Isomorphous Substitution of Boron in Mordenite and Zeolite Y - ACS

Chapter 26

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on April 9, 2016 | http://pubs.acs.org Publication Date: July 31, 1989 | doi: 10.1021/bk-1989-0398.ch026

Isomorphous Substitution of Boron in Mordenite and Zeolite Y Thomas R. Gaffney, Ronald Pierantozzi, and Mark R. Seger Air Products and Chemicals, Inc., Allentown, PA 18195

Synthetic limits on the extent of isomorphous substitution often limit our ability to modify the catalytic properties of zeolites. Here we describe two methods of substituting boron into mordenite, which together allow a wide range of substitution levels to be effected. Direct crystallization of mordenite from gels which contain borates results in low levels of boron incorporation into the framework structure. The level of boron substitution is primarily determined by the SiO /Al O ratio of the gel. Boron cannot compete with aluminum for a site in a growing crystallite, but boron will incorporate into mordenites crystallized from aluminum deficient gels. Using a post-synthetic treatment of dearuminated mordenite and zeolite Y, we obtained higher levels of boron substitution for aluminum. Boron-11 NMR experiments and silicon-29 cross polarization NMR experiments indicate that borate anions condense with hydroxyl nests and become part of the framework structure. The level of substitution is limited by the hydroxyl nest content of the dealuminated mordenite. The alpha test establishes that the n-hexane cracking activity of boron substituted mordenites is determined exclusively by the aluminum content. 2

2

3

When applied to z e o l i t e s the term "isomorphous substitution** refers to the replacement of s i l i c o n or aluminum atoms by elements with ionic r a d i i and coordination requirements which are compatible with the Τ (tetrahedral) s i t e s of the z e o l i t e structure. One method of preparing isomorphously substituted z e o l i t e s i s to include a reactive source of the replacement 0097-6156/89A)398-0374$06.00/0 ο 1989 American Chemical Society Occelli and Robson; Zeolite Synthesis ACS Symposium Series; American Chemical Society: Washington, DC, 1989.


26. GAFFNEYETAL.

375

Isomorphous Substitution of Boron

element i n the synthesis g e l . I f the solution chemistry of the reactive species i s compatible with gel formation and c r y s t a l l i z a t i o n processes, then an isomorphously substituted z e o l i t e may c r y s t a l l i z e . For t h i s to occur, the substituting element must be able to compete with S i or A l f o r addition to the growing c r y s t a l l i t e . The substituting element does not need to be involved i n the nucleation process, but i t must not i n t e r f e r e with nucleation f o r c r y s t a l l i z a t i o n to occur. When isomorphous substitution can be controlled, i t can be used to enhance desirable properties into z e o l i t e catalysts and adsorbents. A review on isomorphous substitution and i t s p o t e n t i a l c a t a l y t i c implications recently appeared (1). We now report our attempts to prepare boron substituted mordenites and z e o l i t e Y d i r e c t l y from gel precursors, and by post-synthetic treatment of c r y s t a l l i n e mordenite and z e o l i t e Y. Experimental Hexane (99% n-hexane) was obtained from A l d r i c h . S i l i c a was obtained as a 30% C o l l o i d (LUDOX-HS, 30 wt% S1O2) from DuPont. Aluminum hydroxide, boron oxide, and sodium metaborate were obtained from A l f a Products. Boric acid was obtained from J . T. Baker. Commercial samples of mordenite and z e o l i t e Y were obtained from the Norton and Union Carbide Companies, respectively. A s p e c i f i c example which i l l u s t r a t e s the t y p i c a l procedure used f o r preparing mordenite and z e o l i t e Y from gels i s given below. Solid state and S i MAS NMR spectra were obtained using a Bruker CXP-200 spectrometer and a Doty multinuclear double a i r bearing MAS probe. The samples were run as powders loaded into sapphire rotors with Kel-F endcaps, and were spun at the magic angle at approximately 3 kHz. Boron and s i l i c o n NMR chemical s h i f t s are r e l a t i v e to boron t r i f l u o r i d e etherate and tetramethylsilane, respectively. The ^-B NMR spectra were collected with and without using a s o l i d echo pulse sequence ( 2 ) with a 1 0 ms delay between 90° pulses. The r e p e t i t i o n times u t i l i z e d f o r B and S i MAS NMR were 1 s and 1 0 s, respectively; S i CPMAS NMR u t i l i z e d 1 . 5 s r e p e t i t i o n times and decoupling f i e l d strengths exceeding 50 kHz. The contact times u t i l i z e d f o r S i CPMAS were varied from 1 0 0 ys to 10 ms, with 1 ms giving strong CP i n t e n s i t y . 2 9

X1

2 9

2 9

2 9

Direct Synthesis. A solution was prepared by d i s s o l v i n g A 1 ( 0 H ) 3 , B2O3 , and NaOH i n deionized water. This solution was slowly added to s i l i c a s o l (30 wt%) with s t i r r i n g , to give a gel with the composition 22.9 Si02:AI2O3:3.2 Na O:0.91 B O :460 H2O. The gel was s t i r r e d f o r 1 0 min, and then heated at 165°C f o r 48 h i n a s t i r r e d , Teflon lined, s t a i n l e s s s t e e l pressure vessel. After 48 h the reactor was cooled to room temperature, and the product was separated by f i l t r a t i o n and washed with copious quantities of water and dried at 120°C. The X-ray powder pattern of the product i s c h a r a c t e r i s t i c of mordenite, with no extraneous peaks. 2

2

3

Occelli and Robson; Zeolite Synthesis ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

376

ZEOLITE SYNTHESIS


Zeolite Y gels were prepared by the same procedure outlined above. These gels were aged f o r 24 h at room temperature p r i o r to c r y s t a l l i z a t i o n . They were c r y s t a l l i z e d at 95°C f o r 64 h i n polypropylene bottles without s t i r r i n g (3). The products were separated by f i l t r a t i o n , washed with copious amounts of water, and dried at 120°C. Post-Synthetic Boron Incorporation. The acid form of mordenite (Zelon 900H) was heated at 90*C i n 14 M n i t r i c acid f o r 24 h with stirring. The product was washed with water to a neutral pH. The samples were analyzed f o r Na, S i , and A l content by x-ray fluorescence spectroscopy. The extent of dealumination was found to vary f o r d i f f e r e n t lots of Zelon 900H. For a given l o t , the extent of dealumination increased with acid extraction time. The extent of dealumination was controlled by adjusting the extraction time. Boron oxide (1.50 g B 2 O 3 ) was dissolved i n 140 mL of 0.25 M KOH. Dealuminated mordenite (25 g; S i / A l = 64) was added to the solution, and the pH was adjusted to 13 by addition of KOH. The r e s u l t i n g suspension was s t i r r e d at 80°C for 24 h i n a Teflon lined Parr pressure vessel. The product was separated by f i l t r a t i o n , thoroughly washed with deionized water, and dried at 120°C. The composition of the product was determined by XRF and atomic absorption spectroscopy. Zeolite Y was dealuminated with EDTA according to a published procedure (4). Boron oxide (0.15 g) was dissolved i n 140 mL of 0.22 M KOH. Z e o l i t e Y (5 g, S i / A l = 9.1) was suspended i n the solution. The suspension was heated at 80°C f o r 24 h i n a Teflon vessel. Products were washed thoroughly with deionized water and dried at 120 C. The chemical composition of products prepared by d i r e c t synthesis and by post-synthetic methods are l i s t e d i n Table 1. e

Table 1.

Chemical Composition of Boron Substituted Mordenite Samples

Method of Preparation Direct Direct Direct Direct Direct

Weight % (Dry Basis) A1 0 S1O2

Synthesis Synthesis Synthesis Synthesis Synthesis

Post-Synthetic Post-Synthetic Post-Synthetic Post-Synthetic

Substitution Substitution Substitution Substitution

2

3

80,.7 85,,7 88,.1 87,.9 88,.3

12,.1 8,.9 7..1 6,,7 5,.9

76..7 91,.1 94,.0 88,.9

11..9 8,.3 4..7 1,.26

Na 0

K0

B 0

7,.81 5,.91 5,.38 5,.03 4,.18

0.04 0.01 0.01 0.01 0.01

0.02 0.11 0.29 0.66 0.57

2

--

2

10.2 a a 6.1

a) Sample analyzed i n the acid form.


2

3

0.07 0.52 1.22 3.68

26.

GAFFNEY ET AL.

377



Mordenite samples that were not i n the acid form were exchanged into the ammonium ion form by three successive treatments with 0 . 2 M NH4NO3 at room temperature f o r 4 h p r i o r to c a t a l y t i c testing. The acid form of the z e o l i t e s was generated by heating the ammonium form at 540°C i n a i r f o r one hour. The published procedure f o r the alpha test (5) was used with modification. Hexane ( 2 0 % i n helium) was passed over the catalysts (lg) at a s u p e r f i c i a l flow rate (F) of 3 4 . 1 mL/min. The effluent gases were sampled a f t e r the catalyst was on stream for 3 . 0 min. F i r s t order rate constants and alpha values were calculated as previously described (5). Results and Discussion Direct Synthetic Substitution Mordenite. I t has been reported that p a r t i a l substitution of boron i n mordenite occurs when boric acid i s added to the synthesis gel (6.7). Evidence f o r boron substitution f o r aluminum i n mordenite samples prepared i n b o r o s i l i c a t e glass reactors has also appeared ( 8 ) . We sought to determine the extent of boron substitution into the mordenite framework which can be affected by d i r e c t synthesis. Mordenite can be synthesized from gels comprised of N a 0 , A I 2 O 3 , S i 0 , and H 0 (9-13). More s i l i c e o u s products are formed as the S 1 O 2 / A I 2 O 3 and S i 0 / N a 0 r a t i o s are increased (13). The range of mordenite compositions which can be prepared as pure phases by t h i s method are represented i n Formula 1. 2

2

2

2

2

Na [(Si0 ) _ (A102) J-yH 0 x

2

4 8

x

x

2

χ = 3-8

(1)

Gels i n the composition range S 1 O 2 / A I 2 O 3 = 10-25 and S i 0 2 / N a 2 0 = 0 . 1 - 0 . 3 c r y s t a l l i z e to give pure mordenite. We carried out syntheses i n which Β 0 replaced varying amounts of A I 2 O 3 i n the gels. Figure 1 correlates the S 1 O 2 / A I 2 O 3 r a t i o of the products with the r a t i o i n the gel. The straight l i n e p l o t shows that the s i l i c a to alumina r a t i o of the product i s determined by the s i l i c a to alumina r a t i o of the g e l . Inclusion of reactive borates i n the gel ( c i r c l e s i n Figure 1 ) has no e f f e c t on the amount of aluminum incorporated into the mordenite structure. Apparently boron cannot successfully compete with aluminum f o r l a t t i c e s i t e s i n the growing c r y s t a l l i t e s . Boron does not i n t e r f e r e with the c r y s t a l l i z a t i o n of mordenite when 50% or less of the A I 2 O 3 i n the gel i s replaced by B 2 O 3 . However, higher levels of 2°3 - t i n c o - c r y s t a l l i z a t i o n of non-microporous phases, such as aluminum borate and sodium b o r o s i l i c a t e s . Although boron cannot successfully compete with aluminum for s i t e s i n the mordenite structure, the more s i l i c e o u s products contain s i g n i f i c a n t amounts of boron. Figure 2 shows the B 03 content of the products as a function of the S i 0 / A l 0 3 r a t i o i n the gel and i n the product. I t i s clear that i f enough aluminum i s available i n the gel to form an 2

3

r e s u i

B

2

2

2



378

ZEOLITE SYNTHESIS

0.6

r

10

15

20

25

30

SiOo / A l o O q Figure 2 . Boron content versus boron substituted mordenites.

S1O2/AI2O3

ratio for



379


GAFFNEY ET AL.

A l saturated mordenite (x = 8 i n Formula 1), the product contains only trace amounts of boron. As the S 1 O 2 / A I 2 O 3 r a t i o of the gel increases, and i n s u f f i c i e n t A l i s present to form the aluminous mordenite end-member, the boron content of the product r i s e s dramatically. The upper l i m i t on the S 1 O 2 / A I 2 O 3 l e v e l at which pure mordenite forms, which i s approximately 25, determines the upper l i m i t on boron s u b s t i t u t i o n l e v e l . For the most s i l i c e o u s preparations, the mole % boron i n the product i s proportional to the mole % boron i n the g e l . Figure 3 shows the c o r r e l a t i o n between wt % boron i n the gels and products. The amount of boron which substitutes into the mordenite structure increases with increasing boron concentration i n the g e l , and decreases with increasing aluminum content i n the gel. The l e v e l of substitution i s much more dependent on the aluminum content i n the gel than on i t s boron content. The explanation f o r t h i s i s found i n the higher a f f i n i t y of the structure f o r aluminum. Boron and aluminum compete f o r a limited number of s i t e s i n the mordenite structure as i t forms. When enough aluminum i s available to form an aluminum saturated mordenite structure (x = 8, Formula 1) then boron substitution does not occur. The most highly substituted material prepared by t h i s method i s N a ^ ^ K S i i ^ ) ^ . 5 ^ 1 0 2 ) 3 . 9 ( Β θ 2 ) . 6 } · Similar r e s u l t s have been reported (14) f o r c r y s t a l l i z a t i o n of ZSM-5 from gels which contain aluminum and boron. υ

Faujasite. Z e o l i t e Y was prepared from N a 2 0 , A I 2 O 3 , S 1 O 2 , and H 0 derived gels using Breck*s method ( 3 ) . We prepared four gels i n which 0 , 5 , 1 0 , and 25% of the molar amount of A 1 0 used to prepare z e o l i t e Y was replaced by B 2 O 3 . The gels were crys t a l l i z e d under i d e n t i c a l conditions. The y i e l d of c r y s t a l l i n e product decreased with increasing boron content i n the g e l . The major phases formed from the boron doped gels were α-Αΐ2θ3·ΟΗ (boehmite) and sodium aluminum s i l i c a t e hydrate; however, the undoped parent gel yielded highly c r y s t a l l i n e z e o l i t e Y as the only c r y s t a l l i n e product. In contrast to the mordenite system, addition of B2O3 to the gel i n h i b i t s nucleation or c r y s t a l l i z a t i o n of z e o l i t e Y. E a r l i e r work showed that another synthetic f a u j a s i t e , z e o l i t e X, can be c r y s t a l l i z e d from borate containing gels, but that borate i s occluded i n the z e o l i t e pores, and i s not incorporated into the framework structure (15,16). 2

2

3

Characterization of Substituted Boron. We used s o l i d state ^-B NMR and X-ray d i f f r a c t i o n data to d i s t i n g u i s h occluded borates from boron substituted into the z e o l i t e framework. When an element replaces aluminum or s i l i c o n i n a z e o l i t e structure, the l o c a l coordination environment changes to accommodate the new ion. Since B^ i s a much smaller ion than Al***, the u n i t c e l l axes contract when boron replaces aluminum i n the framework. The ionic r a d i i of t r i v a l e n t Β and A l i n a tetrahedral environment are 0.25 Â and 0 . 5 3 Â, respectively (1). The magnitude of the contraction i s dependent upon the l e v e l of substitution (17). The l a t t i c e constants of a boron substituted mordenite sample with S i 0 / A l 0 3 = 25.3 and S i 0 / ( A 1 0 3 + B 0 ) = 2 2 . 5 +

2

2

2

2

2

3



380

ZEOLITE SYNTHESIS

Figure 3. Boron substitution l e v e l versus gel composition for boron substituted mordenites.


26.

381


GAFFNEYETAL.

were refined from corrected XRD data, obtained using corundum as an internal standard. Table 2 shows the v a r i a t i o n i n the unit c e l l dimensions of mordenite as a function of the S i 0 / A l 0 3 r a t i o of the sample (13.18). The unit c e l l values f o r the boron containing sample, designated "synthetic", are well below the values of mordenite and aluminum d e f i c i e n t mordenite samples. In each dimension, the value i s s i g n i f i c a n t l y lower than would be expected i f the structure were comprised of S i and A l with e x t r a - l a t t i c e boron occluded i n the pore system. 2

2


Table 2:

Unit C e l l Dimensions of Mordenites and Boron Substituted Mordenites

00

b(A) 20.53

c(A) 7..528

V(A ) 2798

12..5

00

18.11

20.38

7,.49

2764

12..7

90

18.05

20.34

7 .46

2741

60,.0

00

18.06

20.26

7 .46

2730

59,.9

14

17.84

20.03

7 .36

2627

Si/Al 5,.0

a Synthetic b PostSynthetic

3

a(A) 18.11

Sample a

Si/B

a) Literature values from (13)

b) Literature value from (18)

X 1

Solid state B NMR i s a very sensitive probe of the s i t e symmetry of boron (8,19). For s o l i d samples, the quadrapolar interaction gives r i s e to resonance lineshapes that are very dependent on the symmetry of the boron environment, even i f Magic Angle Spinning (MAS) techniques are used. Boron atoms located at s i t e s of tetrahedral symmetry have vanishingly small e l e c t r i c f i e l d gradient (EFG) anisotropics, and thus have very narrow MAS NMR lineshapes. Trigonal boron s i t e s however, have large EFG anisotropies and give r i s e to very broad MAS lineshapes i n powder samples. Figure 4a i s a ^ B MAS NMR spectrum of a mordenite sample prepared from an aluminum d e f i c i e n t gel which contained B2O3. The sharpness of the peak indicates a tetrahedral boron location, and the chemical s h i f t agrees with previously reported values for boron i n a z e o l i t e framework (8). In contrast, extra- l a t t i c e boron i n mordenite (vide infra) gives a broad resonance, as shown i n Figure 4b. The NMR and X-ray d i f f r a c t i o n data are only consistent with substitution of boron into the framework structure of the mordenite. Although we prepared boron substituted mordenite d i r e c t l y from modified gels, d i r e c t synthesis has severe limitations. The solution chemistry of the substituting element can interfere with z e o l i t e nucleation and c r y s t a l l i z a t i o n , as


382

ZEOLITE SYNTHESIS observed f o r boron i n faujasite-type syntheses. In syntheses where there i s no apparent interference with the c r y s t a l l i z a t i o n processes, the amount of substitution that occurs may be limited by the s o l u b i l i t y of the replacement element i n the gel during c r y s t a l l i z a t i o n . The i n t r i n s i c a f f i n i t y of the c r y s t a l l i z i n g s o l i d f o r depleting one gel species r e l a t i v e to others may also l i m i t substitution. Because of these l i m i t a t i o n s , i t was only possible to prepare mordenites which contain low levels of boron by d i r e c t synthetic methods.


Post-Synthetic Substitution Methods f o r isomorphously substituting preformed z e o l i t e structures have the potential f o r allowing a wider range of compositions to be prepared. When the z e o l i t e structure i s already formed, the compatibility of the solution chemistries of the replacement element with aluminum and s i l i c o n i s not a concern, and nucleation and c r y s t a l l i z a t i o n are not required. Several methods were attempted f o r e f f e c t i n g isomorphous substitution i n preformed c r y s t a l l i n e z e o l i t e s . Decationized samples of mordenite and z e o l i t e Y were refluxed with boric acid, sodium metaborate, and B 2 O 3 i n aqueous solutions with pH varying between 4 to 12. The products exhibit broad B NMR signals which are c h a r a c t e r i s t i c of boron i n a t r i g o n a l environment (Figure 4b). After thorough washing only trace amounts of boron (ca. 0.03%) remain i n the product. Boron apparently cannot displace aluminum or s i l i c o n from the framework of a preformed structure under these conditions. When treated under the proper conditions, aluminum can be removed from z e o l i t e s without loss of c r y s t a l l i n i t y (20). I t has recently been demonstrated that f o r z e o l i t e Y the e x t r a - l a t t i c e aluminum i s incorporated into the framework when the sample i s treated with KOH (21). We sought to determine i f dealuminated z e o l i t e frameworks are reactive, and are isomorphously substituted when treated with borate s a l t s i n basic medium. After we completed our studies, a report of boron incorporation into z e o l i t e Y using similar methods appeared (22). 1 1

Mordenite. We dealuminated mordenite with acid and treated the product with B 2 O 3 i n basic solution. The products obtained contain high levels of boron. The amount of boron i n the product i s dependent upon the level of dealumination of the mordenite. Figure 5 i s a plot of the boron content of the product versus the S i / A l l e v e l of the s t a r t i n g material. More heavily dealuminated samples contain more boron a f t e r a KOH/B2O3 treatment. The data point o f f of the curve ( c i r c l e i n Figure 5 ) i s f o r a s y n t h e t i c a l l y prepared s i l i c e o u s mordenite, whereas the data points on the curve are f o r acid dealuminated ( s i l i c e o u s ) mordenites. I t i s clear that materials which have the same aluminum content prepared by these two methods react d i f f e r e n t l y with borate anions. An acid dealuminated mordenite with S i / A l 12 takes up almost ten times as much boron than i t s s y n t h e t i c a l l y prepared analogue.


GAFFNEYETAL.



26.

Figure 5. Boron uptake of synthetic and dealuminated mordenites.


383

384

ZEOLITE SYNTHESIS The observed r e a c t i v i t y difference between z e o l i t e s and acid dealuminated z e o l i t e s i s d i r e c t l y attributable to the presence of i n t r a z e o l i t i c hydroxyl nests i n the acid treated samples. Two resonances are present i n the S i MASNMR spectra of acid dealuminated mordenites (Figure 6), which we assign to s i l i c o n with no nearest neighbor aluminum atoms (-112 ppm), and to SiOH "nests" (-103 ppm). Cross p o l a r i z a t i o n (CP) NMR techniques corroborate the assignment of the resonance at -103 ppm (vide i n f r a ) . The spectra i n Figure 6 demonstrate that the -103 ppm resonance i s enhanced by the cross p o l a r i z a t i o n technique, confirming the presence of hydroxyl nests i n our dealuminated samples. Figure 7 shows the spectra of the same sample after treatment with B2O3/KOH. The resonance at -111 ppm i s broadened, presumably due to the presence of boron and aluminum atoms i n s i t e s adjacent to S i , and the CP resonances have almost disappeared. These two changes are consistent with boron reacting at the internal hydroxyl nest position as shown i n Scheme 1.


2 9

Scheme 1:

Framework Substitution

Si / 0 H

S i —OH

Si / 0 HO — S i

B(0H)4

Si—0-B-0 —Si

H 0

0

\

Si

+ 4H 0 2

\.

Si

i L

The B NMR spectrum of t h i s sample contains a single narrow resonance centered at -3.2 ppm, which i s c h a r a c t e r i s t i c of boron i n a tetrahedral coordination environment i n the framework structure. The S i NMR spectra of a s y n t h e t i c a l l y prepared s i l i c e o u s mordenite with the same S i / A l r a t i o i s shown i n Figure 8. No CP resonances are present, which indicates that hydroxyl nest concentration i n t h i s material i s very low compared to the acid treated sample. These data confirm that hydroxyl nests, generated by the removal of A l from the z e o l i t e structure, are reactive s i t e s f o r isomorphous substitution. Aluminum d e f i c i e n t , preformed z e o l i t e s which do not contain hydroxyl nests, i . e . s y n t h e t i c a l l y prepared samples, do not undergo isomorphous substitution when treated i n a similar fashion. The amount of boron substitution achieved using this post-synthetic method i s approximately s i x times higher than the most heavily substituted mordenite prepared by d i r e c t synthesis from gels (vide supra). I f t h i s post-synthetic treatment 2 9


GAFFNEYETAL.



26.

Figure 7. Single pulse excitation and cross p o l a r i z a t i o n S i NMR spectra of boron mordenite prepared from dealuminated mordenite. 2 9


385


386

ZEOLITE SYNTHESIS

F i g u r e 8. S i n g l e p u l s e e x i t a t i o n and c r o s s p o l a r i z a t i o n S i NMR s p e c t r a o f s y n t h e t i c m o r d e n i t e . 2

9


GAFFNEY ET AL.

387


produces a s t r u c t u r a l l y substituted z e o l i t e , then the u n i t c e l l axes should exhibit a large contraction upon KOH/B2O3 treatment. The l a t t i c e constants of the material with K3 9 [ ( S i 0 2 ) 4 4 . i ( A l 0 2 ) o . 7 ( 8 0 2 ) 3 . 2 ^ refined from corrected XRD data, and are l i s t e d i n Table 2 along with the constants f o r mordenite and s y n t h e t i c a l l y prepared boron substituted mordenite. The large contraction of the u n i t c e l l observed f o r the heavily borated material i s strong evidence f o r s t r u c t u r a l substitution of boron i n these materials. The u n i t c e l l volume of the highly substituted structure (K+ ion form) i s 6.1% smaller than the c e l l volume of the aluminum analogue (H " form). Only a 1.6% contraction i s a t t r i b u t a b l e to the removal of aluminum. The major reduction i n u n i t c e l l volume (-4.5%) r e s u l t s from boron p u l l i n g the oxygen atoms i n hydroxyl nests closer together, as depicted i n Scheme 1. I t must be noted that although t h i s comparison i s between d i f f e r e n t ion forms, the potassium form should have a larger u n i t c e l l than the acid form i f no s t r u c t u r a l substitution occurred (23). w

e

r

e


4

Z e o l i t e Y. We also substituted boron into dealuminated z e o l i t e Y. We dealuminated z e o l i t e Y by EDTA treatment using standard methods (4). The presence of hydroxyl nests i n the product was confirmed using S i CPMAS NMR spectroscopy. The dealuminated material incorporated 33 times more boron than z e o l i t e Y when treated with KOH/B2O3. These data are summarized i n Table 3 . The boron substituted f a u j a s i t e exhibits a single sharp resonance i n the ^ B NMR spectrum, consistent with s t r u c t u r a l s u b s t i t u t i o n . Since the s u b s t i t u t i o n l e v e l was low and would not be expected to cause large s h i f t s i n the d i f f r a c t i o n pattern, no corrected XRD data were obtained on substituted z e o l i t e Y. Z e o l i t e Y does not r e c r y s t a l l i z e i n KOH solutions (24)· r e s u l t s are i n agreement with t h i s f o r Y, but f o r dealuminated Y z e o l i t e there i s a decrease i n the S i / A l r a t i o when treated with B2O3 i n KOH solution (Table 3 ) . We found a s i m i l a r trend i n the mordenite system. Apparently these z e o l i t e structures are more susceptible to r e c r y s t a l l i z a t i o n when dealuminated. Preparation of boron substituted z e o l i t e Y by post-synthetic substitution demonstrates that t h i s method may be used to prepare materials which are not r e a d i l y a v a i l a b l e by d i r e c t synthetic procedures. 2 9

0

Table 3 :

Sample Zeolite Y Dealuminated Y

Substitution Levels of Z e o l i t e Y and Dealuminated Z e o l i t e Y Starting Material Si/Al 2.6 9.1

Product Si/Al ppm Β 2.7 15 5.9

500


u

r

ZEOLITE SYNTHESIS

388

NMR Methods. The cross p o l a r i z a t i o n technique was used to transfer magnetization from to S i . Spin temperature a l t e r a t i o n ensures that only s i l i c o n atoms with a strong dipolar (through-space) interaction with protons w i l l appear i n the CPMAS spectrum. Only s i l i c o n atoms near immobile hydrogens ( i . e . , not exchanging i n the NMR time scale) w i l l be detected. Comparison of the FTMAS and CPMAS spectra i n Figure 6 shows a strong enhancement of the -103 ppm peak with only a weak enhancement of the -112 ppm peak assigned to Si(OAl). Application of the same experimental method to mordenite p r i o r to acid dealumination f a i l e d to detect any CP s i g n a l , indicating that the hydrogen atoms present i n the hydrated sample were a l l involved i n rapid chemical exchange. Furthermore, i t also demonstrates that the protons involved i n the cross p o l a r i z a t i o n to s i l i c o n i n the acid dealuminated sample are not from the adsorbed water of hydration. The v a r i a t i o n of the S i CPMAS spectrum was examined as the contact time f o r the cross p o l a r i z a t i o n experiment was changed. The most intense CP signal was obtained with r e l a t i v e l y short contact times, i n d i c a t i v e of short silicon-hydrogen internuclear distances. Maximum CP intensity i n the -103 ppm peak occurred at about 1 to 2 ms, suggesting *H to S i distances of a few angstroms or less f o r t h i s type of s i l i c o n . The Si(OAl) peak at -112 ppm showed a less intense CP signal, but with maximum enhancement occurring at longer contact times, i n d i c a t i v e of a greater average distance to immobile hydrogen atoms. The presence of the broad signal a r i s i n g from trigonal e x t r a - l a t t i c e boron species can only be r e l i a b l y detected by using a s o l i d echo pulse sequence (2). The broad lineshape i n Figure 4b, attributed to t r i g o n a l boron, was not detected by the simpler one-pulse NMR method. There are reports i n the l i t e r a t u r e that t r i g o n a l framework boron atoms may be present i n some dehydrated calcined z e o l i t e s (25). The broad signals attributed to t r i g o n a l framework boron r e v e r s i b l y disappear upon rehydration of the z e o l i t e sample, indicating conversion back to tetrahedral symmetry (25) As the samples analyzed by MAS NMR were a l l hydrated, no t r i g o n a l framework boron species were observed.


2 9

2 9

2 9

C a t a l y t i c Properties. The Bronsted acid s i t e s i n z e o l i t e s are associated with bridging hydroxyl groups adjacent to aluminum atoms (5). Substitution of other t r i v a l e n t elements for aluminum greatly affects the strength of these Bronsted acid s i t e s , and there i s a growing body of experimental data (1) which indicates that aluminosilicates are stronger acids than b o r o s i l i c a t e s . Previously, a quantitative test based on n-hexane cracking (alpha test) was developed and used to assess the strong acid properties of z e o l i t e s (5). We used the alpha test to compare the strong a c i d i t y of substituted and unsubstituted mordenites. We found the a c i d i t y of chemically dealuminated mordenites shows a large v a r i a t i o n with aluminum content (Figure 9). The alpha values decrease over f i v e orders of magnitude as the aluminum content decreases from four aluminum atoms per unit c e l l down to 0.8 A l / u n i t c e l l . I f the Bronsted acid s i t e s associated with boron are of equal strength as those associated with


26. GAFFNEYETAL.

Isomorphous Substitution ofBoron

389

aluminum, then the alpha values of unsubstituted and substituted samples i n which S 1 O 2 / A I 2 O 3 (unsubstituted) i s equal to Si02/(Al203+B 03) (substituted) should be equal. The a c t i v i t y measured f o r the heavily substituted material i s almost i d e n t i c a l to the sample with the same aluminum content, and i s over four orders of magnitude lower than would be expected i f boron s i t e s are as a c i d i c as aluminum s i t e s ( c i r c l e i n Figure 9). This c l e a r l y i l l u s t r a t e s that only aluminum contributes to the hexane cracking a c t i v i t y of these materials under the experimental conditions used. The presence of framework boron has no a f f e c t on the strong acid character of the catalyst. The presence of framework boron may s t i l l increase the number of weak and medium acid s i t e s , but the alpha test w i l l not be sensitive to these s i t e s , e s p e c i a l l y when strong acid s i t e s associated with aluminum are present. The alpha values of s y n t h e t i c a l l y prepared mordenites are within the range of ΙΟ^-ΙΟ^. There i s only a small variation i n a values of s i l i c e o u s mordenite as the aluminum content of the material varies. This i s i n contrast to dealuminated mordenites, which exhibit a much larger v a r i a t i o n i n alpha values as the aluminum content varies. These c a t a l y t i c results on synthetic and acid dealuminated mordenites indicate that factors other than the t o t a l aluminum content must contribute to the v a r i a t i o n i n a c t i v i t y of the catalysts. The linear c o r r e l a t i o n of alpha versus aluminum content reported f o r some z e o l i t e s (26.) does not apply to the acid dealuminated samples. Evidence has been presented f o r the presence of both Bronsted s i t e s and Lewis s i t e s enhancing the strong a c i d i t y of z e o l i t e catalysts (27). The presence of e x t r a - l a t t i c e aluminum i n acid dealuminated mordenite samples was confirmed by ^A1 NMR spectroscopy. The presence of both framework aluminum and e x t r a - l a t t i c e aluminum i n the acid treated materials may account f o r the wide v a r i a t i o n i n alpha values as a function of aluminum content. The synthetic boron substituted mordenites have alpha values s i m i l a r to the aluminosilicate analogues. This i s not surprising since boron replaces only 10% of the aluminum i n these materials. For boron to a f f e c t the alpha values of these samples, the acid strength of a B-OH proton would have to be much greater than f o r a A1-0H proton, which i s c l e a r l y not the case.


2

2

Conclusions We prepared boron substituted mordenite by d i r e c t synthesis from gel precursors and by post-synthetic substitution into dealuminated mordenite. Direct substitution i s favored i n aluminum d e f i c i e n t gels, but exacting c r y s t a l l i z a t i o n requirements f o r mordenite formation l i m i t the amount of boron that can be incorporated into the framework structure. Higher substitution levels were achieved using a post-synthetic treatment. Boron substituted z e o l i t e Y could not be prepared by a s i m i l a r d i r e c t synthetic method, but post-synthetic methods were e f f e c t i v e at providing low substitution l e v e l s . This demonstrates the more general u t i l i t y of post-synthetic substitution methods. The hexane cracking a c t i v i t y of



390

ZEOLITE SYNTHESIS

Figure 9 .

n-Hexane cracking a c t i v i t y of mordenites.


GAFFNEY ET AL.

Isomorphous Substitution ofBoron

391

substituted and unsubstituted mordenites i s consistent with strong acid s i t e s associated with aluminum, but not boron. Acknowledgments T. A. Braymer carried out much of the synthetic work described. C. C e c c a r e l i i carried out the l a t t i c e constant refinements. We thank A i r Products and Chemicals, Inc. f o r permission to publish t h i s work.


Literature Cited 1. Tielen, M.; Geelen, M.; Jacobs, P. A. Acta Phys Chem, 1985, 31 (1-2), 1-18. 2. Boden, N. In Determination of Organic Structure by Physical Methods; Nachod, F.C.,Ed.; Academic: New York, 1971, Vol.4, p 51-137. 3. Breck, D. W. U.S. Patent 3 130 007, 1964. 4. Kerr, G. T. J. Phys. Chem. 1968, 72(7), 2594-6. 5. Miale, J. N.; Chen, N.Y.; Weisz, P. B. J. Catal., 1966, 6, 278-87. 6. lone, K. G.; Duc Tien, N.; Klyueva, Ν. B.; Vostrikova, L. A. Catalysis Sci. Tech., 1985, 281-6. 7. Klyueva, Ν. V.; Duc Tien, N.; lone, K. G. React. Kinet. Catal. Lett.. 1985, 29 (2), 427-32. 8. Gabelica, Z.; Nagy, J. B.; Bodart, P.; Debras, G. Chem. Lett.. 1984, 1059-62. 9. Bajpai, P. K. Zeolites, 1986, 6(1), 2-8. 10. Bajpai, P. K.; Rao, M. S.; Gokhale, Κ. V. G. K. Ind. Eng. Chem. Prod. Res. Dev., 1978, 17(3), 223-7. 11. Domine, D.; Quobex, J. In Molecular Sieves; Barrer, R. Μ., Ed.; Soc. of Chem. Industry: London, 1968;p78. 12. Sand, L. B. In Molecular Sieves; Barrer, R. M., Ed.; Soc. of Chem. Industry: London, 1968;p71. 13. Itabashi, K.; Fukushima, T.; Igawa, K. Zeolites, 1986, 6(1), 30-4. 14. Bodart, P.; Nagy, J. B.; Gabelica, Z.; Derouane, E. G. Appl. Catal., 1986, 24, 315-8. 15. Barrer, R. M.; Freund, E. F. J.C.S. Dalton Trans., 1974, 1049-53. 16. Shannon, R. D. Acta Crystallogr., 1976, 32(5), 751-67. 17. Taramasso, M.; Perego, G.; Notari, B. Proc. 5th Int. Conf. Zeolites, 1980,p40. 18. Olsson, R. W., Rollmann, L. D. Inors. Chem., 1977, 16, No. 3, 651-4. 19. Gabelica, Z.; Debras, G.; Nagy, J. B. Stud. Surf. Sci. Catal., 1984, 19, 113-21. 20. Scherzer, J. ACS Symp. Ser. 1984, 248 (Catal. Mater. Relat. Struct. React), 157-200. 21. Liu, X.; Klinowski, J.; Thomas, J. M. J. Chem. Soc. Chem. Commun. 1986, 582-584.


392

ZEOLITE SYNTHESIS

22. Klinowski, J.; Hamdan, H.; Sulikowski, B.; Man, P. P.; Carr, S. W. 194 ACS National Meeting, New Orleans, LA, August 30 - September 4, 1987. 23. Mortier, W. J., Compilation of Extraframework Sites In Zeolites; Butterworth & Co.: Great Britain, 1982, pp 55-6. 24. Barrer, R. M. Hydrothermal Chemistry of Zeolites; Academic: London, 1982. 25. Kessler, H.; Chezeau, J. M.; Guth, J. L.; Strub, H.; Coudurier, G. Zeolites. 1987,7(4),360-6. 26. Haag, W.O.; Lago, R. M.; Weisz, P. B. Nature, 1984, 309, 589-91. 27. Beyerlein, R. Α.; McVicker, G. B.; Yacullo, L. N.; Ziemiak, J. J. J. Phys. Chem., 1988, 92(7), 1967-70.


th

RECEIVED January 25, 1989


Isomorphous Substitution of Boron in Mordenite and Zeolite Y - ACS

Recommend Documents