Surface Characterization of Chromium Oxide-Zirconia Catalyst

temperatures due to the interaction between chromium oxide and zirconia. ... chromium oxide, a-Cr~Os was observed only at the calcination temperature ...
2 downloads 0 Views 642KB Size
Langmuir 1993,9, 126-131

126

Surface Characterization of Chromium Oxide-Zirconia Catalyst Jong Rack Sohn' and Sam Gon Ryut Department of Industrial Chemistry, Engineering College, Kyungpook National University, Taegu 702- 701,Korea Received April 20,1992. In Final Form: September 15,1992 Chromium oxidezirconia catalysta were prepared by dry impregnation of powdered Zr(OH)4 with an aqueous solution of (NHJ2Cr04. The surface characterization of prepared catalysta was performed by using Fourier transform infrared (FTIR),X-ray photoelectronspectroecopy(XPS),and X-ray diffraction (XRD),and by the measurement of surface area. The addition of chromium oxide to zirconia shifted the phase transitions of Z r o z from amorphous to tetragonal and from tetragonal to monoclinic to higher temperatures due to the interaction between chromium oxide and zirconia. Since the Zroz dispersed chromium oxide, a-Cr~Os was observed only at the calcinationtemperatureabove 1173K. In the treatments by oxygen, hydrogen, carbon monoxide, and n-hexane the reduction-oxidation between Cr(VI)and Cr(III) is reversible, as observed by X P S and IR. Upon the addition of only a small amount of chromium oxide (1 w t % Cr) to ZrO2, both the acidity and acid strength of catalyst increased remarkably, showing the presence of Br6nsted and Lewis acid sites on the surface of CrOJZrOo. used. In particular, in previous papers from this laboratory,it has been shown that Ni&ZrO2 and ZrOz modified with sulfate ion are very active for acid-catalyzed reactions, even at room temperature.'em The high catalyticactivities in the above reactions were attributed to the enhanced acidicproperties of the modified catalysts, which originate from the inductive effect of Sgo bonds of the complex formed by the interaction of oxides with the sulfate ion. It is well-knownthat the dispersion,the oxidation state, and the structural features of eupported species may strongly depend on the support. Structure and phyaicochemical properties of supported metal oxides are considered to be in different states compared with bulk metal oxides because of their interaction with supports. Thie paper describesthe surface characterization of chromium oxide supported on zirconia. The characterization of the samples was performed by means of Fourier t r d o r m infrared (FTIR),X-ray diffiaction (XRD), and X-ray photoelectron spectroscopy (XPS), and by the measurement of surface area.

Introduction Supported chromium oxide catalysts are being used for the polymerization,hydrogenation,dehydrogenation,and oxidation-reduction reactions between environmentally important molecules such as CO and NO.'+ Recently, many efforts have involved the characterization of these catalysts in an attempt to fiid the appropriate reaction mechanisms. Volumetric titrations to determine the oxidation state of the chromium used in conjunctionwith infrared and electron paramagnetic resonance spectroscopy have provided much information dealing with these questions. So far, however, they have been studied mainly on silica and alumina,(+* and only a small amount of work was done for the ZrO2 s u p p ~ r t . ~ J ~ Zirconia is an important material due to its interesting thermal and machanical properties and so has been investigated as a support and catalysts in recent years. Different papers have been devoted to the study of ZrO2 catalytic activity in important reactions such as methanol and hydrocarbon synthesis from CO and H2 or C02 and H211J2 or alcohol dehydration.13J4 Zirconia has been extensively used as a aupport for metals or incorporated in supports to stabilize them or make them more resistant to sintering.l"l7 Z r 0 2 activity and selectivity highly depend on the methods of preparation and the treatment

Experimental Section Catalyst Preparation. The precipitate of Zr(OH)r was obtained by adding aqueous ammonia slowly to an aqueous solution of zirconium oxychloride at room temperature with stirring until the pH of mother liquor reached about 8. The precipitate thus obtained was washed thoroughly with distilled water until chloride ion was not detected d was dried at room temperaturefor 12h. The dried precipitatewas powdered below

* To whom all correspondence should be addressed.

+ Present addreas: Agency for Defense Development, P.O.Box 35. Yuseone. Taeion 305-600. Korea. '(1) Hog&; J. P: J. Polym. Sci. 1970,8,2637. (2) Myers, D. L.; Lunsford, J. H. J. Catal. 1986,99, 140. (3) C h k , A. Catal. Reu. 1969, 3, 145.

100 mesh.

The catalysts containing various chromium content were prepared by dry impregnation of powderedZrfOH)rwith aqueous solution of (NH&$304 followedby calciningat high temperature for 1.5 h in air. This series of catalysts are dew@ by their weight percentage of chromium. For example, l-CrO,/ZrOz indicates the catalyst containing 1 wt 7% chromium. Surface Characterization. FTIR spectra were obtained in a heatable gas cell at room temperature using Mattson Model GL 6030E spectrophotometer. The self-supporting catalyst wafers contained about 9 mg/cm2. Prior to obtaining the spectra, the samples were heated under vacuum at 673-773 K for 1.5 h.

(4)Groeneveld, C.; Wittgen, P. P. M. M.; van Kersbergen, A. M.; Meetrom, P. L. M.; Nuijten, C. E.;Schuit, C. E. J. Catal. 1979,59,153. (5) Shelef, M.; Otto,K.; Gandhi, H. J. Catal. 1968,12, 361. (6) McDaniel, M. P. Adu. Cotol. 1986,33, 47. (7) Ghiotti, G.;Garrone, E.; Zecchina, A. J. Mol. Catal. 1988,46,61. (8) Hill, W.; bhlmann, G. J. Cotal. 1990, 123, 147. (9) Cimino, A.; Cordisch, D.; Febbraro, S.;Gauoli, D.; Indovina, V.; Occhiuzzi, M.; Valigi, M. J. Mol. Catol. 1989,55, 23. (10) Yamaguchi,T.;Tan-No, M.;Tanabe, K. ReparationojCatalysts V; Eleevier: Amsterdam, 1991; p 567. (11) He, M. Y.; Ekerdt, J. G.J. Catal. 1984,90, 17. (12) Maehashi, T.; Maruya, K.; Domen, K.; Aika, K.; Oniehi, T. Chem. Lett. 1984, 747. (13) Yamaguchi, To;Saeaki, H.; Tanabe, K. Chem. Lett. 1978, 1017. Ganesan, P. Id.Eng. Chem. R o d . Res. Deu. 1979, (14) Davis, B. H.; 18. . -, 191. - - -. (15) Iizuka, T.; Tanaka, Y.; Tanabe, K. J. Catal. 1982, 76, 1. (16) Turlier, P.; Dalmon, J. A.; Martin, G. A. Stud. Surf. Sci. Catal.

(17) Szymaneki, R.; Charcoeeet, H.; Gallezot, P.; Mlleeardier, J.; Tournayan; L. J. Cotal. 1986,97,366. (18) Sohn,J. R.; Kim. H. J. J. Catal. 1986,101,428. (19) Sohn,J. R.; Kim. H. W.; Kim. J. T. J. Mol. Catol. 1987,41,379. (20) Sohn,J. R.; Kim. H. W. J. Mol. Catol. 1989,52, 361.

1982, 11, 203.

0743-7463/93/2409-0126$04.00/0

Q

1993 American Chemical Society

hngmuir, Vol. 9, No. 1, 1993 127

Chromium Oxide-Zirconia Catalyst Cr 2p3 I

2

I

I

1

I

584

580

576

572

Elndiny EnerSy. e V

595

590

585

580

575

570

Bindins Energy.eV

Figum 1. Cr 2p XPS of 3-CrO,/Zro~treated under various conditione: (a) uncalcined sample; (b) sample calcined at 873 K; (c) after reduction of sample b with HP at 823 K; (d) after reoxidation of sample c with 02 at 823 K; (e) after reaction of sample b with n - h e m e at 823 K. Catalysta were checked in order to determine the structure of the support ae well as that of chromium oxide by means of a Jeol Model JDX-Bosodiffractometer,employing Cu Ka (Ni-filtered) radiation. X-ray photoelectronspectra were obtained with a VG Scientific Modal ESCALAB h4K-11spectrometer. Al Ka and M g Ka were wed tu the excitation source, usually at 12 kV, 20 mA. The analysis chamber was 1o-B Torr or better and the spectra of samples, as fine powder, were analyzed. However, to examine the redox behavior of catalysta,some samples were preesed onto a plate, treated with Hz and 02 flowing at 10 mL/min for 6 min at 823 K in a Beparate gas cell, and transferred into a analysis chamber without exposure to air. Binding energiee were refere n d to the CI, level of the adventitious carbon at 286.0 eV. The specific surface area was determined by applyingthe BET method to the adsorption of N2 at 77 K. Chemisorption of ammonia waa also employed as a measure of the acidity of catalysts. The amount of chemisorption wae obtained ae the irreversible adsorption of gmmonia21~~

Results and Dimusmion X-ray Photoelectron Spectra. The Wiculty in the study of supported chromium oxide comes from the eimultaneoupresence of oxidationstatee. Figure 1shows the Cr 2p spectra of 3-CrO,/ZrO2 treated under various conditions. The shape of peaks and binding energies of 2p electron are different depending on the treatment (21) Soh, J. (22) Soh, J.

R;ozaki,A. J. Cotal. 1980,61, 29. R.;Jang, H.J. J. Mol. Catal. 1991,64, 349.

Figure 2. Cr 2 ~ 3 1 2fitted X P S of 3-CrOJZrO2 treated under various conditione: (a) uncalcined sample; (b)samplecalcinedat 873 K (cT after reduction of sample b with H2 at 823 K (d) after reoxidation of sample c with 0 2 at 823 K, (e) after reaction of sample b with n-hexane at 823 K.

conditions,indicatingthat the oxidation stateof chromium varies with the treatment pro". To obtain further information on the oxidation state, the spectrum in the Cr 2p3p region was analyzed by appropriate curve fitting and the presence of two or three componentswas confirmed, as shown in Figure 2. For the noncalcined and calcined samples, we have obtained a Cr 2p3p binding energy of 579.3 eV due to Cr(VI) and a binding energy of 576.7 eV identified as Cr(III). Cimino and co-worker@ have measured Cr 2p binding energiea for a variety of different chromium compounds. The correepondingCr 2p3p binding energies for CrOs and Cr& are 579.9 and 576.8 eV, reapectively. The reaults of quantitative d y a i s for the chromium oxidation state are listed in Table I. It is noted that the number of Cr(III) increases to some extent by the calcination at 873 K. However, Cr(VI) concentration of the sample calcined at 873 K is quite large, being 54% of the total Cr concentration, indicating that the ZrO2 support stabilizes sup ported chromium oxide as follows Zr-

a-0'

0, Cr

40

*o

However, when b sample in Figure 2 was treated w i t h H2 (10 mL/min) at 823 K for 6 min, Cr(VI) species was d y reduced to Cr(III>species, as illustrated in Figure 2c and Table I. C sample was reoxidized w i t h 0 2 (10 mL/min) at 823 K for 6 min and ita XPS result is shown in Figure (23) Cimino, A,; DeAngelis, B. A; Luchetti, A.; Minelli, C. J. Catal. 1976,4.5,316.

Sohn and Ryu

128 hngmuir, Vol. 9, No.1, 1993 I

n 01s

I 1800

I

1800

I

1

1400

1200

wavenumher/cm I

I

536

I

I

532

I

528

524

Einding Energy.eV

Figure 3. 01,XPS of 3-CrOJZrOz treated under various conditions: (a) uncalcined sample; (b)sample calcined at 873 K (c) after reduction of sample b with H2 at 823 K (d) aftar reoxidation of sample c with 0 2 at 823 K (e) after reaction of sample b with n-hexane at 823 K. Table I. Percentage of Chromium Species from the Area of the Fitted Bands in the Cr 2pa/z XPS Region treatment condition crs+ crs+ cro 66 35 uncalcined sample 54 46 calcined in air at 873 K after reduction with H2 at 823 K 19 81 after oxidation with 0 2 at 823 K 59 41 after reaction with n-hexaneat 823 K 3 89 7

2d. The Cr(VI)/Cr(III) ratio increased, which indicates that reduction-oxidation process is reversible. For a n-hexane reaction a pulse-type microreactor, constructed of in. stainless steel was used. Ten pulses of 1 pL of n-hexane were injected successively into a He gas stream which passed over 0.08 g of sample calcined at 873 K at 16 mL/min. The reaction temperature was 823 K. XPS of catalystafter reaction is illustrated in Figure 2e, resulting in the increase of Cr(II1) concentration. In this case we have also obtained a Cr 2~312binding energy of 674.7 eV assigned to chromium metal.24 This again suggests reduction from Cr(V1) to Cr(II1) or Cr(0) by n-hexane. Figure 3 shows 01, spectra of 3-CrOxlZr02treated under various conditions. In all the spectra the 01, peak at lower binding energy, 629.2 eV, corresponds to lattice oxide ions.2s However, for the sample reacted with n-hexane, the 01, peak at higher binding energy, 533 eV, appeared. This peak can be attributed to a carbonate species formed during the catalytic reaction of n-hexane. Recently Gonzalez et al.= have dealt with the characterization by X P S of the surface state of some metal oxides. In their work a second 01, peak appearing at the high binding (24) Merryfield,R.;McDaniel, M.; Parks, G. J. Catol. 1982,77,348. (25) C o d q ~ - E l i p eA., R.;Enpin-, J. P.; Fernandez,A.; Munuera, G. Appl. Surf. Scr. lSW,45,103.

Figure 4. IR spectra of CO-reduced l-CrO,/ZrOz: (a) after evacuation at 773 K for 1h; (b) aftar reduction with 160 Torr of CO at 623 K followed by evacuating at 623 K; (c) after reduction with 160 Torr of CO at 773 K followed by evacuating at 773 K.

energy side of the main peak corresponding to the oxide anions has been attributed to carbonate species. IR Spectra. To obtain further confiiation on the formation of carbonate species, IR spectra of the COreduced catalysts were investigated and some carbonate groupswere observed as shownin Figure 4. For the sample reduced with 160 Torr of CO at 623 K for 1h followed by evacuating at 623 K for 20 min, several bands at 1614, 1668,1480,1441,1370, and 1350 cm-l were observed. On the other hand, for the sample reduced at 773 K for 1 h followedby evacuatingat 773 K for 20 min, more complex bands at 1610, 1651,1488-1470,1463, 1446,1370,1366, and 1278 cm-' were observed. Zecchina et a1.N studied the adsorption of C02 on a-chromiaandamigned IR bands of several carbonate species. On the basis of the above assignments,the bands at 1614-1610,1463-1441, and 1273 cm-l are assigned to bicarbonategroups, the bands at 16681661 and 13661360 cm-l are due to the bidentate carbonate groups, and the monodentate carbonate groups are responsible for the 1488-1470 and 1370 cm-1 bands. The presence of several types of carbonate groups reveals a surface heterogeneity. Figure 6 shows the IR spectra of 1-CrOz/zrO2 treated under various conditions. After evacuation at 773 K for 1 h, three ban& at 1032,1018, and 920 cm-l were oherved. The doublet at 1032 and 1018 cm-1 is aeeigned to the asymmetric and symmetricstretching mod-, respectively, of surface chromates.0 The broad band around 900 cm-1 is typical of chromium-oxygen groups of a lower doublebond character. However, upon evacuation at 773 K for 2 h the intensities of doublet banda decreased remarkably due to the removal of surface oxygen. h x i d a t i o n of sample b in Figure 6 was performed by the addition of 02 (50 Torr) at 773 K for 0.6 h and then IR spectrum was taken at room temperature (Figure6c). Heating in oxygen (26) z%cchina,A.; Coluccia,9.; Gugliehinotti, E.;Ghiotti, 0.J. Phya. Chem. 1971,76,2790.

Chromium Oxide-Zirconia Catalyst

Langmuir, Vol. 9, No.1, 1993 129

973 K

723 K

20

60

40

80

w/o

I

10

I

1000

1

I

800

Figure 8. IR spectra of l-CrOJZr02 treated under various conditions: (a) after evacuation at 773 K for 1 h; (b) after evacuation at 773 K for 2 h; (c) after oxidation of sample b with 02 (50 Torr) at 773 K for 1 h; (d) after reduction of sample c with CO (50 Torr) at 773 K for 0.5 h; (d) after reoxidation of sample d with 0 2 (50 Torr) at 773 K for 0.5 h.

at 773 K nearly restored the situation observed in Figure 5a and an additional band at 1004cm-l assigned to C 4 species was observed. The weak and sharp band at 920 cm-l can be interpreted as due to the blue shift of the Zr-OGr mode caused by Cr oxidation. Upon reduction of oxidized c sample in Figure 5 with CO at 773 K, the intensitiesof allbands in 1100-900cm-l decreasedas shown in Figure 5d. Upon introduction of 0 2 (50 Torr) to the sample reduced with CO (50 Torr) and heating at 773 K for 0.5 h, the bands at 1032,1018,1004,and 920cm-l were also recovered. Completereversibility was observed after reduction-o.idation procegsesand these results are in good agreement with those of XPS described above. Crystalline Structure of Catalyst. The crystalline structure of catalysts calcined in air at different temperatures for 1.5 h was examined. Z r 0 2 was amorphous to X-ray diffraction up to 573 K, with a two-phase mixture of the tetragonal and monoclinic forma at 623-873 K and a monoclinic phase at 973 K. Three crystal structures of ZrO2, tetragonal, monoclinic,and cubic phases, have been reported.n*m However, in the case of supported chromium oxide catalysts the crystalline structures of samples were dif(27) Torralvo, M.J.; Alario, M.A.; Soria, J. J. Catal. 1984,86,473. (28) Clearfield, A. Inorg. Chem. 1964,3,146.

Figure 6. X-ray diffraction pattems of 5-CrOJZrO2 calcined at different temperatures for 1.5 h 0, tetragonal phase ZrOz; 0, monoclinic phase ZrOn; X, a-Cr203.

ferent from that of support, ZrO2. For the 5-CrOZ/Zr02, as shown in Figure 6,ZrOz was amorphous up to 723 K. Inother words, the transition temperature from amorphous to tetragonal phase was higher by 100 K than that of pure Zr02. X-ray diffraction data indicated a tetragonal phase of ZrO2 at 773 K, a two-phase mixture of the tetragonal and monoclinic ZrO2 forms at 873-1073 K,and a threephase mixture of the tetragonal and monoclinicZrO2 forma and a-CrzO3 at 1173 K. It is assumed that the strong interaction between chromium oxide and ZrO2 hinders the transition of ZrO2 from amorphousto tetragonal phase. The presence of chromium strongly influences the development of textural properties with temperature in comparison with pure Zr02. Moreover, for the sample of 10CrOJZrOz the transition temperature from amorphous to tetragonal phase was higher by 200 K than that of pure ZrO2. That is, the more the content of chromium, the higher the transition temperature. 10-CrOZ/Zr02was amorphous to X-ray diffraction up to 773 K, with a tetragonalphase of ZrO2 at 873-973 K,a twephase mixture of the tetragonal and monoclinic forms at 1073 K, a threephase mixture of the tetragonal and monoclinicZrO2 forms and a-Cr20~ at 1173-1273,and a two-phase mixture of the monoclinic ZrO2 form and a-CrzO3at 1373 K. Chromium oxide was not observed in any phase up to 1073 K of calcination temperature, indicating a good dispersion of chromium oxide on the surface of Zr02 support due to the interaction between them. As shown in Figure 6, a-CrgOs phase was observed only in the samples calcined above 1173 K, indicating that the Zr02 support stabilizes supported chromium oxide, and chromium oxide is well dispersed on the surface of ZrO2 under the calcination temperature below 1173 K.

130 Longmuir, Vol. 9, No.1, 1993

Sohn and Ryu

i.

/I

il 120

*E

100

L-J

40 20 0

2

4

C r

6

8

10

content/wt%

20

-

Figure 7. Variation of surface area of CrOJZrO2 calcined at 873 K with chromium content.

It is also of interest to examinethe influence of chromium oxide on the transitiontemperature of ZrO2 from tetragonal to monoclinicphase. In view of X-ray diffractionpatterns, the calcination temperatures at which monoclinic phase is observed initially are 623 K for pure ZrO2,673 K for 1-CrOZ/ZrO2,873K for 5-CrOZ/Zr02,and 973 K for 10CrO,/ZrO2, respectively. That is, the transition temperature increases with increasing chromium oxide content. This can be also explained in terms of the delay of transition from tetragonal to monoclinic phase due to the interaction between chromium oxide and zirconia, in analogy with the delay of transition from amorphous to tetragonal phase described above. Variation of Surface Area. It is necessaryto examine the effect of chromium oxide on the surface properties of catalysts, that is, specific surface area, acidity, and acid strength. The specific surface areas of samples calcined at 873 K for 1.5 h are plotted as a function of chromium content in Figure 7. The presence of chromium oxide strongly influences the surface area in comparison with thepureZrOz. SpecificsurfaceareaeofCrOJZrO2samples are much larger than that of pure Zr02 calcined at the same temperature, showing that surface area increases gradually with increasing chromium content. It seems likely that the interaction between chromium oxide and Zroz protects catalysts from sintering. The dependence of antisintering effect on chromium oxide content is clear from Figure 7. These results are ale0 in agreement with those of phase transitions of ZrO2 observed in X-ray diffraction patterns as described in crystalline structure. In addition to chromium oxide content, calcination temperature influencesthe surfacearea value. When pure ZrOz and some CrO,/ZrOz samples are subjected to calciningin air at a given temperature, the results obtained are plotted in Figure 8 as a function of calcination temperature. The antisintering effect of chromium oxide is greater at lower calcination temperature than at higher temperature. As illustrated in Figure 8, the difference of surface area between pure Z r 0 2 and CrO,/ZrO2 calcined at 1173K is very small compared with the cases of samples calcined at lower temperatures. For the CrO,/ZrOz samples, a-CrzO3 crystalline was observed at the calcinationtemperatureabove 1173Kasdescribedintheresults of X-ray diffraction. Therefore, it seem likely that the small antisintering effect of chromium oxide at the calcination temperature of 1173 K is related to the formation of a-Cr~O3and consequently the weak interaction between a-Cr~O3and zirconia.

01

I

I

I

I

I

Temperature. K

Figure 8. Variations of surface area for Zr02 and some CrOJ Zr02samples against calcinationtemperature: (a) 10-CrOJZr02; (b) 5-CrOZ/ZrO2;(c) l-CrO,/ZrOz; (d) ZrO2. Table 11. Acid Strength of 0.1-CrOdZrOr and ZrO2 pK,value CrOJ CrO,/ Hammett indicator of indicator ZrOf ZrOzb 2 1 0 2 -3.0 + + + dicinnamalacetone benzalacetophenone -5.6 + + + -8.2 + + anthraquinone -12.4 + + nitrobenzene -14.5 + + 2,4-dinitrofluorobenne Calcined in air at 873 K. Oxidized with 02 at 773 K.

Acidity and Acid Strength. The acid strength of the catalysts was examined by a color change method, using Hammett indicator%in dried benzene. Since it was very difficult to observe the color of indicators adsorbed on catalysts of high chromium oxidecontent, a low percentage of chromium content (0.1 wt %) was used in this experiment. The results are listed in Table 11. In this table, (+) indicates that the color of the base form was changed to that of the conjugated acid form. Z r O z evacuated at 673 K for 1h has an acid strength HOI -6.6, while O.l-CrO,/ZrOz was estimated to have a HOI -14.5, indicating the formation of new acid sites stronger than those of single oxide components. The acid strength of O.l-CrO,/ZrOz oxidized with 02 at 773 K was also found to be HOI-14.5. Acids stronger than HOI -11.93, which corresponds to the acid strength of 100% HzSOl, are superacids.30 Consequently, CrOJZrO2 catalysts would be solid superacids. The superacidicproperty is attributed to the double bond nature of the C d in the complex formedby the interaction of ZrOz with chromate,in analogy with the case of ZrOz modified with sulfate ion.lS2O The acidity of catalysts, as determined by the amount of NH3 irreversibly adsorbed at 503 K,2l is plotted as a function of the chromium content in Figure 9. Although pure Z r 0 2 showed the acidity of 0.06 mequiv/g, l-CrO,/ ZrOz resulted in a remarkable increase in acidity (0.1 mequiv/g). As shown in Figure 10, the acidity increases abruptly upon the addition of 1wt % chromium to ZrO2, and then the acidity increases very gently with increasing (29) Hammett, L. P.;Deyrup, A. J. J. Am. Chem. Soc. IOSZ,54,2721. (30)Olah, F. G. A.; P r h h , G. K. S.; Sommer, J. Science 1979, BM, 13.

Chromium Oxide-Zirconia Catalyst

Langmuir, Vol. 9, No. 1, 1993 131

/510

01 0

I

I

I

I

L

2

4

6

8

10

Y

1605

I

1443

C r Content. wt96

Figure 9. Acidity of CrOJZrOz against chromium content.

chromium oxide content. Many combinations of two oxides have been reported to generate acid sites on the surface.31-33 The combinationof Zr02 and CrO, probably generatesstronger acid sites and more acidityas compared with the separate components. A mechanism for the generation of acid sites by mixing two oxides has been propoeed by Tanabe et al.sl They suggest that the acidity generation is caused by an excess of a negative or positive charge in a model structure of a binary oxide related to the coordination number of a positive element and a negative element. Infrared spectroscopic studies of pyridine adsorbed on solidsurfaceshave made it poesible to distinguishbetween Brtlnsted and Lewis acid sites.% Figure 10 shows the IR spectra of pyridine adsorbed on l-CrO,/ZrOz evacuated at 773 K for 1h. There were peaks at 1443,1488,1497, 1664,1680,1696,and 1606cm-l comprisingthe vibrational modes of pyridine after evacuation at room temperature. Many peaks were weakened after evacuation at 523 K. Consequently, this set of disappeared absorption peaks can be assigned to hydrogen-bondedpyridine.% The bands at 1654 and 1497 cm-l are the characteristic peaks of pyridinium ion, which are formed on the Brtlnsted acid sites.36 The other set of absorption of peaks at 1443,1482, 1680, and 1606 cm-l is contributed by pyridine coordinatively bonded to Lewis acid sites. It is clear that both Br(lnsted and Lewis acid sites exist on the surface of CrO,/ ZrO2 catalysts calcined at 873 K. Conclusions Thispaper hae shownthat a combination of FTIR,XPS, and XRD can be used to perform the surface character(31) Itoh, M.; Hattori, H.; Tannbe, T. J. Catal. 1974,35,225. (32) M k o , V. A. h o c . Int. Congr. Catal., 3rd 1964, Vol. 1, No. 19. (33) Miura, M.; Kubota, Y.; Iwaki, T.; Takimoto,K.;Muraoka, Y. Bull. Chem. Soc. Jpn. 1969,42,1476. (34)Parry, E. P.J. Catal. lW, 2,371. (36)Kung, M.C.; Kung,. H. H. Catal. Rev. Sci. Eng. 1986,2,425. (36)Connell, G.; Dumeeic, J. A. J. Catal. 1987,106, 285.

I

10

1600

I

I

1500

wavenrimber/cm-

1400 1

Figure 10. IR spectra of pyridine adsorbed on l-CrO,/ZrOz: (a) background of 1-CrOJZrOz after evacuation at 773 K for 1 h; (b) pyridine adsorbed on sample a followed by evacuating at room temperature for 1 h; (c) pyridine adsorbed on sample a followed by evacuating at 523 K for 1 h.

ization of CrO,/ZrOz prepared by dry impregnation. The interaction between chromium oxide and zirconia influencesthe physicochemicalproperties of prepared catalysta with calcinationtemperature. The presence of chromium oxide delays the phase transitions of zirconia from amorphous to tetragonal and from tetragonal to monochic, and the specific surface area of catalysts increaeea in proportion to the chromium oxide content. Cr(VI) concentration of l-CrO,/ZrOz calcined at 873 K is quite large, being 54% ! of the totalCr concentration, and arCr203 is observed only at the calcination temperature above 1173 K,indicating that the Z r O z support stabilizes supported chromium oxide, and chromium oxide is well dispersed on the surface of ZrO2. In the treatments by oxygen, hydrogen, carbon monoxide, and n-hexanethe reductionoxidation between Cr(VI) and Cr(II1) is reversible, as observed by XPS and IR. Upon the addition of only a small amount of chromium oxide (1wt 95 Cr) to 2102, both the acidity and acid strength of catalmt increase remarkably, showingthe presence of two kindsof Brt5nst.d and Lewis acid sites on the surface of CrO,/ZrO2.

Acknowledgment. Thispaper was supported by NON DIRECTED RESEARCH FUND, KO= Reesarch FOUdation, 1990. Registry No. Zr0~ 1314-23-4;CO, 630-08-0;chromiumoxide, 11118-57-3;n-hexane, 110-54-3.