Determination of framework aluminum content in zeolites X, Y, and dealuminated Y using unit cell size

Determination of framework aluminum content in zeolites X, Y, and dealuminated Y using unit cell size

LETTERSTO THE EDITOR Determination of framework aluminum content in zeolites X, Y, and dealuminated Y using unit cell size Sohn et al. l found a very ...

170KB Sizes 0 Downloads 13 Views

LETTERSTO THE EDITOR Determination of framework aluminum content in zeolites X, Y, and dealuminated Y using unit cell size Sohn et al. l found a very good correlation between unit cell size and the framework aluminum content for. a normal zeolite Y and seven dealuminated zeolite Y samples. They compared their results with those of Breck and Flanigen~who studied the relationship between Si/Ai ratio and cell parameter of 37 hydrated sodium zeolites X and Y samples. Prior to the Sohn el al. paper, Fichtner-Schmittler et al. 3 and Beyer et al. 4 published the results of their studies of dealuminated samples very similar to those of Sohn and co-workers. 100-

/ A









(Figure 1).








i ~

NH4Y, used for dealumination, as mentioned in letter (not used in calculatingslope and intercept)


All 29 data points of Dempsey et al. and Kerr et al. lie between the straight lines reported by Sohn et al. and Breck and Fhmigen. The Mobil data obey a linear relationship with a correlation coefficient of .9965 compared with .9991 for the Sohn et al. data. This difference is not surprising since the Mobil data fit five different straight lines better than one straight line. Breck and Flanigen did not disclose a correlation coefficient. The data of Fichtner-Schmittler et al. and Beyer et al. lie between the line of Sohn et al. and the Mobil line. The slopes, m, and the intercepts, X, for the equation N A I = m(ao - X) are summarized, where N A I = framework aluminums per unit cell and ao = cell parameter in/~:

/eTi~Sample after HC~ treatment used in calculating /" / ( n o t / /slope and intercept)


/ 7/q~HC£- treated 0 i i 23.80 24!00 24.20 24.40 24~60 24'.80 25100 25120 o ao,A Figure 1 In the above figure, Line A represents the data of Breck and FlanJgen2 while Line B represents the data of Sohn et a l ) The open circles (C)) indicate the experimental points of Dempsey, et al. s and the closed circles (0) indicate the experimental points of Kerr, et al. ~


The purpose of this letter is to extend the comparisons to include studies conducted at the Central Research Division of Mobil Research and Development Corporation. Dempsey e/ al."-' studied 20 samples in the same compositional range as did Breck and Flanigen but with quite different resuhs: while Breck and Flanigen reported that one straight line described their correlation, Dempsey et al. found that three distinct straight lines were required to fit their data. This is similar to the findings of Breck and Flanigen on 20 samples of calcium zeolites X and Y. K~ihl~ reported that the data for 18 gallosilicate faujasites, containing from about 52-80 Ga/unit cell, also fit three straight lines better than one line. The large Ga-O bond length (relative to AI-O) effects more distinctive steps between the lines and thus strongly supports the veracity of the Dempsey et al. plots. The work of Kerr et al." on nine dealuminated Y samples covers a range of fi'amework aluminum similar to that of Sohn and co-workers. However, the latter effected dealumination by the SiCLI vapor technique of Beyer and Belenkaja, s whereas Kerr et al. used the EDTA procedure. In addition, Sohn et al. founds that their data fit one straight line, whereas the Kerr et al. data best fit two distinct straight lines

ZEOLITES, 1989, Vol 9, July

Breck and Flanigen Kerr et al. + Dempsey et al. Sohn et al.

Slope, m

Intercept, X

115.2 112.1 107,1

24.191 24.222 24.238

It is interesting to note that the Mobil nearly equal the averages of the other values. There are significant compositional between the Sohn et aL and the Kerr et

values very two sets of differences aL samples

Letters to the editor

that might f u r t h e r explain some of the differences observed above. T h e Sohn et aL samples contain relatively large amounts of nonfi'amework alunfin u m , some of which may occupy cation-exchange sites. T h e Kerr et al. samples appear to contain only framework aluminum, in which the cations are predominantly a m m o n i u m ion with a small a m o u n t of h y d r o n i u m ion. One final observation: the data point for the a m m o n i u m zeolite Y used by Kerr el al. ft," their dealuminations lies significantly outside the three linear plots described above, that is, to the right of the Sohn et al. plot. However, this point corresponds almost exactly to the data very recently reported by Fritz et al. ° for a highly exchanged a n m m n i u m zeolite Y in which "a u.c. parameter of ca. 24.78 was found to be the largest attainable for a Si/AI ratio equal to 2.55."

REFERENCES "1 Sohn, J.R., DeCanio, S.J., Lunsford, J.H. and O'Donnell, D.J. Zeolites 1986, 6, 225 2 Breck, D.W. and Flanigen, E.M. in Molecular Sieves, Society of Chemical Industry, London, UK, 1968, p. 47 3 Fichtner-Schmittler, H., Lohse, U., Engelhardt, G. and PatzeIova, V. Cryst. Res. Tech. 1984, 19, K1 4 Beyer, H.K., Belenykaja, I.M., Hange, F., Tielen, M., Grobet, P.J. and Jacobs, P.A.J. Chem. Soc., Faraday Trans. 1 1985, 81, 2885 5 Dempsey, E., K0hl, G.H. and Olson, D.H.J. Phys. Chem. 1969, 73, 387 6 K0hl, G.H., in "Molecular Sieve Zeolites-l," Adv. Chem. Ser. 1971, 101,199 7 Kerr, G.T., Chester, A.W. and Olson, D.H. Acta Phys. Chem. 1978, 24, 169 8 Beyer, H.K. and Belenkaja, I., in Catalysis byZeolites (Eds. B. Imelik et al.) Elsevier, Amsterdam, 1980, p. 203 9 Fritz, P.O., Lunsford, J.H. and Fu, C.-M. Zeolites 1988, 8, 205

George T. Kerr 10 Pin Oak Drive Lawrenceville, NJ 0 8 6 4 8 USA

On the correlation between the results of 129Xe n.m.r. line shift and 1H n.m.r, pulsed-field-gradient measurements on zeolites In the last few years, with the n.m.r, pulsed-fieldgradient techn!,que I and the I~'Xe n.m.r, line-shift measurements,- two novel spectroscopic methods with special sensitivity to the nficrostructure- and structure-related mass transfer have been introduced into zeolite research. Leading to a numl)er of completely new concepts for the u n d e r s t a n d i n g of zeolite structure and zeolitic diffusion, both methods have been subjected to a series of crucial experiments d e m o n s t r a t i n g their self-consistency as well as their compatibility with a large series of results ohtained by conventional, well-established techniques. :s'~ Today, the a m o u n t of accumulated data in a few cases already allows the direct correlation of the infi)rmation provided by either of these novel techniques.

O n the location o f residual water molecules In Ref. 5, a s u m m a r y of the v-'"~Xen.m.r, chemical shifts of xenon adsorbed on zeolite NaY with coadsorbed water molecules is provided. With respect to the water content, the representations of the chemical shift (~) vs. the xenon concentration (c) yield three different types of dependencies. For the first two to three water molecules per Vs unit cell, the chemical shifts are found to increase linearly with increasing xenon concentration, with plots completely coinciding for different water

amounts. One has to conclude, therefore, that these first water molecules are captured by the sodalite cages ~ that are unaccessible by xenon. F u r t h e r coadsorbed water molecules lead to a parallel shift of the 8 vs. c plots to higher values. This means that the propagation of the xenon molecules is now progressively reduced by the zeolite structure with the included water molecules (increasing intercept of the 6 vs. c plot with the ordinate), whereas the rate of mutual encounters between the xenon molecules is unchanged (constant slope of the <5vs. c ph)t).'-' It is evident that the most probable explanation of this situation is the assumption that the water molecules u n d e r consideration will lead to the formation of water cation complexes in the windows between the adjacent supercages, but which Ieave the microdynamic situation within the individual supercages, that is, the probability of mutual encounters of the xenon molecules, unchanged. Only for the largest water concentrations does the increase of the slope of the c5 vs. c plots indicate an increase in the rate of molecular encounters between the xenon molecules, that is, a perceptible reduction of the intracrystalline free space. N.m.r. self-diffusion studies of short-chain-length hydrocarbons in NaX-type zeolites with coadsorbed water molecules v have led to just the same conclu-

ZEOLITES, 1989, Vol 9, July