solve environmental problems caused by the use of extra lye [11,12].
It led to an entirely new line of BTA synthesis to use these kinds of bi-
functional catalysts. Meanwhile, alkali metals doped Pd catalysts have
always been one of the hot research topics [13,14]. However, this signif-
icant research is just getting started. There is still room for study and
improvement.
drogenation bifunctional catalysts by doping different kind of alkali
metals on Pd/γ-Al2O3. Does different kind of alkali metals produce the
same or similar impact on Pd/γ-Al2O3? What is the relationship be-
tween basic amount of these bifunctional catalysts and their catalytic
abilities? What is the key factor in the catalytic hydrogenation of NAB
to BTA over alkali metals doped palladium bifunctional catalysts? In
order to answer these questions, alkali metals (Na, K, Rb, Cs) doped
Pd/γ-Al2O3 were prepared and their catalytic performances were evalu-
ated. Moreover, these catalysts were characterized by Fourier transform
infrared spectroscopy (FT-IR), X-ray diffraction (XRD), Brunauer-
Emmer-Teller (BET) surface area measurement, temperature pro-
grammed desorption of carbon dioxide (CO2-TPD), transmission elec-
tron microscopy (TEM), and elemental mapping and temperature
programmed desorption of ammonia (NH3-TPD) to investigate the
structure-activity relationship.
In this work, further research on solid base-hydrogenation bifunc-
tional catalysts was carried out to explore the effect of different alkali
metals on Pd/γ-Al2O3. Although our previous work has confirmed that
K doped Pd catalysts can efficiently hinder the formation of amino by-
2. Experimental
2.1. Preparation of catalysts
Pd/γ-Al2O3 was prepared by the method involved in our previous
work [12]. The detailed process was put into the supplementary mate-
rial, and Pd/γ-Al2O3 doped by NaNO3 was named after Pd/γ-Al2O3-
NaNO3. The rest catalysts were named by same method.
⁎
Corresponding author at: School of Chemical Engineering and Technology, Tianjin
University, Tianjin 300350, PR China.
E-mail addresses: liyang777@tju.edu.cn (Y. Li), lgchen@tju.edu.cn (L. Chen).
http://dx.doi.org/10.1016/j.catcom.2016.11.015
1566-7367/© 2016 Elsevier B.V. All rights reserved.

36
L. Si et al. / Catalysis Communications 90 (2017) 35–38
2.2. Experimental procedure
The catalytic hydrogenation reaction was accomplished in a fixed-
bed reactor with an inner diameter of 15 mm and a length of 650 mm,
which was filled up with 40 mL catalysts with diameter of 3 mm and
height of 3 mm averagely. The solution of NAB in toluene (the concen-
tration of NAB is 5 wt%) was dosed into the reactor by the tranquil
flow pump. The hydrogenation reaction was proceeding at temperature
60 °C, H2 pressure 2.5 MPa, liquid hourly space velocity (LHSV)
0.23 h−1. The samples for analysis were collected continuously and
the frequency was once an hour. The composition of the reaction mix-
ture was analyzed by high performance liquid chromatography
(HPLC) with a column of Extend C18 (250 mm × 4.6 mm, 10 μm, Agilent
technologies, USA).
3. Results and discussion
3.1. FT-IR
Fig. 2. X-ray powder diffractograms of (a) γ-Al2O3, (b) Pd/γ-Al2O3, (c) Pd/γ-Al2O3-NaNO3,
(d) Pd/γ-Al2O3-KNO3, (e) Pd/γ-Al2O3-RbNO3 and (f) Pd/γ-Al2O3-CsNO3. (○), Na2O; (*) Al-
O-K; (Δ) K2O; (◊) Rb9O2; (□) RbNO3; ($) CsNO3; (&) Pd (111); (#) γ-Al2O3.
The FT-IR spectra of γ-Al2O3, Pd/γ-Al2O3, Pd/γ-Al2O3-NaNO3, Pd/γ-
Al2O3-KNO3, Pd/γ-Al2O3-RbNO3 and Pd/γ-Al2O3-CsNO3 are shown in
Fig. 1. As shown in Fig. 1, the peaks at 3450 cm−1 and 1632 cm−1
were caused by the O\\H stretching and bending vibration of adsorbed
water [12,15,16]. Compared with curves of γ-Al2O3 and Pd/γ-Al2O3,
there were two new peaks at 1561 cm−1 and 1401 cm−1 in the spectra
of curves of Pd/γ-Al2O3-NaNO3, Pd/γ-Al2O3-KNO3 and Pd/γ-Al2O3-
RbNO3. The two bands were attributed to CO23– ions weakly bounded
with M+ (M = Na, K, Rb) present on the surface of the catalyst. The
CO23– ions might come from K2O·CO2 species, which formed in the cata-
lyst calcination [12,16–19]. However, intensity of the two absorption
bands both weakened from curves of Pd/γ-Al2O3-NaNO3 to Pd/γ-
Al2O3-RbNO3 in sequence. The result was partly caused by which
RbNO3 was only partly decomposed into Rb9O2 [20]. The spectra of
curves of Pd/γ-Al2O3-CsNO3 did not display the two peaks because
CsNO3 loading in the Al2O3 did not decompose at 500 °C [21]. The
bands at 1113 cm−1 and 785 cm−1 in the all catalysts were caused by
AlO4 groups, and the bands at 581 cm−1 arose from AlO6 groups.
These results indicated the presence of Al-O-Al framework [12,22–24].
The shoulders at 1113 cm−1 and 785 cm−1 decreased gradually from
curves of γ-Al2O3 to curves of Pd/γ-Al2O3-CsNO3 and shifted to a
sharp wavelength due to the effect of M+ (M = Na, K, Rb, Cs) on the
Al\\O bonds (AlO4 groups) [22,25]. This phenomenon also proved the
all nitrates successfully load in the supports.
3.2. XRD
The X-ray powder diffractograms of γ-Al2O3, Pd/γ-Al2O3, Pd/γ-
Al2O3-NaNO3, Pd/γ-Al2O3-KNO3, Pd/γ-Al2O3-RbNO3 and Pd/γ-Al2O3-
CsNO3 are shown in Fig. 2. The typical diffraction peaks at 2θ value of
37°, 46°, and 67° were attributed to γ-Al2O3 in all catalysts [11,12,26–
28] (JCPDS 23-1009). The peaks at 2θ value of 40° in all curves cannot
categorically be attributed to the (111) plane of face-centered cubic
structure of Pd because the peaks of Al2O3 phases existed in the same
position and the intensity of Pd diffraction peak was too weak due to
the low loading [12,24,26,29]. Compared with Pd/γ-Al2O3, the three
peaks of Al2O3 in other catalysts were relatively weak, which indicated
that the interaction between alkali metals and Al2O3 affects the crystal
of Al2O3. The peak at 2θ value of 32° in the curve of Pd/γ-Al2O3-NaNO3
was ascribed to Na2O phases (JCPDS 23-0528). The peaks at 2θ value
of 21° and 29° in the pattern of Pd/γ-Al2O3-KNO3 were attributed to or-
thorhombic α-KAlO2 species formed on the catalyst surface [12,16,30].
The peaks at 2θ value of 31° were clearly observed in the curve of the
Pd/γ-Al2O3-KNO3 catalyst, which were ascribed to the K2O species [12,
19,25,31]. The peaks at 2θ value of 25° in the curve of the Pd/γ-Al2O3-
RbNO3 catalyst may be attributed to the Rb9O2 groups (JCPDS 28-
0935). The peaks at 2θ value of 28° and 30° in the curve of the Pd/γ-
Fig. 1. FT-IR spectra. (a) γ-Al2O3; (b) Pd/γ-Al2O3; (c) Pd/γ-Al2O3-NaNO3; (d) Pd/γ-Al2O3-
KNO3; (e) Pd/γ-Al2O3-RbNO3; (f) Pd/γ-Al2O3-CsNO3.
Fig. 3. CO2-TPD curves of (a) γ-Al2O3, (b) Pd/γ-Al2O3, (c) Pd/γ-Al2O3-NaNO3, (d) Pd/γ-
Al2O3-KNO3, (e) Pd/γ-Al2O3-RbNO3 and (f) Pd/γ-Al2O3-CsNO3.

L. Si et al. / Catalysis Communications 90 (2017) 35–38
37
Fig. 4. TEM image of (a) Pd/γ-Al2O3-KNO3 and elemental mapping of (b) Pd/γ-Al2O3-KNO3.
Al2O3-RbNO3 catalyst were ascribed to the RbNO3 phases [20] (JCPDS
78-0112). These results were compliance with that the RbNO3 did not
completely decompose at 500 °C [20]. The peak at 2θ value of 29° in
the curve of the Pd/γ-Al2O3-CsNO3 catalyst were attributed to CsNO3
phases (JCPDS 32-0252), indicating that the CsNO3 did not decompose
at 500 °C [21]. The peak was weak because the content of CsNO3 was
too low. The results were consistent with the FT-IR results.
hydrogenation of NAB to BTA was awful. Meanwhile, it was noteworthy
that palladium particles also partly aggregated on Pd/γ-Al2O3-RbNO3
catalyst (Fig. S2c, d). Nevertheless, the distributions of palladium parti-
cles on Pd/γ-Al2O3-KNO3 and Pd/γ-Al2O3-NaNO3 were much better
than those on Pd/γ-Al2O3-RbNO3 and Pd/γ-Al2O3-CsNO3. From Fig.
S2b, it was obvious that sodium element evenly distributed on the sup-
port, however the sodium particles did not strictly distribute around the
palladium particles. In the case of Pd/γ-Al2O3-KNO3 sample, the distri-
bution of potassium elements was similar with that of palladium parti-
cles [12], demonstrating that the basic sites were strictly dispersed
around the active sites of hydrogenation.
3.3. CO2-TPD
The CO2-TPD profiles of γ-Al2O3, Pd/γ-Al2O3, Pd/γ-Al2O3-NaNO3, Pd/
γ-Al2O3-KNO3, Pd/γ-Al2O3-RbNO3, Pd/γ-Al2O3-CsNO3 are presented in
Fig. 3, which were employed to probe the alkaline information of all cat-
alysts. From Fig. 3, the CO2-TPD signal of Pd/γ-Al2O3 was rather weak
due to its acidic nature [12,32], however, all alkali metals doped cata-
lysts displayed remarkable and broad desorption peaks, indicating the
presence of basic sites on the surface of Pd catalysts. Our previous
work has demonstrated that the basicity of catalysts endowed by mod-
ification with alkali metals can effectively prevent the formation of by-
products AC [12], which was further confirmed by this work. Among
all bifunctional catalysts, Pd/γ-Al2O3-NaNO3 exhibited the highest ba-
sicity amount, most probably because more Na atoms could support
on Pd/γ-Al2O3 under the same mass loading, compared with K, Rb,
and Cs. Moreover, the profile of Pd/γ-Al2O3-CsNO3 was completely dif-
ferent from other catalysts, because CsNO3 cannot decompose at 500 °C
[21]. The above analyses were consistent with FT-IR and XRD results.
Hence, in order to further understand the structure-activity relation-
ship, TEM and elemental mapping were carried out.
3.5. Catalytic performance
Pd/γ-Al2O3, Pd/γ-Al2O3-NaNO3, Pd/γ-Al2O3-KNO3, Pd/γ-Al2O3-
RbNO3 and Pd/γ-Al2O3-CsNO3 were employed for the catalytic hydroge-
nation of NAB to BTA without additional base in order to evaluate their
catalytic performance. As shown in Table 1, 2-amino-p-cresol, o-
phenylenediamine (AC) [35], 2-(2′-hydroxy-5′-methylphenyl)benzo-
triazole
N-oxide
(NO)
and
tetrahydro-2-(2′-hydroxy-5′-
methylphenyl)benzotriazole (THB) together with BTA were all detected
in the reaction mixture, which was confirmed by our previous work
(Scheme 1) [11,12]. As expected, the conversion of NAB was satisfactory
(N92%) over all catalysts, and the selectivity of by-products AC over Pd/
γ-Al2O3-MNO3 (M = Na, K, Rb, Cs) catalysts decreased sharply com-
pared with Pd/γ-Al2O3, indicating the importance of alkaline environ-
ment to this reaction [11,12,35]. Among them, the AC selectivity over
Pd/γ-Al2O3-KNO3 was only 4.78% which is the lowest. For the selectivity
of BTA, another important evaluation criterion, these four catalysts pre-
sented distinct catalytic performance. When Pd/γ-Al2O3-NaNO3, Pd/γ-
Al2O3-RbNO3, and Pd/γ-Al2O3-CsNO3 were used for this reaction, the se-
lectivity of BTA were 30.11%, 36.05% and 19.59%, respectively, which
were lower than that of Pd/γ-Al2O3. However, 84.79% selectivity of
BTA was obtained with Pd/γ-Al2O3-KNO3 as the catalyst. Meanwhile,
compared with Pd/γ-Al2O3-KNO3, even with Pd/γ-Al2O3, N17% THB se-
lectivity and 21% NO selectivity were provided over Pd/γ-Al2O3-NaNO3,
Pd/γ-Al2O3-RbNO3, and Pd/γ-Al2O3-CsNO3. Nevertheless, the NO selec-
tivity and THB selectivity over Pd/γ-Al2O3-KNO3 were only 6.54% and
0.59%, respectively. Obviously, Pd/γ-Al2O3-KNO3 exhibited the best
3.4. TEM
TEM image and elemental mapping of Pd/γ-Al2O3-KNO3 are shown
in Fig. 4. It could be found that palladium particles were present as
black dots on these bifunctional catalysts (Figs. 4a and S2a, c, e) [11,
12,33]. Among them, the palladium particles on Pd/γ-Al2O3-CsNO3 ap-
peared serious agglomeration, which was further confirmed by elemen-
tal mapping (Fig. S2f). As everyone knows, the agglomeration of
palladium particles tremendously restricts the catalytic activity [34].
Therefore, the performance of Pd/γ-Al2O3-CsNO3 in the catalytic
Table 1
Catalytic hydrogenation of NAB over catalysts.
Catalyst
Conversion (%)b
Selectivity (%)
AC
NO
BTA
THB
Othersc
a
Pd/γ-Al2O3
100.00
92.49
100.00
99.89
99.88
0.00
0.43
0.00
0.10
0.17
25.39
19.21
4.78
12.72
16.80
0.38
0.19
0.19
0.69
0.38
9.64
21.16
6.54
29.06
22.58
0.63
0.33
0.25
0.50
0.54
59.08
30.11
84.79
36.05
19.59
0.55
0.85
0.49
0.63
0.52
1.08
19.34
0.59
17.89
37.03
0.11
0.51
0.06
0.65
0.40
4.81
10.18
3.30
4.28
4.00
0.36
0.62
0.44
0.13
0.35
a
Pd/γ-Al2O3-NaNO3
Pd/γ-Al2O3-KNO3a [12]
a
Pd/γ-Al2O3-RbNO3
Pd/γ-Al2O3-CsNO3
a
a
Reaction conditions: temperature: 60 °C; hydrogen pressure: 2.5 MPa; liquid hourly space velocity (LHSV): 0.23 h−1. Each data point is an average of three or more runs.
The conversion of NAB.
Including contamination in raw material, errors of measurement instrument and trace impurities produced in the reaction.
b
c

38
L. Si et al. / Catalysis Communications 90 (2017) 35–38
Scheme 1. Catalytic hydrogenation of NAB.
catalytic performance on the catalytic hydrogenation of NAB to BTA.
Furthermore, it was found that no notable deactivation of Pd/γ-Al2O3-
KNO3 in activity was observed in 100 h (Fig. S1) [12].
Appendix A. Supplementary data
Supplementary data to this article can be found online at doi:10.
1016/j.catcom.2016.11.015.
Furthermore, it was interesting that the doping elements of the
same group influenced the catalytic performance of Pd/γ-Al2O3 diverse-
ly. According to Fig. 3 and Table 1, it was easy to find that the mild
strength basicity produced by Pd/γ-Al2O3-KNO3 is important for the re-
action. Combined with the TEM and the results of Table 1, the selectivity
of BTA over Pd/γ-Al2O3-KNO3 was much better than that over Pd/γ-
Al2O3-NaNO3 in the catalytic hydrogenation of NAB to BTA. So the distri-
bution of basic sites was important for this reaction.
Therefore, it can be speculated that there are two key factors in the
catalytic hydrogenation of NAB to BTA over alkali metals doped palladi-
um bifunctional catalysts without additional base. One is the distribu-
tion of palladium particles. The other one is the dispersion of basic
sites that must strictly distribute around the active sites of hydrogena-
tion. Only then can the active sites of hydrogenation and basic sites syn-
ergistically catalyze NAB to BTA in high selectivity.
References
[1] J.C. Crawford, Prog. Polym. Sci. 24 (1999) 7–43.
[2] J. Catalán, M.R. Torres, J.D. Tornero, J. Chem. Soc. Faraday Trans. 93 (1997)
1691–1696.
[3] B.M. Francis, US Patent 3018269.
[4] R.J. Boyle, US Patent 3230194.
[5] A.S. ALzaydien, J. Appl. Sci. 4 (2004) 136–139.
[6] L. C. Vishwakarma, US Patent 5675015.
[7] S. Tanimoto, T. Kamano, Synthesis 1986 (1986) 647–649.
[8] H.L. White, US Patent 4041044.
[9] J. Rody, A.C. Rochat, US Patent 4001266.
[10] B.H. Kim, S.K. Kim, Y.S. Lee, Y.M. Jun, W. Baik, B.M. Lee, Tetrahedron Lett. 38 (1997)
8303–8306.
[11] B. Wang, W. Fang, L. Si, Y. Li, X. Yan, L. Chen, S. Wang, ACS Sustain. Chem. Eng. 3
(2015) 1890–1896.
[12] B. Wang, L. Si, Y. Yuan, Y. Li, L. Chen, X. Yan, RSC Adv. 6 (2016) 16766–16771.
[13] V. Vishwanathan, V. Jayasri, P.M. Basha, N. Mahata, L. Sikhwivhilu, N.J. Coville, Catal.
Commun. 9 (2008) 453–458.
4. Conclusions
[14] W. Kim, Appl. Catal. A-Gel. 268 (2004) 77–82.
[15] W. Xie, H. Peng, L. Chen, Appl. Catal. A-Gel. 300 (2006) 67–74.
[16] X. Li, D. Yu, W. Zhang, Z. Li, X. Zhang, H. Huang, Appl. Catal. A-Gel. 455 (2013) 1–7.
[17] C. Murugan, H. Bajaj, R. Jasra, Catal. Lett. 137 (2010) 224–231.
[18] Y. Amenomiya, J. Catal. 55 (1978) 205–212.
[19] W. Xie, H. Li, J. Mol. Catal. A Chem. 255 (2006) 1–9.
[20] N. Uvarov, P. Vaněk, Y.I. Yuzyuk, V. Železný, V. Studnička, B. Bokhonov, V. Dulepov, J.
Petzelt, Solid State Ionics 90 (1996) 201–207.
In summary, Pd/γ-Al2O3 catalysts modified by alkali metals (Na, K,
Rb, Cs) were prepared and employed in the catalytic hydrogenation of
NAB to BTA. The results indicated that Pd/γ-Al2O3-KNO3 exhibited the
best catalytic performance among these bifunctional catalysts and
84.79% BTA selectivity was obtained.
All catalysts were further characterized by FT-IR, XRD, BET, CO2-TPD,
TEM, elemental mapping and NH3-TPD. The results of FT-IR and XRD in-
dicated that Pd/γ-Al2O3 catalysts were successfully doped by alkali
metals. The profiles of CO2-TPD demonstrated that the basicity of cata-
lyst endowed by modification with alkali metals can effectively hinder
the formation of by-products AC. And the mild strength basicity pro-
duced by Pd/γ-Al2O3-KNO3 is important for the reaction. Moreover,
the characterizations of TEM and elemental mapping revealed that
there are two key factors in the above mentioned reaction. One is the
distribution of palladium particles, and the other is the dispersion of
basic sites that must strictly distribute around the active sites of
hydrogenation.
[21] P. Campbell, M. Ortner, C. Anderson, Anal. Chem. 33 (1961) 58–61.
[22] L. Fernández-Carrasco, E. Vázquez, Fuel 88 (2009) 1533–1538.
[23] Y. Ono, J. Catal. 216 (2003) 406–415.
[24] S. Hinokuma, K. Murakami, K. Uemura, M. Matsuda, K. Ikeue, N. Tsukahara, M.
Machida, Top. Catal. 52 (2009) 2108–2111.
[25] H. Liu, L. Su, F. Liu, C. Li, U.U. Solomon, Appl. Catal. B-Environ. 106 (2011) 550–558.
[26] A. Maione, F. André, P. Ruiz, Appl. Catal. B-Environ. 75 (2007) 59–70.
[27] J.H. Zhu, Y. Wang, Y. Chun, X.S. Wang, J. Chem. Soc. Faraday Trans. 94 (1998)
1163–1169.
[28] C. Shi, P. Zhang, Appl. Catal. B-Environ. 115 (2012) 190–200.
[29] W. Xu, Z. Zhan, L. Di, X. Zhang, Catal. Today 256 (2015) 148–152.
[30] S.L. Husheer, J.G. Thompson, A. Melnitchenko, J. Solid State Chem. 147 (1999)
624–630.
[31] M. Kotwal, P. Niphadkar, S. Deshpande, V. Bokade, P. Joshi, Fuel 88 (2009)
1773–1778.
[32] W. Wang, Y. Li, X. Zhang, W. Fang, Y. Yang, Catal. Commun. 69 (2015) 104–108.
[33] X. Wen, G. Li, Q. Chen, H. Zhang, X. Ba, G. Bai, Ind. Eng. Chem. Res. 53 (2014)
11646–11652.
[34] C. Na-Chiangmai, N. Tiengchad, P. Kittisakmontree, O. Mekasuwandumrong, J.
Powell, J. Panpranot, Catal. Lett. 141 (2011) 1149–1155.
[35] W. Whitmore, A. Revukas, J. Am. Chem. Soc. 62 (1940) 1687–1693.
Acknowledgements
Financial support by the National Natural Science Foundation of
China (Grant No. 21476163) are gratefully acknowledged.