Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Cu-doped zeolites for catalytic oxidative carbonylation: The role of Brønsted acids
Shouying Huang, Yue Wang, Zengzhu Wang, Bing Yan, Shengping Wang, Jinlong Gong, Xinbin Ma ∗
Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China
a r t i c l e i n f o a b s t r a c t
Article history: The aim of this work was to establish the role of Brønsted acidic sites in Cu-doped zeolites for the oxidative
Received 7 November 2011 carbonylation of ethanol to diethyl carbonate (DEC). In order to eliminate the influences of other factors
Received in revised form such as channel structure, faujasite (FAU) with different SiO2 /Al2 O3 ratios and Beta before and after passi-
20 December 2011
vation were investigated. Fourier Transform infrared spectroscopy (FTIR) of pyridine adsorption indicated
Accepted 28 December 2011
that all of Brønsted acidic sites were exchanged with CuCl to form effective active sites for oxidative car-
Available online 8 January 2012
bonylation. Characterization data showed that passivation on Beta zeolites has little effect on surface
area and pore volume distribution except Brønsted acidic sites. The relationship between Brønsted acidic
Keywords:
Oxidative carbonylation
sites and catalyst activity was built on different types of zeolite supports. Diffuse reflectance infrared
Cu-doped zeolite Fourier transform spectroscopy (DIRFTS) of CO adsorption revealed that the location of Cu active sites
Brønsted acids was related to the distribution of Brønsted acidic sites, which also influenced catalytic performance of
Dialkyl carbonate oxidative carbonylation.
© 2012 Elsevier B.V. All rights reserved.
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.12.043
S. Huang et al. / Applied Catalysis A: General 417–418 (2012) 236–242 237
environment of Cu species in the catalysts. We note that the local mixture of ethanol/CO/N2 /O2 with a molar ratio of 4.8/10/6.5/1
environment and concentration of Cu [21], residual Brønsted acid- and a total flow rate was 89.2 ml/min. The reaction products were
ity [22,23], and channel structure [24] of the zeolite supports are directly analyzed by an on-line coupled gas chromatography (Agi-
all crucial factors to determine catalytic performance. Therefore, lent 7890A GC), which was equipped with DB-624 capillary column
architecture of channel and acidity are always considered in par- connected to a FID and two Porapack Q and one Molecular sieve
allel. In order to solely determine the role of Brønsted acidity, it 13X columns connected to a TCD detectors at 30 min intervals. The
is necessary to design experiments for eliminating the influences byproducts observed including acetal and acetaldehyde (Eqs. (4)
of other factors such as channel structure. This paper, there- and (5), respectively), and the selectivity of DEC was determined
fore, describes an investigation exploring the relationship between by Eq. (6):
Brønsted acidity of zeolite supports and catalytic performance for
oxidative carbonylation. 3CH3 CH2 OH + 1/2O2 = (CH3 CH2 O)2 CH2 CH3 + 2H2 O (4)
0.1 a 0.5 b
SiO2/Al2O3=2.5 SiO2/Al2O3=2.5
SiO2/Al2O3=7.1 SiO2/Al2O3=7.1
Absorbance
SiO2/Al2O3=10.1 SiO2/Al2O3=10.1
SiO2/Al2O3=19 SiO2/Al2O3=19
Fig. 1. FTIR spectra of pyridine adsorbed on H-FAU (a) and Cu-FAU (b) with different SiO2 /Al2 O3 ratios.
which catalytic behaviors were examined for oxidative carbonyla- Lewis sites. The bands at ∼1450 cm−1 and ∼1540 cm−1 are often
tion of ethanol. Generally, zeolites with SiO2 /Al2 O3 ratios greater used to determine the number of acid sites quantitatively, which
than 20 possess the similar acidity [26], thus we chose SiO2 /Al2 O3 attributed to the C C stretching vibration of pyridine complex on
ratios of 2.5, 7.1, 10.1, and 19 for detailed examination. The effects Lewis sites and pyridinium ion on Brønsted sites. With increasing
of pore distribution and specific surface area have been taken into of SiO2 /Al2 O3 ratio, intensity of the band at 1540 cm−1 increased
account. The results of N2 adsorption at low temperature (not pre- first followed by a decrease. The NH3 -TPD profiles of protonated
sented) showed that pore width and pore volume calculated using FAU with different SiO2 /Al2 O3 ratios were shown in Fig. 2 and
the Horvath–Kawazoe method varied in a small range and the trend the total peak areas were listed in Table 2. In aluminosilicate-type
was not associated with the catalytic performance. zeolites, the substitution of Si atoms for Al atoms at tetrahedral
To investigate the nature of acid sites in zeolites, several basic positions creates excess negative charge and then performs the
probe molecules with different sizes were employed such as nature of acid sites, and hence the concentration of acid sites deter-
ammonia, d3 -acetonitrile, trimethylamine, and pyridine. Pyridine mined by NH3 -TCD signal varies linearly with the reduction of
was most used because of its sensitivity in IR spectrum. However, SiO2 /Al2 O3 ratio. The trend is somewhat different from the result
not all of acidic sites are accessible to absorb pyridine considering from FTIR spectroscopy of pyridine adsorption due to inaccessibil-
the steric hindrance. The framework structure of FAU zeolite con- ity of acidic sites in sodalite cages and hexagonal prism. Due to
sists of supercage with 12-ring window (7.4 Å), sodalite cage with the steric hindrance, the amounts of Brønsted acidic sites detected
6-ring (2.2–2.6 Å), and hexagonal prism, and the kinetics diame- by pyridine is less than the total quantity of Brønsted acidic sites
ter of pyridine is about 5 Å. It has been shown that pyridine can over HY. For synthesis of organic carbonates by oxidative car-
only probe the acidic sites in supercages of FAU zeolites, while the bonylation, the accessibility to active sites on various locations in
acid sites located in the sodalites and prisms units do not interact Cu-doped zeolite catalyst is also a crucial factor of consideration.
with this probe molecule. On the other hand, FAU has three gen- Upon proton exchange with Cu ions via the SSIE process, the band
eral cation-exchange sites which can be occupied by Cu. Among at 1540 cm−1 and 1630 cm−1 (Fig. 1b) characteristic of Brønsted
them, only those copper cations in supercages are accessible to sites disappeared completely. Bell and co-workers [17] reported
reactants, and hence contribute to the activity of catalyst in synthe- that complete exchange of Cu+ cations with Brønsted acid pro-
sis of DMC [21]. For oxidative carbonylation of ethanol, the reactant tons was achieved on ZSM-5 zeolite only when SSIE temperature
and product molecules are both relatively larger, so that geometric
constraints could be expected. Therefore, pyridine is an appropriate
probe molecule to detect the acidic properties of protonated FAU
in this work, although the number of Brønsted acidic sites deter-
mined in this manner is fewer than that expected from the Si/Al SiO2/Al2O3=2.5
TCD Signal (a.u.)
Table 1
The chemical composition of Cu-FAU with different SiO2 /Al2 O3 ratios.
Table 2
Amount of acid sites of FAU zeolites and apparent activation energies for oxidative carbonylation of ethanol over corresponding Cu-FAU.
is over 1023 K and Cu/Al ratio reached approximately 1. Compara- has been shown that the deposition of thin silica layers by CVD/CLD
tively, Anderson et al. [23] considered that Cu-Y has a small amount from TEOS can neutralize external sites without modifying the
of residual acidity even when the Cu/Al ratio is 1. In our case, internal surface of zeolite [31]. For this reason, influence of Brønsted
the absence of 1540 cm−1 band indicated that complete exchange acidity can be investigated by reducing Brønsted acidic sites. Upon
between H proton in supercages of H-FAU and CuCl occurred at passivation, no obvious differences were observed in position and
823 K. EDX analysis was carried out to analyze the chemical com- intensity of the characteristic peaks in FTIR (Fig. 4) and XRD spectra
position of Cu-FAU catalyst. We put forward the results from two (Fig. 5), which reveals that zeolite framework keeps intact under the
different points on catalyst surface for reliability of data. As shown treatment. A slight increase in the intensity of the peak at 2 = 7.6◦
in Table 1, Cu/Al molar ratios are between 0.97 and 1.13, which indi-
cated that bridging hydroxyl groups functioning as Brønsted acid
Ethanol conversion,DEC yield (%)
was all occupied by Cu+ upon SSIE no matter where they located 2.5
(supercages, sodalite cage or hexagonal prism). It should be noted Ethanol conversion
that the content of Cu increased with the reduction of SiO2 /Al2 O3 , 2.0
which shows a good agreement with the acidity evaluation by NH3 -
TPD, but not FTIR spectroscopy of pyridine adsorption, since partial 1.5
Cu species enters into sodalite cage and hexagonal prism [21]. Con- DECyield
sidering accessibility and agglomeration of Cu species, it is not 1.0
reasonable to relate all exchanged-Cu sites to the formation of DEC.
Additionally, the band around 1440–1450 cm−1 became sharp and 0.5
a blue shift was observed upon the Cu exchange, indicating the
strong pyridine-Cu+ interaction, as also shown in previous inves-
0.0
tigations [28,29]. Thus, the intensities of 1450 cm−1 could reflect 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
the degree of proton exchange for Cu+ . Concentration of Brønsted Amount of B acid sites(mmol/g)
and Lewis acidic sites obtained from the intensity of corresponding
bands is summarized in Table 2. Fig. 3. The relationship between Brønsted acidic sites on H-FAU and catalytic activ-
Based on the mechanism of CuCl exchange with H-zeolite ity of Cu-FAU catalysts.
reported earlier and the results of characterization presented
above, we presumed that it is more appropriate to establish a cor-
relation between Brønsted acidic sites detected by pyridine and
activity rather than Cu content in catalyst. Catalytic activities were 0.2
tested using Cu-FAU with various SiO2 /Al2 O3 ratios in a continu-
Absorbance
Table 3
Amount of acid sites of H and TEOS-treated H and catalytic activity of Cu- on oxidative carbonylation of ethanol.
is due to the formation of Si O Si bonds by silanol groups react- compared to the sample without treatment. The variation was also
ing with adjacent ones. The similarity of N2 adsorption/desorption in line with the reduction of Brønsted acidic sites. It indicates that
curves of H and H-passivated zeolites indicates that the passiva- Brønsted acidic sites on both internal and external surface of H
tion has little effect on surface area and pore volume distribution could form active sites for oxidative carbonylation.
(Fig. 6). These results suggest that TEOS was finely dispersed on From these results, it is apparent that Brønsted acidity of zeolite
zeolite surfaces and formed silica layer was thin and uniform. Thus, supports, which presents in large cavities and accessible to pyri-
inner pore system could be still accessible for diffusion of reactants dine molecule absorbed, dominates the catalytic performance of
and products after passivation [31]. Fig. 7 presents the pyridine Cu-exchanged zeolites on oxidative carbonylation quantitatively.
adsorbed IR spectra of untreated and passivated H at 423 K. A dra- Therefore, it is rational to calculate turnover frequencies (TOFs)
matic decrease in band intensity at 1540 cm−1 and 1490 cm−1 was of DEC formation based on the amount of Brønsted acidic sites
observed, which correspond to Brønsted acidic sites and overlap of determined using pyridine adsorption infrared spectroscopy.
both Brønsted and Lewis acidic sites, respectively. Comparatively, The temperature dependence on reactivity in the form of the
the band at around 1450 cm−1 increased to a certain extent, which Arrhenius behavior is shown in Fig. 8, which is comparable to that
might be ascribed to the formation of tricoordinated framework of Cu-Y catalysts for the synthesis of DMC as reported previously
silicon [32]. According to a study by Kunkeler et al. [31], a rough [21]. The calculated apparent activation energies (Ea ) (Table 2) of
external surface makes microcrystalline  less favorable for the the DEC formation were in a range of 30–60 kJ/mol, consistent with
formation of Si layer compared to macrocrystalline . However, we those reported for relevant systems [6,33,34]. Under the same reac-
show here that the TEOS-treatment successfully reduced external tion conditions, Ea decreased in the order of Cu-FAU (10.1) < Cu-FAU
Brønsted surface acidity although the sample is microcrystalline (7.1) < Cu-FAU (2.5) < Cu-FAU (19), which follows the same trend as
particle. Therefore, the change of acidity is consistent with the dis- order of amounts of Brønsted acidic sites on the surface of zeolites.
tribution of Brønsted and Lewis acidic sites on the H zeolite [30]. We should also mention that distribution and location of
The acid properties of H before and after TEOS deposition and the Brønsted acidic sites have effects on the formation of metal active
results of catalytic performance tests are shown in Table 3. Passiva- sites and consequently the catalytic activity and stability in metal-
tion resulted in a loss of ∼50% of ethanol conversion and DEC yield exchanged zeolite catalyst. CO prefers to interact with Cu(I) cations
at ambient temperature, but not adsorb on Cu(II) sites and form sta-
ble Cu–CO species [35]. In our case, Cu+ cations have been proved
to exchange with Brønsted acidic sites, therefore we used DIRFTS of
CO adsorption to discriminate the location and state of active sites
in different positions inside metal-exchanged zeolites. The FTIR
spectra of CO adsorbed on FAU zeolites with varying SiO2 /Al2 O3
Intensity
were presented in Fig. 9. After purging with helium for 30 min, the
bands in the range of 2140 cm−1 to 2160 cm−1 did not vanish, which
demonstrated the presence of Cu(I)–CO species. CO adsorption
Hβ-passivated spectra of all samples lead to the appearance of two overlapping
band at ∼2162 cm−1 and 2148 cm−1 , which can be assigned to Cu+
Hβ cations positioned in sites II* and II of FAU, respectively [36]. The
relative intensity at the ∼2160 cm−1 band increased first and then
decreased with an increase of Si2 O3 /Al2 O3 ratio obviously. The
10 20 30 40 50
trend is in good agreement with the concentration of Brønsted
2θ (º)
acidic sites and catalytic activity. It is very likely that, upon CO
Fig. 5. XRD patterns of H zeolite before and after passivation.
0.07
0.1
0.06
dv/dp(cm3/g·Å)
0.05
Absorbance
Hβ
0.04
Hβ-passivated
0.03
0.02
Hβ
0.01
0.00 Hβ-passivated
0 20 40 60 80 100 1750 1700 1650 1600 1550 1500 1450 1400 1350
-1
dp (Å) Wavenumber (cm )
Fig. 6. Effect of passivation on H pore volume distribution. Fig. 7. FTIR spectra of pyridine adsorbed on H before and after passivation.
S. Huang et al. / Applied Catalysis A: General 417–418 (2012) 236–242 241
1.2E-3
1E-3
8E-4 Cu-Y(SiO2/Al 2O3=2.5) Cu-Y(SiO2/ Al 2O 3=7.1)
6E-4
4E-4
TOF (s-1)
2E-4
1.2E-3
1E-3 Cu-Y(SiO2/Al 2O3=10.1) Cu-Y(SiO2/ Al 2O 3=1 9)
8E-4
6E-4
4E-4
2E-4
2.35 2.40 2.45 2.50 2.55(2.35) 2.40 2.45 2.50 2.55
-1
1000/T (K )
Fig. 8. Arrhenius plot of turnover frequencies of DEC formation obtained over Cu-FAU catalysts.
2250 2200 2150 2100 2050 Financial support from the National Natural Science Founda-
wavenumber (cm-1) tion of China (20876112, 20936003), Specialized Research Fund
for the Doctoral Program of Higher Education (SRFDP) (grant no.
Fig. 9. FTIR spectra of CO adsorption on Cu-FAU with varying Si2 O3 /Al2 O3 ratios. 20090032110021), the Program for New Century Excellent Tal-
ents in University (NCET-04-0242), Seed Foundation of Tianjin
University (60303002), and the Program of Introducing Talents of
adsorption, Cu+ migrates from sites II to more-exposed positions Discipline to Universities (B06006) are gratefully acknowledged.
(site II*) near the supercage [37]. Therefore, it indicated that the
more intensive distribution of Brønsted acidic sites in supercage
made Cu migration more easily, which was also relative to activ- References
ity of synthesis of diethyl carbonate. Furthermore, two shoulders
[1] A.A.G. Shaikh, S. Sivaram, Chem. Rev. 96 (1996) 951–976.
at 2111 cm−1 and 2190 cm−1 were also registered on the sample [2] I.T. Horváth, Green Chem. 10 (2008) 1024–1028.
with Si/Al ratio of 19. The band at 2111 cm−1 was also observed on [3] G. Rebmann, V. Keller, M.J. Ledoux, N. Keller, Green Chem. 10 (2008) 207–213.
[4] Y. Cao, J.C. Hu, P. Yang, W.L. Dai, K.N. Fan, Chem. Commun. (2003) 908–909.
partly reduced CuO samples [38]. Jana et al. assigned this band to
[5] S.T. King, Catal. Today 33 (1997) 173–182.
Cu+ species in oxide-like ambiance. Excess CuCl agglomerates to [6] Y.H. Zhang, I.J. Drake, D.N. Briggs, A.T. Bell, J. Catal. 244 (2006) 219–229.
form CuOx in FAU (19), due to its low concentration of Brønsted [7] J.L. Gong, X.B. Ma, S.P. Wang, Appl. Catal. A 316 (2007) 1–21.
acidic sites, which cannot offer enough exchangeable protons [36]. [8] P.B. Zhang, S.Y. Huang, Y. Yang, Q.S. Meng, S.P. Wang, X.B. Ma, Catal. Today 149
(2010) 202–206.
The results of CO adsorption indicated that the nature of Brønsted [9] D.N. Briggs, K.H. Lawrence, A.T. Bell, Appl. Catal. A 366 (2009) 71–83.
acidic sites influenced the state and location of Cu active sites and [10] D.N. Briggs, G. Bong, E. Leong, K. Oei, G. Lestari, A.T. Bell, J. Catal. 276 (2010)
thus catalytic activity. 215–228.
[11] S. Telalović, J.F. Ng, R. Maheswari, A. Ramanathan, G.K. Chuah, U. Hanefeld,
Chem. Commun. 47 (2008) 4631–4633.
[12] Y.J. Gong, T. Dou, S.J. Kang, Q. Li, Y.F. Hu, Fuel. Process. Technol. 90 (2009)
5. Conclusions
122–129.
[13] N. Katada, K. Suzuki, T. Noda, G. Sastre, M. Niwa, J. Phys. Chem. C 113 (2009)
In summary, we have established a quantitative relationship 19208–19217.
[14] H. Huo, L. Peng, C.P. Grey, J. Phys. Chem. C 115 (2011) 2030–2037.
between amount of Brønsted acidic sites and catalytic activities
[15] N. Keller, G. Rebmann, V. Keller, J. Mol. Catal. A: Chem. 317 (2010) 1–18.
for oxidative carbonylation. Considering aggregation and accessi- [16] C. Lamberti, A. Zecchina, E. Groppo, S. Bordiga, Chem. Soc. Rev. 39 (2010)
bility of Cu species on zeolites, it is more reasonable to relate the 4951–5001.
Brønsted acidic sites to catalytic activity rather than Cu contents. [17] I.J. Drake, Y.H. Zhang, A.T. Bell, Chem. Mater. 18 (2006) 2347–2356.
[18] Z. Li, K.C. Xie, R.C.T. Slade, Appl. Catal. A 209 (2001) 107–115.
Brønsted acidic sites on both internal surface and external surface [19] A.Z.G. Spoto, S. Bordiga, G. Ricchiardi, G. Martra, Appl. Catal. B 3 (1994) 151–172.
of zeolite, which exchange with Cu+ during the SSIE process, form [20] Y. Kuroda, T. Mori, Y. Yoshikawa, Chem. Commun. (2001) 1006–1007.
242 S. Huang et al. / Applied Catalysis A: General 417–418 (2012) 236–242
[21] Y.H. Zhang, I.J. Drake, D. Briggs, B. Lim, T. Chau, T. Alexis Bell, J. Phys. Chem. B [30] J.C. Jacobus, E.J. Creyghtonb, S.L. Njoa, H.V. Koningsveld, H.V. Bekkum, Catal.
110 (2006) 11654–11664. Today 38 (1997) 205–212.
[22] T. Blasco, M. Boronat, P. Concepción, A. Corma, D. Law, J.A. Vidal-Moya, Angew. [31] P.J. Kunkeler, D. Moeskops, H.V. Bekkum, Microporous Mesoporous Mater. 11
Chem. Int. Ed. 46 (2007) 3938–3941. (1997) 313–323.
[23] S. Anderson, S. Manthata, T. Root, Appl. Catal. A 280 (2005) 117–124. [32] J. Cejka, A. Corma, S. Zones, Zeolites and catalysis, Synthesis, Reactions and
[24] Y.H. Zhang, D.N. Briggs, E. Desmit, A.T. Bell, J. Catal. 251 (2007) 443–452. Applications, Wiley, 2010, 527.
[25] C.A. Emeis, J. Catal. 141 (1993) 347–354. [33] I.J. Drake, K.L. Fujdala, A.T. Bell, T.D. Tilley, J. Catal. 230 (2005) 14–27.
[26] G. Ertl, H. Knözinger, F. Schüth, J. Weitkamp, Handbook of Heterogeneous Catal- [34] X.B. Zheng, J. Phys. Chem. C 112 (2008) 5043–5047.
ysis, Wiley, 2008, 1103. [35] K. Hadjiivanov, H. Knözinger, Phys. Chem. Chem. Phys. 3 (2001) 1132–1137.
[27] N. Žilková, M. Bejblová, B. Gil, S.I. Zones, A.W. Burton, C.Y. Chen, Z. Musilová- [36] J. Engeldinger, C. Domke, M. Richter, U. Bentrup, Appl. Catal. A 382 (2010)
Pavlačková, G. Košová, J. Čejka, J. Catal. 266 (2009) 79–91. 303–311.
[28] M. Jiang, H.G. Karge, J. Chem. Soc. Perkin Trans. 91 (1995) 1845–1851. [37] G.T. Palomino, S. Bordiga, A. Zecchina, G.L. Marra, C. Lamberti, J. Phys. Chem. B
[29] F. Benaliouche, Y. Boucheffa, P. Ayrault, S. Mignard, P. Magnoux, Microporous 104 (2000) 8641–8651.
Mesoporous Mater. 111 (2008) 80–88. [38] B. Guido, J. Mol. Catal. 43 (1987) 225–236.