Characterization and catalytic behavior of potassium-modified ZrO₂ base catalysts

Introduction

ZrO₂ has attracted more and more attention as a promising catalyst and catalyst support because of its high thermal stability, amphoteric nature, and redox properties [

[] T. Yamaguchi, Catal. Today 20 (1994) 199–218.

]. Until now, ZrO₂ has been transformed into strong, solid acid catalysts by modification with sulfate [

[] M. Hino, K. Arata, J. Chem. Soc. Chem. Commun. (1980) 851–852.

[] D.A. Ward, E.I. Ko, J. Catal. 157 (1995) 321–333.

[] V. Quaschning, J. Deutsch, P. Druska, H.J. Niclas, E. Kemnitz, J. Catal. 177 (1998) 164–174.

[] X.M. Song, A. Sayari, Catal. Rev. -Sci. Eng. 38 (1996) 329–412.

], WO₃ [

[] W.M. Hua, F. Zhang, Z. Ma, Z. Gao, Catal. Lett. 65 (2000) 85–89.

], MoO₃ [

[] J.C. Yori, C.L. Pieck, J.M. Parera, Catal. Lett. 64 (2000) 141–146.

], or B₂O₃ [

[] B.Q. Xu, S.B. Cheng, X. Zhang, S.F. Ying, Q.M. Zhu, Chem. Commun. (2000) 1121–1122.

[] P.T. Patil, K.M. Malshe, P. Kumar, M.K. Dongare, E. Kemnitz, Catal. Commun. 3 (2002) 411–416.

]. These ZrO₂-based solid acid catalysts show high activity and selectivity in many reactions. However, few efforts have been made to turn ZrO₂ into a base catalyst by taking advantage of its natural basicity, although several studies have been done on a highly effective alkali-modified ZrO₂ catalyst for the oxidative coupling of methane [

[] A.Z. Khan, E. Ruckenstein, J. Catal. 139 (1993) 304–321.

[] S. Sugiyama, K. Shimodan, H. Hayashi, N. Shigemoto, K. Miyaura, K. Saitoh, J.B. Moffat, J. Catal. 141 (1993) 279–286.

]. In general, it is common to prepare base catalysts by modifying or supporting alkali metal oxides on various supports [13]. Various alkali metal oxides have been loaded on different supports, such as magnesium oxide [

[] P. Thomasson, O.S. Tyagi, H. Knözinger, Appl. Catal. A 181 (1999) 181–188.

], zeolites [22,25,40], alumina [117,118,

[] J.H. Zhu, Y. Wang, Y. Chun, X.S. Wang, J. Chem. Soc. Faraday Trans. 94 (1998) 1163–1169.

], silica [

[] R. Bal, B.B. Tope, T.K. Das, S.G. Hegde, S. Sivasanker, J. Catal. 204 (2001) 358–363.

], by the decomposition of alkali compounds. These catalysts have proved to be excellent solid base catalysts for numerous vapor-phase probe reactions, such as isopropanol dehydrogenation [22,197,198], the isomerization of 1-butene [26,

[] J.H. Li, R.J. Davis, Appl. Catal. A 239 (2003) 59–70.

] and cis-but-2-ene [119], the methylation of phenol [198], toluene [

[] W.S. Wieland, R.J. Davis, J.M. Garces, Catal. Today 28 (1996) 443–450.

], and catechol [

[] Y. Fu, T. Baba, Y. Ono, Appl. Catal. A 176 (1999) 201–204.

] with methanol. However, alkali metal oxides supported catalysts used in liquid phase base-catalyzed reactions have been investigated less. It is not clear whether this kind of catalyst is a real heterogeneous catalyst in liquid phase reactions. Recently, Wang et al. [119] used a dry impregnation process to prepare KNO₃/ZrO₂ superbases, which possessed a base strength of H_-= 27.0 and were very active for cis-but-2-ene isomerization under mild conditions. In this chapter, potassium-modified ZrO₂ were prepared by the calcination of hydrous zirconia and anhydrous zirconia after impregnation with potassium compounds. The catalytic activity of this system was studied in the vapor-phase isomerization of 1-butene and in the liquid-phase Michael addition of 2-methylcyclohexane-1,3-dione to methyl vinyl ketone [

[] Z.-J. Li, H.A. Prescott, J. Deutsch, A. Trunschke, H. Lieske, E. Kemnitz, Catal. Lett. 92 (2004) 175–180.

Preparation and characterization of potassium-modified ZrO₂

Catalyst preparation

Hydrous zirconia, ZrO(OH)₂·aq, was prepared from ZrOCl₂⋅8H₂O (Fluka, 99%). Pure anhydrous ZrO₂ was obtained by calcination of hydrous zirconia at 600 °C in air for 4h. The supported catalysts were prepared by wet impregnation of ZrO(OH)₂ (ZRH samples)

Sample

Precursor

Modifying

agent

XRD

S_BET

(m²/g)

V_P ^a

(cm³/g)

d_p ^b

(Å)

bc

ac

ZR

ZrO(OH)₂

-

A

M

26.0

0.08

127

KC-ZRH

ZrO(OH)₂

K₂CO₃

A

T+M

7.4

0.04

212

KHC-ZRH

ZrO(OH)₂

KHCO₃

A

T+M

6.3

0.03

281

KAC-ZRH

ZrO(OH)₂

KOAc

A

T+M

9.5

0.04

185

KN-ZRH

ZrO(OH)₂

KNO₃

A

M+T

4.8

0.04

322

KC-ZRO

ZrO₂

K₂CO₃

M

M

10.0

0.06

219

KHC-ZRO

ZrO₂

KHCO₃

M

M

11.7

0.07

236

KAC-ZRO

ZrO₂

KOAc

M

M

11.1

0.06

220

KN-ZRO

ZrO₂

KNO₃

M

M

3.0

0.02

266

Characteristics of different potassium-modified zirconia catalysts (II)

Sample

X-content (wt%)^c

K/Zr

(mol) ^d

bc

ac

ZR

C:0.23, H:1.19

C:0.04, H:0.08

-

KC-ZRH

C:1.23, H:1.28

C:0.45, H:0.13

0.18

KHC-ZRH

C:1.19, H:1.33

C:0.46, H:0.14

0.19

KAC-ZRH

C:3.45, H:1.33

C:0.51, H:0.10

0.17

KN-ZRH

C:0.37, H:1.14, N:2.21

C:0.26, H:0.06,N:1.00

0.18

KC-ZRO

C:1.01, H:0.12

C:0.80, H:0.12

0.15

KHC-ZRO

C:1.30, H:0.22

C:0.72, H:0.16

0.16

KAC-ZRO

C:3.00, H:0.52

C:0.87, H:0.11

0.14

KN-ZRO

C:0.14, H:0.10, N:2.57

C:0.13, H:0.09, N:1.61

0.17

Table 4.1 Characteristics of different potassium-modified zirconia catalysts (I)

and ZrO₂ (ZRO samples) powders with aqueous solutions of KHCO₃, K₂CO₃, KOAc and KNO₃ (in most cases, about 1.0 mL solution for 1.0 g powder). The theoretical K/Zr ratios were 0.2. After impregnation, the excess solution was evaporated at room temperature. The samples were then dried at 110 °C (for the samples modified with KHCO₃, 80 °C was used instead), calcined at 600 °C in flowing air for 4h, and stored in closed containers until used.

Characterization of potassium-modified ZrO2

Table 4.1 gives the notations and characteristics of the samples. Fig. 4.1 shows the XRD patterns of some catalysts after calcination. Amorphous, hydrous zirconia formed monoclinic ZrO₂ after calcination at 600 °C (Fig. 4.1: a). However, after the modification of hydrous zirconia using C-containing potassium compounds mentioned above and calcination, the main phase of KC-ZRH, KHC-ZRH and KAC-ZRH (Fig. 4.1: b) was metastable, tetragonal ZrO₂. Both the tetragonal and monoclinic phases were found in KN-ZRH (Fig. 4.1: c). It has been postulated that the K⁺ ions are incorporated in the vacant sites on the surface of hydrous zirconia and stabilize the tetragonal ZrO₂ phase [119,

[] Z. Liu, W.J. Jie, L. Dong, Y. Chen, J. Solid State Chem. 138 (1998) 41–46.

]. All of the ZRH samples, except for KN-ZRH, also have sharper and more intense XRD peaks than those of the pure ZrO₂ and the ZRO samples; this indicates higher sample crystallinity of the ZRH samples. Almost no phase or intensity change was found in the XRD patterns of the impregnated, anhydrous ZrO₂ samples after modification with potassium compounds (Fig. 4.1: d, KAC-ZRO).

Fig. 4.1XRD patterns of the catalysts after calcinations, a) ZR; b) KAC-ZRH; c) KN-ZRH; d) KAC-ZRO

The N₂ adsorption/desorption isotherms of all the samples (not shown) are of Type IV, according to IUPAC classification. The hysteresis loops, due to the capillary condensation associated with large mesopores, can be ascribed to the H2 Type [

[] G. Leofani, M. Padovan, G. Tozzola, B. Venturelli, Catal. Today 41 (1998) 207–219.

]. There is not much difference between the adsorption/desorption isotherms of pure ZrO₂ and modified ZrO₂, but the specific surface areas of the modified samples are much lower than that of ZrO₂, whereas the pore volumes are only slightly smaller (Table 4.1). The decrease in surface area may be attributed to 1) rearrangements during the formation of the networks, 2) surface covering by potassium oxide produced at higher temperatures, or 3) the presence of bulk potassium compounds on the surface. On the other hand, the average pore diameters of the samples were somewhat larger (Table 4.1). The mesopore size distributions of some samples are shown in Fig. 4.3. The curves all show sharp maxima with narrow distributions. The pore diameters of pure anhydrous ZrO₂ and the modified samples were centered at about 70 and 100-180 Å, respectively (Fig. 4.2). Except for the KNO₃-modified samples with the lowest pore volume, the anhydrous ZrO₂ supported samples (KC-ZRO, KHC-ZRO, and KAC-ZRO) had pores with higher pore volumes than that of the hydrous ZrO₂ supported samples, KC-ZRH, KHC-ZRH, and KAC-ZRH. The carbon and nitrogen contents of the calcined samples indicate that the potassium compounds have not decomposed completely, which may be one reason for the decrease in surface area. The K/Zr mole ratios of calcined samples measured by ICP-OES were close to the nominal value of 0.2 (Table 4.1).

Fig. 4.2 Pore size distributions of the indicated catalysts

TG-DTG-DTA was used to investigate the decompositon of potassium compounds on ZrO₂. The profiles of KHC-ZRH and KHC-ZRO are shown in Fig. 4.3. From the profiles of KHC-ZRH, a broad, endothermic effect, two DTG peaks, and 13% weight loss were observed from 100 to 400 °C, which can be ascribed to the evolution of water in hydrous zirconia and the decomposition of KHCO_3. This was comfirmed by an increase in the ionic currents of m/z =18 (H₂O⁺) and 44 (CO₂ ⁺) with maxima at around 135 and 228 °C. A sharp exothermic effect at about 516 °C was observed without weight loss for the crystallization of amorphous ZrO₂.

Fig. 4.3 TG-DTG-DTA profiles of KHC-ZRH (above) and KHC-ZRO (below) dried at 80 °C before calcination

The exothermic effect for the crystallization of pure ZrO₂ is at about 410 °C in DTA curve [189]. Thus, the introduction of potassium ions in ZrO₂ shifts the crystallization temperature of ZrO₂ to higher temperatures. The sulfation of zirconia has the same effect [189]. In the measurement of KHC-ZRO, two endothermic effects and corresponding DTG peaks at around 133 and 159 °C clearly indicate the decomposition of KHCO₃ on ZrO₂ with about 3% weight loss between 100 and 200 °C, which is different from that of KHC-ZRH. The carbon contents of KHC-ZRH and KHC-ZRO after thermal analysis up to 700 °C were 0.33% and 0.63%, respectively, which corresponds with the elemental analysis of these samples after calcination at 600 °C (0.46 and 0.72%, respectively, in Table 4.1). These results indicate that the loaded potassium compounds do not completely decompose; this is also reported by Wang et al. for KNO₃/ZrO₂. For a sample with 20% KNO₃ on ZrO₂, the percentage of decomposed KNO₃ was only 7.6% [119]. This and our results could be explained by the formation of layers of thermally stable potassium compounds on the zirconia surface.

Fig. 4.4 Temperature-programmed desorption of CO₂

The CO₂-TPD results (Fig. 4.4) show that pure ZrO₂ has basic sites indicated by two desorption maxima at about 135 and 335 °C. The potassium-modified zirconia samples, like ZrO₂, show an initial desorption maximum at around 140 °C. A second step of CO₂ desorption at higher temperatures starts at around 400 °C for KC-ZRH and KHC-ZRH, but at a lower temperature, 250 instead of 400 °C, for the ZRO samples. This increase in the CO₂ profile intensity continues up to 600 °C and lasts for 5–10 min under isothermal conditions. These results suggest that modified anhydrous zirconia (ZRO samples) has a wider distribution of basic sites than modified hydrous zirconia (ZRH samples). The second step of desorption of the modified samples indicates their stronger basic sites in comparison to those of ZrO₂.

CO₂ desorption at higher temperatures could be attributed to strong basic sites formed by K₂O extra-fine particles on the surfaces of ZrO₂ during decomposition. This has been reported by Zhu et al.[197]. He also excluded the possibility of some unusual oxides of potassium as the main basic sites on KNO₃/Al₂O₃. Hathaway et al. [22,23] and Kim et al. [40] also suggested in the case of Cs-modified zeolites that the decomposition product, cesium oxide, forms the active sites.

Catalytic behavior of potassium-modified ZrO₂

Catalyst

Weight

(mg)

Yield of 2-butene (%)

Yield/mass

(%/mg) at

30 min

Average cis/trans ratios

10 min

30 min

60 min

90 min

ZR

330

24.1

21.7

19.1

16.7

0.066

4.0

KC-ZRH

395

69.7

54.2

42.4

35.2

0.137

5.3

KC-ZRH^a

395

75.2

62.9

49.6

40.7

0.159

5.1

KHC-ZRH

361

62.7

44.3

30.2

22.8

0.123

5.3

KAC-ZRH

411

74.1

63.8

52.1

44.7

0.155

5.0

KAC-ZRH^a

411

63.2

58.5

50.2

44.8

0.142

5.2

KN-ZRH

379

9.1

6.8

5.6

4.7

0.018

3.9

KC-ZRO

380

51.9

40.8

30.8

25.1

0.107

5.1

KHC-ZRO

350

53.1

36.8

25.9

21.0

0.105

5.3

KAC-ZRO

316

56.5

41.3

29.6

23.4

0.131

5.0

KN-ZRO

329

4.1

3.6

2.7

2.1

0.011

2.5

a: second run (the same reaction procedure was repeated again)

Table 4.2 The results of double-bond isomerization of 1-butene at 150 °C over calcined catalysts

Both acid and base catalysts can catalyze the double-bond isomerization of 1-butene. Generally, the cis/trans ratio of 2-butene produced by isomerization of 1-butene is higher for base-catalyzed reactions [40,199]. The results of the double-bond isomerization of 1-butene are given in Table 4.2 with the normalized yields of 2-butene at 30 min for the catalysts. Cis- and trans-2-butene are the exclusive products found at 150 °C (100% product selectivity); skeletal isomerization or alkylation did not occur. Unmodified ZrO₂, which was inactive at 60 °C in [

[] K. Parida, V. Quaschning, E. Lieske, E. Kemnitz, J. Mater. Chem. 11 (2001) 1903–1911.

], had a low catalytic activity at 150 °C. About 21% yield of 2-butene with a cis/trans ratio of 3.9 was obtained after 30 min. The potassium modification of ZrO₂ resulted in an increase in the yield of 2-butene, i.e., from 21.7 up to (at the most) 63.8% after 30 min, except for the KNO₃-modified samples (Table 4.2). The improved activity is indicated by a) the yield/mass ratio at 30 min of 0.066 (pure zirconia) and 0.10–0.16 (potassium-modified ZrO₂) and b) the higher cis/trans ratios (more than 5.0). This can be explained by the stronger basic sites on catalysts produced by modification and confirmed with CO₂-TPD (Fig. 4.4). According to the literature [26,199], strong active basic sites are essential for the catalysis of 1-butene isomerization. Why the KNO₃-modified samples give yields of 2-butene lower than that of ZrO₂ and lower cis/trans ratios is not clear. The catalysts prepared from the modification of hydrous zirconia are slightly more active than the corresponding catalysts from anhydrous zirconia; this could be a result of the difference in the distribution of basic sites indicated by CO₂-TPD. At the same time, the locations of the basic sites on the surface of the different zirconia (tetragonal or monoclinic) may also affect the reaction. The catalysts, KC-ZRH and KAC-ZRH, were reused without much decrease in conversion after treatment at 600 °C (Table 4.2).

The liquid-phase Michael addition of 2-methylcyclohexane-1,3-dione to methyl vinyl ketone used for the catalytic experiments is an important reaction for the synthesis of steroid pharmaceutical products.

Catalyst

Yield of 2-methyl-2-(3-oxo-butyl)-cyclohexane-1,3-dione (%)^a

2h

4h

6h

8h

24h

Leaching test

Before separation

After separation of the catalyst

KC-ZRH

17

24

-

48

99

73 (14h)

95 (12h)

95 (24h)

KHC-ZRH

18

32

-

56

98

82 (14h)

96 (12h)

96 (24h)

KAC-ZRH

16

30

41

52

96

-

-

-

KC-ZRO

39

63

82

93

89

66 (4h)

88 (16h)

85 (24h)

KHC-ZRO

59

79

90

86

86

75 (4h)

89 (16h)

89 (24h)

KAC-ZRO

36

62

81

92

96

-

-

-

a: related to 2-methylcyclohexane-1,3-dione

Table 4.3The results of Michael addition and leaching tests at room temperature

Pure ZrO₂ was catalytically inactive in this reaction, but the modification of ZrO₂ with potassium salts was successful in producing catalytic activity. Thus, the target product, 2-methyl-2-(3-oxo-butyl)-cyclohexane-1,3-dione, was formed selectively on the catalysts, KC-ZRH, KHC-ZRH, and KAC-ZRH, within 24 h in yields of more than 95% (Table 4.3). In contrast, KN-ZRH and KN-ZRO gave yields of only 72 and 55%, respectively (not given in Table 4.3). The rate of product formation was highest on KC-ZRO, KHC-ZRO, and KAC-ZRO. Yields of about 80–90% were achieved within 6 h.

Leaching tests performed by filtration of the catalyst from the reaction mixture after 4 or 14 h show that the reaction continues after removal of the solid potassium-modified zirconia catalyst. The presence of potassium ions in solution was checked and confirmed by ICP-OES. It can be concluded that potassium compounds leach from the catalyst surface into the methanolic reaction mixture. This leaching and consequent homogeneous catalysis may be explained by the weak interaction of the potassium compounds with ZrO₂ and the solubility of potassium compounds in methanol. Consequently, exclusive heterogeneous catalysis can be ruled out for the liquid-phase reaction investigated on potassium-modified zirconia in methanol. However, the catalytic results differ from those obtained for the soluble base catalyst, KOH. When KOH is used, low product yield (40%) and a mixture of various products is found. This indicates that different K species are present in the methanol solution, and that the reaction may not be a straightforward example of homogeneous catalysis.

These results show that this catalyst system is less suitable for liquid-phase reactions, although the potassium-modified ZrO₂ samples do produce both higher selectivities and higher yields than KOH, and catalysts with potassium or alkali metal cations have been used successfully in vapor-phase reactions. The nature of the interaction between the support, ZrO₂ or Al₂O₃, with potassium or other alkali metal cations is open to further discussion [18].

Conclusions

In summary, the properties and the activities of the catalysts can be influenced by a variety of parameters, such as the zirconia precursor and the potassium modifying agent. High product yields and selectivities were also obtained for the liquid-phase Michael addition of 2-methylcyclohexane-1,3-dione to methyl vinyl ketone in methanol, but leaching could not be prevented. The following can be concluded:

Modified hydrous zirconia forms the metastable, tetragonal ZrO₂ phase after calcination; the phase of modified anhydrous zirconia, on the other hand, is monoclinic. The specific surface areas of the modified samples are much lower than that of ZrO₂. The potassium compounds on hydrous zirconia decompose in a broader temperature range than those on anhydrous zirconia.

Stronger basic sites are produced after modification, and modified anhydrous zirconia has a wider distribution of basic site strengths than modified hydrous zirconia.

Potassium-modified zirconia, except for that modified with KNO₃, give higher yields and cis/trans ratios of 2-butene in the double-bond isomerzation of 1-butene and is reusable.

Potassium-modified zirconia is less suitable as a solid base catalyst for liquid phase reactions in methanol because of potassium leaching effects.