9411
J . Phys. Chem. 1991, 95, 941 1-9415 TABLE IV: Structural Parameters for the Reference Compounds and tbe EXAFS Panmeters Reported for the Preparation of the Reference Filesa ~~
reference
compd shell N R , A ref n Pt-Pt 12 2.77 9 3 Pt foil 3 1.87 14 3 Ir,(CO),2 I r C
lr4(CO),2 1r-O'
3
3.01
Fourier transform k range, R range, A-I A ref
1.95-19.97 2.80-16.45 14 3 2.80-16.45
1.90-3.02 1.06-1.98 1.98-3.30
21 21 21
ONotation: n, power of k used in weighting of Fourier transform; k range, limits of forward Fourier transformation; R range, limits of shell isolation [lr-lr contribution (N = 3, R = 2.69 A) previously subtracted]. Source in Daresbury, U.K., where it was loaded into an EXAFS cell inside a drybox purged with N2. The loaded EXAFS cell was purged with He for 3 min at room temperature and closed under a positive pressure of He. A fraction of the sample was returned to Delaware and checked with infrared spectroscopy to confirm its stability during the handling. The spectrum showed no measurable change. The EXAFS data were collected at X-ray beamline 9.2 with Si(220) crystals in the monochromator. The sample was scanned at the Ir Lllledge (1 1 2 15 eV), and the data from three scans were averaged. The data were collected after the monochromator had been detuned by 50% to minimize the effects of higher harmonics
present in the X-ray beam. The sample contained 0.96 wt % Ir, as estimated from the step height in the X-ray absorption spectrum. Reference Materials. Pt foil and [Ir4(CO)12]were used as references for analysis of the EXAFS data. The reference data were collected at beamline 9.2 at Daresbury. The Ir-Ir contributions were characterized with an experimentally determined reference file obtained from EXAFS data for the Pt foil. The interactions between the Ir atoms and the terminal and bridging carbon atoms, Ir-Ct and Ir-Cb, respectively, were analyzed with the reference file characterizing the Ir-C interaction in [Ir4(CO)12]. The interactions between Ir and O* were also analyzed with a reference file obtained with [Ir4(CO)12]. The details of the generation of these reference files are described elsewhere.22 The structural parameters of the reference materials and the EXAFS parameters reported for the preparation of the reference files are shown in Table IV.
Acknowledgment. We thank Professor W. L. Gladfelter for helpful comments about the infrared spectra and ion pairing and for providing unpublished infrared spectra. This research was supported by the National Science Foundation (CTS-89 10633 and CTS-9012910) in the U.S.A., by the NWO in the Netherlands, and by a NATO travel grant. ~~~~
(22) van Zon, F. B. M.Ph.D. Thesis, Eindhoven University of Technology, The Netherlands, 1988.
Cation Location in La,Na-Y Zeolites by Two-Dimensional 23NaNutation NMR Chiung-Fang Lin and Kuei-Jung Chao* Department of Chemistry, Tsinghua University, Hsinchu, Taiwan, Republic of China (Received: December 3, 1990)
The variation of the cation distribution with the lanthanum-exchanged level in hydrated La,Na-Y zeolites was clearly monitored by two dimensional nutation sodium-23 NMR, in which the chemical shift and the quadrupole interaction could be separated. From a comparison of Z3NaNMR at different hydrated states and temperatures it is concluded that the mobility of water molecules and sodium ions can be reduced by decreasing sample temperature and the migration of lanthanum ions from supercages to small cages causes the redistribution of sodium ions after dehydration at 350 O C .
Introduction The acidity and thermal stability of synthetic Na-Y zeolite can be dramatically improved by ion exchange with lanthanum ions, and the amount of improvement is controlled by the degree of exchange and therefore the location of the cations.'.* The structural characteristics and the catalytic application of La,Na-Y zeolites have been extensively investigated; most of the published structural studies were done on a lanthanum-exchanged Y zeolite under different heat treatment^.^^ Few studies were found on ( I ) Ikemoto, M.;Tsutsumi, K.; Takahashi, H. Bull. Chem. Soc. Jpn. 1972, 45, 1330. ( 2 ) Jacobs, P.A. Carhniogenic Actiuiry ojZeolire; Elscvier: Amsterdam, 1977; p 125. (3) Costenoble, M. L.; Mortier, W. J.; Uytterhoeven, J. B. J . Chem. Soc., Faraday Trans. I 1978, 74, 466; 1976, 72, 1877. (4) Bennett, J. M.; Smith, J. V . Mater. Res. Bull. 1968, 3, 865; 1969, 4, 7; 1969, 4. 343. (5) Cheetham, A. K.; Eddy, M.M.;Thomas, J. M. J . Chem. Soc., Chem. Commun. 1984, 1337. (6) Jacobs, P. A.; Uytterhoeven, J. B. J . Chem. Soc., Faraday Trans. I 1973, 69, 373. ( 7 ) Scherzer, J.; Bass, J. L.; Hunter, F.D. J . Phys. Chem. 1975, 79,1194. (8) Rotssner, F.;Steinberg, K. H.; Winkler, H. Zeolites 1987, 7, 47.
the effect of the degree of cation exchange on the cation distribution in Y zeolites.I*l2 The determination of the cation distribution has involved a number of techniques, e.g., X-ray and neutron diffractions,ss infrared,+* and NMR.+I3 With use of diffraction methods, the stationary cations can be located, and about half of cations can often be located for hydrated samples by X-ray diffraction. With the IR method, the vibrational modes of framework and the adsorbed molecules are usually investigated, however, only limited information was obtained for the distribution of cations in zeolites. Our previous 29SiNMR study9,10showed that the variation of the cation location and occupancy at different sites led to the change of the electric field on framework silicon atoms around cation sites and hence the change of corresponding (9) Chao, K. J.; Chern, J. Y. J . Phys. Chem. 1989, 93, 1401. (10) Chao, K. J.; Chern, J. Y.; Chen, S. H.; Shy, D. S. J . Chem. Soc., Faraday Trans. 1990, 86, 3167. (1 I ) Welsh, L. B.; Lambert, S. L. In Characterization and Catalyst Development. ACS Symp. Ser. 1989, 411, 262-273. (12) Hayashi, S.; Hayamizn, K.; Yamamoto, 0. Bull. Chem. Soc. Jpn. 1987, 60, 10s. (1 3) Welsh, L. B.; Lambert, S. L. Perspectives in Molecular Sieve Science. ACS Symp. Ser. 1988, 368, 33-47.
0022-3654 /91/2095-94l1%02.50/0 0 1991 American Chemical Society
9412 The Journal of Physical Chemistry, Vol. 95, No. 23, 1991
29SiNMR. The correlation of the NMR spectrum of the cationic nucleus with its content may be more direct and more useful for probing the distribution of cations in zeolites. Using high-field magic angle spinning (MAS) technique, Welsh and Lambertiisi3observed the change of line shape with the cation content on the 23NaNMR spectra in the partially ion exchanged Na-Y zeolites. However, for the nucleus with spin number I > I /2, e+., 23Na(I=3/2)and '39La(I=7/2), only the central transition line is usually investigated and broadened by the second-order quadrupole interaction, and hence the discrimination of cations in different sites via chemical shift is difficult and complicated. These problems may be overcome by using the two-dimensional nutation technique, by which the quadrupole interaction and the chemical shift'"" can be separated, as demonstrated by the recent studies on zeolite Na-Ai8*I9and sodalite.20*2iIn these studies, the effects of water content, temperature variation, and phase transition at the sodium sites were followed. In the present study, we have investigated the 23NaNMR spectra of a series of partially lanthanum-exchanged Na-Y zeolites; the distributions of the sodium ion as the function of lanthanum exchange level are studied. The static two-dimensional nutation spectra were compared with one-dimensional magic angle spinning and static spin-echo data on hydrated, dehydrated, and rehydrated La,Na-Y zeolites.
Lin and Chao MAS
Static
b
t
I
150
0
-150
I
150
0
-150
m iv Figure 1. 23Naspin echo and MAS NMR spectra of hydrated Na-Y and
La,Na-Y samples. 31 L
i z' I
Experimental Section A binder-free Na-Y zeolite was obtained from Strem Chemical Co. To avoid possible deficiency in cations, the zeolite sample was washed with I N NaCl solution, and deionized water subsequently then dried at room temperature. La,Na-Y zeolites were prepared by exchanging Na-Y zeolite with LaC13solution at room temperature. The percentage of metal ion exchanged in the zeolite was determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES) and the number is used as a prefix for the samples, e g , a La-exchanged level of 60% is represented as 60La,Na-Y. The bulk and framework Si AI ratios of 2.29 and 2.4 were determined by ICP-AES and 9Si MAS NMR, respectively. No extraframework octahedral aluminum was detected by 27AIMAS NMR on hydrated Na-Y and La,Na-Y zeolites. Dehydrated samples were evacuated under shallow-bed conditions and heated from room temperature to 150 OC with a heating rate of 0.2 "C/min, and maintained at 150 "C for 4 h; then the temperature was raised to 350 OC at 0.2 OC/min and held at 350 "C for -30 h to a final equilibrium pressure of lV5Torr. Samples were then cooled under vacuum and sealed in a 10-mm glass tube. The rehydrated samples were prepared by exposing the dehydrated samples under water vapor for at least 3 days over saturated NHICl solution at room temperature. The spectra of hydrated and rehydrated samples were obtained from samples placed in a 7-mm MAS rotor. Most of the water molecules were desorbed during heating treatment from room temperature to 350 "C; weights are 23.5-25.0% and 24.2-24.8% of the hydrated and rehydrated samples, respectively, detected by thermogravimetric analysis under Ar flow of 50 cm3/min and heating rate of 10 OC/min. The total weight loss up to 800 OC is 24.4-25.9% for hydrated samples.
i
(14) Kundla, E.; Samoson, A.; Lippmaa, E. Chem. Phys. Lett. 1981,83, 229. (15) Samoson, A.; Lippmaa, E. Phys. Rev. B 1983,28,6567; Chem. Phys. Lett. 1983, 100. 205. ( W ) Fenzke, D.; Freude, D.; Frohlich, T.; Haase, J. Chem. Phys. Lett. 1984. / / -/ .- 171. -, (17) Kentgens, A. P. M.; lemmens, J. J. M.; Geurts, F. M. M.; Veeman, W. S . J . Magn. Reson. 1987, 71, 62. (18) Tijink, G.A. H.; Janssen, R.; Veeman, W. S. J. Am. Chem. SOC. 1987, 109, 7301; J . Chem. Phys. 1988, 8, 518. (19) Janssen. R.;Tijink, G. A. H.; Veeman, W. S.; Maesen, Th. L. M.; van Lent, J. F. J. Phys. Chem. 1989, 93, 899. (20) Janssen. R.; Breuer, R. F. H.; de Boer, E.; Geismar, G. Zeolites 1989, 9, 59. (21) Engelhardt, G.; Buhl, J.-Ch.; Felsche, J.; Foerster, H. Chem. Phys. Lett. 1988. 153, 332. (22) Kentgens, A. P. M. Ph.D. Thesis, Universiteit te Nijmegen, The Netherlands, 1987.
----.
I
i
/
t I
0
20
60
40
La'' exchange
80
X
Figure 2. Variation of line widths of static (A,A) and MAS (0, 0) NMR spectra with lanthanum content on hydrated (---) and re-
hydrated (-)
La,Na-Y zeolites.
23Na NMR spectra were recorded on a Bruker MSL-200 spectrometer. A duration of 0.5 s between scans was allowed for nuclear spin relaxation. A minimum of 500 scans were accumulated for each spectrum. The chemical shifts were calibrated against a saturated aqueous solution of NaCl with high-field shift as negative. MAS experiments were performed at a spinning speed of 3-5 kHz. For static experiments, one- and two-dimensional spectra were collected by spin-echo and nutation techniques, respectively. The two-dimensional nutation experiments were carried out with optimum radio frequency fields and t , values ranging from 0 to 126 ps in 2-ps steps. Typically, 64 rows of 1K data were collected with the first row being zeroed data. Each twodimensional experiment took about 6-16 h of spectrometer time. Data were plotted as white-wash stacked plots. The rehydrated and dehydrated samples showed no degradations by 23NaNMR after 1 month and 1 week, respectively.
Results One-Dimensional NMR. The 23Na static and MAS NMR spectra of hydrated La,Na-Y are shown in Figure 1. Spectra obtained from both techniques show a broad peak at E O ppm corresponding to the sodium-23 central transition -I/*). An increase of N M R line width with increasing degree of La exchange is evident on the hydrated samples, while the 23Naline width of rehydrated La,Na-Y decreases slightly with their lanthanum content (Figure 2). The line width of the spin-echo spectrum is larger than that of the MAS spectrum, especially on the highly La-exchanged sample. The center of the 23NaMAS NMR peak was found to shift to the high field slightly with the
-
The Journal of Physical Chemistry, Vol. 95, No. 23, 1991 9413
Cation Location in La,Na-Y Zeolites
I
I
I
0
300
-300
I
300
ppn
0 Fm
-300
Figure 3. 23Na spin echo spectra of hydrated (a), dehydrated (b), and rehydrated (c) Na-Y and 69La,Na-Y at 4.73 T.
LL 200
2O0C
-200
0 Fm
Figure 4. 23Naspectra of hydrated Na-Y zeolite as a function of tempera t u re.
lanthanum content, which is consistent with the study of Welsh and Lambert. However, the spectral deconvolution or simulation of quadrupole spins with the cation distribution at different sites is difficult because the NMR profile is broad and because of the function of nuclear quadrupolar interactions and the mobility of cations. The difference of line width between the static and MAS spectra of hydrated zeolites is large with La > 50% shown in Figure 2. The line widths at half-height, A v , , ~of , the static and MAS spectra of Na-Y and La,Na-Y (La exchange level > wQ, which results in a single line with nutation frequency of 1 wrf; (2) w,f
