【正文】 thermodynamics as methanol synthesis is an extremely exothermic reaction. 1,2 For example, at 573 K and 50 bar, it is calculated by thermodynamics that theoretic maximum onepass CO conversion is around 20% for flowtype reactor when H2/CO=2. Also it is reported that the onepass CO conversion in the industrial ICI process is between 15 and 25%, even if H2rich gas is used (H2/CO =5,523573 K).3 Therefore, developing a lowtemperature process for methanol synthesis, which will greatly reduce the production cost and utilize the thermodynamic advantage at low temperature, is challenging and If conversion is high enough in methanol synthesis, recycling of the unreacted syngas can be omitted and air can be used directly in the reformer, instead of pure oxygen. Generally, lowtemperature methanol synthesis is conducted in the liquid phase. The BNL method first reported by Brookhaven National Laboratory (BNL), using a very strong base catalyst (mixture of NaH, acetate), realized this continuous liquidphase synthesis in a semibatch reactor at 373403 K and 1050 bar. However, a remarkable drawback of this process is that even a trace amount of carbon dioxide and water in the feed gas or reaction system will deactivate the strongly basic catalyst soon,4,5 resulting in high cost ing from the plete purification of the syngas from reformer, and reactivation of the deactivated catalyst. This is the main reason stopping the mercialization of this lowtemperature methanol synthesis method. Liquidphase methanol synthesis from pure CO and H2 via the formation of methyl formate has been widely studied, where carbonylation of methanol and successive hydrogenation of methyl formate were considered as two main steps of the 外文資料原文 2 Palekar et al. used a potassium methoxide/copper chromite catalyst system to conduct this liquidphase reaction in a semibatch reactor at 373453 K and 3065 Although the mechanism of BNL method is still controversial, a lot of researchers think that it is similar to the mechanism However, similar to that in the BNL method, in this process CO2 and H2O act as poisons to the strong base catalyst (RONa, ROK) as well and must be pletely removed from syngas, making mercialization of lowtemperature methanol synthesis difficult. Tsubaki et al. proposed a new method of lowtemperature synthesis of methanol from CO2/H2 on a Cubased oxide catalyst using ethanol as a kind of “catalytic solvent”, by which methanol was produced in a batch reactor at 443 K and 30 This new process consisted of three steps: (1) formic acid synthesis from CO2 and H2。 (2) esterification of formic acid by ethanol to ethyl formate。 but no difference in reaction behavior was observed. So using passivated catalyst reduced separately had no influence. In the reaction, a closed typical batch reactor with inner volume of 80 mL and a stirrer was used. The stirring speed of the propellertype stirrer was carefully checked to eliminate the diffusion resistance between gas, liquid, and solid phases. A desired amount of solvent and catalyst was added into the reactor. Then the reactor was closed and the air inside the reactor was purged by reactant gas. A pressurized mixture gas of CO (%), CO2 (%), and H2 (%) was introduced and then the reaction took place at the desired temperature. Ar of % in the feed gas was used as inner standard. After reaction, the reactor was cooled by icewater and then the gas inside the reactor was released very slowly and collected in a gasbag for analysis. The standard reaction conditions were as follows: catalyst= g。 reaction temperature=443 K。 initial pressure=30 bar。 catalyst g。 the difference ing from the influence of molar numbers of different alcoholic solvents can be ignored. Concerning the alcohols with the same carbon number but different structure, the second alcohol had highest activity, as shown in the reactions in 2propanol, 2butanol, and 2pentanol separately. 2Propanol exhibited highest activity among these three 2alcohols. For example, at 443 K, the total conversion in the solvent of 2propanol was high up to %, among which methanol and 2propyl formate yields accounted for % and %, respectively. For alcohols with larger spatial obstacle, the reaction had lower activity, as shown in the cases of isobutanol, tertbutyl alcohol, and cyclopentanol. In addition, for ethylene glycol and benzyl alcohol, no activity was observed. But the reason is not very clear now. On the reasons for different behaviors of the alcohols with the same carbon number but different structure, it is considered that different alcohol type affected step(2) by both the electronic effect and spatial effect. For 1butanol, the electron density of oxygen atom in ROH is lower. As a result, ROH attacked the carbon atom of HCOOCu, the intermedia