1358
J. Phys. Chem. 1980, 84, 1358-1360
(12) E. Ruckensteln and R. Nagarajan, J . Phys. Chem., 79, 2622 (1975); R. Nagarajan and E. Ruckensteln, J. Collo/dIntedace Scl., 60, 221 (1977). (13) P. Mukerjee, Adv. Colloid Interface Sci., 1, 241 (1967); J . Phys. Chem., 76, 565 (1972); in "Micellization, Solubilization and Microemulsions", K. L. Mittal, Ed., Plenum Press, New York, 1977, p 171. (14) A. A. Abramzon, Russ. J . Phys. Chem., 52, 678 (1978). (15) A. S. Kertes, H. Gutrnann, 0. Levy, and G. Markovits, Isr. J. Chem., 8, 421 (1968). (16) N. Muller, J. Phys. Chem., 79, 287 (1975). (17) N. Muller, J . Colloid Interface Sci., 63, 383 (1978). (18) A. Bondi, J. Phys. Chem., 58, 929 (1954), Figure 7.
(19) A. Bondl, "Physical Properties of Molecular Crystals, Liquids and Glasses", Wlley, New York, 1968. (20) W. J. Taylor, J. Chem. Phys., 18, 257 (1948). (21) R. Foster, "Organic Charge Transfer Complexes", Academic Press, New York, 1969. (22) R. C. Llttle and C. R. Singleterry, J. Phys. Chem., 88, 3453 (1964). (23) The reduction of the translatlonal freedom of motion always provides a positive contributlon to the total free-energy change associated with aggregation. However, a large positive quantity representing a part of the above contribution is already included in the entropy term of eq 3. Consequently, the remalning part which re resents the contrlbution to the standard free-energy change AG, may be elther positive or negative.
8
Acid-Site Distribution on Faujasite-Type Zeolltes Determlned by n-Butylamlne Titration. 1. Comparison of the Use of Hammett and Arylmethanol Indicators on Zeolltes X Dlanne Atklnson and Geoffrey Curthoys" Department of Chemistry, Unlversb of Newcastle, Newcastle 2308, Australia (Received August 23, 1979) Publlcatlon costs assisted by the Universky of Newcastle
The determination of the acidity of the cation-exchanged forms of zeolite X (Na+,Ca2+,La3+,H+)was undertaken by the nonaqueous n-butylamine titration method, the endpoint being determined visually by using a series of Hammett indicators for the combined Bronsted and Lewis acidity and a series of arylmethanol indicators to distinguish the Bronsted acidity. Comparison of the visual results from the two types of indicators shows that the arylmethanol indicators were unsatisfactory for visual use, probably because of unreliable color changes due to physical adsorption on the zeolite surface. The distributions of the acid sites found from the recommended visual Hammett indicators were reliable and showed a greater heterogeneity of acid strengths than previous studies. The number and strength of acid sites were increased with a high exchange of sodium ions by multivalent cations, especially for the more highly polarizing La3+cation. The acidity of the H form was slightly less than the La form and was not thermally stable. Introduction It is important to be able to determine the nature, strength, and distribution of acid sites on the surface of solid catalysts in order to be able to tailor a catalyst to a specific reaction. The catalytic activity of many surfaces, for example, in petroleum refining has been linked strongly with the presence of acid sites. These may be Bronsted sites (proton donors), such as acidic structural hydroxyl groups, or Lewis sites (electron acceptors), such as the tricoordinated aluminum atoms of a zeolite surface. Not only the number of acid sites but also their strength is important in many catalytic conversions. For example, in the cracking of aromatic molecules, sites of high strength are required whereas the isomerization of alkenes is catalyzed by weaker sites.1,2 Emphasis in zeolite catalytic studies has been placed on the role of Bronsted-acid sites in the production of carbocations by proton addition to the adsorbed hydrocarbons. Many correlations between known maximum Bronsted acidity of a surface and maximum conversion of hydrocarbons have been made. The roles of Lewis-acid sites and cation sites are not very clear. Lewis-acid sites have been reported to have a modifying effect on the strength of Bronsted-acid sites that are present3i4or may act directly by forming charge-transfer complexes. The framework cations may be involved in the catalysis of such reactions as the high-temperature cracking of alkanes over alkali-metal forms of zeolites X and Y,5!6 and it has been postulated that a free-radical mechanism applies on these surfaces which do not possess Bronsted acidity. Zeolites are finding rapidly increasing use as catalysts because of their many possible structures and hence acidity modifications. This investigation was undertaken to de0022-3654/80/2084-1358$01.OO/O
termine the acidity of a series of cation-exchanged zeolites: Na-X, Ca-X, La-X, H-X, dehydrated at 400 "C. The procedure chosen was that of n-butylamine titration in benzene, the endpoints being found for the distribution of Bronsted- and Lewis-acid sites by using a series of increasingly basic Hammett indicators and for the distribution of Bronsted-acid sites only by a series of increasingly basic arylmethanol indicators. Experimental Section Na-X zeolite was obtained from Union Carbide (Type 13X) and purified by repeated exchange with 1M NaCl solution. The composition of the zeolite was (Na20)o.95(A1203)(Si02)2,5.5.7H20. The cation-exchanged forms were prepared by repeated exchange of a pH 5 buffered suspension of the Na form with a solution of the appropriate chloride salt to give the percentage exchange shown in the parentheses. CaCl,. 2H, 0 Lac13* 7 H,0 NH,Cl
Ca(84.2)-X La( 8 6 .O )-X NH,( 69.2)-X
The ammonium-exchanged form was hented at 400 "C for 2 h to convert it into the decationated form H-X. All samples were dried overnight at 140 "C and then calcined in air at 400 "C for 2 h. The acidity of the surface was determined by following the procedure described by Benesi' with visible Hammett indicators as recommended by Drushel and Sommers8and arylmethanol indicators as introduced by Hirs~hler.~ The indicators used and their visible color changes are given in Table I. Samples (200-mg) of the zeolites were transferred in a drybox to each of 25 preweighed screwcap vials of approximately 15-cm3 volume. The solid was 0 1980 American Chemlcal Society
Acid-Site Dlstrlbutlon on FauJaslte-TypeZeolltes
The Journal of Physical Chemistry, Vol. 84, No. 11, 1980 1359
TABLE I
Satisfactory Visible Hammett Indicators wt % acid compound H,, H,SO, color
42
base
+8
+4
HR 0
-4
-8
-12
-16
1.41
color
6.8 8 X 10-8 yellow blue 4.6 yellow green benzeneazodimethylamine 3.3 3 x lo-' red orange bromthymol blue -1.5 red brown bromcresol green - 3.7 red brown
bromthymol blue bromcresol green
hylmethanol Indicators wt %
compound
HR
4,4',4"-trimethyltriphenylmethanol
triphenylmethanol diphenylmethanol
4,4',4"-trinitrotriphenylmethanol
- 4.04 -6.63 -13.3 - 16.3
%SO4 36 50 77 88
HR
12
8
+4
0
-4 -
-8
-12
-16
Ho Figure 2. Ackcstrength distribution of zedlte La-X 400 O C pretreatment (a) Hammett Indicators; (b) arylmethanol Indicators.
$0.4
1
t12
+8
+4
HR 0
-4
-8
-12
-16
-1.2
L +8
+6
+4
+2
0
-2
-4
-6
-8
HO
Flgure 1. Acld-strength distribution of zeolite Ca-X; 400 OC pretreatment: (a) Hammett Indicators; (b) arylmethanol indicators.
\
b
covered with 1 cm8 of dry benzene. Out of the drybox increasing aliquots of 0.04 M n-butylamine in benzene were added to the vials from a 10-cm3buret. The samples were equilibrated with shaking overnight and small amounts of each suspension transferred to small vials and tested with a few drops of a 0.1 7% solution of each indicator in benzene. The vials in which the appropriate color change occurred were noted, and the acidity was determined from the number of millimclles of n-butylamine which had to be added before the appropriate color change occurred. The results are shown in Figures 1 , 2 , and 3 for each of the cation forms and include both types of indicators.
(a) Hammett Indicators; (b) arylmethanol indicators.
Discussion The visual determination of the endpoint using colored indicators has the advantage of rapidity of operation as compared with spectrophotometric observations. For the indicator to be reliable visually, the color change must occur at the point of neutralization of all the surface acid sites of that strength to which the indicator refers. This study did not use those Hammett indicators such as benzalacetophenone ( I f o -5.6) and anthraquinone (Ho -8.2) which Drushel and Sommers8pointed out as leading to an overestimation of the surface acidity at these strengths because of faulty surface endpoints that are due to color changes upon physical adsorption on the surface as well as adsorption on the acid sites. However, Hammett indicators which are colored in both acid and base forms were found to be quite satisfactory. Previous studies on cracking catalysts with these indicators found that results obtained by using both visual and spectrophotometric determinations of the endpoint are in satisfactory agree-
ment.8 The results obtained in this investigation of zeolites X by using Hammett indicators with visual determination of the endpoint show a more heterogeneous distribution of strengths of acid sites than those found in previous studies with similar surfaces.l0-l2 This is possibly due to these authors' visual use of the aforementioned unreliable Hammett indicators which correspond to stronger acid sites. The Na-X zeolite exhibited only negligible and very weak Hammett acidity (3.3 < HoI7) which is catalytically unimportant. This was expected from the poor polarizing power of this monovalent cation either in the formation of acidic hydroxyl groups or by its acting as a Lewis acid, especially in the low electrostatic field environment of zeolite X with its low Si/Al ratio. The expected rise in the number and the strength of acid sites after a high-percentage exchange with the more highly polarizing multivalent cations was observed. The exchange for both Ca-X (84.2%) and La-X (86.0%) was sufficient for these cations
4
-6 HO Figure 3. Acid-strength distribution of zeolite H-X 400 O C pretreatment:
1300
The Journal of physlcal Chemistry, Vol. 84, No. 11, 1980
TABLE I1 acidity (Ho < zeolite 3.3), m m o l g-' Na-X Ca-X
La-X
H-X
pretreatment
0 0
4 0 0 " C in air 500 "C in air 0.3 (Ho < 5) 400 "C in N, 0.20 4 0 0 "C in air 0.35 500 "C in air 0.35 (H,< 5) 4 0 0 "C in N, 0.9 4 0 0 "C in vacuo 2.8 600 "C in vacuo 0.98 4 0 0 "C in air 1.2 600 "C in air 2.4 600 "C in vacuo 0.75 4 0 0 "C in air 0.50 (Ho
