SILYLATION 0F SILICA GEL AND LAYERED SILICATES Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY ALLAN MATT ALAN KO 1976 a IIIII IIII IIIIIIIIIIIIIIIIL 3129730069 71745 ‘ ABSTRACT SILYLATION OF SILICA GEL AND LAYERED SILICATES BY Allan Matt Alanko Coupling agents are ambifunctional silanes contain- ing organic functionality attached to a silane via a hydrocarbon linkage. Their dual reactivity allows the inorganic silanol to react with silanols of a silaceous surface while the organic group is free to perform its designated function. Monolayers of y-aminopropylsilanes are prepared by washing a treated silica surface contain- ing multilayers of coupling agent with H20. Metal cations react with the functionalized surface and result in immobilized metals on the silica surface. Immobilized Cu (II) on silica gel whose surface was treated with a monolayer of an aminofunctional silane was characterized by esr. Comparison of esr parameters of immobilized Cu (II) to those obtained with ethylene- diamine (en) complexes of Cu (II) in a methanol glass showed copper (II) in different environments depending on the particular coupling agent employed. Cu (II) immobilized on silica that had been functionalized with Allan Matt Alanko y-aminopropylsilyl was complexed by two nitrogen atoms. However,when copper (II) was immobilized on silica which had been treated with E-(B-amino-ethyl)-y-aminopropylsily the central metal ion was chelated with four nitrogen atoms. Washing the latter complex with boiling water gave 2 (en) and Cu+2 (en)2 on the surface. Comparison both Cu+ of esr spectra at 300°K and 77°K showed little change occurred, and spectra at both temperatures showed Cu (II) on amino functional silica gel with distinct g and gL H components. This indicates the copper was not free to tumble or translate along the surface. This was not observed when Cu (II) was adsorbed on nonfunctional silica gel. Quaternary ammonium forms of layered silicates (montmorillonite and hectorite) were treated with E: (B-amino-ethyl)-Y-aminopr0pyltrimethoxysilane (Z-6020) and the copper complexes were characterized by esr. Analysis of Cu (II) on silylated Bu N+ saturated hectorite 4 showed copper in an environment similar to that obtained on silica gel with four nitrogen atoms around the copper atom. Analysis of oriented films of the silylated clay/ Cu.aI)complexes by scanning electron microprobe showed silylation occurred on the edge of the layered silicate crystal. SILYLATION OF SILICA GEL AND LAYERED SILICATES BY Allan Matt Alanko A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1976 DEDICAT ION To my family ii ACKNOWLEDGMENTS I wish to thank Dow Corning Corporation, my many friends and coworkers, and Dr. Thomas J. Pinnavaia for their help and support during the course of this work. iii LIST LIST II. III. IV. TABLE OF CONTENTS OF TABLES O O O C O I O O O O O I O 0 OF FIGURES O O O O O O O O O O O O 0 INTRODUCTION . . . . . . . . . . . . DISCUSSION OF RESULTS . . . . . . . A. B. C. D. Copper Amine Complexes in Methanol Copper Immobilized on Silica Gel COpper on Clays . . . . . . . . Trimethylsilylation of Clays . . CONCLUSIONS 0 O O O O O O O O O O 0 EXPERIMENTAL . . . . . . . . . . . . A. B. C. LTJ CEO "*1 Fractionation of Montmorillonite Preparation of Cation Exchange Forms Of Clay 0 O O O O O O O O O O Silylation of Pr N+ Montmorillonite with N-(B-Amino-Ethyl -y y-Aminoqfiopyltri- Z6020 methoxysilane (Dow Corning Silane) I O O O O O O O I O O Sil lation of Bu N+ Hectorite and Its Cu+ Complex . . . . . . . . . Silylation of Silica Gel with Aminoalkylsilanes . . . . . . Copper (II) Complex of Silylated Silica Gels . . . . . . . . . . Copper Amine Complexes . . . . . Trimethylsilylation of Clay . . l. Na+ Montmorillonite with Me3SiCl and (Me Si) 2NH in Dioxane . 2. Na+ Mongmorillonite with Me3 and (Me 81) NH in Bulk . . . SiCl 3. Na+ MongmorIllonite with Me SiCl in Pyridine . . . . . . . § 4. Pr N+ Montmorillonite with Me3SiCl in4MeCN O O O 0 O O O 0 O I iv Page vi vii 16 25 32 38 39 39 39 40 41 42 42 43 43 43 44 44 45 I. J. BIBLIOGRAPHY. . . . . . . . Intercalation of Silica in Na+ Montmorillonite . . . . . . . General. . . . . . . . . . . . LIST OF TABLES Table Page 1. Esr parameters at 77°K for Cu+2 complexes of amines in methanol glass . . . . . . . . . 5 2. Esr parameters for Cu+2 immobilized on silica gel and a layered silicate . . . . . . 18 vi LIST OF FIGURES Figure Page 1. Esr spectrum at 77°K for a O.5:l.0 molar ratio of H NCH CH NH and CuCl -2H 0 in a methafiol glags ? . . . . g . g . . . . . 6 2. Esr spectrum at 77°K for a l.0:1.0 molar ratio of H NCH CH NH2 and CuCl -2H 0 in a methafiol glags . . . . . g . g . . . . . 8 3. Esr spectrum at 77°K for a 2.0:l.0 molar ratio of H NCH CH NH and CuCl '2H 0 in a methafiol élags ? . . . . g . g . . . . . 9 4. Esr spectrum at 77°K for a 4.0:l.0 molar ratio of H NCH CH NH2 and CuClZ-ZHZO in a methafiol élags . . . . . . . . . . . . . 10 5. Esr spectrum at 77°K for a 10.0:1 molar ratio of HZNCH CHZNH2 and CuClz'ZHZO in a methanol glass . . . . . . . . . . . . . ll 6. Esr spectrum at 77°K for a 0.5:1 molar ratio of (MeO)3Si CH CH CHZNHCHZCHZNH2 and CuC12°2H20 in a fietfianol glass . . . . . 12 7. Esr spectrum at 77°K for a 1.0:l.0 molar ratio Of (MeO)3Si CH CH CH NHCH CH NH2 and CuC12°2H20 in a fietfiangi gigss2 . . . . . 13 8. Esr spectrum at 77°K for a 2.0:l.0 molar ratio of (MeO)3Si CH CH CHZNHCH CHZNH2 and CuC12°2H20 in a fietganol glgss . . . . . l4 9. Esr spectrum at 300°K for a 2.0:1.0 molar ratio Of (MeO)3Si CH CHZCHZNHCHZCHZNHZ and CuClz’ZHZO in meghanol . . . . . . . . . 15 10. Esr spectrum at 77°K for a 3 x 10-3 M CuCl °2H O in a methanol glass . . . . . . . 17 2 2 11. First derivative esr spectra of Cu+2 on silica gel at 77°K . . . . . . . . . . . . l9 vii Figure Page 12. First derivative esr spectra of Cu+2 on untreated silica gel at 300°K . . . . . . 20 13. First derivative esr spectra of Cu+2 on silica gel treated with Z6020 . . . . . . 21 14. First derivative esr spectra at 300°K of Cu+2 on silica gel treated with Z6020 and washed with H O at 100°C after complexation . .2. . . . . . . . . . . 22 15. Oxygen (o) and hydroxylnio) network of layered silicates. M ° x H O are the solvated interlayer exchange cations . . . . 26 16. First derivative esr spectra for Cu (II) 26020 complex on Pr N+ form of montmorillonite at IA) 300°K and (B) 77°K and the same complex on BuN+ form of hectorite at (C) 300°K and (D) 77°K . . . . . 28 17. Microprobe scan on edge of oriented films of montmorillonite . . . . . . . . . . . . . 30 18. Microprobe scan on surface of oriented films of montmorillonite . . . . . . . . . . 31 19. SEM of Pr N+ form of montmorillonite which wgs treated with 26020 and complex with Cu+2 at 2600 x . . . . . . . . . 33 20. Infrared spectrum of montmorillonite silylated with Me SiCl and (Me Si)2NH at a molar ratio 8f 2.0 in dioxane . . . . . 35 21. Infrared spectrum of montmorillonite silylated with Me SiCl and (Me Si) NH at a molar ratio 8f 0.5 in dioxane2 . . . . . 36 22. Infrared spectrum of montmorillonite silylated with Me3SiCl in pyridine . . . . . 37 viii I . INTRODUCTION Coupling agents are ambifunctional silanes usually of the formula RSi(OMe)3, where R is an organic radical attached to silicon by a hydrolytically stable silicon- carbon bond. Reactive groups such as -NH2, -NHCH2CH2NH2, -SH, -CH9CH or C1 are usually part of the organic radical. 2 Hydrolysis of the coupling agent gives a silane triol which can condense to oligomers that contain both silanol and organic reactivity as shown below. 1 RSi(OMe)3 + H O + HO(R Si O)xH + MeOH 2 Because of this dual reactivity, coupling agents have been used to increase adhesion between inorganic oxides and plastics, immobilize metals on silica gel, and numerous other applications. The oligomers formed from coupling agents readily adsorb on oxide surfaces from aqueous solution. Deposits of the coupling agents vary in thickness depending on the solution concentration during application, on the conditions which are employed to deposit the coupling agent, and the nature of the oxide surface. Water desorption studies of 14C labeled y-aminopropyltriethoxysilane from glass surface indicate three types of layers are present.2 The first type consisted of 270 monolayers obtained from the initial deposition. This thick surface coating could be reduced to 10 monolayers after 3-4 hrs of washing at room temperature. Additional washing at 100°C showed the adsorbed silane on the surface as a monolayer.2 Adsorption of coupling agents in general on E- glass gave increased thickness as a function of solution concentration, however, a monolayer formed on silica gel regardless of the solution concentration of coupling agent.3 These data suggest monolayers of coupling agents form on silica gel whereas multilayers are deposited on E-glass. Stability of RSi(OH)3 in aqueous solution is relatively independent of the substituent group on silicon except when the group is substituted at the y-position. These latter silanes form more stable solutions, especially when an amine is three carbon atoms removed from silicon. This unique stability has been attributed to cyclicization of the dimer in aqueous solution as shown below.1 R a bi? CHz-CH2 O"°°NH 6-) // \ 0H / \ CH2 Si ‘——— 01— 51 CH / / 2 :NH° . ,0 HO CHz-CH2 R I R Silica gels and clays have been used as supports for heterogeneous catalysts and adsorbents for a variety of 4-9 Catalytic activity of metals is greatly deter- metals. mined by its chemical environment, whether the metal ion is in homogeneous solution or immobilized on a solid sup- port. Hectorite and montmorillonite are layered silicates which have exchangeable metal cations in the interlayer. COpper exchange forms of these clays and copper on silica gel have been characterized by electron spin resonance 5’7 which has provided fundamental information about (esr). the chemical environment of the cation. Treatment of silica gel or porous glass beads with aminofunctional coupling agent followed by complexation with a metal ion has been described as a method of concentrating cations and pre- paring catalysts.8’9 The purpose of the present study was to characterize immobilized copper cations and gain information about its chemical environment and the nature of the silylation reaction. COpper which was immobilized on silica and clays by N-(B-amino-ethyl)-y-aminopropyltrimethoxysilane (Z-6020) and y-aminopropyltriethoxysilane (A-llOO) was characterized by electron spin resonance spectrosc0py and electron microprobe analysis. II. DISCUSSION OF RESULTS A. Copper Amine Complexes in Methanol A solution of 3 X 10_3 M methanolic CuClz-ZHZO and its complexes of ethylene diamine (en) or N-(B-amino-ethyl)- y-aminopropyltrimethoxysilane (26020) at various molar ratios were analyzed by esr to obtain 9 values and coupling constants. These data are summarized in Table 1. Copper at these dilutions should not interact with the magnetic dipoles of electronic and nuclear spins of neighboring copper atoms. Since the esr spectra were obtained in a methanol glass at 77°K, exchange interaction with neigh- boring unpaired electrons and tumbling should be minimal. These factors should minimize line broadening. In each case the 9" component is resolved into four components which is characteristic of tetragonal Cu (II) with a nuclear spin of 3/2. However, hyperfine splitting of the g‘l component or super hyperfine splitting by N is not observed. As can be seen in Figure l, at a molar ratio of 0.5 en/Cu (II) at least two 9" components are observed. These are attributed to uncomplexed copper (g” = 2.40, A = 0.0112 cm-l) and copper complexed by one en molecule 4 N N N N mommo N N N N mz mo momz mo am mxomzc u omoma an no no m2 u cm mH.~ om 0H.~ mvam Had sm.~ amen .c o . mH.~ om mo.m omam «as mm.~ mmsm H m mom: a xamam AHHV so .0 mo.m mma mo.~ ovum ems ma.m moon o.~ .m NH.~ ONH mo.m ommm mma m~.~ ommm o.H .s HH.~ oma mo.m oamm ama mm.~ ommm m.o .m ma.~ om mo.m omam «as o¢.~ omsm AHHV so\omooN .m HoE\Hoe mo.~ OGH mo.m memm ems ma.~ oaom o.oa .m mo.m oma mo.~ mvmm nmfl mH.~ oaom 0.9 .9 mo.~ mma mo.~ osmm sea ma.~ maom o.~ .m NH.N mm mc.~ oamm «Ga o~.~ mamm o.a .m ma.~ om mo.~ oeam was ov.~ omsm m.o .H mH.N OMH mo.~ comm mmH em.~ momm AHHV so\cm .a HOE\HOE mmsaw mmomu so = mmoao 96m .32 4o mm mafia a mum .mmmam Hocwnumfi ca mmswfim mo mwxmamsoo N +50 How Mosh um mumumEmnmm Hmm11.H manna I 8.013. - #000 5 Tan: Lontf = 0,5 A“? 0.53: a“ . GM" ‘3 1.11 lo" arr” r“... “"‘* 0.3 S I." 17"" Shun 127" r 2 710K F'C’KM(, ' i'Zz, 4""‘ 1:45 (‘8?an t‘w H H .500 G U V . 44;” 17.15173 ‘ T/nu? unfit/,0 3 . (“wad-19'" K4”)! :— 20006 ‘rlmc Lu bfi\.-t l Rama: . 1:49.09 9 I Fig. l.--Esr spectrum at 77°K for a 0.5:1.0 molar ratio of HZNCHZCHZNH2 and CuClZ-ZHZO in a methanol glass. (g" = 2.27 and A = 0.0159 cm-l). Stability constants for the following equilibria are large. Cu+2 + en ;: Cu+2 - en K1 2 1011 Cu+2 ° en + en == Cu+2 ' Zen K2 2 109 Since K1 >> K2, the complexation takes place in a stepwise fashion. The assignment of the g“ component = 2 2.27 to a 1:1 Cu+ :en complex is verified by the presence 2 of a single gH resonance (2.26) at a ligand to Cu+ ratio of 1.0. At en:Cu+2 1 1 2:1 a single g‘I value of 2.19 and A = 0.0187 cm- is observed. This resonance is character- istic of the Cu (en)2+2 complex as shown in Figures 2-5. Analysis of N—(B-amino-ethyl)-y-aminopropy1tri- methyoxysilane copper (II) complexes by esr in a methanol glass gave 9H and A values that were almost identical with those observed for the en Cu+2 complex as shown in Figures 6-9. The value for the 1:1 complex are g‘l = 2.23, A = 0.156 cm-l. The 2:1 complex gave 9" = 2.19, A = 0.0184 cm-l. These results show that both the coupling +2 agent and en behave in a similar manner in forming Cu amine complexes. A slight difference was noticed at a molar ratio of 1.0 in 9", A, and AHl. This might indicate a slight difference in the 1:1 complex, however the evi- dence strongly suggests the complex is primarily CuII° w AI Fig. 2.--Esr spectrum at 77°K for a 1.0:l.0 molar ratio of HZNCHZCHZNH2 and CuC12'2H20 in a methanol glass. INN Fig. 3.--Esr spectrum at 77°K for a 2.0:1.0 molar ratio of HZNCHZCHZNH2 and CuClz-ZHZO in a methanol glass. 10 500 5 NIVIM I Fig. 4.--Esr spectrum at 77°K for a 4.0:1.0 molar ratio of H2NCH2CH2NH2 and CuClz'ZHZO 1n a methanol glass. 11 J 500 G ‘— ’-‘—o M) m J Fig. 5.--Esr spectrum at 77°K for a 10.0:1 molar ratio of H2NCH2CH2NH2 and CuC12°2H20 in a methanol glass. 12 500 6 " NJ v‘I I Fig. 6.--Esr spectrum at 77°K for a 0.5:1 molar ratio of (MeO) Si CH CH CH NHCH CH NH and CuCl °2H O in a methaRol glgss? 2 2 2 2 2 2 13 A Fig. 7.--Esr spectrum at 77° K for a 1. 0:1.0 molar ratio of (MeO) Si CH CH 2CHZNHCHZCHZNH2 and CuClz'ZHZO in a methaaol g1 ss. 14 Fig. 8.—-Esr spectrum at 77°K for a 2.0:1.0 molar ratio of (MeO)3Si CH CH CH NHCH CH NH and CuCl '2H 0 in a methanol 91 ss? 2 2 2 2 2 2 15 g/V“V\fiqN/w Fig. 9.--Esr spectrum at 300° K for a 2. 0:1.0 molar ratio of (MeO) Si CH 2CH ZCHZNHCHZCHZNHZ and CuC12°2H20 in methaaol 16 26020. These data show that both the 26020 and EH1 behave in a similar manner toward Cu (II). Analysis of copper chloride in a methanol glass by esr shows two different forms of copper are present. The coupling constants vary only slightly for the parallel com- ponent, but the different gH values of nonequivalent copper ions are well resolved. A possible explanation for different forms of COpper would be the presence of both Cull-ZHZO and CuII- Hzo-MeOH. The latter form of c0pper would result from displacement of water of hydration by the solvent. Alternatively the following equilibrium could account for the spectrum observed in Figure 10. Cu+2 + 2C1-(->- CuCl+ + C1-“ CuCl2 B. Copper Immobilized on Silica Gel Silica gel was treated with a 5% aqueous solution of aminoalkylsilane and washed with boiling water to obtain a monolayer of silane. The resulting aminofunctional silica gels and an untreated silica gel were mixed with 1 M CuCl2 and washed exhaustively with distilled water to rid. the silica gel of any free CuClZ. Analysis of the silica gels by esr are summarized in Table 2 and Figures. 11-14. Comparison of Cu (II) immobilized on a silica sur- face with the Cu (II) amine complexes immobilized in a methanol glass at 77°K shows distinct amine complexes are formed on the silica surface. The environment of the 17 II WJ N Range=4000G Ran e=2000G Fig. 10.--Esr spectrum at 77°K for a 3 x 10"3 M CuC12-2H20 in a methanol glass. 18 N m2 Naommommo am mxoumv u ooaama. ~=z~mo~momz mowmo so am mxomze n omomu. o.m~ um pwnmm3 mufluouomm no show +zsm may no omomu. mo.m he oma mo.~ mmmm mad mH.~ smom -AHHV so .m Hmw moflawm ~H.~ com oHH mo.~ comm sea s~.~ mmmm no ooaa¢.. oa.~ he mos mo.~ ommm mus m~.~ mmmm IAHHC so .9 HH.~ mus v~.~ maom = mo.~ oom oma so.~ mmmm mma mH.~ omom = sum moflaflm oa.~ and m~.~ comm co omomu. mo.~ he oma mo.~ mvmm «ma mH.~ moon IAHHV so .m U 0 mm N um saws o m awe moflaflm no omomN so.m he ops No.m onmm «ma ea.~ osom IAHHC no .N . . . awe messam as N be om no N «mam and mm m poem no AHHV so .2 s mmsmo mmsaw so mmoao cmummmH m was How MOHHHm so UmNHHNQOEEH N .wumowaflm +50 How mumumamnmm Hmm11.~ manna Daofl.f32 O 560 6 55M a. 30 loo 6. . J‘Ias '0 3006 gal!" I“ Fig. 11.--First derivative esr spectra of Cu+ gel at 77°K. 19 Silica gel alone. Water washed at 100°C. Silica gel treated with 2—6020 and the Cu(II) complex water washed at 25°C. Same as B except washed at 100°C. Silica gel treated with A-llOO and the Cu(II) complex water washed at 100°C. 2 . . on s1l1ca 20 5006 I; Fig. 12.--First derivative esr spectra of Cu+2 on untreated silica gel at 300°K. 21 j Fig. 13.--First derivative esr spectra at 300°K of Cu+2 on silica gel treated with A-llOO. 22 ( 900 G‘ Fig. 14.-—First derivative esr spectra at 300°K of Cu+2 on silica gel treated with 26020 and washed with H20 at 100°C after complexation. 23 copper depends on the silylating agent used and the con- ditions of complexation. Without any coupling agent present the g'. com- ponent of Cu+2 at 77°K is resolved into a quartet with a value close to that of CuC12'2H20 in a methanol glass. Comparison of Figure 11A with Figure 12 shows that a poorly resolved.g“component is observed at 300°K, where a quarter was observed at 77°K. This indicates that although Cu (II) is not removed by the washing technique, it is able to spin and/or slowly tumble on the silica surface at 300°K. Copper immobilized with N-(B-amino-ethyl)-y- aminopropyltrimethoxysilane on silica gave esr parameters which show the copper is immobilized as his (N-B-amino— ethyl-y-aminOpropylsilyl) complex surrounded by four nitrogen atoms. Analysis of the immobilized Cu (II) at 300°K shows little change in the esr spectra from 77°K. Although the lines have broadened slightly, the 9" component is clearly observed. When the copper complexed with his (N-B-amino- ethyl-y-aminopropylsilyl) treated silica surface is washed at 100°C a change is observed in the esr spectrum. Inter- pretation of the spectrum obtained at 77°K showed copper in two different forms, the original form and copper sur- rounded by two nitrogen atoms. Values of g“ for the new form of copper agree with the one to one complex between 24 CuII and N-(B-amino—ethyl)-Y-aminopropyltrimethoxysilane. Analysis of the sample at 300°K showed little change in the II was not free to tumble. It appears that the water wash in the presence of Cu+2 at esr spectrum indicating the Cu 100°C caused desorption of the coupling agent. The same treatment of silica gel with y- aminopropyltriethoxysilane gave Cu (II) which was complexed by his (y-aminopropylsilyl) groups. Analysis at both 77°K and 300°K gave esr parameters which show the copper com- plexed with two nitrogen atoms and restricted mobility at 300°K (g“ = 2.27, A = 0.0167 cm"1 ). The esr parameters are consistent with axial symmetry. Smaller 9“ values and larger values of A relative to Cu (II) in a methanol glass are indicative of increased covalency of the amine complexes. Predominately bis (alkylammoniumsilyl) com- plexes with y-amin0propylsilyl and N-(B-amino-ethyl)-Y- aminopropylsilyl on silica gel support the proposed dimeric structure of y-aminopropylsilane in aqueous solution. H RR \ \IH CHz-CH2 O O""N \ \/ +\ CH2 Si - O - Si CH2 \+ "/\ \ / N"°'O OH CHZ-CH2 /I\ R H R 25 C. Copper on Clays Layered silicate minerals known as smectites have a layered structure which is shown in Figure 15. An octa- hedral layer is positioned between two tetrahedral silica sheets and a negative charge originates from a positive charge deficiency in the octahedral layer. Cations, which balance the negative charge and occupy a layer between the silicate sheet, can be exchanged by treating an aqueous suspension of the silicate with an excess of another cation.9 Hectorite and montmorillonite were fractionated by a sedimentation technique to give a fraction with a particle size <2u. This fraction was isolated from sus- pension by centrifuging and freeze drying. Montmorillonite (Upton, Wyoming) and hectorite (Hector, California) were converted to the tetrapropylammonium form and tetrabutyl- ammonium form respectively. Idealized unit cell formulas are given below: Montmorillonite (Pr4N)0.64[A13.06Fe0.32M90.66](A10.1OSi7.90)020(0H)4 Hectorite (3“4N)0.42[M95.42Lio.68A10.02](318.00’020(0H'F)4 Both forms were silylated with N-(B-amino-ethyl-y- aminopropyl)-trimethoxysilane (Z6020), washed at 25°C and a copper complex prepared. Analysis of the tetra- propylammonium form of montmorillonite gave broad signals 26 Fig. lS.--Oxygen (o) and hydroxyl (0) network of layered silicates. Mn - x H O are the solvated inter- layer exchange catioHs. 27 at both 300°K and 77°K. However, the hectorite sample gave an esr spectrum which was clearly resolved into a g'| and gl component (Figure 16) at both 77°K and 300°K. Significant amounts of iron are present in montmorillonite and probably cause line broadening by decreasing the relaxation time of the free electron of Cu (II) in the interlayer. The iron signal at g = 4.6 is much more intense in the esr spectrum of montmorillonite than in the hectorite sample. Esr parameters for Cu (II) immobilized on hectorite with N-(B-amino-ethyl)-y-aminopropyltrimeth- oxysilane are almost the same as those given in Table 2 for silica gel. This suggests that copper is in a similar environment when immobilized on a microcrystalline silicate or an amorphous silica gel and is probably in an environ- ment similar to that shown below. ,CH CH OSiCHZCHZCHZNHR EgzzNH2 COpper immobilized with ; 0 kCu.\* N-B-amino-ethyl-y-amino- 4 OSiCHZCHZCHZNH /NH2 propylsilyl \CH CH 2 2 OSiCH CH CH NH X Copper immobilized with O 2 2 2‘KECui'12 y-aminopropylsilyl OSiCHzCHZCHZNHS' "x Another property of the swelling clays is their ability to form oriented films when susPensions are allowed to evaporate. The covalent bonding theory of coupling agents requires the formation of a covalent bond between the metal oxide surface and the coupling agent.1 The only 28 Fig. 16.--First derivative esr spectra for Cu (II) - 26020 complex on Pr4N+ form of Montmorillonite at (A) 300°K and (B) 77°K and the same complex on BuN+ form of Hectorite at (C) 300°K and (D) 77°K. 29 accessible silanols present in the silicate sheet are located at the fractured edge of the silicate crystal, therefore, one would expect silylation to take place at the edge of the microcrystal. Oriented films of mont- morillonite which had been silylated with N-(B-amino- ethyl)-y-amin0propyltrimethoxysilane and treated with Cu (II) were fractured and analyzed for Cu on the surface and the edge by microprobe. These analyses are compared II saturated montmorillonite and to oriented films of Cu the results are given in Figure 17 and 18 for the edge and surface of the films respectively. Microprobe analysis is a surface analysis technique where a sample is bombarded with an electron beam and characteristic x-rays which fluoresce from the sample are measured. When copper is being analyzed by this technique the intensity of Cu Ka and La can be monitored as the sample is scanned. Copper saturated montmorillonite gave about the same average x-ray fluorescence on the edge as the silylated montmorillonite indicating the edge capper concentrations are approximately equivalent. Differences between the CuKa and CuLa fluorescence suggest a deeper section is being analyzed in the copper saturated mont- morillonite. Analysis of the surface of each film shows the intensity of x-rays caused by copper are much more intense in the copper saturated montmorillonite than the sample where copper is immobilized with a silylating agent. 30 EDGE OF FILM 1...: '\-) C) 911:th - “‘“Kon I-‘Iontmorillonits 5 L1 .. . ; . ;. . :. .~-;--‘——j—‘- :-:E 100 «Cu—26020 Com—.1..- S... ...'K‘,;.. .':.--.-. ‘59-'57 ,:-;--£-._ Pleron Pr N+ 5...".-- " .‘i 1:1... : Saturated 4 " 31.....-1..__TL¢L " ‘ I ‘ "7 fl ' . I‘ I‘l'orlt'morillonite,--'*-T ‘ . ; 1 .-.. . . . ;. I . ' . \ . ‘ . .. . l . . .: I. , . ,. . _ . . . a E , . '-. . -- I 1 . ..---l. '. I.;.. '.l ' ....-. ”1...-.. . 4.:- . 1 ' . ‘ . . . -. c. .. o. f. —. a . _-—-.o ‘-—-u .-o.- F4 u.-.’ -na. _ - ~-..— ._—..o-- ---....—.. .n -A- .- .—.—.m’-- .A-- - a- o- n l . I 9 . . _ ; - ,... . . - - .- - o. . ., a .. ..- .....- - . ..- -. . - -.. . . 7 . I . ‘ Counts Per Second Microns Scanned Fig. l7.--Microprobe scan on edge of oriented films of montmorillonite. Counts Per Second 120 100 40 20 31 SURFACE OF FILM f “ I. :13; :3: Cu(II) Saturated "d—‘M iii 3"": -_ [E'- Montmorillonite —.I§’._Ld. A _ [I _‘P' _‘?j __ ‘Ij_ .Cu-Z602O ICom- _._.E_... I I .j- III-'7' ALI . WK“ _.-_.--..-;. .1- .._.I.....-'...-_...-. . I...“ gplex on Pr N+ __2.- Saturated 4 lat Montmorillonite .‘IR .. . , . . . 1 _ . . I u o c .. . . ' . . . o | . ,. .. . . - u - o-q - ~ ' - I - _ - . . . . .- p .o— .-u— . , . . . - -‘ ' o , . Microns Scanned Fig. 18.--Microprobe scan on surface of oriented films of montmorillonite. 32 X-ray diffraction patterns of both samples show first order diffraction in the 001 planes of 14.5 g and 12.4 g for the tetrapropylammonium montmorillonite and copper montmorillonite respectively as would be expected. Total copper analysis showed 2.45% Cu in the copper saturated montmorillonite and 0.34% Cu in the montmorillonite which was treated with the silylating agent. These data are consistent with silylation taking place at the edge of the montmorillonite and the capacity of interlayer copper is six times greater than the capacity of copper immobilized on the edge by silylation. The relative amount of inter- layer c0pper to edge copper will depend on the particle size of the clay and the thickness of the copper complex at the edge. D. Trimethylsilylation of Clays Trimethylsilylation of silica and silicates under acidic and basic conditions have been described.10 Con- ditions of high or low pH in the system cause degradation of the silicate structure. Attempts to preserve the layered structure by trimethylsilylating the clays at near neutral pH resulted in poorly defined products. Weak infrared absorptions near 1250, 1395 and 795 cm.1 are assigned to Me3Si-O. Attempts to increase the degree of silylation by use of polar solvents were complicated by side reactions of the solvent. The purpose of this silylation was to deactivate the polar silanol groups on Surface lOu 1" *f’?. Edge 1, 3, _ 7 7_ ______, ¥_ —-—-w— _ __ ,, .__..__._.. . >-... ._ 7 ~— Fig. l9.--SEM of Pr N+ form of montmorillonite which was treated With 26020 and complex with Cu+ at 2600 X. 34 the edge of the silicate structure with trimethylsilyl groups while the layered structure of the silicate struc- ture was to be preserved. This would require silylation at mild pH. Trimethylsilylation of montmorillonite with mix- tures of Me3SiCl and (Me3Si)2NH in dioxane gave an unex- plained carbonyl band at 1610 cm.1 (Figures 20 and 21). When the reaction was attempted with Me3SiC1 and pyridine a band at 1490 cm”1 suggested pyridine was intercallated between the silicate sheets. However weak bands at 1250 cm-1, 795 cm-1 and 1395 cm"1 suggested silylation took place (see Figure 22). These results are too inconclusive for any lengthy discussion but might be useful for future studies. Smaller particle size clay would be better for such a study since the amount of edge silanol is greater and the products would be easier to characterize by infrared. 35 N m m .ocmxowv cw o.m mo owumu HMHOE m um m2 Adm 02V can Howw oz Spas cmumHMme onwaoaawuoeucofi mo asuuommm commumcH33.oN .mam _ “ “L233: 3.3... . ...... m. J . 3 \ ’0 ..n .2 ...... 3n... ac... n.I-..u I. .ullu'iJnruruqu “gut-)z‘HJF ' Us .. r..l.\~.. no...” _ .3 '0 t.‘ ... .. I33: I 3 . § . . . .. . . . .34... s .0 \ .m'r-orfi ....HIH-I"I l ..Ii-.. II“..- .. 3"! «I .. ..ukO-Iu .I‘ A . m - . . . . ...c.":ln.oo' p... . E‘s! .Il 0| .1lnl .I. I I.’. . . OI-ll. .. ..ll .IDUv’ObI...-l0’lnn.ll-0." ....._ (...me 00m... 000m 003 000m Hm. 4. 3. . .... J . \ 5133.21 " . . x) \ 44... I - - 0‘ I gin-In.-. I: . ' 0.: I. I v. I, O a. 3., . . _ .. m . - — . c. / . H . a n _ i .. - 3 31.33.33 , - :33..- 3.3.3.333--- -l..3.--I.-333.3I..- 33 .- .. . - ; r . u...“ . n.1,. . ., . .1 n . . . w a . h ‘ — . . . ” --. _ - ......h .. 1 . 3.3J33I.133.33.51.23!. .. I . J . . _ . \ _ .3. .w .. _ i .. n . i _ r . ..L .l I 0 I c .. ~ u — - . n . . . . n . . . . a . — . “U a 0 w m n n I ‘ I“ . w .. OII - I, “I I ll). an. n .'I—. n.5«1c m: .:..!“ .0 y . - u / a . n u . . . . . .. . w \ . L L . . n _ l u . _ u . i h 2 w "3. -. 3 3 33:... - 1 .. .. 333. - ...Il! IlliI I- .3323: . -- 3 -..... ... .--3 - .V .3 _ . .. . n w m . . . m . . J . H b m m C ...) . . . . . . . . . . OI . . . . I|,I. 5 III.- I. 0!-l 3.!" ll.ll.n'll .I.I ....3 0|. .1 . Ola . a. . . . 3. 1.. a \l i . . I. M . .. u . . . . .. . i i , u . n . . . i 3.3 3 .r . u n m . . _ u M , . . — IL. 3 . II... II! I u I ..._ Iv Alllll . . . a .p . . . . .'. .I'Illll,l.i3I.OIl.3 II. 333' L “J _ u .. u . H . u . .. . ’ . \\ _ .n . n .. u. .—l l I In 01 -...” ..‘HIIIII .IL I. filllfil'l I. . w. .3.... . -n . . n. . H l m N _ _ n _ u . a m . _ _ fl .3 \J. - .13. m- .3. . _ ..... ..l. (0.4.. . / a . _ . . . . .-I m . _ 3 . - i _ us . . . . . . . . . . . . . . . _ . . . . . . o . u . . . 4 . . . . . q u . 0.0.4 to . . . u I . I I 3.. .n . -. O. . u .. . . 3 - .II c I ... . . c . . . . . . - . n . . . . . . . n u . . . . . r . . . . . i . . 3 . . . . . . . . . r .43. . . . . . . . l .31.. .. ..I.-. 3 nu . 30' u 3 o .l-.‘.': 3" .I" I‘ll. . . . . h . . . . l . . . . . . . . . . . . . . . . a . . _ . . . . . . u . . . . v . u I... . . . . . . . . . ... u v . U.- o I I ... c . .. o... . o O. o . u . . . . . o . . . . . . . . l n . . . . . . . _ .. . 3.3. .. q. .. 3.. . H _ v _ . . 1 . a \l I .... .M I I ill U” II II. ll" 1 IIIIOOA _ , _ I 31-..! ...I'I'I-l ' .Cw . . 1 u . _ .1- . H . . . . M s o.'l. . I. Ill"... 0!... III |" I ‘IM IIIbII.‘| .l 'l . . . . . — m _ . . _ . u. a . m .:...m :3 . I... m. _ n. . _ . u . . _ . n . . . . . w . . , . ~ . _ , . . . . . . . | .3 I Q... .Fl -3.“ .-ll: ' '0. l l luvlllm 0|...u.L nirhltlorllllu." ”llol.o|i|mlo'u .rllWIuIIIIIIII rulll. .PI . 3“. . — I I i- 0' ”/0." 36 N m m .mamxoww cw m.o mo owumu Hades a pm :2 flaw m2. can HUflm mz spas cmumamaflm wuacoaafluoeuaos mo Eauuommm cmumumcmu3.am .mam Q SCIIKCLuhC _ ..\ \. ~ . .3 3 . 3. ..I (I. .- 3333. ‘ . o n I.II.II3. I . I .. I. . .3 I ..3. . \N . Q «15.5mm Iu.J.r..3.30.:m 000— ooom 00mm 000m 000m mum» . .. _ . n _ .. a M _ . U . . . u . . . . o . . . I. a v 30. . . ..lI-lc . 3” ..II33IIIIliIIWII'3IIIIbgtIIIll 4| ? .IIIII'I... 3| .. .u . .03...I.II . . ... u, n. H . w . . . . . _ . . . q u . . . . . n _ . _ . . _ . - .3-3.-:- . . . 3.3-:33-33 - -- 3--- _om . . . . . . o_ . . H h h u . . _ - 3 . . 3 . . . 3, 333:3-h333 3- 3- m . L . . . p a l . . a L- - ._ “3.3.3 3.. _ _ - u - . Nu . . — . . I7 0 m . . . _ \V .. 3. I3 3 3 1'33»... _ ..l .333 3....3Lll33!3.-3|33.3.3- I. I! I . .3IJ . .3J . . _ . . . _ .o.N . _ _ . OJ )/ — c u c . . . v . ..., - 3 .. .3 _ - .-.. - .. . .. .. h . . _ . . . .. . U3 . f . n . . u , . . H” . H _ m . . m _ . . . . . .3. .333.- ..l-.3.-3 .3.- 3 3T . . . fl . . . _ _ . :I. . .3333.. 3.3. [lull-3333.333.3.3.. 3 ._ 3|. . H ‘ . . m n . n . m H h m V . . ..3..33“. .. . . 3. r." . H..- .3.. . m H . m w _ . _ M . _ w 3|3ll3 llulu 3 I. . . . _ . . ll. .I33 L-l3. “3333 .33”.ll.|-..-..3|3.. 3. . 23.0 II M . I... . . .. _ H3 a . ( . ..w. ._ . _ . u - _ . a _ . . M. . u _ . _. ( . . .mafloflumm a. Howmm w: suw3 cmumHMHHm muwcoaaflHOEucoe mo Eduuommm vmumnmcH33.- .mflm .33 n .. .. ~IZU wtfi%U . 000. new. 02... com. . . 0 8m 9m.- — . _ . _ _ . . . . a . _ . . . . . . . a . . . ‘ _ . . . . .. . . 3h 7 \ J .3 J” )x 3. \I\ no a m .n j. 0W *4 D . . . J. . __ 5n» 5 . .. _ /. ... ~ .... _. _ . . .. \ .. u . DV\ 3“ on. w . s . w 33-. #, ...... _, r. . . 3 . .... ..h r a. - . . . f III. CONCLUSIONS The environment of copper on clays and silica gel depends on the method of deposition. Esr is a useful tool for detecting the difference in the metal ion's environment. Cu (II) amine complexes in a methanol glass at 77°K resemble immobilized Cu (II) amine complexes on silica surfaces. Silylation of clays occurs on the edge of the silicate crystal where silanols are present, as suggested by the covalent bonding theory of coupling agents. On montmorillonite Cu interlayer/Cu edge is equal to 6.3. 38 IV . EXPERIMENTAL A. Fractionation of Montmorillonite About 6 kg of a 2.5 wt % aqueous suspension of Na- montmorillonite (Upton Wyoming, Source Clay) was allowed settle in graduated cylinders. After 16 hours of sedimenta- tion the top 17.5 cm of suspension (50 volume %) was siphoned, centrifuged at 9250 rpm and the buff colored product freeze dried to give 29.5 g of clay. Analysis by infrared showed vOH 1, VHOH 1626 cm.1 and 1 850-1200 cm- . BET surface area of 9 mz/g was 3200-3700 cm' “81031 obtained by N2 adsorption after 2 hrs of degassing at 180°C. B. Preparation of Cation Exchange Forms of Clays Cation exchange forms of the <2u fraction of montmorillonite were prepared by adding excess halide salt to a 2.5% aqueous suspension of <2u montmorillonite, removing the clay by centrifuging, washing the clay twice with a l M solution of the halide salt and finally washing with distilled water until no halide was detected in the supernate. The exchange form of the clay was then freeze dried. 39 40 Thus, n-Pr4N + montmorillonite (2.56 g) was pre- pared from 3.5 g of <2u fraction of montmorillonite. . . -1 Analy31s by 1nfrared showed VOH free 3620 cm , vCH 2890, l 1 2950 and 2980 cm” , VCH 1360, 1380, 1460 and 1480 cm” , and a weak band at 1630 cm.1 caused by H20 deformation. Heating the clay in air to 110°C for 1 hr did not cause disappearance of the band at 1630 cm-1. Analysis of the powder by x-ray diffraction gave a first order spacing of the 001 plane of 14.2 A. The copper exchange form (2.38 g) was prepared from 3.00 g of <20 fraction of montmorillonite. The infrared spectrum of the product showed VOH OH 850-1200 cm'l. Chemical analysis showed 3620, v 1622 and VSiOSi 2.15% Cu present or 67.6 meg/100 g. Analysis of an oriented film by x-ray diffraction gave a first order spacing in O the 001 plane of 12.4 A. The <20 fraction of baroid hectorite (2 g) was treated with excess Bu4NI in methanol (400 ml). The pro- duct was washed free of halide, filtered, and air-dried. Analysis of the product by x-ray diffraction gave a first order spacing in the 001 plane of 15.2 A. C. Silylation of Pr N + Montmorillonite with N-(B- Amino-Ethyl)-y-Aminopropy1- ® trimethoxysilane (Dow Corning Z6020 Silane) The Pr4N+ exchange form of montmorillonite (1.225 g) was suspended in 75 g of a 0.4% aqueous solution of Dow 41 ® Corning Z6020 Silane, stirred for 15 min and freeze dried. A portion of the product (0.946 g) was washed with 50 m1 portions of H 2 20 until the supernate gave no color change with Cu+ (total of 4 washes). The washed clay was sus- pended in 50 ml of 0.10 M CuCl for 10 minutes, washed 2 until no Cu+2 was detected in the supernate and freeze dried. The product (0.706 9) gave infrared spectrum similar to the starting clay. However the absorptions v 2850-2970 cm-1 were broader and more intense. An CH oriented film was prepared by evaporation of a suspension of the product. Analysis by scanning electron microsc0py and microprobe are presented in Figures 17, 18, and 19. Analysis of Cu+2 by esr is presented in Figure 16. Chem- ical analysis showed 0.34% Cu present or 10.8 meq/lOO gm of clay. Analysis of x-ray diffraction gave a first order spacing of 14.5 A. D. Silylation of Bu Hectorite and Its Cu+2 Complex 4N+ The Bu4N+ exchange form of Hectorite (0.831 g) was dispersed in 60 ml of H20. A 3% aqueous solution of Dow Corniné§ Z6020 Silane (10 ml) was added, and the mix- ture was stirred for 30 min. The clay was washed until the supernate gave a negative amine test with CuCl2 solution, and the resulting clay was suspended in 50 m1 of 0.10 M CuCl solution for 10 minutes. Excess COpper was removed 2 by repeated washing with distilled H20. After air drying, 42 the product (0.592 g) was analyzed by esr. The results can be found in Table 2. E. Silylation of Silica Gel with Aminoalkylsilanes Silica gel (10.0 g, BET surface area 540 mZ/g) was mixed with a 5% aqueous solution (42 g) of Dow Corning Z6020 Silane or Union Carbidé®.A1100 Silane for 1 hr at room temperature. The solution was decanted, and the silica gel was washed twice with H O (100 ml), once with 2 boiling H O (100 ml for 30 min) and finally at room tempera- 2 ture with 100 ml H20. The products were dried for 10 hrs at 95°C to give 9.0 and 8.7 g of product respectively. F. Copper (II) Complexes of Silylated Silica Gels Silica gel (1.00 g) which contained the function- alizing group indicated below was stirred with aqueous 0.10 M CuClz, washed twice with 100 m1 of H O and allowed 2 to air dry. Temp of Recovered Functional Group Wash Silica Gel Color ESi(CH2)3NH(CH2)2NH2 100 C 0.90 9 Dark Blue = a O -Si(CH2)3NH(CH2)2NH2 25 C 0.80 g Blue ESi(CH2)3NH2 100°C 0.80 g Light Blue None 100°C 0.83 g Greenish White Esr data for each sample is presented in Table 2. 43 G. Copper Amine Complexes GD Dow Corning Z6020 Silane and ethylenediamine were added to 3 x 10'3 M CuC12°2H20 in methanol to give various molar ratios of amine to Cu+2 as summarized in Table 1. The solutions were analyzed by esr. The spectra are shown in Figures 1-10 and Table 1. H. Trimethylsilylation of Clay 1. Na+ Montmorillonite with MeQSiCl and (MeqSi)2§§ in Dioxane.--A round bottom flask fitted with a stirrer, reflux condenser and N2 purge was loaded with 1.00 g of <20 montmorillonite and 24 g of dioxane. After stirring the mixture for 20 min, Me3SiCl (1.072 g, 9.88 mole) and (MeBSi)2NH (0.799 g, 4.96 mole) were added and the mixture was refluxed for 4 1/2 hrs. The cooled mixture was stirred an additional 17 hrs, and 1 M NaCl (70 ml) was added to the slurry. The product was washed with distilled H 0 until 2 a negative halide test was obtained with AgNOB. The infra- red spectrum of the freeze dried.lclay (0.655 g) 3630, 3360 and 1630 cm’l, ”CH 1 975-1150 cm‘ . A weak band at (Figure 20) showed v0 2960 H (weak), 1425 and vSiOSi 1730 cm-1 (possibly C=O) was also present. The procedure was repeated with Me3SiC1 (0.715 g, 6.59 m mole) and (Me3Si)2NH (2.121 g, 13.17 m mole). Analysis of the product (0.877 g) by infrared showed 3625, and 3400 cm’l, v 2940, 1500, 1420 and 1350, V CH OH 44 VC=O 1610 and 1720 cm-1. The band at 1610 cm.1 was very strong as shown in Figure 21. 2. Na+ Montmorillonite with MeQSiC1 and (Me,Si)2§§ in Bulk.--The <20 fraction of Na-montmorillonite (1.00 9) Me SiCl (4.48 g, 62.2 m mole) and (Me3Si)2NH (20.0 g, 124 3 m mole) were refluxed for 18 hrs as previously described. The clay was removed by centrifuging, washed twice with hexane (30 ml) and once with 1 M NaCl (30 ml). After washing the clay free of chloride it was freeze dried to give 0.6943 g of product whose infrared spectrum was the same as montmorillonite, except that weak bands at v 1 CH 2910 and 2850 cm‘ , v 850 cm“1 were detected. No band due to SiMe deformation (1250-1270 cm-l) was detected. 3. Na+ Montmorillonite with Me3SiC1 in Pyridine.-- The formation of a light blue color was observed when Na- montmorillonite (1.00 g) and pyridine (23 ml) were stirred for 3 hrs in a N2 atmosphere. After refluxing the mixture with Me SiCl (2.74 g, 25.2 m mole) for 18 hrs the clay 3 turned dark blue. The product was recovered by decanting the supernate. It was washed with EtOH (50 m1), then with H20 (100 ml), and finally with EtOH (100 m1). Oven drying at 100°C for 1 1/2 hrs gave a white product (0.950 9) whose 1 , VHZO 1625 950—1200 cm-l. Weak infrared showed vOH 3600 and 3100-3500 cm- -1 cm , 1490 cm'1 and v vpyridine SiOSi 45 1 bands at 1250 cm- , 795 and 1395 cm—1 could be attributable to SiMe (Figure 22). 4. PrAN+ Montmorillonite with MeBSiCI and (Me,Si)2NH in MeCN.--A flask as previously described was loaded with the Pr N+ exchange form of montmorillonite 4 (0.400 g) and acetonitrile (9.0 ml). After 30 min of stirring, Me3SiC1 (0.543 g, 5.00 m mole) and (Me3Si)2NH (0.808 g, 5.02 m mole) were added, and the mixture was refluxed for 11 hrs. The product was washed with water until no chloride was detected with AgNO After drying 3. at 110°C for 4 hrs analysis of the product (0.321 g) by infrared showed the same absorptions observed with the starting clay, but weak absorptions at 1255 and 840 cm-1 could be attributed to SiMe. I. Intercalation of Silica in Na+ Montmorillonite Two separate aqueous suspensions (2.4 wt %) of Na-montmorillonite and CabosiflE)S-l7 fumed silica were prepared by mixing with distilled water and dispersing in an ultrasonic disruptor for 4 min at 55 watts. About 14.2 g of a 1.41% solution of [(MeO)BSi(CH NMe3J+c1‘ was added 2’3 to the silica dispersion and it was treated again with the ultrasonic disruptor. Both suspensions were diluted to 2.00 wt % and mixed as shown below. After 6 hrs the pro- ducts were centrifuged at 9000 rpm for 30 min, decanted and freeze dried. 46 ma whm mm mm OOH hma bmH vmm m\mE mmud momeSm 9mm 05H.o mwN.o mmH.o mmm.o v5m.o hmm.o mmm.o mnm.o mcflmuo mummum Hmumd madam meaoao ummHU snsoao snsoao HmmHU HmmHU HmmHo Hmoau mwmcuwmsm mo mocmummmmm wooa mo wom mom who wom wmm woa smao w o.oa m.ma o.mH mmao UmNHHMGOHpossm mm as mHQEmm 47 I. General A varian E-4 electron spin resonance spectrometer fitted with a liquid nitrogen dewar was used to obtain esr spectra. Typical conditions for obtaining spectra are given in figure 1. Copper Ka and La radiation was measured with an ARL scanning electron microprobe instrument at 15.5 kv and 0.018 p A. 10. BIBLIOGRAPHY P. Pleuddemann in Composite Material, Vol. 6, L. J. Broutman and R. H. Krock, eds., Academic Press, New York (1974). E. Schrader, I. Lerner, and F. J. D'Oria, Mod. P1as., 45, 195 (1967). K. Johanson, F. O. Stark, G. E. Vogel, and R. M. Fleischman, J. Comp. Material, 1, 278 (1967). Tominaga, Y. Ono, and T. Keii, J. of Catalysts, 40, 197 (1975). L. Burwell, Chemtech, 370, June (1974). M. Clementz, T. J. Pinnavaia, M. M. Mortland, J. Phys. Chem., 11, No. 2, 198 (1973). E. Leyden and G. H. Luttrell, Anal. Chem., 41, 1612 (1975). E. Leyden and G. H. Luttrell and T. A. Patterson, Anal. Letters, 8, 51 (1975). M. Barrer and K. Brummer, Trans. Faraday Soc., 59, 959 (1962). W. Lentz, Inorg. Chem., 3, 574 (1964). 48 ICHIGRN STRTE UNIV. LIBRRRIES 31293006971745