The solid-state ion exchange (SSIE) of CuCl with solid acid supports to produce CuI supported catalysts is an important reaction. CuCl has a low melting point and is easily dispersed over the support surface when heated, with the spontaneous formation of either a mono or multilayer [1, 2, 3, 4, 5]. He et al. [2] modified mesoporous silica (SBA-15) by incorporating alumina and found that the resulting material could be used to support CuCl. The same study determined that isolated cuprous species generated by SSIE between CuCl and Brönsted acid sites formed on the silica host were effective at promoting the dispersion of CuCl. The ion exchange of the CuI in solid CuCl with the H+ in HY zeolite has been shown to occur at temperatures in excess of 300 °C, with the maximum ion exchange rate achieved at 340 °C. During this process, HCl gas is produced and any excess CuCl sublimates from the solid catalyst surface [6, 7].
King [8, 9] reported that a CuI/Y catalyst prepared by the SSIE of CuCl and HY zeolite was more active and more stable than catalysts prepared by the solution ion exchange of Cu(NO3)2 with HY zeolite during the oxidative carbonylation of methanol to dimethyl carbonate (DMC). Anderson et al. [10, 11] prepared CuI/X and CuI/ZSM-5 catalysts by the SSIE of CuCl and HX or HZSM-5 zeolites and found that the reduced adsorption of CO on CuI/X as compared with CuI/ZSM-5 favored the synthesis of DMC. The Cu-O species formed on these materials were proposed as the active species during the catalytic reaction [12, 13, 14]. DMC was the primary product over CuI/Y, whereas dimethoxy methane (DMM) was the majority product generated over CuI/ZSM-5 and CuI/MOR. The higher activity and selectivity of the CuI/Y catalyst were attributed to the weaker adsorption of CO onto the CuI cations in this material. Huang et al. [15] prepared a series of CuI/FAU catalysts with varying SiO2/Al2O3 molar ratios and found that the locations of the Cu active sites were related to the distribution of Brönsted acid sites, which also influenced the performance of the catalyst during oxidative carbonylation. In addition to the H-type zeolite, other Brönsted acids have also been used to prepare heterogeneous CuΙ catalysts by SSIE with CuCl; studies have investigated the use of SiO2-Al2O3, SiO2-TiO2, SO42-/ZnO, and S2O82-/ZnO solid acids to prepare CuI/SiO2-Al2O3 [16], CuI/SiO2-TiO2 [17, 18], CuI/SO42-/ZnO, and CuI/S2O82-/ZnO [19] catalysts, respectively. The obtained catal ysts exhibited excellent catalytic activity during the oxidative carbonylation of methanol to DMC.
Notwithstanding the substantial utility of CuCl, it is well known that this material readily oxidizes in air to form Cu(II), and thus it is difficult to be produce commercially. CuCl2, however, is more stable in air and thus easier to obtain and less expensive than CuCl. In addition, CuCl2 quickly decomposes to CuCl and Cl2 gas when heated to high temperatures. We have therefore previously attempted to prepare a CuI catalyst by the SSIE method using CuCl2 as the Cu source instead of CuCl [20]. In this prior work, the catalyst formation process as well as the microstructure and active site locations of the catalyst were not completely elucidated, and consequently a series of systematic experiments were performed in the present study to obtain further information with regard to these aspects of the catalyst.
NaY zeolite (Si/Al = 8.1, Qilu Branch Research Institute, Sinopec, China) was twice subjected to ion exchange with a NH4NO3 solution (0.5 mol/L) for a 4 h time span and then calcined at 500 °C in air to produce the HY zeolite. CuCl2·2H2O was subsequently thoroughly mixed with either the HY or NaY zeolite (at a 1:1 mass ratio), and the mixture was heated in a tube furnace (Φ 60 mm x 1000 mm) at a rate of 5 °C/min to 650 °C and held at that temperature for 4 h under N2. After cooling to ambient temperature, the solid materials, designated as either the CuI/Y or CuI/NaY catalyst, were obtained. The same procedure was also applied to mixtures of CuCl2·2H2O or CuCl with HY zeolite (again at a 1:1 mass ratio), which were heated at various temperatures under N2 to obtain either Cu/Y(CuCl2) or Cu/Y(CuCl) catalyst.
Thermogravimetric (TG/DTG) analysis was carried out using a Netzsch STA409C instrument (Selb, Germany), applying a heating rate of 10 °C/min under N2 (50 mL/min). X-ray photoelectron spectroscopy (XPS) was performed with an ESCALAB-250 spectrometer at a chamber pressure of 7.0 × 10-8 Torr with an X-ray beam generated by bombarding an Al target with electrons at 1486.6 eV. Samples were fixed onto double-sided adhesive tape and then placed into a specimen holder. A C 1s binding energy of 284.6 eV was adopted as the internal reference. Elemental analysis (EA) of Cu in the solid materials was carried out by the iodometric method using an SP-723P spectrophotometer while the mercury thiocyanate spectrophotometric method provided in the DZG-93-01 standard was applied for analysis of the Cl content of the catalysts.
The synthesis of DMC via the vapor phase oxidative carbonylation of methanol was investigated at atmospheric pressure in a fixed-bed reactor (Φ 6 mm x 450 mm). In a typical process, 0.45 g catalyst sample (approximately 1.0 mL) was packed into the reactor tube and placed in the center of the reactor furnace. A thermocouple was inserted into the middle of the furnace to ensure accurate control of the reaction temperature. The reactor was heated from room temperature to 140 °C over 40 min and then held at that temperature for the reaction. The furnace temperature was controlled with a precision of ± 0.5 °C by a time programmed temperature controller (SK-II, Shanxi Institute of Coal Chemistry, Chinese Academy of Sciences). Liquid methanol was pumped at a rate of 0.02 mL/min using a constant flux micro-pump (2PBOOC, Beijing Weixing Co., China) and vaporized in a pre-heater at 130 °C. The resulting constant flux of methanol vapor was thoroughly mixed with CO (28 mL/min) and O2 (2.8 mL/min, standard atmosphere), each of which was controlled individually by mass gas flow controllers (MFC, Seven Star Huachuang Co., China). The hot gas mixture subsequently flowed into the catalyst bed where the catalytic oxidative carbonylation of methanol to DMC took place. The effluent gas was routed to an Agilent 6890 gas chromatograph (GC) for online analysis of the products. The GC was equipped with an HP-INNOWAX capillary column connected to a flame ionization detector (FID) as well as a series consisting of a PORAPAK-Q packed column and HP-PLOT/Q carbon sieve and HP-PLOT Molesieve/5A capillary columns connected to a thermal conductivity detector (TCD). Methanol, DMC, methylforma te (MF), DMM, and traces of dimethyl ether (DME) in the products were detected by the FID, whereas CO, O2, and CO2 were detected by TCD. The gaseous products were sampled and analyzed automatically every 20 min throughout the reaction. The catalytic activity was reported as methanol conversion as well as the space-time yield (STY) of DMC, while product selectivity (S) was measured as the sum of the quantities of DMC, DMM, and MF produced.
The TG/DTG plots obtained for the mixtures of HY or NaY zeolite with CuCl2·2H2O under N2 are shown in Fig. 1. To avoid the effects of moisture contained in the zeolites, samples were dehydrated by heating at 300 °C for 1 h prior to analysis. The anhydrous samples were then heated from 300 to 1000 °C at a rate of 1 °C/min. In the case of both materials, there are two obvious mass loss peaks in the DTG profile, occurring at approximately 438 and 780 °C. With regard to the NaY mixture, the first peak can be attributed to the release of Cl2 during the decomposition of CuCl2 to CuCl, while the second peak results from the sublimation of CuCl. In contrast, the HY mixture exhibits a more pronounced mass loss between 300 and 550 °C and a lower mass loss between 550 and 1000 °C. This is due to the SSIE reaction of CuCl formed from the decomposition reaction of CuCl2·2H2O with the HY zeolite [14]. The release of HCl results in the additional mass loss between 300 to 550 °C while the lower mass loss is caused by the subsequent sublimation of CuCl. Calculations based on these plots indicate that the extent of ion exchange that CuCl underwent with HY and the amount of CuI loaded on the Y zeolite during the heating process were 53.3% and 6.3 g/g, respectively.
The catalytic performance of the CuI/Y and CuI/NaY catalysts during the gas phase oxidative carbonylation of methanol to DMC were evaluated in the fixed-bed reactor, and the resulting STY and SDMC values are shown in Fig. 2. Both STY and SDMC increased in the initial reaction stages and reached equilibrium levels after approximately 3 h. The CuI/Y catalyst exhibited higher catalytic activity than the CuI/NaY catalyst; STY and SDMC were 100 mg/(g·h) and 72.56% over CuI/Y and 70 mg/(g·h) and 67.71% over CuI/NaY. In addition to the highly dispersed CuCl present in the Y zeolite structure, the CuI/Y catalyst had also been loaded with ion-exchanged CuI. It therefore appears that both the CuCl and the ion-exchanged CuI are active during DMC formation and that the ion-exchanged CuI improves the catalytic activity.
The catalytic activities of the Cu/Y materials prepared from different Cu salts are compared in Fig. 3, which summarizes the STY and SDMC values obtained from catalysts prepared at different temperatures. These data suggest that the activities of the two Cu/Y catalysts increased with increasing processing temperatures. At a processing temperature of 250 °C, the Cu/Y(CuCl2) catalyst shows lower activity than Cu/Y(CuCl) although the former has a faster rate of increase of both STY and SDMC. At 450 °C, Cu/Y(CuCl2) exhibits better catalytic activity than Cu/Y(CuCl). The STY and SDMC of the Cu/Y(CuCl2) catalyst were 38.16 mg/(g·h) and 60.27%, respectively, both of which are higher than those of Cu/Y(CuCl). With further increases in processing temperature, Cu/Y(CuCl2) exhibits an improved STY and similar SDMC compared with the Cu/Y(CuCl) catalyst. Analysis of X-ray diffraction of the catalysts (not shown) indicated that higher processing temperatures resulted in improved dispersion of CuCl2 or CuCl on the catalyst surfaces. The specific surface areas of the prepared catalyst also increased, leading to the observed improvements in the catalytic activity of the Cu/Y materials.
Figure 4 presents the Cu 2p XPS spectra of the Cu/Y(CuCl2) catalyst prepared at various temperatures. These results demonstrate that the Cu 2p3/2 and Cu 2pl/2 binding energies for these catalyst samples were within the ranges of 930-938 and 950-957 eV, while the divalent Cu satellite peak was at 940-945 eV. Through curve-fitting of the Cu 2p3/2 spectra, two peaks attributed to CuI (932.4 eV)and CuII (at 934.4 eV) were identified, and the results are shown in Table 1. It indicates that with increasing processing temperatures the CuI content on the catalyst surfaces increased while the CuII decreased. As an example, the atom percentages of CuII on the surfaces of the catalysts prepared at 450 and 550 °C were 56.75% and 29.90%, respectively. Thus CuI rather than CuII is the dominant species on the catalyst prepared at 650 °C. The Cl content is also seen to decrease while the molar ratio of Cu to Cl increased with rising processing temperatures. The apparent presence of CuII on the surfaces of catalysts prepared at 450 and 550 °C could be due to incomplete decomposition of CuC12 because these materials exhibit surface Cu/Cl atomic ratios below 1. On the surface of the Cu/Y(CuCl2) catalyst prepared at 650 °C, only CuI was present and the Cu/Cl atomic ratio was 1.08, indicating that the Cu on the surface of this catalyst is primarily in the form of CuCl.
The Cu/Y samples prepared from different Cu salts were analyzed by elemental analysis, with the results shown in Table 2. The Cu content in Cu/Y(CuCl) catalyst is higher than that of Cu/Y(CuCl2). However, both have Cu/Cl atom ratios above 1, and this ratio increases further with higher heating temperatures. Comparison of the Cu/Cl atom ratios indicates that the ratio calculated on the basis of the XPS surface elemental analysis is less than 1, while the ratio in the bulk phase of the catalyst is much higher. The dominant compounds on the catalyst surface are therefore CuCl and CuCl2, while the ion-exchanged CuI is in the internal cage structure of the zeolite. Thus Cu exists in the form of both CuCl and CuCl2 on the external surface of the Y zeolite, while the ion-exchanged CuI and small amounts of adsorbed CuCl are in the Y zeolite cage structures [21]. The catalyst prepared at 650 °C has only CuI derived from the small amount of adsorbed CuCl (0.09 mol/g) and ion-exchanged CuI (0.1 mol/g).
Figure 5 shows the dependence of the steady-state catalytic activity on the Cu loading of the Cu/Y(CuCl2) catalysts. With increases in the CuCl2·2H2O content, the STY and selectivity for DMC increased, although this effect decreased at CuCl2·2H2O contents above 30 wt%. Both STY and SDMC eventually plateaued and did not increase any further with higher CuCl2·2H2O content, likely owing to the saturation of Brönsted acid sites with Cu+ from dispersed CuCl in the zeolite.
Figure 6 depicts the dependence of the catalytic performance on reaction time over Cu/Y(CuCl2) catalyst. It shows that the formation rate of DMC increases with the on-stream time, reaching an equilibrium value after 3 h. After 200 h of reaction time, the STY and selectivity for DMC were approximately 85 mg/(g·h) and 80%, respectively. About 3 h was required for the MF and DMM formation rates to plateau, at which point their selectivities were 15% and 2%, respectively.
A Cu/Y catalyst was prepared by the SSIE method using a combination of CuCl2 and HY zeolite. The highly dispersed CuCl and ion-exchanged Cu+ in the zeolite were both active in DMC synthesis from methanol. CuCl and CuCl2 were found to co-exist on the external surface of the catalyst, while ion-exchanged CuΙ or small amounts of adsorbed CuCl were contained in the internal cage structure of the Y zeolite. Compared with the CuI/Y catalyst prepared by heating a mixture of CuCl and HY, the Cu/Y catalyst synthesized by heating a mixture of CuCl2 and HY showed higher catalytic activity during the oxidative carbonylation of methanol, even though the latter material had lower Cu and Cl contents.
2014, Vol. 35 Issue (1): 134-139
PDF (473KB)
2014, Vol. 35 Issue (1): 134-139
PDF (473KB)