A study on deactivation of Cu–Zn–Al catalyst for higher alcohols synthesis

Abstract

A Cu–Zn–Al catalyst without promoters was prepared using a complete liquid-phase method and tested for a deactivation study in higher alcohols synthesis from syngas. Results showed that the selectivity of higher alcohols first increased from 36.0% to 68.6% then gradually decreased to 14.1% with time on stream. Characterization results showed that Cu species and Zn species had little changes whereas the phase of Al species changed after reaction. It was found that the Al species of the Cu–Zn–Al catalyst changed from AlOOH to Al₂O₃. The phase change weakened CO dissociation and chain growth which led to the decrease of higher alcohols selectivity with time on stream. It was suggested that AlOOH had a function of CO dissociation and chain growth, which favored the formation of higher alcohols, whereas Al₂O₃ had no function of CO dissociation, which caused the formation of methanol.

---

1. Introduction

Higher alcohols have received considerable interest recently because of their potential as fuels and substitutes for gasoline.^1,2 Currently, Cu-based catalysts for higher alcohols synthesis (HAS) from syngas are Cu–Co or Cu–Fe bimetallic catalysts and alkali modified methanol synthesis catalysts.^3–5 For the Cu–Co or Cu–Fe bimetallic catalysts, the Cu–X (X = F–T elements) center is thought to be the active site (dual site) for higher alcohols synthesis.^6,7 And alkalis are considered to play a key role in the synthesis of higher alcohols using alkali modified methanol synthesis catalysts.⁵

It is well known that Cu–Zn–Al catalysts are usually used to synthesize methanol from CO or CO₂ by hydrogenation.^8,9 Nonetheless, in our previous study, it was found that the selectivity of ethanol over Cu–Zn–Al catalysts without promoters could reach an unexpected point, however, the result is very difficult to reproduce.¹⁰ At first, this novel phenomenon of ethanol formation over Cu–Zn–Al catalysts without promoters was ascribed to the synergism of Cu⁰ and Cu⁺ through experimental and theoretical studies.^11,12 But presently we have found that Al species also play a key role in the formation of higher alcohols, not just as the role of a carrier in the synthesis of methanol.

Therefore, in this paper, the Cu–Zn–Al catalyst without promoters was prepared using a complete liquid-phase method and tested for a deactivation study in the higher alcohols synthesis from syngas. To clarify the reason of catalyst deactivation, the structure of each metallic component of the catalyst was investigated before and after reaction. The correlation between the change of catalytic performance and the structural change was also discussed.

2. Experimental

2.1. Catalyst preparation

A Cu–Zn–Al slurry catalyst with a composition of Cu/Zn/Al = 2/1/0.8 (atomic ratio) was prepared using a complete liquid-phase method. Typically, aluminum isopropylate [(C₃H₇O)₃Al] was dissolved in a mixture of deionized water with a certain amount of citric acid at 323 K and maintained for 3 h, then the temperature was raised to 368 K and held for 1 h. Next, Cu(NO₃)₂·3H₂O and Zn(NO₃)₂·6H₂O were dissolved in glycol and the mixture was slowly added to the Al solution. The resulting Cu–Zn–Al solution was stirred at 368 K until a homogeneous sol was obtained. The sol was aged at room temperature for 10 days to obtain a gel. Finally, the gel was dispersed in liquid paraffin, heated under a N₂ atmosphere from 333 K to 573 K with a heating rate of 5 K min⁻¹ and held for 8 h at 573 K. A slurry catalyst was subsequently obtained.

2.2. Catalyst characterization

The slurry catalyst was centrifuged, extracted using petroleum ether and dried at room temperature before characterization.

Powder X-ray diffraction (XRD) patterns were recorded on a Rigaku D/MAX-2500 diffractometer in a 2θ range of 5–85°. Fourier transform infrared spectra (FTIR) were obtained on an AVATAR 370 spectrometer. NH₃ temperature-programmed desorption (NH₃-TPD-MS) was performed to measure the basicity of the catalyst. ²⁷Al-MAS-NMR measurements were recorded on a Bruker Avance DSX 500 spectrometer with a ²⁷Al frequency of 130.4 MHz. Graphite furnace atomic absorption spectrometry (GF AAS) was performed to quantify the composition of the catalyst using SpectrAA-220 AAS equipment. X-ray photoelectron spectroscopy (XPS) measurements were recorded using an ESCALAB 250 spectrometer. Thermogravimetric mass spectrometry (TG-MS) analysis was performed using a Setaram SETSYS TGA coupled with a hiden HPR20 QIC R&D mass spectrometer.

2.3. Catalytic activity test

CO hydrogenation to higher alcohols was carried out in a slurry reactor with a mechanical magnetic agitator. The syngas (H₂/CO = 2) was introduced into the reactor at a feed flow rate of 150 mL min⁻¹ under 523 K and 4.5 MPa. The steady-state activity measurement was taken after at least 24 h on stream. The gaseous products were analyzed online with a gas chromatograph equipped with a flame ionization detector (FID) to detect gaseous hydrocarbons and a thermal conductivity detector (TCD) to detect gaseous inorganic substance. The liquid products were collected daily and analyzed offline using the gas chromatograph.

3. Results and discussion

3.1. Catalytic performance

The catalytic performance of the Cu–Zn–Al catalyst for CO hydrogenation at 523 K with time on stream (TOS) was listed in Tables 1 and 2. CO conversion initially increased from 19.8% at 24 h to 38.9% at 48 h, then it gradually decreased to 32.0% at 120 h with TOS. Meanwhile, the selectivity of total alcohols first decreased from 31.4% to 14.0% at 72 h and then increased to 22.1% at 120 h. Notably, the selectivity of dimethyl ether (DME) increased gradually with TOS, and the selectivity of higher alcohols (C₂₊OH/ROH) increased from 36.0% at 24 h to 68.6% at 48 h and then gradually decreased to 14.1% at 120 h. Moreover, the selectivity of CH₃OH and CH₄ first decreased and then increased overall, which illustrated that the function of catalyst for CO dissociation and propagation was weakened with TOS. The catalytic performance indicated that a deactivation of the Cu–Zn–Al catalyst for higher alcohols synthesis happened and the reason for the catalyst deactivation could be clarified by combining activity results with the characterization results.

Table 1 Catalytic performance of the Cu–Zn–Al catalyst^a

Time/h	CO conversion (%)	Selectivity (wt%)				C2+OH/ROH (%)
		ROH	HC	DME	CO2
a Notes: reaction conditions: T = 523 K, P = 4.5 MPa, H2/CO = 2, feed low rate = 150 mL min^−1, ROH for total alcohol and HC for hydrocarbon, C2+OH/ROH for higher alcohols selectivity.
24	19.8	31.4	18.4	1.37	48.8	36.0
48	38.9	15.4	28.7	4.53	51.3	68.6
72	35.2	14.0	32.9	12.6	40.6	56.7
96	33.8	19.1	25.2	17.6	38.1	28.6
120	32.0	22.1	25.7	17.9	34.2	14.1

---

Table 2 Product distributions of the Cu–Zn–Al catalyst

Time/h	ROH distribution (wt%)					HC distribution (wt%)
	MEOH	ETOH	PrOH	BuOH	PeOH	C1	C2	C3	C4	C5^+
24	64.0	20.3	5.8	7.3	2.6	29.7	33.8	17.5	15.2	3.8
48	31.4	38.7	13.3	12.8	3.8	36.0	28.6	16.2	13.5	5.7
72	43.3	25.8	10.9	13.7	6.3	43.0	22.3	15.3	12.2	7.1
96	71.4	11.5	4.6	9.0	3.5	45.4	23.4	14.9	11.1	5.2
120	85.9	3.6	2.0	7.0	1.5	44.3	24.2	14.6	11.3	5.7

---

3.2. XRD analysis

As discussed above, it can be seen from the catalytic performance that deactivation of the Cu–Zn–Al catalyst happened during the formation of higher alcohols. It was speculated that some structural changes occurred during the reaction. In order to confirm this speculation, the catalyst was subjected to analyze using XRD, FT-IR, NH₃-TPD-MS, ²⁷Al-MAS-NMR, XPS and TG-MS. Fig. 1 presented the XRD patterns of the Cu–Zn–Al catalyst before and after reaction (meaning after a 120 h reaction throughout the whole manuscript). Only diffraction peaks of Cu⁰ and ZnO could be detected, revealing that no new species were formed during 120 h reaction. The XRD results illustrated that the structure of Cu species or Zn species had little changes, which suggested that the deactivation of the catalyst was not caused by the structural changes of Cu species and Zn species.

Fig. 1 XRD patterns of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

3.3. FT-IR analysis

Due to the low content of Al in the catalyst, the diffraction peaks of the Al species could not be detected using XRD. So the phases of the Al species of the Cu–Zn–Al catalyst before and after reaction were confirmed using FT-IR. As indicated in Fig. 2a, the absorption bands at 3414, 1629, 1423, 1327, 1089, 939, 769, and 586 cm⁻¹ were in agreement with the reported values of AlOOH, which confirmed the formation of AlOOH.^13,14 The intense bands at 3414, 1089 and 1327 cm⁻¹ belonged to the ν_as (Al)O–H, ν_s Al–O–H and ν_as Al–O–H vibrations of AlOOH, respectively. The three strong bands at 939, 769, and 586 cm⁻¹ were ascribed to the vibration mode of AlO₆. The peak at 1423 cm⁻¹ corresponded to OH stretching vibrations. The shoulder at 1629 cm⁻¹ was the feature of the bending mode of absorbed water. In Fig. 2b, the intense bands at 688 and 540 cm⁻¹ were ascribed to Al–O stretching vibrations, which was the principal feature of Al₂O₃.^14,15 The intense band centered at 3434 cm⁻¹ and the weak band at 1598 cm⁻¹ were attributed to the stretching vibrations of OH groups in the hydroxide structure and physically adsorbed water, respectively. All these facts confirmed that the Al₂O₃ phase was obtained after reaction. The FT-IR analysis illustrated that the structure of the Al species had changed from AlOOH to Al₂O₃ after 120 h reaction.

Fig. 2 FTIR spectra of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

3.4. NH₃-TPD-MS analysis

Fig. 3 showed the NH₃-TPD-MS profiles of the Cu–Zn–Al catalyst. As seen in Fig. 3, the amount of weak acid decreased after 120 h reaction. Busca et al.¹⁶ reported that the dehydration of the hydroxyl groups would induce some free and defective Al³⁺ ions and then increase the density of surface acid sites. This result indicated that the structure of the Al species changed from AlOOH → Al₂O₃, which was accompanied by the dehydration reaction: 2AlOOH → Al₂O₃ + H₂O.

Fig. 3 NH₃-TPD-MS profiles of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

3.5. ²⁷Al-MAS-NMR analysis

Further evidence for the structural change of the Al species before and after reaction could be obtained from the ²⁷Al-MAS-NMR spectra. Fig. 4 showed the ²⁷Al-MAS-NMR spectra of the Cu–Zn–Al catalyst before and after reaction. As seen in Fig. 4, the resonance at approximately +8 ppm corresponded to the octahedral AlO₆ site of AlOOH^17,18 and the resonance at +65 ppm was ascribed to the tetrahedral AlO₄ groups of Al₂O₃.¹⁸ This transformation in the ²⁷Al NMR spectrum illustrated that a proportion of Al in octahedral sites (AlOOH) had transformed to tetrahedral sites (Al₂O₃).

Fig. 4 ²⁷Al-MAS-NMR spectra of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

3.6. AAS analysis

In order to confirm that the deactivation of the catalyst was not caused by the change of catalyst composition, the composition of the Cu–Zn–Al catalyst was analyzed using GF AAS and the results were listed in Table 3. As we described in the catalyst preparation, the atomic ratio of Cu : Zn : Al was kept at 2 : 1 : 0.8 theoretically. However, from the GF AAS results, it was calculated that the atomic ratio of Cu : Zn : Al was 1.8 : 1 : 0.3 before reaction and after reaction the atomic ratio of Cu : Zn : Al was 1.7 : 1 : 0.3. It could be seen that the composition of the catalyst before and after reaction was almost unchanged, which indicated that the catalyst deactivation was not caused by the change of the catalyst composition. Since the slurry catalyst was centrifuged and extracted using petroleum ether, both before and after reaction, the aluminum concentration was lower than the theoretical composition, which might be the main reason why we could not detect either AlOOH or Al₂O₃ using XRD.

Table 3 The composition of the Cu–Zn–Al catalyst before and after reaction

Catalyst	GF AAS (mg L^−1)
	Cu	Zn	Al
Before reaction	0.430	0.238	0.029
After reaction	0.408	0.240	0.032

---

3.7. XPS analysis

In order to further confirm that the deactivation of the catalyst was the result of the structural change of the Al species rather than Cu species and Zn species, the Cu–Zn–Al catalyst before and after reaction was analyzed using XPS and the binding energy (BE) of each component of the Cu–Zn–Al catalyst before and after reaction was presented in Table 4. As shown in Table 4, the Cu 2p_3/2 BE values were about 932.0 eV and there was no shake-up satellite peak between 940.0–945.0 eV (Fig. 5), which indicated the absence of Cu²⁺, and the Zn 2p_3/2 BE values were less than 1022.0 eV which could be assigned to ZnO.^19,20 This was consistent with the XRD results. However, the Al 2p BE value before reaction was 74.4 eV, corresponding to Al–OH bonds, which was in agreement with Al in AlOOH.²¹ After reaction, the BE of Al 2p shifted to 73.7 eV, corresponding to Al–O bonds, which was considered to be Al₂O₃.²¹

Table 4 XPS parameters of the Cu–Zn–Al catalyst before and after reaction

Metal component	Binding energy (eV)
	Before reaction	After reaction
Cu 2p3/2	932.0	932.1
Zn 2p3/2	1021.7	1020.6
Al 2p	74.5	73.7

---

Fig. 5 Cu 2p XPS spectra of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

3.8. TG-DTG analysis

Fig. 6 presented the TG-DTG profiles of the Cu–Zn–Al catalyst before and after reaction. As shown in Fig. 6a, the thermal analysis of the Cu–Zn–Al catalyst showed four decomposition steps before reaction. The first step at around 353 K was ascribed to the desorption of physically adsorbed water. The weak DTG step at about 563 K was responsible for the loss of layered structure water of AlOOH. The strong step at about 713 K was attributed to the decomposition of AlOOH to γ-Al₂O₃,²² and the last step at about 1073 K was considered to be the further conversion of γ-Al₂O₃ to other Al₂O₃. But after reaction, it could be found that the rate of weight loss decreased from DTG (curve b), especially for the step of AlOOH to γ-Al₂O₃, which illustrated that a proportion of AlOOH had converted to Al₂O₃ during the reaction. Considering the reaction temperature was only 523 K, further studies were required to figure out what caused such structural change, which might be beneficial to improve the stability of catalysts in higher alcohols synthesis for our future work.

Fig. 6 TG-DTG profiles of the Cu–Zn–Al catalyst. (a) Before reaction and (b) after reaction.

In order to directly prove that the formation of Al₂O₃ could deactivate the catalyst for the higher alcohols synthesis, a Cu–ZnO–Al₂O₃ catalyst was prepared using Al₂O₃ as Al source and compared with the present study to investigate the formation of higher alcohols. The results were listed in Tables 5 and 6. As seen in Tables 5 and 6, the selectivity of higher alcohols was only 7.2%, which was lower than that of the Cu–ZnO–AlOOH catalyst. This work further indicated that Al₂O₃ was not beneficial for the formation of higher alcohols. In addition, F. Schüth et al.²³ prepared ternary Cu–ZnO–Al₂O₃ catalysts to obtain the greatest methanol synthesis activity. Wang et al.³ also prepared a series of alumina-supported copper–cobalt catalysts to investigate the formation of higher alcohols and the results showed that methanol was the major product over monometallic Cu/γAl₂O₃.

Table 5 Catalytic performance of a Cu–ZnO–Al₂O₃ catalyst^a

Catalyst	CO conversion (%)	Selectivity (wt%)				C2+OH/ROH (%)
		ROH	HC	DME	CO2
a Notes: reaction conditions: T = 523 K, P = 4.5 MPa, H2/CO = 2, feed low rate = 150 mL min^−1, ROH for total alcohol and HC for hydrocarbon, C2+OH/ROH for higher alcohols selectivity.
Cu–ZnO–Al2O3	24.7	48.1	5.5	24.6	21.8	7.2

---

Table 6 Product distributions of a Cu–ZnO–Al₂O₃ catalyst

Catalyst	ROH distribution (wt%)					HC distribution (wt%)
	MEOH	ETOH	PrOH	BuOH	PeOH	C1	C2	C3	C4	C5^+
Cu–ZnO–Al2O3	92.8	2.9	1.0	2.2	1.1	31.7	22.4	20.8	16.8	8.2

---

It was widely accepted that a bi-functional (dual site) catalyst was required for higher alcohols formation, in which one site acted for CO non-dissociative adsorption and insertion, while the other acted for CO dissociation and chain growth, with the synergism between the two sites benefiting the synthesis of higher alcohols.^4,24,25 In the stability study of this paper, the results showed that the CO conversion and the selectivity of higher alcohols first increased at an early stage and then gradually decreased with time on stream, which was considered to be the result of the structural change of the catalyst according to the characterization results. It could be speculated that AlOOH had a function of CO dissociation and chain growth, which favored the formation of higher alcohols, whereas Al₂O₃ had no function of CO dissociation. As a result, the main reason for the decrease of higher alcohols selectivity was ascribed to the structural change of AlOOH → Al₂O₃.

4. Conclusion

The deactivation process and the structural change of Cu–Zn–Al catalyst before and after reaction were investigated. CO conversion and the selectivity of higher alcohols first increased at an early stage and then gradually decreased with time on stream in HAS. During the reaction, a proportion of Al species of the Cu–Zn–Al catalyst changed from AlOOH to Al₂O₃. Such structural change was the main reason for the deactivation of the Cu–Zn–Al catalyst in the formation of higher alcohols. This result suggested that AlOOH had a function of CO dissociation and chain growth, which favored the formation of higher alcohols, whereas Al₂O₃ had no function of CO dissociation using the Cu–Zn–Al catalyst without promoters.

Acknowledgements

This research was supported by Natural Science Foundation of China (21176167); the Key Project of Natural Science Foundation of China (21336006); and the Doctoral Program of Higher Education Priority Development Areas (20111402130002).

References

1. J. J. Spivey and A. Egbebi, Chem. Soc. Rev., 2007, 36, 1514–1528.
2. V. Subramani and S. K. Gangwal, Energy Fuels, 2008, 22, 814–839.
3. J. J. Wang, P. A. Chernavskii, A. Y. Khodakov and Y. Wang, J. Catal., 2012, 286, 51–61.
4. K. Xiao, Z. H. Bao, X. Z. Qi, X. X. Wang, L. S. Zhong, M. G. Lin, K. G. Fang and Y. H. Sun, Catal. Commun., 2013, 40, 154–157.
5. E. Heracleous, E. T. Liakakou, A. A. Lappas and A. A. Lemonidou, Appl. Catal., A, 2013, 455, 145–154.
6. M. Gupta, M. L. Smith and J. J. Spivey, ACS Catal., 2011, 1, 641–656.
7. K. G. Fang, D. B. Li, M. G. Lin, M. L. Xiang, W. Wei and Y. H. Sun, Catal. Today, 2009, 147, 133–138.
8. M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abild-Pedersen, S. Zander, F. Girgsdies, P. Kurr, B. L. Kniep, M. Tovar, R. W. Fischer, J. K. Nørskov and R. Schlögl, Science, 2012, 336, 893–897.
9. Y. Yang, J. Evans, J. A. Rodriguez, M. G. White and P. Liu, Phys. Chem. Chem. Phys., 2010, 12, 9909–9917.
10. W. Huang, L. M. Yu, W. H. Li and Z. L. Ma, Front. Chem. Eng. China, 2010, 4, 472–475.
11. Z. J. Zuo, L. Wang, Y. J. Liu and W. Huang, Catal. Commun., 2013, 34, 69–72.
12. Z. J. Zuo, L. Wang, L. M. Yu, P. D. Han and W. Huang, J. Phys. Chem. C, 2014, 118, 12890–12898.
13. Z. Tang, J. L. Liang, X. H. Li, J. F. Li, H. L. Guo, Y. Q. Liu and C. G. Liu, J. Solid State Chem., 2013, 202, 305–314.
14. L. M. Zhang, W. C. Lu, R. R. Cui and S. S. Shen, Mater. Res. Bull., 2010, 45, 429–436.
15. Z. J. Wang, H. Du, J. H. Gong, S. G. Yang, J. H. Ma and J. Xu, Colloids Surf., A, 2014, 450, 76–82.
16. G. Busca, Catal. Today, 2014, 226, 2–13.
17. F. J. Berry, R. L. Bilsborrow, A. J. Dent, M. Mortimer, C. B. Ponton, B. J. Purser and K. R. Whittle, Polyhedron, 1999, 18, 1083–1087.
18. S. Jiansirisomboon and K. J. D. MacKenzie, Mater. Res. Bull., 2006, 41, 791–803.
19. Z. H. Gao, W. Huang, L. H. Yin and K. C. Xie, Fuel Process. Technol., 2009, 90, 1442–1446.
20. P. Gao, F. Li, H. J. Zhan, N. Zhao, F. K. Xiao, W. Wei, L. S. Zhong, H. Wang and Y. H. Sun, J. Catal., 2013, 298, 51–60.
21. T. Isobe, M. Shimizu, S. Matsushita and A. Nakajima, Journal of Asian Ceramic Societies, 2013, 1, 65–70.
22. Y. F. Zhou, H. Y. Fu, X. J. Zheng, R. X. Li, H. Chen and X. J. Li, Catal. Commun., 2009, 11, 137–141.
23. C. Baltes, S. Vukojevic and F. Schüth, J. Catal., 2008, 258, 334–344.
24. K. Xiao, Z. H. Bao, X. Z. Qi, X. X. Wang, L. S. Zhong, K. G. Fang, M. G. Lin and Y. H. Sun, J. Mol. Catal. A: Chem., 2013, 378, 319–325.
25. S. Z. Li, W. P. Ding, G. D. Meitzner and E. Iglesia, J. Phys. Chem. B, 2002, 106, 85–91.

---