Chemically Modified Surfaces in Catalysis and Electrocatalysis

13 Derivatized L a y e r e d M(IV) Phosphonates

Downloaded by PURDUE UNIV on August 25, 2016 | http://pubs.acs.org Publication Date: July 2, 1982 | doi: 10.1021/bk-1982-0192.ch013

MARTIN B. DINES, PETER M. DIGIACOMO, KENNETH P. CALLAHAN, PETER C. GRIFFITH, ROBERT H. LANE, and RICCI E. COOKSEY Occidental Research Corporation, Irvine, CA 92713

In contrast to the conventional approach whereby various organic groups are subsequently bound to a previously prepared surface, we have been synthesizing a broad series of anchored, layered-structure solids by precipitating the pre-derived phosphonate salts with tetravalent metal ions. The two-dimensional backbone has the zirconium phosphate structure; however, substituted for hydroxylic groups are the desired organics, oriented away from the basal surfaces in a bilayered fashion in the interlayer region. These crystals can act as packets of modified surface, accessible by intercalation. Our focus has been aimed at the characterization and behavior of these compounds in various ion-exchange, sorption and catalytic reactions which w i l l be described. Of particular interest are the mixed component products, in which two or more different groups are present within the interlayer. The modification of surface properties by covalent bonding of various organic groups i s by now a well-established procedure for a broad range of solids. The use of inorganic layered compounds as substrates for such chemistry i s , however, quite limited i n i t s known scope, p a r t i c u l a r l y when the "surface" i s meant to include both external and internal basal s i t e s . The primary motivation for preparing covalently anchored layered compounds should be apparent: not only would such products enjoy the very significant advantages of other immobilized systems (ease of separations, s t a b i l i t y , concentration of s i t e s , etc.), but i n addition they should have the potential for enhanced s e l e c t i v i t y effects i n their interactions with other molecules, and they present the poss i b i l i t y of trans- or cis-chelation, both as a direct consequence of the two-dimensional situation of the termini of the affixed organics (see Figure 1). Of course, there may be a trade-off i n accessing the internal (bulk) sites to reactants, since this w i l l 0097-6156/82/0192-0223 $6.00/0 © 1982 American Chemical Society Miller; Chemically Modified Surfaces in Catalysis and Electrocatalysis ACS Symposium Series; American Chemical Society: Washington, DC, 1982.


224

CHEMICALLY MODIFIED SURFACES

..llllllll..

illUUlll

Anchored Organic Sites

. ,1 III III ITT ]//////)//////////7777T/

Figure 1.

Ordered array of covalently anchored organic groups on a stable inorganic support with a layered structure.

Miller; Chemically Modified Surfaces in Catalysis and Electrocatalysis ACS Symposium Series; American Chemical Society: Washington, DC, 1982.


13.

DINES

ET

AL.

Derivatized Layered M(IV) Phosphonates

225

have to occur by an intercalation process, whereby the diffusion implied could have significant rate impedence. The f i r s t challenge i n this endeavor was the choice of an appropriate layered substrate. A b r i e f process of elimination led us to the class of tetravalent metal phosphates, t y p i f i e d by the zirconium s a l t , whose a-phase structure was solved by C l e a r f i e l d and h i s group (1). Yamanaka (2) had described a method of preparing anchored alcohols by treating the y-variation of this salt with ethylene oxide, but we found this route to be far too limited in i t s scope. However, the structure of the phosphate was exactly suited f o r our purposes (Figure 2) and an alternative approach toward anchoring on the pendant hydroxyl groups based on chloros i l a n e s (as had been described with s i l i c a surfaces (3)-) was undertaken. This method proved to be fraught with preparative d i f f i c u l t i e s which, though not intractable, quickly yielded to a far superior alternative i n which phosphonic acids, having the d e s i r e d organic group already present p r i o r t o p r e c i p i t a t i v e polymerization of the solid (equation 1), were used. (1)

+4

Z r ( s o l n ) + 2 H 0 PR 2

3

— • Zr(0 PR) (solid) 3

2

As hoped f o r , the products of this very general and simple reaction were found to have the same layered backbone structure as the parent phosphate, only with the organic group (-R) substituted for the hydroxyl. This i s amply borne out by the x-ray d i f f r a c tion powder data on the products, which y i e l d the expected layerlayer distances d i r e c t l y , and by a plot of molecular density vs. interlayer distance, which gives a straight l i n e dependence whose slope corresponds to the common s i t e area, 24A (Figure 3 ) . In the midst of our investigation, A l b e r t i published some similar work leading t o the same conclusions ( 4 ) . Many o f our e a r l y results on the properties of this new class of hybrid inorganicorganic m a t e r i a l s were r e c e n t l y d i s c l o s e d , and experimental details may be found therein (5, 6). In this symposium i t i s our intention to report on some recent progress made i n the areas of surface area and c r y s t a l l i n i t y manipulation, c a t a l y s i s , mixed component products and the effects of " p i l l a r i n g " i n the layered phosphonates. In separate papers we w i l l describe results on the preparation and ion-exchange behavior of anchored sulfonic acids, and on the "magic angle" nmr techniques for characterizing the phosphonates. Surface Area and C r y s t a l l i n i t y Depending on the conditions of the precipitation reaction, including the u t i l i t y of séquestrants such as HF, products of varying c r y s t a l l i n i t y (as assessed by breadth of the x-ray d i f f r a c tion reflections) can be obtained. Similar effects have been found with the phosphates (7, 8). We were primarily interested i n


226



Figure 2.

The structure of a-zirconium phosphate. The area per site on the basal surface is about 24 À . 2


Miller; Chemically Modified Surfaces in Catalysis and Electrocatalysis ACS Symposium Series; American Chemical Society: Washington, DC, 1982. 2

2

3

2

3

2

2

3

3

2

3

3

2

T I (

2

2

3

2

_L 200

2

Mol wt/Density

3

2

P

2

)

3

(cm /mol)

2

3

2

2

300

2

2

2

Figure 3. Molecular volume vs. interlayer spacing, revealing that the various compounds share a common site area (about 24 A from the slope) and are essentially isostructural.

100

3

Zr(0 P-H)

3

o( -Zr(0 POH) CT 3 Z r ( 0 P C H ) y / O Zr(0 PCH2CH PO ) X O OTh(0 P-H)

3

3

2

U(0 PCH CI)

2

3

10

3

3

3

3

Th(0 P//////////

Figure 6. Sequence of events in reacting zirconium ions with a mixture of phenylphosphonic and phosphorous acids.



234

Table I.

Interlayer Distances Found for Crosslinking Groups. Compound

d-Spacing

Zr(03PCH CH PO3) 2

7.8

2

Zr(0 PCH CH CH PO ) 3

2

2

2

3

Zr(0 P(CH ) PO )

no reflect: 17.2

Â

9.6

Â

13.9

Â

Zr^@-P03)

18.5

Â

Zr(0 PCH ^. H Po )

10.8

Â

3

2

10

3

Zr(0 P-^-P0 ) 3

z

3

r(0 P^@-P0 ) 3


Â

3

3

3

2

C

2

3

The only anomaly we encountered, apparent i n the Table, was the finding that the three-carbon bis-phosphonic acid product, which had a good elemental a n a l y s i s , d i d not e x h i b i t any XRD lines. We conjecture that this apparent i n a b i l i t y to form a c r y s t a l l i n e product results from the fact that there i s no obvious way to force a conformation on a three carbon chain so that the terminal phoshonates w i l l be i n a p a r a l l e l c o n f i g u r a t i o n , as required by a layered structure. This, then, may be an example of an i n t r i n s i c a l l y amorphous product. It should be apparent that a p a r t i c u l a r l y desirable class of layered phosphonates i s the mixed composition crosslinked products. In these structures, there would be present some (preferably) small groups such as -H, -OH or a l k y l , together with a b i s phosphonic acid which can serve to " p i l l a r " the layers so as to form microporous voids i n the structure which would allow for the p o s s i b i l i t y of molecular sieving, as seen with the well-known zeol i t e s . A schematic of the structure desired i s shown i n Figure 7, where i t can be seen that the dimensions of the micropores produced can be hopefully controlled by the length of the p i l l a r i n g groups, and t h e i r c o n c e n t r a t i o n . Even i f they are randomly distributed, they should be able to manifest a size exclusion for incoming molecules based mainly on the basal surface separation. Of course, i f the p i l l a r i n g groups are too d i l u t e , one might expect some sort of "roof collapse" to occur, r e s u l t i n g i n a part i a l closing o f f of the internal volume. In the preparation of such mixed component p i l l a r e d compounds, a simple method of estabishing that single phase products having structures such as that given in Figure 7 was required. We chose to use the surface area measurement (nitrogen one point BET) as a means of v e r i f y i n g p i l l a r i n g . This was necessitated by the observation that very poor XRD patterns were usually obtained,



13.

DINES ET

AL.

Derivatized Layered M(IV)

Phosphonates

235

thus obviating this method, and furthermore, that the surface area can give a direct measurement of the new surface which should be opened up i n the process of p i l l a r i n g . The argument adopted was that i n a series of mixed component products, substantial deviations from a straight l i n e connecting the surface areas of the end members was most l i k e l y attributable to opening up of the microstructure, whereas lack of such could either mean that no enhanced access results from forming the mixed component phase, or that phase separation was occurring. The c r i t i c a l assumption i m p l i c i t in this logic i s that there i s a monotonie v a r i a t i o n i n p a r t i c l e size throughout the series; that i s , that any deviation from the straight l i n e was not attributable to changes i n the p a r t i c l e size induced by the presence of the coreacting phosphonate. In Figure 8 i s presented the results for a small series of such products whose end members are ZrCOßPH^ and Zr (O3P-^-©-POß). Note that we have designated the specific surface areas i n units of M^/mmole, so that the mole fraction axis i s linear. SEM revealed no substantial change i n p a r t i c l e size in the series (particles about 0.06 microns i n diameter, appearance non-crystalline). The surface area observed for the intermediate compositions i s about double that expected i f no consequential interaction occurs. An estimate of how much increase could be expected i f each -H s i t e allows one N molecule to incorporate for the 0.5 mole fraction case i s about 100 M (6.02xl0 ^ molecules x 16.2 A^/molecule). The value observed was 80 additional M^/mmol. Additional evidence for the presence of micropores was obtained by p r e - t r e a t i n g the 0.33 mole f r a c t i o n product i n nonane and rerunning the surface area measurement after exhaustive pumping to remove any residual l i q u i d . This method i s described (9) to e f f e c t i v e l y "plug" microporous surface area. We found a diminution of about 75% i n the surface area measured, roughly placing the point on the non-interaction l i n e (Figure 8). On subsequent heating i n a flow of helium, the original area was nearly comp l e t e l y r e s t o r e d . An estimate of the dimension of the pores thought to r e s u l t i s about 8.3 A by 5 A, based simply on the spacing difference between the diphenyl and hydrogen compounds, and an assumption that an average of one -H s i t e separates neighboring p i l l a r s . We have prepared other mixed composition p i l l a r e d compounds which have as their non-pillaring group the hydroxyl moiety, and thus are simply relatives of zirconium phosphate i n which the l a y e r s are spread at a f i x e d distance apart. These products behave as expected i n t i t r a t i o n and ion-exchange experiments, and w i l l not be further discussed here. 2

2

2

Anchored Catalysts One of the most i n t r i g u i n g p o s s i b l e a p p l i c a t i o n s of the derivatized layered phosphonates i s i n the area of heterogenized


236


Non-Pillaring Groups


Pillaring Group

Figure 7. Architecture of micropores in layered phosphonate compounds. Distance A is dictated by the density of pillars; B is determined by the length of the pillar, relative to the size of the nonpillar groups.

150 h CM

5 CO O O CO

100

r 3 CO Ü a> a CD Mole Percent Diphenyl Pillar Figure 8. Variation of the specific surface area of mixed component pillared compounds, whose end members are the hydride and the diphenyl bisphosphonate. The point denoted with a diamond corresponds to the area measured after nonane treatment.


13.

DINES

ET AL.


237

transition metal catalysts. Generally, the idea here i s to prepare a product containing appended organic groups which can then function as ligands f o r the subsequent complexation of suitable metal species which can, i n turn, serve to catalyze reactions of intercalated molecules. We have chosen as a prototype support for this chemistry the anchored pyridine compound (broad XRD r e f l e c tion corresponding to an 18.6A spacing), prepared by precipitation of 2-ethyl(4-pyridyl)phosphonic acid (Equation 2), which can then be used as a substrate for palladium(II).

+


(2) Zr 4 + Pd(ll) Using bis(benzonitrile)palladium dichloride, and s t i r r i n g i n a tetrahydrofuran s l u r r y of the p y r i d y l anchored compound f o r several days, a product having a value for x, after exhaustive e x t r a c t i o n , o f about 0.2 (corresponding to 3% Pd l o a d i n g by weight) was afforded. This catalyst was found to be very active for the hydrogénation of cyclohexene at 80*C with 375 p s i hydrogen. Under these conditions, the conversion to cyclohexane was complete i n less than 20 min. There was evidence, based at f i r s t on color changes of the catalyst (from tan to black) that reduction of the P d ( l l ) had occurred. Later, this was confirmed by ESCA spectra run on the used and fresh catalyst (Figure 9). We cannot be certain whether the Pd(II) or Pd(0) i s the c a t a l y t i c a l l y active species, nor do we yet understand how the Pd(0) i s bound to the s o l i d . The catalyst was also found to be very active for the hydrogénation of benzonitrile to benzyl amine, nitrobenzene to a n i l i n e , diphenylacetylene t o s t i l b e n e s , and benzene t o cyclohexane. We are investigating the incorporation of other metals into s i m i l a r l y anchored s o l i d s , and examining other c a t a l y t i c reactions. In p a r t i c u l a r , we are interested i n attacking the problem of catalyst leaching from the s o l i d . This work i s i n progress and w i l l be reported on i n the future. Transmission Electron Microscopy There i s no more compelling evidence of the microscopic morphology of layered compounds than a direct image as can be afforded only by transmission electron micrography. The current state of the art allows f o r routine l i m i t s of resolution on the order of 5 - 10 A as opposed to about 70 A for scanning electron methods. We turned to the hexyl rather than the methyl analog f o r our prototype compound — Zr^OßPC^H^^ — to gain a factor of about two i n the layer-layer distance (19 vs. 9 A).




238

Figure 9. ESCA spectra of Pd-loaded pyridyl compounds before (A) and after (B) use in catalytic hydrogénation. Note the decrease in Pd3d binding energy: 342.6 eV in A, 341.2 eV in B, consistent with Pd(II) in A, Pd(0) in B. s/t


DINES ET AL.

239



13.

Figure 10. Transmission electron micrograph of the compound Zr(O PC H h, at 540,000 power. The layer-layer stacking is apparent in profile, with a repeat distance of about 15Â. s


6

ls


240

The material used was prepared by a simple precipitation and overnight heating. It had a single point BET surface area of 17.8 m^/g. Shown in Figure 10 is a microcrystal in profile, in which the layer stacking is clearly apparent. The repeat distance e s t i mated from the photographic and instrumental magnification was 15 A, i n satisfactory agreement with the powder diffraction result.


Summary and Conclusions The derivatized layered tetravalent metal phosphonates have proven to be a particularly apt example on which to test many of our hypotheses regarding planar bulk arrays of anchored organics. They are relatively easily prepared, providing the phosphonic acid or ester is available, and they provide for a site area which is perfect for nearly close-packed coverage. We have made substant i a l progress in the detailed characterization of their physical and chemical properties, especially in the areas of crystallinity, surface area and micropore behavior, mixed component phases, and in heterogenization of catalytic sites. Literature Cited 1. Clearfield, A . ; G. D. Smith, Inorg. Chem. 1969, 8, 431. 2. Yamanaka, S., Inorg. Chem. 1976, 15, 2811. 3. Boucher, L. J.; A. A. Oswald; L. L. Murrell, Preprints Petroleum Div. A.C.S. Meeting, Los Angeles, CA, March, 1974, 162. 4. Alberti, G; U. Costantino; S. Alluli; N. Tomassini; J. Inorg. Nucl. Chem. 1978, 40, 1113. 5. Dines, M. B . ; P. M. DiGiacomo; Abstract of Papers, 179th A.C.S. National Meeting, Houston, TX, March, 1980, Inorg. Div. paper no. 168. 6. Dines, M. B . ; P. M. DiGiacomo, Inorg. Chem. 1981, 20, 92. 7. Alberti, G.; E. J. Torracca, Inorg. Nucl. Chem. 1968,30,317. 8. Clearfield, A . ; A. Oskarsson; C. Oskarsson, Ion Exchange and Membranes 1972, 1, 91. 9. Gregg, S. J.; J . F. Langford, Trans. Faraday Soc. 1969, 65, 1394. RECEIVED November 4,

1981.


Chemically Modified Surfaces in Catalysis and Electrocatalysis

Recommend Documents