Report

Selective conversion of syngas to light olefins

See allHide authors and affiliations

Science  04 Mar 2016:
Vol. 351, Issue 6277, pp. 1065-1068
DOI: 10.1126/science.aaf1835

Small olefins from syngas

The conversion of coal or natural gas to liquid fuels or chemicals often proceeds through the production of CO and H2. This mixture, known as syngas, is then converted to hydrocarbons with Fischer-Tropsch catalysts. For the light olefins (ethylene to butylenes) needed for chemical and polymer synthesis, conventional catalysts are mechanistically limited to <60% conversion and deactivate through carbon buildup. Jiao et al. developed a bifunctional catalyst that achieves higher conversions and avoids deactivation (see the Perspective by de Jong). A zinc-chromium oxide creates ketene intermediates that are then coupled over a zeolite.

Science, this issue p. 1065, see also p. 1030

Abstract

Although considerable progress has been made in direct synthesis gas (syngas) conversion to light olefins (C2=–C4=) via Fischer-Tropsch synthesis (FTS), the wide product distribution remains a challenge, with a theoretical limit of only 58% for C2–C4 hydrocarbons. We present a process that reaches C2=–C4= selectivity as high as 80% and C2–C4 94% at carbon monoxide (CO) conversion of 17%. This is enabled by a bifunctional catalyst affording two types of active sites with complementary properties. The partially reduced oxide surface (ZnCrOx) activates CO and H2, and C−C coupling is subsequently manipulated within the confined acidic pores of zeolites. No obvious deactivation is observed within 110 hours. Furthermore, this composite catalyst and the process may allow use of coal- and biomass-derived syngas with a low H2/CO ratio.

Fischer-Tropsch synthesis (FTS) has played an important role as a gas-to-liquid technology, producing synthetic lubricants and synthetic fuels from coal, natural gas, or biomass since it was developed almost a century ago. It is also the only effective technology to date for direct conversion of synthesis gas (syngas) to light olefins—i.e., C2=–C4=—olefins containing two to four carbon atoms, so called Fischer-Tropsch to olefins (FTTO) (15). FTS has been under extensive study for more than 50 years, and a variety of metal catalysts, including iron, cobalt, and ruthenium, have been tested (6). However, application of this technology is still limited by low olefin selectivity and high methane selectivity, as well as severe carbon deposition. These drawbacks arise from the FTS reaction mechanism, which is generally accepted to proceed via surface polymerization of CHx (x = 1, 2, or 3)—i.e., addition of CHx monomer units to the adsorbed alkyl species on open metal surfaces. Thus, a wide distribution of hydrocarbons with different chain lengths is produced. It can be described by the Anderson-Schulz-Flory (ASF) model (7, 8), which predicts that selectivity of C2 to C4 hydrocarbons (C2–C4), including C2=–C4= olefins and C2o−C4o paraffins, does not exceed 58% (6).

Thus, the key challenge of selective formation of light olefins from syngas lies in precise control of C−C coupling while suppressing overhydrogenation and methane formation. We present here a process named OX-ZEO (Oxide-Zeolite), which separates CO activation and C−C coupling onto two different types of active sites with complementary properties. CO and H2 are activated over a partially reduced oxide (ZnCrOx) surface, whereas C−C coupling is controlled within the confined environment of zeolite pores with acidic sites. As a result, a C2−C4 selectivity up to 94% (including 80% C2=–C4= and 14% C2o–C4o) of all hydrocarbons (carbon atom–based) was achieved with only 2% methane at a CO conversion of 17%. This C2−C4 selectivity is far beyond the maximum predicted by the ASF model in FTS.

The composite catalyst contained an oxide (ZnCrOx) that exhibits a typical spinel structure (fig. S1) and a mesoporous SAPO zeolite (MSAPO) exhibiting CHA structure with a hierarchical pore texture (figs. S1 and S2 and table S1) (9). Figure 1A shows that C2=−C4= selectivity reached 74% at CO conversion of 16% over a catalyst with a mass ratio of ZnCrOx/MSAPO = 1.4, under reaction conditions of H2/CO = 1.5, pressure 2.5 MPa, and space velocity 4800 ml/h·gcat. A higher H2/CO ratio benefits CO conversion, which rose to 30% at H2/CO = 3, for instance, whereas a higher space velocity favored the selective formation of olefins. Selectivities varied in the range from 67 to 80% for C2=−C4= and 81 to 94% for C2−C4 in the studied H2/CO (0.5 to 3.0) and space velocity ranges (1285 to 7714 ml/h·gcat) and different ratios of ZnCrOx/MSAPO. This C2=−C4= selectivity is higher than the best value reported for FTTO (61%) (4) and exceeds the maximum predicted for C2−C4 hydrocarbons according to the ASF distribution in typical FTS, as depicted in Fig. 1B. Both CH4 and C5+ selectivities were <5% (Fig. 1A), prominently lower than those in FTTO (in a range of 10 to 40% for CH4 and >10% for C5+) (4). In contrast, the C1 and C5+ products were difficult to suppress simultaneously in FTTO; causing one to decline caused the other to increase (1, 7). The reactivity of this composite catalyst showed good reproducibility, with CO conversion and selectivity fluctuating within 4 and 5%, respectively, among 11 tests (fig. S3). Furthermore, it delivered rather stable performance (Fig. 1C). The total C2−C4 selectivity remained >90%, and that of C2=−C4= ~78% during a 110-hour test at 400°C, 2.5 MPa, and a space velocity of 6828 ml/h·gcat.

Fig. 1 Catalytic process of OX-ZEO.

(A) CO conversion and product distribution at different H2/CO ratios in syngas over a catalyst with a mass ratio of ZnCrOx/MSAPO = 1.4 at a space velocity of 4800 ml/h·gcat. (B) Hydrocarbon distribution in OX-ZEO in comparison to that reported for FTTO (4) and that in FTS predicted by the ASF model at a chain growth probability of 0.46, with the yellow bar representing selectivity of C2−C4 hydrocarbons. (C) A stability test of a composite catalyst with ZnCrOx/MSAPO ratio = 0.9 at 6828 ml/h·gcat and H2/CO of 2.5.

Although composite catalysts containing metal oxides such as Cr2O3-ZnO and CuO-ZnO, and zeolites such as ZSM-5, Y, and β have been attempted previously for syngas conversion (1015), the products were mainly dimethylether (10), liquefied petroleum gas (C3o−C4o) (1113), or gasoline (14, 15). We attribute the efficiency of the OX-ZEO process to the bifunctionality of the catalyst, with two types of active sites exhibiting complementary and compatible properties. In the absence of MSAPO (corresponding to Mode 1 of Fig. 2A), syngas was converted mainly to CH4 (selectivity 53%), and selectivity to C2–C4 hydrocarbons was only 38% over ZnCrOx. Upon combination with MSAPO, which was packed below the oxide and separated by a layer of inert quartz wool (Mode 2), the products shifted and C2−C4 selectivity increased to 69% containing 23% C2=–C4=, whereas CH4 selectivity dropped to 26%.

Fig. 2 Bifunctionality of the composite ZnCrOx/MSAPO catalyst and investigation of the reaction intermediate.

(A) Reaction results over the catalysts with the two functionalities packed in different modes under the same conditions. The picture on the right side displays the catalyst beds after reaction, representing Mode 1 to 4 from left to right: Mode 1, the catalyst contains ZnCrOx only; Mode 2, MSAPO packed below ZnCrOx, separated by an inert layer of quartz wool; Mode 3, MSAPO and ZnCrOx packed in an alternating sequence and separated by quartz wool; Mode 4, MSAPO and ZnCrOx well mixed. (B) In situ NAP-XPS Zn3d spectra of ZnCrOx under different conditions. (C) In situ NAP-XPS C1s spectra of ZnCrOx exposed to H2, CO, and H2 sequentially, and then again to CO atmosphere under different conditions. (D) In situ study of syngas conversion over ZnCrOx by SVUV-PIMS at hυ = 9.72 eV. The insets display the signals of m/z = 42.01 (ketene) and m/z = 42.05 (propene) detected at hυ = 9.72 and 11.40 eV, respectively.

This result shows that reaction intermediates generated over the oxides must have transported in the gas phase toward the active sites of MSAPO, where they were converted to C2−C4 hydrocarbons, rather than being hydrogenated to CH4 over the oxide or in gas phase. Thus, any measures that could facilitate the transport of intermediates in gas phase should benefit selective formation of olefins, for example, increasing space velocities, as demonstrated in Fig. 2A. In addition, Mode 3, where the oxides and MSAPO were packed in an alternating sequence and thus the transport distance was reduced, also leads to enhanced C2=−C4= selectivity (65% at 7714 ml/h·gcat). We observed an even higher selectivity of C2=−C4= (80%), with overall C2−C4 selectivity reaching 94% and only 2% CH4 over the well-mixed composite catalyst (Mode 4). Furthermore, CO conversion could be tuned by varying the relative mass ratio of oxide/zeolite (fig. S4) (9), which lends further support to the above hypothesis that the composite catalyst is bifunctional and that the reaction involves intermediate transport in gas phase.

Activation of the C−O bond has been reported previously for surfaces promoted with oxygen vacancies—for example, Fe3O4 (16), CoOx/TiO2 (17), Cu/ZnO (0001) (18), and ZnGa2O4 (19). However, little is known about subsequent formation of C−C and C−H bonds in the presence of H2. Density functional theory (DFT) calculations using the ZnCr2O4 spinel (111) surface as a model reveal that it is reducible, resulting in a surface with a number of oxygen vacancies (figs. S5 and S6) (9). In situ near-ambient pressure x-ray photoelectron spectroscopy (NAP-XPS) of ZnCrOx surface (Fig. 2B) revealed a new signal with a lower Zn3d binding energy upon exposure to H2 and CO, compared with the initial oxidized surface. CO can be activated on such a surface, leading to formation of CO2, detected by time-of-flight mass spectrometry (TOF-MS) (fig. S7) (9) and surface *C species in absence of hydrogen, which is evidenced by a strong C1s signal emerging at 284.7 eV, even at 350°C (curve II in Fig. 2C). This C1s signal remained even after heat treatment at 300°C under ultrahigh vacuum (fig. S8) (9), but it remarkably attenuated in H2 (curve III) and finally vanished at 450°C (curve IV). Analysis of the effluent by TOF-MS shows that *C species were removed via hydrogenation forming hydrocarbons (fig. S7, b to d) (9). Thus, the precursor of hydrocarbons, CHx species formed over ZnCrOx, are likely the intermediates in the OX-ZEO process.

To detect the intermediate in gas phase, we turned to the highly sensitive synchrotron-based vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS) (9), which has been widely used in the field of combustion to detect active intermediates and radicals with low concentrations in gas phase (20, 21). It provides a tunable photon energy, soft ionization, and superior signal-to-noise ratio. The schemes of SVUV-PIMS and the reactor are shown in fig. S9. When syngas was fed into the reactor in the absence of catalysts (a blank experiment), no other signal was observed. In the presence of ZnCrOx, a signal of mass/charge ratio (m/z) = 42.01 appeared in the effluents at a photon energy hυ = 9.72 eV (Fig. 2D) in addition to stable hydrocarbon products. Further varying the photon energy (fig. S10) (9) confirmed that this signal was unambiguously attributed to ketene CH2CO, considering the m/z ratio and the ionization energy (2224). In comparison, propene (m/z = 42.05) was only distinguishable at hυ > 9.73 eV (fig. S10e) (9), and it became prominent at hυ = 11.40 eV (the black signal in the inset of Fig. 2D). Additional signals in Fig. 2D were assigned to stable products of C4H8 and C6H6, and their derivatives from stepwise addition of CH2 monomers up to m/z = 126.

In addition to ketene and the stable products, signals of methanol (m/z = 32) and its dissociated product methoxyl group (m/z = 31) were also detected at a high hυ (11.40 eV) in the presence of ZnCrOx catalyst. However, the reaction did not appear to go via methanol in this OX-ZEO process because the yield of C2=−C4= dramatically dropped to 3% within 22 hours when feeding methanol (50 mbar in He) directly to the composite catalyst (fig. S11). Figure S12 shows that MSAPO alone as a methanol-to-olefin (MTO) catalyst also deactivated quickly within 30 hours, even at a methanol partial pressure as low as 5 mbar, and 0.18 mol C/gcat was converted before deactivation. By contrast, the composite catalyst delivered rather stable performance in OX-ZEO. No obvious degradation of CO conversion was observed for operation as long as 650 hours (fig. S13), and 9.6 mol C/gcat was converted within this period of time. The total coke deposit was only 11 weight % (wt %) (fig. S14) (9). In contrast, fast deactivation remains a major issue for MTO in that coke deposits can mount up about 10 wt % within 15 min (25). Furthermore, very little water is produced in OX-ZEO, in contrast to the MTO process, where two moles of H2O are produced for each mole of ethylene (assumed as the only product) from methanol (9). In addition, O from CO is removed mainly by producing CO2 with a selectivity of 40 to 45% in OX-ZEO, whereas little CO2 forms in MTO. The above results show that reaction via methanol may not be the dominating pathway in OX-ZEO, although it cannot be excluded completely, whereas ketene likely plays an important role.

Such CH2 species are very active and readily react with CO in the presence of CO, forming a relatively less reactive ketene (26, 27), which is detectable by SVUV-PIMS. Thus, CH2 species are likely the primary intermediates. By forming ketene, the reaction pathway of surface polymerization of CHx is blocked, hence circumventing the ASF limits of FTS. Subsequently, in the presence of confined acidic environment of zeolite pores, CH2CO can be converted to olefins (Fig. 2A). This was validated by feeding ketene directly as the reactant to the MSAPO catalyst, with the effluents monitored by the online SVUV-PIMS. Ketene was synthesized via pyrolysis of acetic anhydride following the reported procedure (fig. S15) (24, 28). Comparison of the results in the blank reactor and in the presence of the catalyst (Fig. 3A and fig. S15) demonstrates the capability of MSAPO catalyzing conversion of ketene to light olefins.

Fig. 3 The role of MSAPO in OX-ZEO.

(A) Catalytic conversion of ketene to olefins over MSAPO at 400°C, with the effluent monitored by SVUV-PIMS. The inset shows the results of the blank experiment with ketene fed to the same reactor containing no catalyst. Spectra are recorded at 10.70 eV. (B) Ratio of olefin/paraffin as a function of the NH3 desorption peak temperature from the medium-strength acidic sites of MSAPO. The dashed line is only for the purpose of guiding the eyes, and the inset displays a typical NH3-TPD profile.

Figure 3B and fig. S16 (9) demonstrate that the product selectivity can be modulated by the medium strength acidity, characterized by temperature-programmed-desorption (TPD) of NH3, with the peak maximum located in the range from 350° to 410°C. For example, a commercially available SAPO-34 exhibiting NH3 desorption at ~394°C had a C2=−C4= selectivity of 43% and an olefin/paraffin ratio (C2=−C4=/C2o−C4o) of 0.9. The olefin/paraffin ratio increased with the decreasing NH3 desorption temperature and reached 4.7 for a MSAPO sample with NH3 desorption peak at 350°C (Fig. 3C). In contrast, a composite catalyst with almost no medium strength acidic sites had a similar product distribution, as the catalyst contained no zeolite (fig. S17 and table S3) (9). Thus, it is reasonable to assume that there may exist an optimum acidity with NH3 desorption temperature <350°C for olefin formation. However, it is still a challenge to synthesize zeolites with a weaker acidity but without reducing the number of active sites (necessary to achieve a reasonable conversion). In addition, the preliminary results in fig. S18 (9) show that the distribution of different-sized hydrocarbons might be tunable through the shape selectivity of zeolites, because a larger pore generally yields higher hydrocarbons. However, further elucidation of the relation between the structure (pore size/acidity) and the activity/selectivity will require much more sophisticated experiments because the acidity frequently varies simultaneously with the pore structure and crystallinity.

The capability of the partially reduced oxide in activating CO but incapability of catalyzing surface polymerization of CHx makes it possible to manipulate C–C coupling with confined acidic zeolite pores. Thus, the CO conversion and selectivity can be tuned at the same time—i.e., CO conversion is manipulated via the surface structure of the oxides and the ratio of oxides/zeolite, whereas the olefin selectivity is controlled by the properties of zeolites, particularly the pore structure and acidity. These findings open up a new avenue for development of syngas-to-olefin technologies, which may allow utilization of coal- and biomass-derived syngas with a low H2/CO ratio.

Supplementary Materials

www.sciencemag.org/content/351/6277/1065/suppl/DC1

Materials and Methods

Figs. S1 to S18

Tables S1 to S4

References (29, 30)

References and Notes

  1. Supplementary materials are available on Science Online.
  2. Acknowledgments: This work was financially supported by the National Natural Science Foundation of China (grant nos. 21425312, 21321002, and 91545204), the Ministry of Science and Technology of China (no. 2013CB933100), the “Strategic Priority Research Program” of the Chinese Academy of Sciences (grant XDA09030101), and Dalian Institute of Chemical Physics Fundamental Research Program for Clean Energy (DICP M201308). The Advanced Light Source is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under contract no. DE-AC02-05CH11231. A Chinese patent and an international patent application under the Patent Cooperation Treaty are pending.
View Abstract

Navigate This Article