544653 Oxygenate Formation over K/Mo2c Catalysts - Role of Preparation and Promotion

Tuesday, June 4, 2019: 2:42 PM
Texas Ballroom EF (Grand Hyatt San Antonio)
Wijnand Marquart1, David J. Morgan2, Graham J. Hutchings3, Michael Claeys1 and Nico Fischer1, (1)University of Cape Town, Cape Town, South Africa, (2)Carfidff University, Cardiff, United Kingdom, (3)School of Chemistry, Cardiff Catalysis Institute, Cardiff, United Kingdom

The Fischer-Tropsch (FT) process, producing long chained waxes and transportation fuels, is competing with fuels derived from crude oil, and its profitability is therefore dependent on the global oil price.[1] However, increasing the value of synthesized products could render the profitability of the FTS independent of fluctuations in the oil price (which are mostly due to global political trends rather than market demand). One way to achieve this, is to tailor the reactions selectivity through catalyst design and/or changes in reaction conditions to yield higher value compounds such as long chained oxygenates. Oxygenates are platform chemicals which can through further processing, be converted to alternatives for synthetic fuels or industrial chemicals and polymers.[2] Oxygenates are a typical by-product of the FT synthesis but due to selectivity limitations of the commonly employed catalysts, no commercial FT based process exists which produces oxygenates at a significant yield.[3]

Typically, transition metals such as Fe, Co, Rh and Ni are active for the FT synthesis. Based on reaction conditions employed, commercial Fe and Co based catalysts have been shown to produce between 6 and 12 C% oxygenates.[4] Rh has been demonstrated to have a higher oxygenate selectivity, but the associated high raw material cost becomes prohibitive for use as a commercial FT catalyst.[5] Catalysts other than the traditionally known FT active transition metals have also shown promising results. Transition metal carbides such as Mo2C, were investigated under Fischer-Tropsch conditions. While the bare carbide produces mainly methane and other hydrocarbons, upon promotion with potassium the selectivity showed a significant shift towards oxygenates.[6]

Figure 1: XPS analysis of the samples carburized at 630, 760 and 1000°C before (1) and after (2) the TPH treatment. Figure on the left displays the C 1s (red), O 1s (grey) and Mo 3d (blue) relative content on the surface of the samples. Figure on the right displays the graphitic (red), carbidic (grey), C-O (blue), C=O (green) and saturated graphitic (purple) carbon relative content on the surface of the samples (from [7]).

In the present study,[7] β-Mo2C was obtained via carburization of MoO3 in a mixture of methane and hydrogen at different temperatures. It could be confirmed that with increasing temperature the formation of carbon deposits is enhanced. Samples treated at 630-650°C did not show any evidence of carbon deposition in TEM or Raman spectroscopy and re-oxidized to MoO2 upon exposure to air at room temperature. A mild passivation treatment in 1% O2 in N2 prevented the bulk re-oxidation. Samples carburized at 760 and 1000°C were apparently stabilized via a layer of carbon. It has to be noted though, that for all air-stable/stabilized samples, Raman spectroscopy evidenced the presence of surface MoO3, suggesting that the carbon layer does not fully protect the Mo2C surface but yields a passivated surface possibly through diffusion limitation.

Detailed XPS analysis of the materials support the trend of an increasing graphitic layer with increasing carburization temperature. However, samples synthesized at 630-650°C also displayed graphitic carbon depositions besides the expected carbidic one. Neither TEM nor Raman did detect this. With a a temperature programmed reduction of the samples carburized at 760 and 1000°C monitored by the methane concentration in the reactor outlet gas by online GC-TCD, reaction conditions were determined to remove the carbon layer. After treatment, the carbon speciation of the high temperature samples resembled that of the low temperature carburized sample.

Both the passivated and the carbon coated catalysts were activated in H2 to remove the passivation/carbon layers before being exposed to syngas conversion conditions (T = 280°C, P = 33 bar, H2/CO = 1). Compared to pristine β-Mo2C (synthesized at 630°C, not passivated but directly exposed to reaction conditions) the CO conversion of the passivated and reduced and of the carbon bearing catalysts is significantly lower. Interestingly, at iso-conversion (adjusted through changes in space velocity and to a lesser extend reaction temperature), the passivated and reduced sample showed an eight times higher oxygenate content in the organic product while at the same time displaying a higher CO2 selectivity (from 31 to 40 C%) and a lower methane selectivity (from 27 to 20 C%). The composition of the oxygenate fraction remained relatively constant at 74-79 C% MeOH. This change in selectivity is possibly linked to the introduced oxidic Mo species evidenced by Raman, supporting a different mechanistic pathway.

Figure 2: Alcohol distribution over the unpromoted samples (figure on the left) and the promoted samples (figure on the right) as a function of the CO conversion, with from left to right per series methanol (red, #1), ethanol (grey, #2), propanol (blue, #3), C4 alcohols (green, #4) and C5+ alcohols (purple, #5) (from [7]).

Passivated samples were subsequently promoted with up to 6.2 wt.-% potassium. Upon promotion the CO conversion of the catalyst decreased (apparent activation energies decreased from 122.4 kJ/mol for the unpromoted sample to 40.8 kJ/mol for the highest promotion level) in parallel to an increase in CO2 and a decrease in CH4 selectivity. The amount of potassium added did not influence the performance within the studied range (1.9 to 6.2 wt.-%). The oxygenate fraction in the organic product also increased upon promotion, either through a secondary formation route in the presence of potassium or rather through the suppression of secondary reaction of oxygenates forming hydrocarbons. The composition of the oxygenate fraction changes drastically. While still dominated by the alcohols, ethanol is now the most dominant product with significant amounts of aldehydes detected at lower conversion levels.

We have been able to demonstrate the mayor effect synthesis conditions have on the performance of a β-Mo2C catalyst in the CO hydrogenation. In addition, our results suggest that amorphous surface oxide speciesinfluence not only on activity but also on selectivity. The oxygenate selectivity and composition could be altered via potassium promotion although the amount of potassium within the studied range had only a minor effect and the increase in oxygenate formation must be balanced with a decreased CO conversion, increased water gas shift activity and decreased methanation.

 

[1]  Dry, M.E. Catalysis Today, 2002, 71, 227-241.

[2]  Surisetty, V.R et al., Applied Catalysis A: General, 2011, 404, 1-11.

[3]  Klimkiewicz, R. Chemistry Central Journal, 2014, 8, 77-85.

[4]  Klerk, A. d. Energy & Environmental Science, 2011, 4, 1177-1205.

[5]  Forzatti, P. et al., Catalysis Reviews - Science and Engineering, 1991, 33, 109-168.

[6]  Woo, H.C. et al., Applied Catalysis, 1991, 75, 267-280.

[7]  Marquart, W. et al., Catalysis Scinence and Technology, 2018, 8, 3806-3817.

 

 

 


Extended Abstract: File Not Uploaded
See more of this Session: Alcohols and DME
See more of this Group/Topical: General Submissions