CI A MOSSBAUER AND ADSORPTION STUDY OF F e 5 7 IN LINDE L ZEOLITE by RAYMOND TROY LASSAU B.Sc, University of B r i t i s h Columbia, 1966 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE i n the Department of Chemistry We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 31, 1972 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree tha t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and s tudy . I f u r t h e r agree t h a t permiss ion f o r ex tens ive copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l ga in s h a l l not be al lowed w i t h o u t my w r i t t e n pe rm iss ion . Department o f The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada ABSTRACT 3+ A s ing le Fe species has been introduced in to Lmde L z e o l i t e and i s i d e n t i f i e d as an i r o n oxyhydroxide from i t s Mossbauer r e l a x a t i o n p r o p e r t i e s . The e f fec t s of , CO^ and C2Hg on the f e r r i c species and on the bulk behaviour of the sample are observed by Mossbauer Spectroscopy and adsorption studies- under outgassing cond i t ions . The f e r r i c oxyhydroxide i n i t i a l l y blocks the pores to gas adsorption but reducing i t O o o to Fe at 943 K and r e o x i d i z i n g i t to a - F e 2 ° 3 i n a i r at 773 K frees the pores . The bulk a -Fe2 0 2 i s concluded to l i e on the ex terna l z e o l i t e surface . Acknowledgements: I would l i k e to thank Dr. J . R . Sams for h i s patience and guidance during the course of t h i s i n v e s t i g a t i o n . I would a lso l i k e to thank Mrs. L . S a l l o s , Dr. J . Ruddick and Mr. J . C . Scott for invaluable ass i s tance . i i i TABLE OF CONTENTS Page ABSTRACT i ACKNOWLEDGEMENTS i i LIST OF FIGURES i v LIST OF TABLES V INTRODUCTION 1 EXPERIMENTAL 15 RESULTS AND DISCUSSION 31 BIBLIOGRAPHY 73 APPENDIX 76 iv List of Figures Figure Page 1. Energy level diagrams for Isomer Shift, Quadrupole Splitting and Zeeman Splitting 5 2. Adsorption system 20 3. Adsorption c e l l 23 4. Mossbauer c e l l 28 5 . Variable temperature Mossbauer Spectra for iron-L-zeolite 35 6. Magnetic measurements on iron-L-zeolite 40 7. Mossbauer Spectra of calcined powder 45 8. Mossbauer Spectra showing change in hyperfine pattern with gas treatment 51 9. Mossbauer Spectra of hydrogen sweeping of iron-L-zeolite 57 10. Mossbauer Spectra showing effect of temperature on hydrogen sweeping of iron-L-zeolite 61 11. Nitrogen adsorption isotherms on solvated-L-zeolite and iron-L-zeolite 65 12. Nitrogen adsorption isotherms over a range of sample treatments on iron-L-zeolite 67 13. Adsorption isotherms of CC^ and C2H6 o n solvated L-zeolite and iron-L-zeolite 68 14. Isosteric heats of adsorption versus surface coverage 71 V L i s t of Tables Table 1. Chemical ana lys i s of d i f f e r e n t L - z e o l i t e samples 2. Volume of system segments 3. d-spacings from x-ray powder photographs 4. Mossbauer parameters for various sample treatments 5 . Mossbauer parameters for evacuated sample and various gas treatments 6. Mossbauer parameters for outgassed sample and var ious gas treatments 7. Mossbauer parameters for various treatments of sample 8. Adsorption data on L - z e o l i t e and i r o n - L -z e o l i t e INTRODUCTION In the study of s o l i d surfaces many physical tools have been employed. Such techniques as are used i n electron microscopy, o p t i c a l and magnetic resonance spectroscopy, Mossbauer Spectroscopy and sorption studies have contributed s u b s t a n t i a l l y to present day knowledge about surface structure and c a t a l y t i c a c t i v i t y . Of the many v a r i e t i e s of surfaces studied by these methods a large measure of attention has been paid recently to the physical and chemical behaviour of porous materials such as z e o l i t e s . l 2 Synthetic z e o l i t e s ' consist of tet r a h e d r a l l y arranged s i l i c a t e s and aluminates bonded by common oxygen atoms. The polyhedra so formed contain large i n t e r s t i t i a l c a v i t i e s r e s u l t i n g i n highly porous s o l i d s with large i n t e r n a l surface areas. The s i l i c o n to aluminum r a t i o varies i n d i f f e r e n t z e o l i t e s and hence the number of cations required to balance the r e s i d u a l charge on aluminum also v a r i e s . The cations occupy d i f f e r e n t s i t e s owing to the random replacement of s i l i c o n by aluminum. Water molecules occupy the c a v i t i e s and form hydration complexes with the cations as well as e l e c t r o -s t a t i c a l l y i n t e r a c t i n g with the framework oxygen atoms. The molecular sieving properties of z e o l i t e s have been well established i n the l i t e r a t u r e . Of more recent i n t e r e s t are studies of the centres of c a t a l y t i c a c t i v i t y on the surface and the changes that occur by including d i f f e r e n t ions i n the z e o l i t i c channels. - 2 -When an iro n ion i s adsorbed from a l i q u i d onto a z e o l i t i c support or ion-exchanged into i t , the valence state and type of bonding of the ion to the surface, the symmetry of the e l e c t r i c f i e l d surrounding the ion, and the presence and magnitude of a magnetic f i e l d are but a few of the many pieces of information 3 that can be obtained by Mossbauer Spectroscopy . I f sample evacuation, thermal treatment or the presence of eithe r p h y s i c a l l y or chemically adsorbed gases a f f e c t any of these properties, the Mossbauer Spectra w i l l r e f l e c t the changes. The isomer s h i f t , 6, arises from the i n t e r a c t i o n between e l e c t r o n i c charges and nuclear charges. The magnitude of the s h i f t depends on the size of the nucleus i n i t s ground and 2 excited states and on the s-electron d e n s i t i e s ( | i f)(o)| ) at the nucleus for the source and absorber. I t i s given by 6 = |E Ze 2[|*a(o)| 2 - |*s(o) l % 2 e x - R 2 g d ] where Z i s the nuclear charge, e the electron charge, R.„ and •R ^ the excited and ground state r a d i i ; "a" and "s" denote source and absorber. P and d-electron density influence the isomer s h i f t only through sh i e l d i n g e f f e c t s of the s-electron density at the nucleus. 3+ 2+ In i o n i c compounds of high spin Fe and Fe , the 2+ presence of an additional d-electron i n Fe shields the 3+ s-electrons from the nucleus to a greater extent than i n Fe , 2+ 2 hence Fe has a lower value of |^(o)| . This, coupled with 2 2 4 57 the fact that (R e x~R g^) i s negative for Fe leads to the 2+ f a c t that Fe should have a more p o s i t i v e isomer s h i f t than 3+ Fe . Thus the valence state of the ion can be i d e n t i f i e d . Quadrupole s p l i t t i n g i s the r e s u l t of the i n t e r a c t i o n between the nuclear quadrupole moment and an e l e c t r i c f i e l d gradient (EFG) at the nucleus established by charges i n i t s , immediate v i c i n i t y . The e l e c t r i c f i e l d gradient can a r i s e from non-spherical d i s t r i b u t i o n of the atom's own valence electrons or from neighbouring ion or atom asymmetries. Nuclei with spins I » 1 have quadrupole moments, for example, 57 3 the f i r s t excited state of Fe where I = /2. 57 For Fe the quadrupole s p l i t t i n g i s the difference i n energy l e v e l s between the magnetic quantum states 3 1 3 m = ± /2 and ± /2. The degeneracy of the I = /2 state i s p a r t i a l l y l i f t e d by the EFG. The quadrupole s p l i t t i n g i s given by AE Q = E Q ( 3 / 2 ) - E Q ( | ) = | v z z e Q [ l + n 2] ± / 2 3 where V„„ i s the Z-component of the EFG tensor, eQ the nuclear quadrupole moment of the f i r s t excited state, and n an asymmetry parameter defined by n = ^ VXX~ VYY^ VZZ" T h e t r a n s i t i o n 1 3 1 1 p r o b a b i l i t i e s ±j •* i j a n < ^ ±2~ ±2 a r e e c l u a ^ - f ° r a powdered sample with random or i e n t a t i o n of the c r y s t a l l i t e s and two l i n e s of equal i n t e n s i t y are obtained. Where c r y s t a l l i t e orientations are not t o t a l l y random the angular dependence of the t r a n s i t i o n s may not average to zero and two l i n e s of unequal i n t e n s i t y w i l l r e s u l t . 3+ For high spin Fe compounds the h a l f - f i l l e d d - s h e l l 5 (3d ) i s s p h e r i c a l l y symmetric and hence where a small quadrupole s p l i t t i n g i s observed i t must ar i s e from the EFG •- 4 -set up by the arrangement and e l e c t r o n e g a t i v i t i e s of i t s nearest neighbour ligands which are e i t h e r not a l l chemically i d e n t i c a l or which do not have cubic symmetry. For high spin 2+ 6 Fe the extra d-electron (3d ) contributes much more to the large quadrupole s p l i t t i n g (the EFG i s proportional to -3 5 < r >) than do the ligands . Figure la) represents isomer s h i f t and quadrupole s p l i t t i n g and Figure lb) Zeeman s p l i t t i n g . Zeeman s p l i t t i n g i s the r e s u l t of an i n t e r a c t i o n between the nuclear magnetic dipole moment (y) and a net magnetic f i e l d (H) at the nucleus. The energy l e v e l s for the i n t e r a c t i o n are E = -yHmT = -gy Hm_ m I ^ n I I where g i s the gyromagnetic r a t i o and y n the nuclear magneton. In the absence of any quadrupole i n t e r a c t i o n there are 21+1 equispaced l e v e l s , the s p l i t t i n g between adjacent l e v e l s being 9Jv n H - Only six spectral l i n e s r e s u l t because of the se l e c t i o n rule Air^ = 0,±1. With quadrupole i n t e r a c t i o n the energies of the d i f f e r e n t m^ states are s h i f t e d . The magnitude of the quadrupole s p l i t t i n g i s determined as the difference between the two pairs of outer l i n e s , that i s | A i 2 - A 5 6 ' a S ""s t h e c a s e w :*- t h the s i x - l i n e spectra of a-Fe202« The sign of the e l e c t r i c f i e l d gradient depends on whether ( A^2~ A56^ ^ s P o s i t ; i - v e o r negative. The asymmetry of quadrupole s p l i t l i n e s for a given oxidation state of the absorbing atom has been attributed to FIGURE 1 Energy l e v e l diagram showing: a. Isomer S h i f t and Quadrupole S p l i t t i n g . b. Nuclear Zeeman S p l i t t i n g and the e f f ec t thereon of Quadrupole Coupl ing . I S O M E R S H I F T M A G N E T I C H Y P E R F I N E S P L I T T I N G M A G N E T I C H Y P E R F I N E S P L I T T I N G + Q U A D R U -P O L E S P L I T T I N G - 6 -two causes: 1) the non-zero averaging of the angular dependence of c r y s t a l l i t e or i en ta t ions (the angle i s a funct ion of the d i r e c t i o n of o r i e n t a t i o n of the EFG and y-ray) as has already been mentioned and 2) the presence of s u f f i c i e n t numbers of c r y s t a l l i t e s con tr ibut ing to the Mossbauer e f fec t with 6 an i so trop ic l a t t i c e v i b r a t i o n s . A t h i r d and current ly very important cause of asymmetry observed i n Mossbauer Spectra i s a t t r i b u t e d to f l u c t u a t i n g e l e c t r i c and magnetic f i e l d s generated e i t h e r as a r e s u l t of the r e l a x a t i o n of paramagnetic ions or by the f l u c t u a t i o n of the environment surrounding the 7 8 nucleus ' . Paramagnetic r e l a x a t i o n phenomenon are the r e s u l t of time-dependent changes i n unpaired e l ec tron spin d i r e c t i o n s . The two mechanisms which cause s p i n - f l i p p i n g are e l e c t r o n i c s p i n - s p i n i n t e r a c t i o n s with neighbouring ions and e l e c t r o n i c s p i n - l a t t i c e i n t e r a c t i o n s . Both involve the t rans fer of energy between i n t e r a c t i n g sp ins , the former through d ipo le and exchange spin in t erac t ions and the l a t t e r through s p i n - o r b i t coupl ing . Sp in - sp in re laxa t ion i s temperature independent and the r e l a x a t i o n time increases as the distance between paramagnetic centres increases . S p i n - l a t t i c e re laxa t ion i s temperature dependent and the r e l a x a t i o n time increases with decreasing temperature. I f the re laxa t ion time i s less than the time taken for a Mossbauer event then the e f f e c t i v e magnetic f i e l d generated at the nucleus w i l l be zero. Where the re laxa t ion time i s equal to or greater than that for a Mossbauer event a non-zero e f f e c t i v e magnetic f i e l d w i l l r e s u l t . Mossbauer spectra show 57 these e f f e c t s i n the following manner. For an Fe nucleus i n a f l u c t u a t i n g magnetic f i e l d and a fi x e d e l e c t r i c f i e l d gradient: a) where the rate of f l u c t u a t i o n i s slower than the nuclear precession frequency a s i x - l i n e hyperfine pattern r e s u l t s , b) for extremely rapid f l u c t u a t i o n the s i x - l i n e pattern collapses inward (motional narrowing) and a symmetrical quadrupole doublet i s seen, and c) for intermediate f l u c t u a t i o n 3 1 rates the -*• ±j t r a n s i t i o n l i n e which makes up one h a l f of the quadrupole doublet i s broader than the other l i n e of the doublet because the relaxation time for the 4 * 4 l _ i ^ 2 * * 2-l a t t e r component i s shorter. Hence an asymmetric quadrupole doublet i s seen (line-widths T Q are inversely proportional to the h a l f - l i v e s t, of the excited s t a t e s ) . 1/2 The same pattern i s generated by relaxation e f f e c t s i n cooperative phenomenon such as ferromagnetism and a n t i f e r r o -magnetism. Techniques other than Mossbauer Spectroscopy (magnetic studies for example) are sometimes necessary to d i s t i n g u i s h between ordered magnetic and disordered paramagnetic hyperfine spectra. The a p p l i c a t i o n of Mossbauer Spectroscopy to the study of surfaces y i e l d s a great deal of information. In p a r t i c u l a r , from the adsorption i n t e r a c t i o n between gas and metal atoms, changes i n the isomer s h i f t and the quadrupole s p l i t t i n g r e f l e c t changes i n the e l e c t r o n i c structure of the Mossbauer - 8 -atom and produce information about the adsorption complex. However, where ions are introduced in to z e o l i t e s the degree to which the ions occupy external or i n t e r n a l surfaces i s not n e c e s s a r i l y d iscernable from Mossbauer parameters i f the s i t e s have s i m i l a r energies and geometries. Also pore s ize r e s t r i c t i o n s on adsorbed gases, surface area and energet ics of the bulk mater ia l are not obta inable . Adsorption isotherms can provide t h i s information and valuable comparisons can be made between the c o l l e c t e d data of both techniques. When the e q u i l i b r i u m concentrat ion of gas molecules at a s o l i d surface i s greater than that i n the bulk gas phase the gas adsorbate i s i n t e r a c t i n g with the s o l i d adsorbent. Adsorpt ion processes are conveniently c l a s s i f i e d as e i t h e r p h y s i c a l or chemical depending upon the nature of i n t e r a c t i o n between adsorbate and adsorbent, and can usua l ly be d i f f e r e n t i a t e d by the fo l lowing c r i t e r i a : 1) The i n t e r a c t i o n energies for p h y s i c a l adsorption are of the order of magnitude of those for condensation processes , whereas for chemisorption energies s i m i l a r to those for bond formation are observed and u s u a l l y an a c t i v a t i o n energy b a r r i e r must f i r s t be overcome. The energies of both types of adsorption can vary appreciably with surface coverage i f the adsorbent surface i s not homogeneous and under pressure condit ions where l a t e r a l gas-gas in t erac t ions occur . Highly porous s o l i d s o f f er a wide v a r i e t y of adsorption s i t e s whose energet ics can be qui te d i f f e r e n t . 2) Chemisorbed gases are one molecular diameter i n thickness over any pressure range whereas m u l t i l a y e r formation i s - 9 -frequent ly observed when gases are p h y s i c a l l y adsorbed at pressures approaching the adsorbate vapor pressure . 3) P h y s i c a l adsorption i s pre ferred at the adsorbate b o i l i n g p o i n t . Owing to the k i n e t i c energy of the gas molecules at higher temperatures the volume adsorbed decreases. The a c t i v a t i o n energy involved i n chemisorption r e s u l t s i n the reverse of t h i s t rend , the higher temperatures being favoured. The rates of adsorption i n both cases may be s i m i l a r i f e q u i l i b r i u m depends on d i f f u s i o n e f f ec t s inherent i n porous adsorbents; otherwise chemical adsorption i s slower. 4) P h y s i c a l l y adsorbed gases are r e a d i l y desorbed when the pressure i s reduced at the temperature at which adsorption took p l a c e , but often high temperatures of outgassing are required to remove a chemisorbed l a y e r . As the determination of z e o l i t e surface areas i s one of the main objec t ives of t h i s work only p h y s i c a l adsorption w i l l be discussed f u r t h e r . The number of moles of gas adsorbed on a s o l i d depends on the i n t e r a c t i o n energy generated between the two phases, on the temperature and pressure and on the area of the s o l i d surface exposed to the gas phase. P lo ts of the amount of adsorbed gas against e q u i l i b r i u m pressure at a constant temperature (adsorption isotherms) y i e l d q u a l i t a t i v e and semi-quant i tat ive information about the adsorption system. From such isotherms estimates of surface area and adsorption energy can be made although the i n t e r p r e t a t i o n of these - 10 -q u a n t i t i e s i s subject to the assumptions contained i n the model chosen to represent the adsorption process . For porous mater ia l s such as z e o l i t e s , i f isotherms are determined below the c r i t i c a l temperature of the adsorbate and over a pressure range from a few t o r r to an atmosphere, m u l t i l a y e r adsorption i s u s u a l l y observed. The mult imolecular 9 adsorption theory of Brunauer, Emmett and T e l l e r (BET theory) provides a use fu l two constant equation from which approximate surface areas and heats of adsorption can r e a d i l y be evaluated. The BET theory i s based on the fo l lowing assumptions: 1) The surface possesses uniform, l o c a l i z e d s i t e s 2) The adsorption i s non-cooperative i n that l a t e r a l i n t e r a c t i o n s between gas molecules are ignored 3) The energy of adsorption i n the f i r s t layer E ^ , i s a constant 4 ) The energy of adsorption i n succeeding layers i s E _ , the energy of adsorbate l i q u e f a c t i o n and 5) The surface area a v a i l a b l e for the nth layer i s equal to the coverage of the ( n - l ) t h l a y e r . As has been pointed out l o by d e t a i l e d c r i t i c i m s of these assumptions, the area and energy parameters have no absolute s i g n i f i c a n c e . However, where surface areas are compared on a r e l a t i v e basis using the same adsorbate the theory i s qui te u s e f u l . The BET isotherm takes the fo l lowing form v C P m (Po-P) ^1+(C-1) C P /Po)j where v i s the volume adsorbed at pressure p, v the adsorbate c m volume required to complete the f i r s t adsorbed l a y e r , Po the adsorbate vapor pressure at the experimental temperature T ( E , - E ) / and C = e . I n order to obtain values for v and C m - l i -the above equation i s rearranged into the following form Po = 1 + Po-P v(Po-P) v v CP m m Thus, p l o t s of p°/ v(p 0_p) against (Po-P)/ should be l i n e a r with 1^ a s the intercept and 1 as the slope. v v C m m The value of the surface area A, of the adsorbent i s obtained by assuming clo s e s t packing of the adsorbate. In f a c t t h i s i s i n opposition to the site-wise l o c a l i z e d adsorption postulated by the BET theory. This would require that the adsorbent be close-packed and defect free and as well that the adsorbate and adsorbent be p r e c i s e l y matched i n s i z e . For hexagonal c l o s e s t packing of spheres A = 9.322 X 1 0 1 5 y ( M \ ^ 3 m 2g _ 1 \4/2Np / where M i s the adsorbate molecular weight, N Avogadro's number a n d p the density of the l i q u i f i e d gas. As the i n t e r p r e t a t i o n of the value of E^, obtained from the BET equation i s hampered by a l l the assumptions inherent i n the model, arguments about energetics are better based on a thermodynamic quantity. Where isotherms f o r the same adsorbate-adsorbent system are avail a b l e at d i f f e r e n t temperatures, d i f f e r e n t i a l heats of adsorption ( i s o t e r i c heats qg T) c a n be obtained from an analogue of the Clausius-Clapeyron equation. qST ^ l / ^ N a f A allows the d i r e c t determination of adsorption heats from experimental isosteres by p l o t t i n g lnp against "^T at constant number of moles adsorbed Na. The necessary thermodynamic - 12 -1 1 - 1 3 r e l a t i o n s have been previously derived from solution thermodynamics. Since the charge-compensation cations of z e o l i t e s l i e i n c r y s t a l l o g r a p h i c a l l y well-defined positions an iron-exchanged molecular sieve should have a high f r a c t i o n of c r y s t a l l o g r a p h i c a l l y well-defined surface Mossbauer atoms. Thus many experimenters have been preparing iron z e o l i t e s and examining the iron products so formed. Depending upon the method of preparation, the z e o l i t e s are eithe r exchanged or act as supports for the growth 14 15 of small p a r t i c l e s ' F e r r i c ions exchanged into z e o l i t e s have led to d i f f e r e n t 1 6 Mossbauer r e s u l t s . Morice and Rees showed that exchange i n X and Y z e o l i t e s produced high spin f e r r i c and ferrous species, 1 7 whereas exchange i n L - z e o l i t e by Wedd and co-workers produced a f e r r i c species only. On outgassing t h i s sample the ferrous species was obtained. F e r r i c ion-exchange i n Y and M z e o l i t e s was studied by 18 Goldanskii and co-workers who obtained a Mossbauer doublet c h a r a c t e r i s t i c of a f e r r i c species. Adsorption studies were interpreted as producing a ferrous species. The doublet due to t h i s species decreased i n i n t e n s i t y with increasing temperature but the f e r r i c doublet appeared impervious to either adsorption processes or temperature changes. They concluded that the f e r r i c ion was strongly bound to the z e o l i t e l a t t i c e but the ferrous ion was not. The asymmetry of the f e r r i c doublet was interpreted as a s p i n - l a t t i c e relaxation e f f e c t . Dehydration of the i n i t i a l - 13 -sample also produced a ferrous species. The following points seem to be c h a r a c t e r i s t i c of these samples: 1) F e r r i c exchange produces a Mossbauer doublet c h a r a c t e r i s t i c of a f e r r i c species, although i n some cases a ferrous doublet i s also observed. 2) Dehydration produces a ferrous species. 3) Adsorption processes appear to a f f e c t the ferrous species but not the f e r r i c species. The above r e s u l t s appear to agree f a i r l y well with the idea that the f e r r i c ions are l o c a l i z e d on the l a t t i c e and that the ferrous ions are not and i n fact appear to be quite mobile. The purpose of t h i s work was to examine a s i m i l a r system 17 (FeCl^ m L - z e o l i t e ) by Mossbauer Spectroscopy and to look at the adsorption isotherms for the same system. If mobile ions e x i s t i n the pores then comparison of the surface area with that of the host z e o l i t e should show a marked change. The r e a c t i v i t y of the surface toward d i f f e r e n t gases could be compared with the Mossbauer parameters. In t h i s way the influence of the oxidation state of iron on adsorbed gases could possibly be determined. If i n f a c t there i s more than one species of iron i n i t i a l l y i n the sample then attempting to convert the species by a reduction process into a single iron species and measuring adsorption isotherms would y i e l d valuable information about the state of aggregation within the pores. The combination - 14 -of Mossbauer Spectroscopy and adsorption techniques could then determine the oxidation state of the mobile species during the conversion process and whether the f i n a l product remained within the pores or on the external surface. - 15 -EXPERIMENTAL (i) M a t e r i a l s : l 7 Wedd and co-workers found that iron-exchange occurred r e a d i l y i n L - z e o l i t e when the PH of the exchange s o l u t i o n was greater than 2. The sample used for t h i s work was prepared by d i s s o l v i n g stock F e C l ^ ' G ^ O i n anhydrous e thy l ether i n proport ions of 33% weight to volume, c e n t r i f u g i n g and decanting the so lu t ion then in troduc ing approximately 10 grams of synthet ic Linde Molecular Sieve L obtained from Union Carbide of Canada. The s l u r r y was s t i r r e d for an hour before i t was f i l t e r e d and washed with ether to a negative KCNS t e s t . The product was again treated i n the same manner with a fresh so lu t ion of FeCl.3-6H.jO i n ether . The f i n a l product had undergone three such treatments before i t was d r i e d i n a i r at 110 C and used for subsequent experiments. Potent iometric i r o n ana lys i s was performed on the samples a f t er each solvent a d d i t i o n . The samples were d i s so lved i n concentrated hydroch lor i c ac id and stannous c h l o r i d e was added to reduce the f e r r i c ion to ferrous i o n . The t i t r a t i o n was c a r r i e d out with a eer ie sulphate so lu t ion prev ious ly standardized against f e r r i c ammonium sulphate. C e l l emf's were measured on a Beckman PH meter using platinum and saturated calomel e lec trodes . The ana lys i s ind ica ted that the quant i ty of i r o n increased from 7.5 to 9.5% by weight a f ter each solvent addi t ion and that fur ther treatment d id not appear to increase the 9.5% value . However, the i d e n t i c a l preparat ion of two such samples a f t er three solvent addi t ions d id not produce i d e n t i c a l i r o n - 16 -contents. The f i r s t sample contained 9.5% iron and the second 8.1% i r o n . Possibly the large quantity of ether required to wash the sample to a negative KCNS tes t was responsible for t h i s v a r i a t i o n . Mossbauer and adsorption studies were conducted on two prepared samples of i r o n - i n - L - z e o l i t e . The second preparation, on which the major portion of t h i s thesis i s based, was analysed by Dr. A. Bernhardt at 5251 Elbach uber Engelskirchen, West Germany. Table 1 l i s t s the r e s u l t s of the analyses performed by Dr. Bernhardt on L - z e o l i t e , solvated L - z e o l i t e (L-zeolite mixed only with anhydrous ethyl ether but prepared i n the above manner) and the second preparation of i r o n - i n - L - z e o l i t e (after three solvent additions). Adsorption experiments were conducted with L-grade nitrogen supplied by Canadian Liquid A i r Co. and research grade carbon dioxide and ethane supplied by Mathesoh Co. Research grade helium supplied by Matheson was used to determine the adsorption system volumes. ( i i ) X-Ray Powder Photographs: Powder photographs were taken with a P h i l i p s powder camera of 11.46 cm diameter. Samples were mounted i n 0.5 mm (inside diameter) glass c a p i l l a r i e s approximately 5 mm i n length. CuKa, X = 1.54050 A, X - i r r a d i a t i o n was used with a Ni f i l t e r to reduce K 0 r a d i a t i o n . The f i l m used was Kodak Medical X-Ray P (Estar safety base) cut into s t r i p s 3.5 X 35.5 cm for use i n the camera. To obtain photographs for l i n e p o s i t i o n measurements a pinhole collimator was used giving photographs with sharp, - 17 -Table 1 SAMPLE % S i A l Fe C l C 1. L - z e o l i t e (no treatment) 23.55 6.89 0.21 2. L - z e o l i t e (after three addi t ions of anhydrous e t h y l ether) 24.53 7.29 0.86 3. L - z e o l i t e (after three addi t ions of F e C l 3 . 6 H 2 0 i n anhydrous e thy l ether) prepar-a t ion #2 18.80 5.68 7.99 4.47 0.57 - 18 -low angle l i n e s . The l i n e s on the photographs were indexed on a l i g h t box provided with a meter s t i c k to which was attached a measuring s l i d e assembly containing a vernier and a magnified cross-hair for l o c a t i o n of the d i f f r a c t i o n l i n e . The d spacings fo r these l i n e s were then calculated. ( i i i ) Magnetic Measurements: Magnetic s u s c e p t i b i l i t y measurements on the samples of i r o n - i n - L - z e o l i t e were made by the Gouy method. The variable temperature s u s c e p t i b i l i t y determinations were made using an electromagnet and semi-microbalance over the temperature range o o 80 to 300 K and at three f i e l d strengths between 4500 and 8000 gauss. A pyrex gouy tube between 3 and 4 mm inside diameter was used and c a l i b r a t e d with g r a v i m e t r i c a l l y analysed HgCo(CNS)^. The s u s c e p t i b i l i t y values reported are the average of two Gouy tube sample packing determinations. Experimental diamagnetic corrections were made for the host z e o l i t e over the same temperature and f i e l d strength ranges. (iv) Adsorption: The amount of gas adsorbed by a surface i s measured as a-difference between two large numbers, the amount of gas o r i g i n a l l y i n jected into the adsorption c e l l under conditions where n e g l i g i b l e gas-surface i n t e r a c t i o n takes place and the amount of gas remaining i n the gas phase a f t e r e q u i l i b r a t i o n over the s o l i d sample when measureable i n t e r a c t i o n occurs. - 19 -In the design of any adsorption system therefore, i t i s necessary to have accurate system segment volumes, pressure measurement and temperature control i n order to minimize the error i n the volume of gas adsorbed. Adsorption studies are generally made using a volumetric 1 9 apparatus which measures the volume of gas uptake by an adsorbent. This apparatus incorporates a dosage system and pressure measuring system, an adsorption c e l l , temperature control u n i t and pumping f a c i l i t i e s capable of obtaining a high vacuum. The design of each of these components i s governed by the range over which temperature and pressure measurements are to be made, the desired p r e c i s i o n of the measurements and the c h a r a c t e r i s t i c s of the materials used as adsorbate and adsorbent. The volumetric apparatus of Figure 2 consists of: a gas p u r i f i c a t i o n t r a i n (A, B, C) i n which impurities could be removed by trap-to-trap d i s t i l l a t i o n ; a p a i r of bulbs (D) used for the storage of gases a f t e r d i s t i l l a t i o n ; and f i n a l l y the adsorption system enclosed within the dotted l i n e s of Figure 2 and described i n the following paragraphs. l 9 a) Dosage System Gas was introduced to the adsorption system d i r e c t l y from the external storage bulbs. Five mercury pipets of volumes 10', 20, 40, 80 and 160 m i l l i l i t r e s surrounded by an ice-mantle (E) were used to i n j e c t gas into the adsorption c e l l (G). The pipet volumes were known from repeated mercury weighings to an accuracy of .02%. These by d i f f e r e n t combinations give Index For Figure 2 A TO GAS LECTURE BOTTLES B DISTILLATION TRAPS C MERCURY MANOMETER D GAS STORAGE BULBS E ICE MANTLE AND PIPETS F IONIZATION GAUGE G ADSORPTION CELL H MERCURY MANOMETER I THREE-STAGE MERCURY DIFFUSION PUMP FIG 2 \ A B V c HIGH VACUUM i g = E LOW VACUUM ADSORPTION SYSTEM O - 21 -volumes of between 0 and 310 ml i n 10 ml increments. Owing to the geometry of the system, gas pressures above 450 t o r r could not be obtained. At times during a run a large f r a c t i o n of the gas was contained i n the p i p e t s . As the temperature c o n t r o l and measurement of an ice-mantle i s v i r t u a l l y free from experimental e r r o r and the p ipet volumes were known accurate ly , the error i n the estimate of the amount of gas i n the adsorption c e l l , was e s s e n t i a l l y independent of t h i s part of the system. b) Pressure Measurement The gas pressure i n the system was measured with a mercury manometer (H) connected i n ser ies with the a d s o r p t i o n . c e l l and p i p e t s , and placed as c lose to the c e l l as poss ib le i n order to minimize the dead space volume. The manometer was mounted i n a wooden platform against a d u l l white background which provided s u f f i c i e n t contrast, to give sharp menisci under condi t ions of normal room i l l u m i n a t i o n . Between the f ixed and free legs was mounted a s t a i n l e s s " S t a r r e t t - s t e e l " meter bar graduated i n 0.5 mm increments which was read with the s l i d i n g microscope of a cathetometer. For low pressure readings the cathetometer (which was securely anchored to the f l o o r to insure maximum r e p r o d u c i b i l i t y of readings) was used to obtain a more accurate estimate of the meter bar reading to ± 0 . 1 mm. Both the f ixed and free legs of the manometer were constructed of 14.6 mm (inside diameter) "Tru-Bor" tubing with the object of minimizing c a p i l l a r y depression e f fec t s and hence of obta in ing we l l -de f ined and r e l a t i v e l y f l a t menisc i . - 22 -c) Adsorption C e l l Figure 3 shows d e t a i l s of the adsorption c e l l . For ease of changing samples the c e l l was f i t t e d with a metal screw cap made of n i c k e l p la ted brass , seated on a Tef lon "0" r i n g . This was attached to the c e l l by means of a Kovar s e a l . The use of t e f l o n tape around the threads assured attainment of the des ired vacuum even a f ter repeated sample changes. One of the side arms was included as part of the c e l l i n order to measure the evacuated weight of the z e o l i t e and subsequently to determine the sample volume from i t s dens i ty . The other was inc luded to allow the passage of hydrogen over the sample when reduct ion experiments were conducted as described l a t e r . These s ide arms have the disadvantage of increas ing the c e l l overhead volume but i n terms of the t o t a l dead space volume of the system, the increase i s smal l . Only part of the c e l l was thermally c o n t r o l l e d during adsorpt ion measurements (the sect ion i n Figure 3 marked by the dotted l i n e ) . The remaining volume was included as part of the c e l l overhead volume and assumed to be at room temperature. The e r r o r i n the volume adsorbed created by t h i s assumption was found to be n e g l i g i b l y smal l . The volume of the c e l l was determined from repeated weighings with d i s t i l l e d water. The temperature at which the isotherms were measured are: O Q l i q u i d n i trogen (77 K ) , dry i ce s lush (195 K ) , chloroform o o s lush (210 K) and carbon t e t r a c h l o r i d e s lush (250 K ) . The temperature of these baths were measured with an iron-constantan thermocouple attached to the adsorption c e l l . Index For Figure A METAL CAP B KOVAR SEAL C CELL VOLUME FIG 3 23 -Under sample outgassing conditions (or when hydrogen was used to sweep the zeolite) a furnace was f i t t e d round the c e l l , and the outgassing temperature was maintained by a previously o c a l i b r a t e d rheostat to ±5 C. d) Associated Equipment The pumping equipment for the high vacuum l i n e consisted of a three-stage mercury d i f f u s i o n pump (I of Figure 2) backed by a Welch rotary o i l pump. This assembly was capable of obtaining a vacuum of 10 ^ t o r r . The low vacuum l i n e which served only to evacuate the mercury pipet reservoirs was also connected to a Welch rotary o i l pump. Beckman thermometers were used to measure room and ice mantle temperatures. Both were c a l i b r a t e d against a standard platinum resistance thermometer. Dead Space Determination and System Er r o r : The overhead volumes of the pipets and c e l l were measured gasometrically with helium at room temperature, and are given i n Table 2. Temperature fluctuations i n the slush baths and ice-mantle were n e g l i g i b l e . Room temperature v a r i a t i o n s were no greater o than 0.3 for any point (P,V) on the adsorption isotherm. The i n i t i a l gas dosage pressure P^ . was about 5 t o r r and depending upon the surface gas uptake the equilibrium pressure P„ could be as small as 0.1 t o r r . Since the error l i m i t s on E the pressure are ±0.1 t o r r the lowest pressure P £ can have an error as large as 100%. However t h i s only creates an error of 7% i n the corresponding value of the volume adsorbed. - 25 - l Table 2 Volume of System Segments P ipet Volume cc Sect ion Volume cc 1 1 5 9 . 1 7 ± . 0 0 5 c e l l 1 2 . 6 2 ± . 0 1 2 7 9 . 6 1 2 ± . 0 0 1 OH pipet 1 3 1 . 0 ± . 5 3 3 9 . 4 1 9 ± . 0 0 1 OH c e l l 2 2 . 3 ± . l 4 18.0031.001 5 9 . 4 0 2 ± . 0 0 1 Equations (1), (2) and (6) i n the Appendix show how the volume adsorbed i s c a l c u l a t e d . As an example, using Pj = 5 ± 0.1 t o r r P„ = 0.1 ± 0.1 t o r r E and the volumes l i s t e d i n Table 1 the error i n i s ±4% and i n n_ i s 100% but the error i n n ' i s only 7%. As the -E A J pressures increase so does t h e i r accuracy and consequently the e r r o r i n n,. becomes smal ler . A The error estimate on the volume adsorbed for the low pressure region i s therefore 7%; and t h i s decreases to approx-imately 2% i n the region of high pressure . Blank adsorption-isotherms (that i s with the c e l l empty) were determined with CG^ and C^R^ i n order to see whether any adsorption of these gases on the remainder of the system occurred. The values obtained for both gases d i f f e r e d by less than 2% from those obtained with hel ium. Hence no appreciable system adsorpt ion needed to be taken in to account. Over the range of temperatures and pressures studied gas-imperfect ion correc t ions were s u f f i c i e n t l y small to be ignored. The magnitude of thermal t r a n s p i r a t i o n correc t ions between the parts of the system at l i q u i d ni trogen and room temperatures were a lso n e g l i g i b l y small and hence ignored. (V) Mossbauer Apparatus: Two types of Mossbauer c e l l were used i n order to accommodate powder and p e l l e t samples. Powder samples were packed to a thickness less than 1mm i n a 1.25 cm diameter brass c e l l having mylar windows. For room temperature runs the c e l l was mounted on an aluminum holder d i r e c t l y i n the y~ray path . For l i q u i d n i trogen runs the c e l l was inser ted i n the top of a copper c o l d - f i n g e r which was immersed i n a dewar of l i q u i d n i t rogen . Styrofoam i n s u l a t i o n was placed around the sample to minimize heat o t rans fer and to obta in a temperature as c lose to 80 K as was f eas ib l e by t h i s geometry. The ni trogen l e v e l was maintained by a Superior A i r Products Company l i q u i d ni trogen l e v e l -c o n t r o l l e r . F igure 4 represents the c e l l used for p e l l e t i z e d samples. Previous c e l l designs cons is ted of a removeable head with rubber "0" r i n g and bery l l ium windows, ne i ther of which permitted the c e l l to hold a vacuum a f t er one or two cyc les between l i q u i d ni trogen and room temperature. In order to make comparisons between Mossbauer and adsorption experiments the Mossbauer c e l l had to maintain a vacuum between room and l i q u i d ni trogen temperatures through many temperature r e c y c l i n g s , and be capable of holding a given pressure of gas under the same temperature cond i t i ons . The B34 j o i n t was subst i tuted for the removeable head to allow sample i n s e r t i o n . The head was constructed from s o l i d n i c k e l - p l a t e d brass and the mylar windows were attached by an indium seal over which was a screw cap that d i s t r i b u t e d the seal evenly. Af ter more than for ty temperature r e c y c l i n g s no leaks were detected and during a given Mossbauer experiment the c e l l was capable of maintaining a high vacuum for several days. The side arms FIGURE 4 Diagram of the Mossbauer C e l l used for p e l l e t i z e d samples FIG 4 MYLAR WINDOW Ni PLATED BRASS HEAD KOVAR SEAL B 3 4 JOINT Pt GAUZE ZEOLITE PELLET SAMPLE HOLDER allowed evacuation and gas "sweeping" of the z e o l i t e . P e l l e t s were made from a Carver Laboratory Press Model B a t pressures around 15,000 p s i and were then shaped to f i t the p e l l e t holder of Figure 4. The p e l l e t remained i n the diagram p o s i t i o n for evacuat ion, gas sweeping and high-temperature outgass ing . For Mossbauer measurements the c e l l was inver ted . In t h i s p o s i t i o n the sample f i t s in to a s l o t i n the head, a l igned with the windows. This c e l l was subjected to three temperatures: room, s l u s h , and l i q u i d n i t rogen . For the low temperature runs the c e l l assembly was bol ted to a copper c o l d - f i n g e r and i n s u l a t e d with styrofoam. One ser i e s of Mossbauer experiments was conducted on an evacuated powder sample which was pos i t ioned i n a Janis model DT-6 "Varitemp" helium c r y o s t a t , f i t t e d with a Cryogenic Research Corp. model RS-1 temperature c o n t r o l u n i t . With o t h i s dev ice , any temperature between about 8 K and room temper-o ature could be maintained to wi th in ± 0 . 0 5 . C a l i b r a t e d germanium and platinum res i s tance thermometers were used to measure the sample temperature. The Mossbauer Spectrometer was of the constant acce l era t ion 20 type and has been described i n d e t a i l . The y -ray source 57 cons i s ted of 25 mC of Co i n a copper matr ix , suppl ied by Nfew England Nuclear Corp. Nat iona l Bureau of Standards c e r t i f i e d sodium n i t ropruss ide and i r o n f o i l were used for isomer s h i f t and v e l o c i t y scale c a l i b r a t i o n s , r e s p e c t i v e l y . In most cases the data were l eas t - squares f i t t e d to lorentzic\n l i n e shapes using a program - 30 -wri t ten by J . C . Scot t . However, i n several instances non-lorentziSln l i n e s were observed (owing to re laxa t ion ef fects ) ; i n these cases Mossbauer parameters were estimated v i s u a l l y from the p l o t t e d spectra . The errors i n the measured parameters are based on the standard dev ia t ion of the computer f i t and the r e p r o d u c i b i l i t y of the spectra . - 31 -RESULTS AND DISCUSSION; In an e a r l i e r Mossbauer i n v e s t i g a t i o n of iron-exchange i n 17 z e o l i t e L the presence of two f e r r i c species was reported . One of the species was presumed coordinated to the l a t t i c e oxygens while the other was thought to be F e C l ^ . As the 2 o observed spectra at l i q u i d n i trogen and room temperature cons is ted of a quadrupole doublet the f e r r i c n u c l e i had to be i n almost i d e n t i c a l environments. The present work was undertaken to f i n d out 1) whether two f e r r i c species coexisted i n the z e o l i t e and 2) the degree to which the z e o l i t e surface was occupied by the occluded i r o n . In preparing the present sample, two pre l iminary pieces of information were required before adsorption and Mossbauer studies could be conducted. The f i r s t was proof that the -zeol i te framework remained i n t a c t a f t er treatment with i r o n c h l o r i d e . The second requirement was a knowledge of the i r o n and c h l o r i n e content of the sample. The r e s u l t s of X-ray powder photography are given i n Table 3. The d-spacings are l i s t e d for the z e o l i t e p r i o r to any treatment, treated with i r o n ch lor ide and a subsequently treated sample which w i l l be discussed l a t e r . Comparison of 21. these numbers with those i n the l i t e r a t u r e for Zeo l i t e L shows that most of the d-spacings, l i e wi th in the range of the reported va lues . Any discrepancies or omissions may be accounted for by the fact that thje photographs were taken using a p in -ho le c o l l i m a t o r and the l i n e s were thus rather weak and sometimes i n d i s t i n c t . - 32 -Table 3 21 d-Spacings from X-Ray Powder Photographs Treated Bulk 34 L i t e r a t u r e L - z e o l i t e i r o n - L - z e o l i t e i r o n - L - z e o l i t e ot-Fe^O^ 16.1 ± . 3 7.52 ± .04 6.00 ± .02 4.57 ± .03 4.35 ± .04 3.91 ± .02 3.47 ± .02 3.28 ± .02 3.17 ± .01 3.07 ± .01 2.91 ± .01 2.65 ± .01 2.46 ± .01 2.42 ± .01 2.19 ± .01 15.8 7.49 5.98 4.62 4.35 3.93 3.67 3.48 3.23 3.18 3.08 2.92 2.66 2.63 2.48 2.42 2.20 15.7 7.43 5.98 5.90 4.62 4.35 3.93 3.64 3.48 3.18 3.06 2.92 2.66 2.64 15.7 7.37 5.90 7.96 2.20 4.52 4.33 3.86 3.62 3.44 3.25 3.14 3.04 2.88 2.65 2.60 2.48 2.17 5.86 5.43 4.77 - 33 -The chemical ana lys i s performed on the sample (Table 1 - #3) shows that both i r o n and ch lor ine are present i n the approximate mole r a t i o of 2:1, and that the s i l i c o n - t o -22 aluminum mole r a t i o of approximately 3:1 for z e o l i t e L i s preserved. Also l i s t e d i n Table 1 (#1 and #2) are the carbon contents for untreated and solvated z e o l i t e L . The very small percent increase i n the solvated sample i s not i n d i c a t i v e of any appreciable e ther-surface complex formation. These r e s u l t s ind i ca te that both i r o n and ch lor ine entered the z e o l i t e without d i s r u p t i n g the l a t t i c e . The Mossbauer spectrum of a powdered sample of i r o n - L -z e o l i t e i n a i r showed a quadrupole doublet . The isomer s h i f t was c h a r a c t e r i s t i c of high spin f e r r i c compounds 3+ 2+ ( t y p i c a l values for Fe are 6 =0.6 mm/sec, whereas for Fe o 6 -1.2 mm/sec and Fe 6 =0.26 mm/sec). At l i q u i d ni trogen temperatures however, a d d i t i o n a l very broad l i n e s i n d i c a t i v e of Zeeman s p l i t t i n g were observed. The magnitude of the i n t e r n a l magnetic f i e l d (approximately 400 kOe) was considerably less than the f i e l d strengths normally observed for f i ve unpaired e l ec trons . (The t h e o r e t i c a l c o n t r i b u t i o n to the f i e l d i s about 110 kOe per unpaired e l e c t r o n , but due to the i n t e r a c t i o n s between an ion and i t s surroundings t h i s value i s usua l ly s l i g h t l y quenched. T y p i c a l observed f i e l d s are approximately 510 ± 20 kOe). The presence of a doublet at room temperature and a hyperfine pat tern superimposed on the doublet at l i q u i d ni trogen temperatures seemed to be t y p i c a l behaviour a t t r i -butable to re laxa t ion e f f e c t s . The very broad nature of - 34 -the outer l i n e s and the extremely low value of the i n t e r n a l f i e l d strength suggested that at lower temperatures the l i n e s may become better resolved as the i n t e r n a l f i e l d increased ' towards i t s saturation value. Figure 5 shows the behaviour o o of the sample between 295 K and 9 K and Table 4 l i s t s the Mossbauer parameters extracted from these spectra, o o The 80 and 9 K parameters were obtained from the "Varitemp" helium cryostat. The remaining temperature runs were made using the Mossbauer c e l l and the appropriate coolant as described e a r l i e r . Figure 5 does not include the o spectra corresponding to the 80 K parameter i n Table 4. However the difference between t h i s spectrum and Figure 5c) was quite apparent. While the doublet s t i l l remained, the outer l i n e s were more intense and better resolved. The l i q u i d nitrogen temperature parameter for the powder and p e l l e t samples are recorded i n the table as approximately o o o 105 K. Several spectra between 80 and 120 K were obtained at known temperatures i n the helium cryostat. The approximate o o temperature range i n which Figure 5c) f i t was 100 - 110 K o o and hence the average value of 105 ± 5 K was chosen. Also included i n Table 4 i s a comparison of the Mossbauer parameters between p e l l e t i z e d and powdered samples. As can be seen from t h e i r isomer s h i f t s and quadrupole s p l i t t i n g s i n a i r the two sample preparations have i d e n t i c a l values. Thus the same trends are expected i n both over the temperature range studied. (Figure 7a) to be discussed l a t e r represents o a powder sample i n a i r at approximately 105 K which could be compared with Figure 5c)) . FIGURE 5 Mossbauer Spectra of o a) p e l l e t i n a vacuum at 295 K o b) p e l l e t i n a vacuum at 195 K o c) p e l l e t i n a vacuum at ^105 K o d) powder i n a vacuum at 9 K J>OPPL£(Z S/ELOCITY fMM SecT'j --8 -6 -4 -2 6 +2 +4 +6 +8 * Table 4 Sample o Temp ( K) <5 a 1) p e l l e t i n a i r r 295 0. 64 + .02 no treatment IM.05 0. 75 + .04 i r 295 0. 63 + .02 i n \ 195 0. 6 + .1 vacuum U l 0 5 0. 69 + .02 2) powder i n vacuum 80 0. 60 + .04 9 0. 65 + .04 3) powder i n a i r a) no treatment 295 0. 63 + .02 •^ 105 0. 76 + .04 b) ca l c ined at 573°K i n a i r 295 0. 63 + .02 c) ca lc ined at 773°K i n a i r 295 0. 73 + .02 ^105 0. 62 + .05 4) C B u l k a -Fe 9 0, 298 0. 62 H 0.73 ± .02 0.96 ± .02 0.79 ± .02 0.75 ± .02 1.2 ± .05 0.68 ± .02 1.1 ± .1 0.59 ± .02 0.74 ± .02 1.2 ± .1 402 ± 25 402 ± 25 0^ 430 ± 20 467 ± 20 0.76 ± .02 1.00 ± .05 0.60 ± .02 1.6 ± .1 402 ± 25 0.98 ± .02 514 ± 20 501 ± 30 516 Continued/ * Table 4 Continued Sample Temp (°K) 6 a A a r a H b 5) Powder washed 295 0.64 ± .03 0.78 ± .02 0.52 ± .04 with hot water *105 0.70 ± .02 1.1 ± .1 1.5 ± .2 d a i n mm/sec r e l a t i v e to sodium n i t ropruss ide • D i n kOe ^ i c Reference #6 outer l i n e s not resolved * o data was obtained from simple p lo t s of spectra at ^105 K - 38 -The most apparent features of the spectra of Figure 5c) are the following: 1) The Zeeman s p l i t t i n g observed at 105°K completely collapses o o at 195 K. 2) The broad outer l i n e s at 105 K are much better resolved at 9°K where the si z e of the i n t e r n a l magnetic f i e l d has obviously increased (represented by an increase i n the s p l i t t i n g between l i n e s one and s i x ) . 3) The central doublet o has v i r t u a l l y vanished at 9 K. 4) The s i x - l i n e pattern at o 105 K i s superimposed on a parabolic base-line and the l i n e -al shapes are non-lorentzidn (this i s also true of the 80 K spectrum). Since one of the probable coordinations of a f e r r i c ion i n t h i s z e o l i t e i s to l a t t i c e oxygens, the above evidence pointed to small p a r t i c l e iron oxide behaviour. Constabaris, 2 3 2 4 Kundig and co-workers ' have examined by Mossbauer spec-troscopy the properties of small p a r t i c l e s of a-Fe^^ supported i n s i l i c a g e l , as a function of temperature and p a r t i c l e s i z e . Their r e s u l t s are now discussed as they apply to the present work. F i r s t , for a given average p a r t i c l e size of a-Fe2C>2 in the support, the temperature dependence shows a doublet o at 295 K and a s i x - l i n e pattern superimposed on the doublet o at 80 K. But the l i n e shapes of the hyperfine pattern are lorentzidn and the observed s p l i t t i n g s between l i n e s one and o o six d i f f e r very l i t t l e between 80 and 12 K. Second, the i n t e n s i t y r a t i o s of the magnetic hyperfine l i n e s are always approximately the same (1:2:3); and are c h a r a c t e r i s t i c of the t r a n s i t i o n p r o b a b i l i t i e s i n the magnetic spectrum of - 3 9 -bulk c* - F e 2 0 3 . Small p a r t i c l e s of a-Fe20 3 e x h i b i t superparamagnetism. The e f f ec t i s observed by Mossbauer Spectroscopy as the co l lapse of a s i x - l i n e spectrum into a doublet . The t r a n s i t i o n occurs at a s p e c i f i c temperature for a given p a r t i c l e s i ze and represents the d i f f erence between magnetic order ( a n t i -ferromagnetic behaviour i n a-Fe20 3 ) and magnetic d i sorder (paramagnetic behaviour) r e s p e c t i v e l y . The s ize of the i n t e r n a l magnetic f i e l d generated by magnetic ordering for a given p a r t i c l e s i ze d i f f e r s very l i t t l e from that of bulk a-Fe 2 0. j . As these e f fec t s are the r e s u l t of p a r t i c l e r e laxa t ion times, the above behaviour i s s p e c i f i c for a given p a r t i c l e s i ze with a given r e l a x a t i o n time. However where a d i s t r i b u t i o n of r e l a x a t i o n times i s exh ib i ted i n a sample a) the t r a n s i t i o n temperature between magnetic order and d i sorder v a r i e s over a range of temperatures b) the l i n e shapes of the hyperfine pattern are usua l ly non-l o r e n t z i d n and c) the magnitude of the i n t e r n a l magnetic f i e l d v a r i e s over a wide range of temperatures. The behaviour pattern of the spectra i n Figure 5 i s s u f f i c i e n t l y d i f f e r e n t from the superparamagnetic behaviour of an average p a r t i c l e s ize of a-Fe2C>2 that the l a t t e r can be ru l ed out on the Mossbauer evidence alone. However, as spin r e l a x a t i o n e f fec t s occur i n simple paramagets as we l l as magnet ical ly ordered paramagnets, measurement of the temperature and magnetic f i e l d dependence of the molar s u s c e p t i b i l i t y should d i s t i n g u i s h between the two. Figure 6 shows the molar s u s c e p t i b i l i t y of i r o n i n FIGURE 6 Magnetic Measurements on i r o n - L - z e o l i t e a) Molar s u s c e p t i b i l i t y of i r o n versus temperature b) E f f e c t i v e magnetic moment versus temperature - 41 -z e o l i t e L as a function of temperature and f i e l d strength. No dependence on an externally applied magnetic f i e l d was observed, as the s u s c e p t i b i l i t i e s at three applied f i e l d s a l l l i e on the same temperature curve. Hence at lea s t down o to 80 K the sample behaves l i k e a simple relaxing paramagnet. The v a r i a t i o n of the e f f e c t i v e magnetic moment ( v e f f ) i s also plotted as a function of temperature i n Figure 6. The s u r p r i s i n g l y low moments and t h e i r v a r i a t i o n with o ° temperature (3.4 - 2.9 B.M. between 300 and 80 K) i s not the t y p i c a l behaviour of a simple paramagnetic f e r r i c compound. However low magnetic moments (2.3 - 5.9 B.M.) i n f e r r i c 2 5 compounds have been attributed to 1) spin exchange i n t e r -actions between ion-ion and ion-ligand and 2) the presence of magnetically non-equivalent s i t e s i n a l a t t i c e . In the l a t t e r o e f f e c t Curie-Weiss behaviour i s s t i l l observed. Below 160 K Curie-Weiss behaviour was also exhibited by the present sample. Eit h e r of these e f f e c t s , however could account for the present r e s u l t s . Although the sample appears to behave l i k e a simple 0 paramagnet down to 80 K, f i e l d dependence studies at temperatures lower than t h i s are required i n order to c a t e g o r i c a l l y state that the f e r r i c species of Figure 5 i s a simple paramagnet. Two further pieces of information can be obtained from Table 4 and Figure 5. The disappearance of the parabolic base-line and appearance of a lorentzidln s i x - l i n e spectrum at 9°K are d e f i n i t e indications of magnetic ordering with long s p i n - l a t t i c e relaxation times. The inner doublet does n t ap ear to have completely vanished wh ch could b the - 42 -reason for the s l i g h t l y broadened nature of the l i n e s i n the spectrum. However, the trend that i s observed i n the d i s -appearance of the c e n t r a l doublet and an approach toward a maximum i n t e r n a l f i e l d strength with decreasing temperature ( i f indeed the maximum has not already been reached) i s i n d i c a t i v e of a s ing le f e r r i c species wi th in the sample. This does not preclude the p o s s i b i l i t y of more than one f e r r i c s i t e wi th in the sample whose geometries may be s u f f i c i e n t l y s i m i l a r that Mossbauer r e s o l u t i o n would be imposs ible . The range of temperature over which the r e l a x a t i o n o e f f ec t s occur (approximately 40 ) s trongly suggest that d i f f e r e n t parts of the system re lax at d i f f e r e n t ra te s . C o r r e l a t i o n s have been made between p a r t i c l e s i ze and re laxa t ion 26 time ; and i t i s known that as the p a r t i c l e s i ze i s decreased the r e l a x a t i o n time becomes shorter . Hence the sample may contain a range of p a r t i c l e s i z e s . A l t e r n a t i v e l y t h i s may simply be a funct ion of the distance between i r o n centres . As w i l l be discussed l a t e r , the measured decrease i n surface area between solvated L - z e o l i t e and i r o n - L - z e o l i t e shows d e f i n i t e l y that the pores are blocked by the f e r r i c species . T h i s , coupled with the very low magnetic moments obtained from magnetic measurements tends to r u l e out p a r t i c l e s ize v a r i a t i o n . The presence of ch lor ide ion from chemical ana lys i s could imply that f e r r i c c h l o r i d e was i n the sample. In order to show that the observed Mossbauer spectra were not the r e s u l t of r e l a x a t i o n e f fec t s i n f e r r i c ch lor ide an a d d i t i o n a l experiment was performed. A sample was washed with hot water - 43 -and the f i l t r a t e was tested for ch lor ide and f e r r i c ions . A very p o s i t i v e tes t with s i l v e r n i t r a t e resu l ted for the c h l o r i d e ion but with KCNS only trace amounts of the f e r r i c ion could be detected. Washing continued u n t i l no further evidence for e i t h e r ion was evident . A comparison of the room and l i q u i d ni trogen temperature Mossbauer parameters for the washed sample with those of the o r i g i n a l untreated sample (Table 4) show that there i s no change i n the p a r a -meters. Neither i s there any appreciable a l t e r a t i o n of the Mossbauer spectrum. The very p o s i t i v e nature of the ch lor ide ion tes t ind ica ted that a large percentage of i t had been removed from the z e o l i t e . The almost negative t e s t for f e r r i c ion ind ica ted that f e r r i c c h l o r i d e was not being washed out as a species . Also i f f e r r i c c h l o r i d e ex i s ted i n the sample i n the amount suspected by chemical a n a l y s i s , washing i t out would have markedly changed the Mossbauer per cent e f f e c t . This also d i d not happen. However, f e r r i c c h l o r i d e may be present as an impurity but as a s u f f i c i e n t l y small percentage of the o v e r a l l f e r r i c content that i t i s not observeable by Mossbauer Spectroscopy. The bulk of the ch lor ide ion may be present as a sodium or potassium c h l o r i d e . Having e l iminated f ine p a r t i c l e a-Fe20^ and poss ib ly FeCl^ as sources of the Mossbauer spectra and cons ider ing the evidence for a s ing le f e r r i c species occluded i n the z e o l i t e , one other p o s s i b i l i t y cons is tent with f e r r i c coordinat ion to l a t t i c e oxygens would be an oxy-hydroxide of i r o n . The bulk behaviour of d i f f e r e n t types of iron-oxyhydroxides - 44 -have been studied by many authors . Dezsi and co-workers have looked i n p a r t i c u l a r at the bulk behaviour of p -FeOOH. Relaxat ion e f fec t s in/?-FeOOH were followed by O o Mossbauer Spectroscopy from 120 K to 295 K. The r e l a x i n g s i x - l i n e spectra showed a parabo l i c b a s e - l i n e . The outer l i n e s were very broad and there was no d e f i n i t e r a t i o of the outer to inner l i n e i n t e n s i t i e s . The Neel po int (the temper-ature at which the t r a n s i t i o n occurs from paramagnetic to o ordered magnetic behaviour) was found to be 295 K and the sa tura t ion magnetic f i e l d at zero temperature was 475 kOe. Of fur ther i n t e r e s t were the spectra obtained on the sample at d i f f e r e n t temperatures of c a l c i n a t i o n i n a i r . The Mossbauer o Spectra consis ted of a high spin f e r r i c doublet between 295 o and 670 K. 7Above t h i s temperature a t r a n s i t i o n occurred and the s i x - l i n e hyperfine pat tern o f - F e 2 0 3 developed. S i m i l a r behaviour i s observed i n 6-FeOOH Table 4 and Figure 7 show the r e s u l t s of c a l c i n i n g the o i r o n - L - z e o l i t e sample i n a i r . Af ter heating i n a i r at 573 K the room temperature spectrum i s a doublet and the l i q u i d n i trogen temperature spectrum i s a non- lorentz ian s i x - l i n e p a t t e r n . The quadrupole s p l i t t i n g of the room temperature doublet has increased from that of the untreated sample, which i s cons is tent with the removal of water molecules from the f e r r i c coordinat ion sphere. F i n a l l y a f ter heating at o 773 K the spectrum of Figure 7b) i s obtained. For comparison the parameters of bulk <*-Fe 20 3 are l i s t e d i n the t a b l e . Although t h i s i s by no means conclusive evidence for a FIGURE 7 Room t e m p e r a t u r e Mossbauer s p e c t r a o f powder c a l c i n e d i n a i r a) u n t r e a t e d powder sample o b) powder sample c a l c i n e d a t 773 K FIG 7 o u 4 5 -1001 98 96 * ° = B a o 4 "V a _ e 3 a 1B# LU 1001 981 964 3 0 -0 9 S a 3 " S 3 a B B O a o b B # -a a 3 * a # ' n o a^ c^ p 3 » 3 % (% B Q C B O ° a 8 a a a a a a s o a * oa a 1.B° a . * B O B O , . aa j _ a a a , a a 1 9 a " a a a a 10 -8 -6 -4 -2 0 +2 +4 +6 +8 +10 DOPPLER VELOCITY [MM Sec'] - 46 -change between oxy-hydroxide and ot-Fe^O^ i n the present sample i t i s c e r t a i n l y not negative evidence. Recent evidence has been reported for the presence of small p a r t i c l e a-FeOOH 3 1 i n n a t r o l i t e which goes to a-Fe202 when c a l c i n e d i n a i r The room temperature spectrum of Figure 7b) i s a lso a s i x - l i n e hyperf ine p a t t e r n . The evidence reported then for t h i s sample of i r o n - L -z e o l i t e suggests that only one f e r r i c species i s present and that t h i s species may occupy more than one type of geometry wi th in the z e o l i t e l a t t i c e . The sample exh ib i t s simple O paramagnetic behaviour down to 80 K (magnetic measurements) and with only the Mossbauer evidence t h i s may continue down 0 to 9 K. I f lower temperatures were studied by Mossbauer Spectroscopy extrapolat ions could be made to determine the o magnitude of the i n t e r n a l magnetic f i e l d at 0 K. This would not neces sar i l y i d e n t i f y the f e r r i c ion bonding (there are many compounds of i r o n whose i n t e r n a l magnetic f i e l d s are very s imi lar ) but would be conclus ive evidence for the presence of only one f e r r i c species . I f magnetic measurements were to be conducted at lower temperatures the state of magnetic order or d i sorder could o o be shown between 8 0 and 4 K. I f the r e l a x a t i o n e f fec t s observed are the r e s u l t of varying distances between i r o n n u c l e i i such an experiment would c e r t a i n l y support i t . Lowering of the bulk property Neel temperature i s known to 6 accompany small p a r t i c l e behaviour . This type of evidence would lend i t s e l f to i d e n t i f y i n g the f e r r i c spec ies . However, i n the absence of any more d e f i n i t i v e informat ion , - 47 -the f e r r i c species within the zeolite appears to behave like an iron oxyhydroxide. The presence of hydroxyl groups on the lat t i c e framework of zeolites i s well known and certainly is consistent with f e r r i c oxyhydroxide formation. If the internal o o f i e l d at 9 K (468 kOe) is a good approximation to the 0 K f i e l d then small particle 6-FeOOH may be the iron species present here. Mossbauer Spectra were used to determine the effects on the f e r r i c ion created by the following sample treatments 1) room temperature outgassing and subsequent adsorption of nitrogen, carbon dioxide and ethane 2) outgassing in a o vacuum at 573 K followed by adsorption of the same gases and 3) passing hydrogen over the sample at different temperatures. 1) Table 5 l i s t s the Mossbauer parameters for the room temperature evacuation of the sample and subsequent adsorption of gases. The differences in the Mossbauer parameters between the pellet in air and the room temperature evacuated pellet l i e primarily within the experimental errors in the data. Hence the evacuation process has not noticeably affected the coordination about the fer r i c ion. o The differences between the room temperature and 105 K parameters are registered as an increase in the magnitude of the quadrupole spli t t i n g and a broadening of the line-width of the central doublet. Hence the EFG at the ferric ion is o ° smaller at 295 than at 105 K. The cause could be a result of the thermal energy available at room temperature. As the temperature i s lowered inhomogeneities in the environment Treatment of p e l l e t evacuated 440 t o r r N 2 reevacuated 44 0 t o r r C0 2 reevacuated 440 t o r r C~H Temp ( K) 6 295 0.63 195 0.6 ^105 0.69 295 0.66 ^105 0.6 295 0.6 ^105 0.66 295 0.6 195 0.71 295 0.64 ^105 0.70 295 0.64 195 0.69 Table 5* A a r a H b 02 0.79 + .02 0.59 + .02 1 0.75 + .02 0.74 + .02 02 1.2 + .05 1.2 + .1 402 ± 25 02 0.75 + .02 0.50 + .02 -05 1.1 + .05 1.2 + .1 c 05 0.73 + .02 0.7 + .05 -04 1.0 + .05 1.0 + .1 c 05 0.76 + .02 0.68 + .02 -02 0.76 + .02 0.61 + .04 -04 0.71 + .04 0.45 + .05 -02 0.86 + .02 0.85 + .05 c 02 0.75 + .02 0.52 + .02 -02 0.76 + .02 0.55 + .03 — Continued/ Table 5* Continued Treatment of p e l l e t Temp (°K) 6 a A a r a H b reevacuated 295 0.55 ± .05 0.66 ± .03 0.55 ± .05 -vl05 0.60 ± .05 0.90 ± .02 1.1 ± .05 c ! O a l l data from ^105 K spectra were taken from simple p l o t s i i n mm/sec r e l a t i v e to sodium n i t ropruss ide i n kOe outer l i n e s very weak or not resolved - 50 -about the f e r r i c ion would become more apparent as the system was "frozen down" and hence the EFG would increase. Line broadening i s attributed to the changes i n relaxation processes with temperature. The gas pressures admitted to the Mossbauer c e l l were chosen to correspond to the maximum gas pressure studied on the adsorption isotherms (to be discussed l a t e r ) . After each gas was adsorbed and a Mossbauer spectrum taken the gas was removed and a spectrum was taken again to insure that the i n i t i a l Mossbauer parameters remained the same. As can be seen from Table 5, the Mossbauer parameters for the f e r r i c ion are not s i g n i f i c a n t l y changed by the presence of any of the gases and hence these gases are not d i r e c t l y involved with the coordination sphere of the i r o n . Nor was there observed any change i n the parameters with d i f f e r e n t gas pressures over the p e l l e t . Figure 8 shows the e f f e c t on the hyperfine l i n e s of the Mossbauer spectra a f t e r gas adsorption and desorption. As i s apparent from a) to c) the outer l i n e s get weaker and are eventually unresolved. However the parabolic background s t i l l remains. A study of the e f f e c t of the degree of hydration on the hyperfine structure of f e r r i c ions on a sulfonate type r e s i n 32 has been conducted by Goldanskii et a l . Their findings showed that at successive stages of dehydration the magnetic hyperfine l i n e s became weaker and eventually unresolved as the dehydration was completed. These r e s u l t s were explained by s p i n - l a t t i c e relaxation e f f e c t s . FIGURE 8 o Mossbauer spectra of p e l l e t at ^105 K showing change i n hyperfine pat tern with gas treatment a) p e l l e t i n a i r b) p e l l e t i n 440 t o r r N 2 c) p e l l e t i n vacuum RELATIVE COUNT RATE 5 8 o CO Y 99 a 98 97 1 0 1001 99 98 971 8 -6 -4 -2 0 +2 +4 +6 +8 DOPPLER VELOCITY [MM Sec"'] - 62 -spectrum, some of the central doublet i s s t i l l suspected to be present. o In order to get from the Fe spectrum of Figure 10b) to a spectrum c h a r a c t e r i s t i c of bulk a-Fe20^ i t was found necessary o to c a l c i n e the p e l l e t i n a i r . Outgassing the Fe sample at o 773 K i n a vacuum did not a l t e r the spectrum of 10b) at a l l . Presumably the mechanism for the conversion requires surface water. This most probably would not be available because the sample had been heated to a higher temperature and most of the chemisorbed water driven o f f . The spectrum of 10c) was obtained a f t e r c a l c i n i n g the 0 o sample at 773 K i n a i r and outgassing i t at 573 K. The Mossbauer parameters for the ce n t r a l doublet and the hyperfine component are l i s t e d i n Table 7. The parameters of the doublet are quite s i m i l a r to those att r i b u t e d to the doublets of the other spectra. From the hyperfine parameters, bulk a-Fe20.j appears to.be the predominant f e r r i c species. The outer l i n e s do not appear to f i t the points properly and the p o s s i b i l i t y of the superposition of a second hyperfine pattern s t i l l e x i s t s . Passing hydrogen over the sample at elevated temperatures causes the reduction of the f e r r i c species to Fe°. The mechanism o for the conversion of Fe to bulk a.-Fe2®2 seems reasonably c l e a r . Perhaps the mechanism involved i n the conversion of the "iro n " species i n Figure 9b) to the hyperfine spectrum of 9c) also depends on the a v a i l a b i l i t y of surface water. 2+ It appears that at intermediate temperatures Fe i s produced though t h i s cannot be said with any degree of assurance about 2+ Figure 9b). The role played by Fe i s not altogether c l e a r . - 63 -Since the system i n i t i a l l y exhib i ted a v a r i e t y of re laxa t ion times cons i s tent with the i n t e r p r e t a t i o n of a magnet ical ly d i l u t e system, the f e r r i c centres affected f i r s t would be those with the longest bond distances and the weakest i n t e r a c t i o n s . The f e r r i c species l e f t contr ibut ing to the c e n t r a l doublet would probably have shorter re laxa t ion times and therefore not o n e c e s s a r i l y contr ibute to the hyperfine pat tern at ^105 K (which was the observed e f f e c t ) . The high temperature required to lessen the i n t e n s i t y of the c e n t r a l doublet would be cons is tent with the increas ing energy of i n t e r a c t i o n between f e r r i c centres . 2+ The Fe species formed would probably r e a d i l y ox id ize to produce the observed magnetic hyperfine pattern at room temperature. A great deal more work needs to be done i n t h i s area i n order to determine exact ly the processes that are involved i n the conversion of the i n i t i a l f e r r i c species to bulk a-Fe20 3 . In order to insure that the temperature treatment involved i n the conversion of the f e r r i c ion to bulk a-Fe20 3 d i d not d i s r u p t the z e o l i t e l a t t i c e an X-ray powder photograph was taken ( l i s t e d i n column 4 of Table 3) . Comparison with the other X-ray data shows that some s t r u c t u r a l rearrangements may have occurred but that the d i f ferences are not so great as to imply the t o t a l co l lapse of the z e o l i t e framework. Column 5 i n Table 3 l i s t s the d-spacings for bulk a-Fe2C>2-The appearance of the 5.90 l i n e may or may not ind ica te the presence of a-Fe20.j. A s ing le magnetic measurement was made by the Gouy method on the reduced and outgassed sample of Figure 10c). The t y p i c a l -3 range for the value of the gram s u s c e p t i b i l i t y was x ^ (6-+10) X 10 - 6 4 -This number i s at l e a s t an order of magnitude greater than the - 4 o r i g i n a l x (^ 1 X 10 ) and since hysteresis e f f e c t s were also 9 observed i t was concluded that the sample showed magnetically ordered behaviour. There was no change i n the Mossbauer parameters of Figure 10c) on nitrogen adsorption. The nitrogen isotherm of Figure l i b ) was obtained on a o powder sample of i r o n - L - z e o l i t e that had been outgassed at 573 K. The difference between t h i s isotherm and that of Figure 11a) o (the nitrogen isotherm of the 573 K outgassed solvated-L-zeolite) i s quite remarkable. From Table 8 i t can be seen that the 2 2 surface area drops from 375 m /gm to 98 m /gm. This i s an extremely good i n d i c a t i o n that the iron species occluded within z e o l i t e - L i s blocking the pores. Afte r adsorption isotherms of C0 2 and C2Hg were measured; t h i s sample was subjected to the passage of hydrogen over i t at various temperatures, c a l c i n i n g i n a i r and f i n a l l y outgassing o i n a vacuum at 573 K. The nitrogen isotherm of Figure 11c) was obtained. The surface area a v a i l a b l e for nitrogen adsorption has increased twofold over the i n i t i a l area of i r o n - L - z e o l i t e . C l e a r l y then t h i s i s a good i n d i c a t i o n that the bulk of the ct-Fe20.j that i s formed resides on the external surface of the z e o l i t e . The question s t i l l remains as to how i t gets there! o o One possible explanation i s that at 943 K the Fe formed may be s u f f i c i e n t l y mobile that i t d i f f u s e s out of the pores and 1 7 forms aggregates on the external surface This powder sample was examined by Mossbauer spectroscopy and a spectrum.whose parameters were s i m i l a r to those of Figure FIGURE 11 Nitrogen Adsorption isotherms at 77 K 8% i r o n content a) on outgassed s o l v a t e d - L - z e o l i t e b) on outgassed i r o n - L - z e o l i t e c) on outgassed i r o n - L - z e o l i t e a f t er producing bulk a -Fe-O. , FIGURE. 11 - 66 -10c) was obtained. Figure 12 represents ni trogen adsorption isotherms on a d i f f e r e n t powder sample of i r o n - L - z e o l i t e . (In t h i s sample chemical ana lys i s ind ica ted that there was approximately a 9.5 percent i r o n content) . The treatment of the sample was: o a) room temperature evacuation b) outgassed at 573 K o c) swept with hydrogen at 573 K and room temperature evacuated o and d) outgassed at 573 K. Table 8 l i s t s the surface areas obtained from BET p l o t s . The surface area d i f ferences between b) and c) are wi th in experimental e rror and hence v i r t u a l l y no change has occurred with treatment. Isotherm d) however does show a small increase i n surface area . For comparison purposes a Mossbauer spectrum was recorded for a sample of d ) . The spectrum obtained was i d e n t i c a l to that of Figure 9c) . The trend toward increased surface area as bulk ct-Fe202 was produced i s a good i n d i c a t i o n that the pores are becoming less congested. However the question s t i l l ar i se s as to the nature of the mobile species that permits d i f f u s i o n out of the pores. Figure 13 represents the adsorption isotherms of CG^ and C 2Hg on both s o l v a t e d - z e o l i t e L and on i r o n - L - z e o l i t e . Compari-son of the surface areas obtained with the same adsorbate between the two z e o l i t e s can be made from Table 8. In both cases the uptake of ethane i s much smaller than that of e i t h e r CC>2 or N 2 « This demonstrates the s i ev ing a b i l i t y of the z e o l i t e s i n that smaller molecules l i k e CC^ and have access to parts of the z e o l i t e which are precluded to bigger molecules. F I G U R E 12 Nitrogen Adsorption isotherms o on i r o n - L - z e o l i t e at 77 K 9.5% i r o n content , a) sample evacuated o b) sample outgassed at 573 K c) sample swept with hydrogen o at 573 K then evacuated o d) sample outgassed at 573 K FIGURE 13 Adsorption isotherms at 195°K: a) C 0 2 on outgassed s o l v a t e d - L - z e o l i t e b) C 2 Hg on outgassed s o l v a t e d - L - z e o l i t e c) CC>2 on outgassed i r o n - L - z e o l i t e c) C 9 H f i on outgassed i r o n - L - z e o l i t e - 69 -The isotherms of C^Hg and N 2 demonstrated purely p h y s i c a l adsorpt ion as can be seen by the magnitudes of the BET energies l i s t e d i n Table 8. This substant iates the Mossbauer evidence. Carbon d iox ide was p h y s i c a l l y adsorbed on solvated L - z e o l i t e but on i r o n - L - z e o l i t e the isotherms behaved qui te unusual ly . As the isotherm temperature was increased the adsorption appeared to be i n c r e a s i n g . However the gas desorbed qui te r e a d i l y . Since the Mossbauer Spectra showed no evidence of C 0 2 a l t e r i n g the coordinat ion sphere about the f e r r i c ion i t appeared u n l i k e l y that chemisorption to i r o n occurred. Further i n v e s t i g a t i o n of t h i s p e c u l i a r i t y i s r equ ired . Since the energies obtained by BET p l o t s are not r e a l l y meaningful , i s o s t e r i c heats for ethane on solvated L - z e o l i t e and i r o n - L - z e o l i t e are shown i n Figure 14. The i n i t i a l drop i n the heat with coverage i s qui te usual because i n the low coverage region s i t e s with s t e a d i l y decreasing energy are f i l l e d . The appearance of an increase and maximum i n the energy on both z e o l i t e s at around monolayer coverage are probably represen-t a t i v e of l a t e r a l i n t e r a c t i o n s between adsorbate molecules occurr ing wi th in the pores . This same trend was observed for carbon d ioxide and ethylene on solvated z e o l i t e - L and has 33 been observed before by Barrer and co-workers In conc lus ion , the present sample contains a s ing le f e r r i c species as an oxyhydroxide of i r o n . The species behaves l i k e o a simple r e l a x i n g paramagnet down to 80 K. Adsorption studies on the untreated sample show that the i r o n oxyhydroxide blocks the pores of the z e o l i t e . Temperature treatment of the sample e i t h e r i n a i r or i n hydrogen frees the pores and produces - 70 -Table 8* Sample and Treatment Gas Temp ( K) Area — gm „ k c a l Energy —*— ^ J mol 1) outgassed solvated L - z e o l i t e N 2 co 2 C 2 H 6 77 195 195 375 382 248 2.3 8.1 5.5 2) i r o n - L - z e o l i t e N 2 co 2 C 2 H 6 77 195 195 98 116 63 2.1 7.0 5.2 3) i r o n - L - z e o l i t e a f t e r producing bulk a - F e 2 0 3 77 220 2.3 4) i r o n - L - z e o l i t e a) evacuated b) outgassed c) H 2 passed over at 573°K then evacuated d) outgassed N, N, 77 77 77 77 75 86 88 110 2.0 2.2 2.1 2.1 * o outgassing temperatures were 573 K a sample contains 8% i r o n b sample contains 9.5% i r o n experimental e rror l i m i t s are ± 5% FIGURE 14 I s o s t e r i c heats of adsorption versus surface coverage for ethane on: a) solvated-L-zeolite b) i r o n - L - z e o l i t e - 72 -bulkac-Fe2C>2 which by some mechanism aggregates on the external surface of the z e o l i t e . The Mossbauer behaviour pat tern observed on t h i s sample i s i n many ways much d i f f e r e n t from that reported i n an e a r l i e r 17 work by Wedd et a l . The primary d i f ference appears to l i e i n the observation of r e l a x a t i o n e f fec t s i n t h i s sample, which they d i d not observe i n t h e i r s . A l s o , they reported the 2+ appearance of an Fe species on outgassing t h e i r sample which was not observed here. Dif ferences i n the temperature at o which Fe spectra were obtained were also apparent. These d i screpancies may a r i s e from the method of preparat ion of the sample which would prove an i n t e r e s t i n g area for further i n v e s t i g a t i o n . - 73 -BIBLIOGRAPHY D.W. Breck, J . Chem. Ed., 4_1, 678 (1964). 2 L.V.C. Rees, Annual Reports on the Progress of Chemistry, A, 67, 191 (1970). 3 •• G.K. Wertheim, "Mossbauer E f f e c t , P r i n c i p l e s and Applications", Academic Press, London, 1964. h L.R. Walker, G.K. Wertheim, and V. Jaccarino, Phys. Rev. Letters, 6_, 98 (1961). 5 R. In g a l l s , Phys. Rev., 133, 3A, 787 (1964). 6 N.N. Greenwood, and T.C. Gibb, "Mossbauer Spectroscopy", Chapman and H a l l Ltd., London, 1971. 7 M. Blume, Phys. Rev. Letters, 14_, 96 (1965) . 8 M. Blume, Phys. Rev. Letters, 18_, 305 (1967). 9 S. Brunauer "Physical Adsorption", Princeton University Press, Princeton, (1945). 1 0 G.D. Halsey, Advances i n Ca t a l y s i s , 4_, 259 (1952) . 1 1 T.L. H i l l , J . Chem. Phys., 1/7, No. 6, 520 (1949). 1 2 T.L. H i l l , J . Chem. Phys., 18_, No. 3, 246 (1950). 1 3 D.M. Young, and A.D. Crowell, "Physical Adsorption of Gases", Butterworth and Co., London, 1962. 1 1 W.N. Delgass, and M. Boudart, Catalysis Reviews, 2_, 129 (1968) . 1 5 M.C. Hobson J r . , Advances i n C o l l o i d and Interface Science, 3, 1 (1971). l 6 J. Morice, and L.V.C. Rees, Trans. Faraday S o c , 6_4, 1388 (1968). l 7 R.W.J. Wedd, B.V. Liengme, J.C. Scott, and J.R. Sams, So l i d State Commun., 7, 1091 (1969). - 74 -1 8 V.I. Goldanskii, I.P. Suzdalev, A.S. Plachinda, and L.G. Shtyrkov, Proc. Acad. S c i . U.S.S.R., Phys. Chem. Sect. E n g l i s h Tr a n s l . , 169, 511 (1966). 1 9 G. Constabaris, J.H. Singleton, and G.D. Halsey J r . , J. Phys. Chem., 63_, 1350 (1959). 20 R.W.J. Wedd, M.Sc. Thesis, University of B r i t i s h Columbia, 1967. 21 U.S. Patent No. 3, 130, 006, A p r i l 21 (1964). 2 2 .. R.M. Barrer, and H. V i l l i g e r , Z e i t s c h r i f t fur K r i s t a l l o g r a p h i e , Bd. 128, S., 352 (1969). 2 3 G. Constabaris, R.H. Lindquist, and W. Kundig, Appl. Phys. L e t t e r s , 7, No. 3, 59 (1965). 2 4 . . . W. Kundig, H. Bommel., G. Constabaris, and R.H. Lindquist, Phys. Rev., 142, No. 2, 327 (1966). 2 5 P.W. Selwood, "Magnetochemistry," 2nd edn., Interscience, New York, 1956. 2 6 D.W. C o l l i n s , J.T. Dehn, and L.N. Mulay i n "Mossbauer E f f e c t Methodology," 3_, (ed. I.J. Gruverman) , Plenum Press, New York, 1967. 27 M.J. Rossiter, A.E.M. Hodgson, J . Inorg. Nucl. Chem., 27, 63 (1965). 2 8 F. van Der Woude, A.J. Dekker, Phys. Stat. Sol. 13, 181 (1966). 29 I. Dezsi, and M. Foder, Phys. Stat. Sol. 15, 247 (1966). 3 0 " I. Dezsi, L. Keszthelyi, D. Kulgawczuk, B. Moinar, and N.A. E i s s a , Phys. Stat. Sol. 22, 617 (1967). 3 1 N. Malathi, and S. P u r i , J . Phys. Soc. Japan, 3_1, No. 5, 1418 (1971). - 75 -32 V . I . G o l d a n s k i i , I . P . Suzdalev, A . S . P lachinda , and V . P . Korneev, Dokl . Akad. Nauk SSSR, 185, 203 (1969). 33 R.M. B a r r e r , and J.W. Sutherland, Proc . Roy. S o c , Sec. A, 237 (1956). 34 W.W. C o l l i n s , and L . N . Mulay, IEEE Transact ions on Magnetics, 4, 470 (1968). APPENDIX - 76 -C a l c u l a t i o n of Volume Adsorbed The fo l lowing symbols are used: Pj i n i t i a l pressure of i n j e c t e d gas P £ e q u i l i b r i u m pressure over sample V volume of p ipets P V volume of overhead pipets op V volume of overhead c e l l oc V volume of c e l l c V m manometer volume c o r r e c t i o n term for a given pressure (V _ or V _) ml mE Tp temperature of p ipets T_. room temperature T c c e l l temperature Tij t o t a l number of moles of gas i n the system n„ number of moles of gas remaining i n the gas phase a f ter e q u i l i b r a t i o n over the sample n A number of moles of adsorbed gas R gas constant With the c e l l and overhead c e l l segments closed o f f from the system, the number of moles of gas i n i t i a l l y injected i s given by ! i (!E + !OR + VE > = n i (i) R T p TR TR The c e l l and overhead c e l l segments are opened to the system, and at equilibrium ( ^ + ^pp + !mE > + V ^ o c + V = n E (2) R T p T R T R R T R T c The number of moles adsorbed i s given by n_ - n„ = n,. at pressure P_-. I E A c E The c e l l and overhead c e l l segments are again closed and the gas pressure increased by a l t e r i n g the combination of the pipet volumes. After expanding the gas into the c e l l a new equilibrium pressure P„^ " i s obtained. The number of moles remaining i n the gas phase a f t e r e q u i l i b r a t i o n over the s o l i d sample i s obtained from P- 1 (V 1 + V + V 1 „) + P- 1 (V + v ) = n- 1 (3) E p op mE E oc c E R T T„ T„ R T_, T p R R R c and the number of moles adsorbed at pressure p„"^ i s 1 1 nA = n i " n E When gas i s again injected into the system from the external storage bulbs (with the c e l l segments closed) the increase i n the i n i t i a l number of moles i s obtained from the difference between the l a s t equilibrium pressure and the new pressure - 78 -AP (V + V + AV ) = An _P_ _2£ "l T T T p X R R and the t o t a l moles of gas i n the system i s given by n t = n j + An. The volume adsorbed at temperature T c and e q u i l i b r i u m pressure P £ i s given by Vads = n . RT (6) A c P E At standard temperature (T g ) and pressure (P g) t h i s becomes V S T P = ¥ E V a d S P S T C = RT C p s A computer programme was wr i t ten to obtain the number of moles of gas adsorbed at each e q u i l i b r i u m pressure and to convert these i n t o Vg T p va lues .