Surface and bulk properties, catalytic activities and selectivities in methane oxidation on near-stoichiometric calcium hydroxyapatites Shigeru Sugiyama,"" Toshimitsu Minami," Toshihiro Moriga," Hiromu Hayashi," Kichiro Koto,b Michie Tanaka" and John B. Moffatd "Department of Chemical Science and Technology, Faculty of Engineering, The University of Tokushima, Minamqosanjima, Tokushima, 770 Japan bFaculty of Integrated Arts and Sciences, The University of Tokushima, Minamijosanjima, Tokushima, 770 Japan "Shikoku Research Institute Inc., 2109 Yashima-nishi, Takamatsu, 761 -01 Japan dDepartmentof Chemistry and the Guelph- Waterloo Centre for Graduate Work in Chemistry, University of Waterloo, Waterloo, Ontario, Canada N2L 3Gl Various calcium hydroxyapatites [Calo~z(HP04)z( PO,)6 -,(OH), -,; z = 0, stoichiometric apatite (Ca/P = 1.67; atomic ratio); 0 < z 6 1, non-stoichiometric apatites ( 1.67> Ca/P > 1.50)], together with the apatites of Ca/P = 1.72, were analysed by powder X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and extended X-ray absorption fine structure (EXAFS) spectroscopy and their catalytic properties in the oxidation of methane at 973 K were examined in the presence and absence of tetrachloromethane as a gas-phase additive.The XRD patterns of each hydroxyapatite were the same. However, the nearest- neighbour distances of the Ca- 0 bond of stoichiometric and non-stoichiometric hydroxyapatites as estimated by EXAFS were 2.41,2.37,2.38 and 2.39 A for each catalyst of Ca/P = 1.72, 1.68, 1.64 and 1.58, respectively.In the oxidation of methane on each hydroxyapatite in the absence of CCl,, the conversion of methane was little influenced by the value of Ca/P but the selectivities to C2H6 and CO showed a maximum and minimum, respectively, at Ca/P = 1.68. On addition of CCl, into the feedstream, the selectivity to C02 on Ca/P = 1.68 and 1.72 decreased as the time on-stream increased while sharp decreases in conversion were observed with time on-stream on non-stoichiometric hydroxyapatites of Ca/P = 1.64 and 1.53, respectively, without suppression of the selectivity to C02. The catalytic activities of a variety of hydroxyapatites for the oxidative dehydrogenation of methane have been investi- gated in our laboratories.'-* Most re~ently,~.~ the oxidation of methane on stoichiometric calcium hydroxyapatite [Calo( PO,),(OH),] in the presence and absence of tetrachloro- methane has been reported.Although it is generally observed that the selectivity to C,, compounds and, in particular, to ethylene in the oxidative dehydrogenation process is increased on the addition of CCl, into the methane oxidation feedstream while that to ethane is decreased,'-13 the introduction of CCl, into the feedstream on stoichiometric hydroxyapatite did not increase the selectivities to C2 compounds but rather resulted in an enhancement to that of carbon monoxide with decrease of that to carbon di~xide.~.' It is well known that calcium hydroxyapatites [Calo-,( HPO,),( P04)6-z(OH)2-z (0dz 6 l)] often display compositionally dependent behaviour in catalytic reactions.For example, the dehydrogenation of alcohols is catalysed by stoichiometric hydroxyapatite (z = 0, Ca/P = 1 .67)14*15 while dehydrogenation and dehydration are observed on non-stoichiometric hydroxyapatites (0< z < 1, 1.67> Ca/P 31.50).14,16However the source of such differences remains unclear since calcium vacancies may be distributed in the hydroxyapatite latticel7-'' a low Ca/P value may result from the formation of a mixture of stoichiometric hydroxyapa- tite and a phosphate having a low calcium content.20~21 The crystallographic structures of calcium phosphate and of the stoichiometric hydroxyapatite are similar, thus making it difficult to detect the presence of these two phases by X-ray diffraction ( XRD).22,23 This is especially so for deficient apa- tite~,'~,~~which are generally poorly crystallized." In part as a consequence of the biological importance of hydroxyapatite, a number of XRD and transmission electron microscopic (TEM) studies have been The surface structure of hydroxyapatites has also been investigated by the appli- cation of atomic force microscopy30 and X-ray photoelectron spectroscopy ( XPS).3-8,31 In the present study, the oxidation of methane in the presence and absence of CCl, has been studied both on the stoichiometric and non-stoichiometric hydroxyapatites as well as on the apatite containing excess calcium (Ca/P = 1.72).XRD and XPS analyses together with extended X-ray absorption fine structure (EXAFS) spectroscopy have also been applied, the latter to clarify the Ca-0 bond distance and the local structure around the calcium atoms in each apatite.Although EXAFS and XRD analyses provide information only on bulk phases, the properties of the surface of each apatite on which the catalysis occurs are undoubtedly strongly influenced by those of the bulk phase. Experimenta1 Catalyst preparation Calcium hydroxyapatites (Ap1.72, AP1.68, Ap1.64 and AP1.53, where the subscripts represent the Ca/P atomic ratio of each apatite) were prepared from Ca(NO,), .4H@ (Wako Pure Chemicals, Osaka) and (NH4),HP04 (Wako) according to the procedure reported by Hayek and Ne~esely.~' In the present study, Ap1.68 is considered to be the stoichiometric hydroxyapa- tite.The resulting solids were calcined at 773 K for 3 h after drying at 353 K overnight. Particles of 0.35-1.75 mm were employed as a catalyst in the present study. Apparatus and procedure The catalytic experiments were performed in a fixed-bed con- tinuous-flow quartz reactor operated at atmospheric pressure. Details of the reactor design and catalyst packing procedure have been described elsewhere." Prior to reaction the catalyst J. Mater. Chem., 1996, 6(3),459-464 459 was calcined zn sztu in an oxygen flow (25 ml min-l) at 1048 K for 1 h The reaction conditions were as follows W=O 1 g, 0 25 g or 0 5 g, F =30 ml min-l, T= 973 K, p(CH,) =28 7 kPa, p(02)=4 1 kPa, and p(CCl,)=O or 0 17 kPa, the balance to atmosphere pressure was provided by helium Analysis and characterization The reactants and products were analysed with an on-stream gas chromatograph (Shimadzu GC-8APT) equipped with a TC detector and integrator (Shimadzu C-R6A) The columns used in the present study and the methods employed in the calculation of conversions and selectivities have been described previously l2 The surface areas of the catalysts (72 9, 65 3, 73 8 and 82 6 m2 8-l for Ap, 72, Ap, 68, Ap, 64, and Ap, 53, respectively) were measured with a conventional BET nitrogen adsorption apparatus (Shibata P-700) X-Ray photoelectron spectroscopy (XPS) (Shimadzu ESCA- 1000AX) used monochromated Mg-Ka radiation The binding energies were corrected using 285 eV for C 1s as an internal standard Argon-ion etching of the catalyst was carried out at 2 kV for 1 min with a sputtering rate estimated as cu 2 nm min-' for SiO, Powder X-ray diffraction (XRD) patterns were recorded with an MXP-18 (MAC Science Co) diffractometer, using monochromated Cu-Ka radiation Patterns were recorded over the range 26 5-60' The concentrations of Ca and P or Cl in each catalyst were determined in aqueous HNO, solutions by inductively coupled plasma (ICP) spectrometry (SPS-1700, Seiko) or ion chroma- tography (Dionex 20101) X-Ray absorption spectra near the Ca K-edge were measured by a laboratory EXAFS spectrometer (Technos, EXAC800) with a molybdenum rotating anode The first-order line diffracted by a Ge monochromater was used X-Ray intensities were monitored with an Si-Li solid-state detector The absorp- tion edge Apt was estimated to be about 10 The fixed time durations for the measurement at each point were 150s for the incident beam and 450s for the transmitted one, so that total photon count was >los Analyses of the EXAFS data were performed using the program (ver 22) provided by Technos Co Ltd In obtaining the EXAFS function ~(k),the background level was subtracted from the observed absorption spectra by using a Victreen fit, and the absorption spectra for the isolated atom was approximated by the cubic spline technique 33 Fourier transformation of ~(k)into real space yielded a cadial structure function $(r),where the k-range from 3 to 8 5 A-' was used For curve-fitting analysis, the first- neighbour distance range for $(r) was filtered with a smooth filtering window and transformed back to k-space ~'(k) Carrying out the non-linear least-squares calculations of Marquardt's method,34 ~'(k)was fitted wit9 an analytical EXAFS function in the k-range from 3 to 8 5 A-' Theoretical amplitudes and phase functions calculated with the spherical wave approach of Teo were used to correct the absorber phase shift and the back-scattering amplitude 35 Results and Discussion Methane oxidation on calcium hydroxyapatites In all catalytic experiments, the products were CO, CO,, C2H4 and C2H6 Water and hydrogen were also produced but are not reported here Carbon balances of 100& 5% were obtained in all experiments The effect of changes in the Ca/P value on the methane oxidation process at 973 K are illustrated in Fig 1 in the presence and absence of CCl, In the absence of CCl,, the conversion of methane changes relatively little with the value of Ca/P While the selectivity to C2 compounds and, in particular, ethane reaches a maximum on Ap168, that to Ca P Fig. 1 The effects of the Ca P ratio of calcium hydroxyapatites on the oxidation of methane at 973 K in the absence (A) and presence (B) of CCl, a, 0 5 h on-stream, b, 6 h on-stream Reaction conditions W=05 g, P=30ml min p(CH,)=28 7 hPa, p(o,)=4 1 kPa and p(CCl,)=O kPa or 0 17 kPa diluted with He (1) C, selectivity (YO),(21) C, selectivity (Y), (iu) CH, conversion and C,(in)0, conversion (YO), yield (YO) carbon monoxide is at a minimum regardless of time on- stream In contrast, in the presence of CC14 the conversions and selectivities are dependent on both the Ca/P value and the time on-stream, but particularly the latter With the non- stoichiometric catalysts, where Ca/P is <1 68, the conversion decreases markedly with increasing time on-stream, but with Ap, 53 the selectivities remain relatively constant on introduc- tion of Ccl4 With Ap164, no C2H4 or C2H6 is formed, after 6 h on-stream, whereas the selectivity to C02 increases signifi- cantly The selectivities to CO, C02, C2H4 and C2H6, obtained with the nominally stoichiometric catalyst, are similar after 0 5 h on-stream regardless of the presence of CCl, in the feedstream However with CCl, present, with the latter catalyst composition, the selectivity to CO has approximately doubled with a corresponding decrease to that of C02,after 6 h on- stream, and additionally, the conversion has decreased some- what With the catalyst containing an excess of calcium, Ap, 72, the conversions and selectivities at both times on-stream are similar with and without CC14 In the presence of CCl, and after 6 h on-stream the selectivity to CO on Ap, 72 increases with decrease in space time (W/F) (Fig 2) A decrease in conversion is particularly evident at the lowest value of W/F In the absence of CCl, no systematic changes are observed with W/F on either Ap, 72 or Ap, 68 In contrast, with Ap, 68 and added eel,, virtually all of the c, formed after 6 h on-stream is CO, regardless of the value of W/F,in contrast with the lower selectivities to CO where CC1, is not present With Ap, 53 and Ap, 64 no remarkable changes in the C1 and C2 selectivities are observed with decreasing W/F (Fig 3) in either the presence or absence of CCl, However, the substantial decreases in conversion after 6 h on-stream where CC14 is present are particularly noticeable with these two catalysts Properties of bulk phase and surface of fresh calcium hydroxyapa ti tes Since the XRD patterns of the four samples of calcium hydroxyapatite with Ca/P values of 153, 164, 168 and 172 460 J Muter Chem , 1996, 6(3), 459-464 W/F/10-2 g rnin rnl-' Fig.2 Effects of W/F in the absence (A) and presence (B) of CCl, on the oxidation of methane at 973 K on Ap,.,, and Ap,,,,. Symbols and reaction conditions are the same as those in Fig. 1 except W=O.1 g at W/F=0.33 x lo-' g min ml-', W=0.25 g at W/F=0.83 x lo-' g min ml-', and W=0.5 g at W/F=1.67 x lo-' g min ml-'. WIFII 0-2 g min ml-' Fig. 3 Effects of W/F in the absence (A) and presence (B) of CCl, on the oxidation of methane at 973 K on Ap,,, and Ap,.,,. Symbols and reaction conditions are the same as those in Fig. 2. are, not surprisingly, essentially indistinguishable (Fig. 4), no correlations with catalytic activity can be deduced. As has been shown, the catalytic activities can frequently be related to the bulk properties of the solid catalysts such as electronega- tivity of cation^^^,^^ or anion^.'^.^' With the catalysts under investigation in the present work, the structures of the catalysts, as demonstrated from the XRD patterns, are similar.Consequently, it may be possible to view each sample as the stoichiometric calcium hydroxyapatite containing an excess of either calcium or phosphorus, the latter present as the phos- phate, or alternatively, containing calcium or phosphorus vacancies. Stoichiometric calcium hydroxyapatite has a hexag- onal structure constructed from columns of Ca and 0 atoms which are parallel to the hexagonal axis as shown in Fig. 5.39 Three oxygen atoms of each PO4 tetrahedron are shared by one column, with the fourth oxygen atom attached to a neighbouring column.The hexagonal unit cell of calcium hydroxyapatite contains ten cations located on two sets of non-equivalent sites, four on site I (Ca,) and six on site I1 (Ca,,). The calcium ions on site I are aligned in columns, while those on site I1 are in equilateral triangles centred on the screw 2tVdegrees Fig. 4 XRD patterns of the fresh calcium hydroxyapatites: (a) Ap1.72; (b)Apl.68; (c)Ap1.64;(d) Ap1.53 Fig. 5 Stoichiometric calcium hydroxyapatite structure projected on the a,b plane39 axes. The site I cations are coordinated to six oxygen atoms belonging to different PO4 tetrahedra and also to three, relatively distant, oxygen atoms. The site I1 cations are coordi- nated to six oxygen atoms belonging to PO4 and one oxygen atom belonging to an hydroxy gro~p.~~,~' The Ca-0 bond distance of the stoichiometric calcium hydroxyapatite as obtained from neutron diffraction are summarized in Table 1,28 together with the coordination numbers based on Fig.5. In the oxidation of methane on the solid catalysts, it is generally accepted that the active sites are oxygen species, although their natures are uncertain and undoubtedly depend on the catalyst, and the activity is also strongly influenced by the electronega- tivity of the cations. Although at this time the location and environment of the excess of calcium or phosphate ions are unclear, it is expected that these ions would exert a perturbing influence on the species contained within the structure and in particular the Ca-0 bond distance which may, in turn, alter the electronegativities of the oxygen species.Therefore EXAFS analyses were applied to clarify the bond distance and the J. Muter. Chem., 1996, 6(3), 459-464 461 Table 1 Ca-0 distances' and coordination numbers in stoichiometnc hydroxyapatite distance/A 2 408 2 454 2 808 CNb 3 3 3 'Ref 28 Coordination number local structure around the calcium atoms Fig 6 shows the X-ray absorption near-edge structure (XANES) spectra of the fresh catalysts with Ca/P 172, 168, 164 and 1 53 The shape of the XANES spectra and the edge position for each catalyst are almost the same, indicating that the electronic configuration and site symmetry of the calcium in the present calcium hydroxyapatites are not significantly different Fig 7 shows the Fourier transforms of the EXAFS oscillation around the Ca K-edge of each fresh catalyst Phase shifts are not corrected in these spectra The strongest peaks in each spectrum correspond to a nea5eest-neighbour distance with separations between 2 3 and 2 8 A (Table 1) It is of interest to note that other-order distances appear to be different from each other It would be expected that the nearest-neighbour Ca-0 dis-tance has the strongest relationship to the catalytic activities Further, the data obtained at the nearest-neighbour distances are more precise owing to the relative independence of the harmonics Thus, the reliability of the phase shift and amplitude functions are tested at approximately the nearest distance by I' A 1 6020 Goso 4100photon energylev ...I ....1 1.1.11--v,= c3 -.......,, 11..a,,., 012345012345 riA Fig. 7 Founer transformation of k3 weighted EXAFS oscillation measured at 300 K near the Ca K-edge of the fresh calcium hydroxyapatites (a) Apl 72, (b)Apl 68, (c) Apl 64, (dl Apl 53 462 J Muter Chem ,1996,6(3), 459-464 2 707 2 358 2 345 2 514 2 384 1 1 2 2 1 fitting the observed EXAFS of each fresh catalyst Fig 8 shows optimum curve fitting for each hydroxyapatite, in which the solid lines represent the expenmental data and the closed circles represent the calculated results The results of the curve fitting analysis are shown in Table 2 The coordination number of calcium, set at 4 8 in the stoichiometric calcium hydroxyapa- tite (Ap, 68), shows the estimated average coordination number of oxygen to CaJ and Call at a distance of Ca-0 between 2 345 and 2 454 A (Table 1) Although cognisance of the esti- mated deviations should be taken, the nearest-neighbour dis- tances of the Ca-0 bond of each catalyst are in the order of Ap, 72 =-Ap, 68 <Ap, 64 <Ap, 53 It may be of interest to com- pare the order of the Ca-0 bond with that of the selectivity to carbon monoxide in the oxidation of methane in the absence of Ccl4 (Ap, 72 >Ap, 68 <Ap, 64 <Ap, 53 as shown in Fig 1) The surface properties of each fresh catalyst were examined by XPS (Table 3) The binding energies of Ca 2p and 0 1s before and after argon-ion etching were virtually identical regardless of the quantity of Ca Furthermore, no correlation is evident between the values of Ca/P and of O/P on the surface and those of the bulk Ca/P This indicates the difficulty in companng and contrasting surface properties of the calcium hydroxyapatites by application of XPS analyses only Properties of bulk phase and surface of the used calcium hydroxyapatites XRD, EXAFS and XPS analyses of the used catalysts were performed to investigate the results of the interaction of the FF 0 2 4 6 8102 4 6 810 klA-' Fig.8 Curve fitting of the fresh calcium hydroxyapatites Experimental Table 2 Results of curve-fitting analysis sample r/k Nb o/A2c Eo/eVd R/(%)" AP112 2 41 55 0 1042 6 741 77 68 2 37 48 0 0999 3 961 81 Apl 64 Apt 53 2 38 2 39 46 42 0 0975 0 0705 4 480 5 018 67 61 a Distance, estimated maximum deviation (+O 01) Coordination number, estimated maximum deviation (+10), except for 4 8 which is fixed Debye-Waller factor Threshold increment Reliability factor Table 3 Binding energies and relative concentrations in the fresh The formation of chlorinated species on the surface of Ap1.72 calcium hydroxyapatites is confirmed by XPS (Table 5).Since the binding energies of relative Cl 1s in Ap1.72, used in the presence of CCl,, together with binding energy/eV concentration those of the other elements, are similar to those found with AP1.68, Ap,.,, and Ap1,53, chlorapatite may be formed in minor Ca amounts on Ap1.72. Note that the quantities of chlorinated species found on the surface of the catalysts previously sample time"/min 2p3/, 2p1,, 0 1s P 2p Ca/P O/Ca employed in the methane reaction, where CCl, was present, AP1.72 0 347.4 351.0 531.3 133.3 1.30 2.40 appear to correlate with results obtained from these reactions. 347.7 351.0 531.7 133.6 1.42 2.36 On Ap,.,, and Ap1.53, with which substantial decreases in the 1 Ap1.68 0 346.8 350.2 530.8 132.6 1.16 2.37 conversion with increasing time on-stream were observed in 1 347.0 350.6 531.2 133.3 1.44 2.26 the presence of CCl,, the amounts of the chlorinated species Ap1.64 0 347.4 350.9 531.5 133.1 1.20 2.49 formed on the surface during the reaction are quite similar 1 348.1 351.5 532.2 134.1 1.38 2.21 Ap1.53 0 347.2 350.7 531.3 133.1 1.20 2.64 (Cl/Ca =0.22 and 0.20, respectively).The quantity of the chlori- 347.8 351.4 531.7 133.7 1.34 2.42 nated species on the surface is found to be a maximum for 1 Ap1.68 with which the largest selectivity to CO was obtained. " Etching time. In contrast, relatively small quantities of chlorinated species were detected on Ap1.72, with which relatively little changes in feedstream components and, in particular CCl,, on the bulk either the conversions or selectivities were observed. Since the and surface properties of these catalysts. order of the amount of the chlorinated species estimated by XRD patterns of the catalysts used in the methane conver- XPS (Ap1.68>Ap1.64 zApl.53 >Ap1.72) is identical to that sion process in the absence of CCl, show that the composition found for the bulk phase and the formation of chlorapatite corresponds to that of calcium hydroxyapatite and no trans- apparently results from ion exchange of the hydroxy group by formations of the catalyst are detected (Table 4).In contrast, C1- it appears that the detected chlorinated species on Ap,.,, complete conversion of the stoichiometric and calcium-is also chlorapatite. Since the quantities of chlorinated species deficient calcium hydroxyapatites (AP1.68, and Ap,.,, and found in the catalysts do not correlate with that of the nearest Ap1.53 to chlorapatite [Gal,(P04)6C12]) is observed after the distance of Ca-0, the ion exchange of the hydroxy group in methane oxidation reaction in the presence of CCl,.the catalyst by C1- ion does not appear to be significantly Surprisingly, chlorapatite is not detected in Ap1.72 although dependent on this distance, and consequently the effects pro- ion chromatographic analyses of C1 show the presence of duced by the addition of CCl, are evidently not dependent on chlorinated species. the Ca-0 separation. Table 4 Properties of bulk phase of the used catalysts" sample SAC phased Cl/Ca" Ca-Of AP1.72 20.1 & 0.8 -2.37( 1) 19.1k0.9 0.026k0.001 2.36( 1) Ap1.68 20.1 & 0.4 -2.37( 1) 15.3k0.3 0.143 0.007 2.36( 1) Apl 64 18.1 kO.1 -2.34( 1) 7.8t0.1 0.122 0.006 2.38(1) Ap1.53 9.4 & 0.2 -2.39( 1) 22.7 k0.3 0.116& 0.006 2.36( 1) 'Previously employed in obtaining the results reported in Fig.1 but after 6 h on-stream. A; absence of CC14. P; presence of CCI,. 'Surface area (m2 g-'). By XRD. Ca2+ by ICP and CI- by ion chromatography. By EXAFS. Values in parentheses are the estimated maximum deviations. Not analysed. Table 5 Binding energies and relative concentrations in the used catalysts' binding energy/eV relative Ca concentrations sample CCI4 t'/min 2P3/2 2Ptp 0 1s p 2P c1 1s Ca/P O/Ca Cl/Ca AP1.72 0 347.2 350.7 531.3 133.2 - 1.30 2.79 - 1 347.6 351.1 531.7 133.7 - 1.40 2.65 - APl.72 0 347.0 350.8 531.1 132.7 199.0 1.23 2.89 0.05 1 347.4 351.2 531.5 133.3 200.2 1.40 2.70 0.07 .68 0 347.3 350.9 531.5 133.0 - 1.37 2.8 1 - 1 347.6 351.1 531.7 133.5 - 1.32 2.57 - 0 347.3 351.0 531.2 133.2 198.8 1.19 2.04 0.35 1 347.8 351.4 531.9 133.7 199.6 1.26 1.96 0.35 Ap1.64 0 1 347.5 347.7 350.9 351.2 531.5 531.9 133.6 133.7 -- 1.27 1.43 2.73 2.54 -- Ap1.64 0 1 347.4 347.8 350.9 351.3 531.3 531.5 133.0 133.7 198.8 198.8 1.18 1.28 2.77 2.56 0.20 0.2 1 Ap1.53 0 1 347.2 347.7 350.8 351.3 531.4 531.7 133.1 133.5 -- 1.16 1.30 2.67 2.48 -- Apl.53 0 1 347.5 347.8 351.0 351.3 531.4 531.7 133.2 133.7 199.1 199.2 1.18 1.30 2.55 2.3 1 0.22 0.19 " Previously employed in obtaining results reported in Fig.1 but after 6 h on-stream. A; absence of CCI,. P; presence of CC14.Etching time. J. Muter. Chem., 1996,6(3), 459-464 463 Analyses of the Ca-0 distance of the used catalysts were also carned out using EXAFS (Table 4) Although the Ca-0 distances of the used catalysts are somewhat smaller than those in the corresponding fresh catalysts, probably owing to sintenng, no systematic variations were found either with or without CCl, Therefore the properties of the catalysts appear to be more strongly influenced by the bond lengths of the fresh catalysts 5 6 7 8 9 10 Y Matsumura, S Sugiyama, H Hayashi and J B Moffat, J Solid State Chem, 1995,114, 138 Y Matsumura, S Sugiyama, H Hayashi and J B Moffat, Catal Lett, 1995,30,235 S Sugyama, T Minami, H Hayashi, M Tanaka, N Shigemoto and J B Moffat, Energy Fuels, in the press S Sugiyama, T Minami, H Hayashi, M Tanaka, N Shigemoto and J B Moffat, J Chem SOC,Faraday Trans, 1996,92,293 S Ahmed and J B Moffat, Stud Surf Sct Catal, 1991,61,57 T Ohno and J B Moffat, Appl Catal, 1993,93,414 Conclusions 11 12 S Sugyama and J B Moffat, Energy Fuels, 1994,8,463 S Sugiyama, K Satomi, N Kondo, N Shigemoto, H Hayashi and J B Moffat, J Mol Catal, 1994,93,53 The conversion of methane in the absence of CCl, was little influenced by the value of Ca/P in each calcium hydroxyapatite, while the selectivity to CO reached a maximum at Ap,,, Over calcium-deficient hydroxyapatites (Ap, 64 and Ap, 53), a sharp decrease in the conversion was observed on addition of 13 14 15 16 17 R Voyatzis and J B Moffat, Energy Fuels, 1995,9,240 H Monma, J Catal, 1982,75,200 J A S Bett, L G Christner and W K Hall, J Am Chem SOC, 1967,89,5535 M Misono and W K Hall, J Phys Chem, 1973,77,791 L Winand and G Duyckaerts, Bull SOC Chim Belg ,1962,71,142 CCI, while a substantial increase of the selectivity to CO occurred on the stoichiometric calcium hydroxyapatite (Ap, 68) On Ap, 72, minor increases in the selectivity to CO on addition of CCl, were observed, particularly at shorter space times The Ca-0 bond distances of the fresh cataly$s as estimated ,by EXAFS folloy the order Aplaz within the estimated maximum deviation of 001 A, while the selectivity to carbon monoxide in the oxidation of methane in the absence of CCl, follows the order Ap, 72>Ap1 68 < Ap, 64<Ap1 53 The order of the atomic ratio of Cl/Ca both in the bulk phase and on the surface of the catalysts previously employed in the methane conversion in the presence of CCl, was found to be Ap, 68 >Ap, 64 M Ap, 53 >Ap, 72 From XRD analyses the chlorinated species formed during the reaction appear to be predominantly chlorapatite Some relationships between the effect of the introduction of CCl, in the feedstream (2 41 A)>Ap, 68 (2 37 A)<Ap, 64 (2 38 A)6Ap, 53 .(2 39 A) 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 E E Berry, J Inorg Nucl Chem , 1967,29,317 E E Berry, J Inorg Nucl Chem, 1967,29, 1585 D McConnel, Arch Oral Biol, 1965,10,421 W E Brown, J P Smith, J R Lehr and S W Franer, Nature, 1962,196,1950 S J Joris and C H Amberg, J Phys Chem ,1971,75,3167 H Monma, Shokubai, (in Japanese, Catalyst),1985,27,237 W E Brown, Nature, 1962,196, 1048 H Ji and P M Marquie, J Muter Sci Lett, 1991,10, 132 W J Landis, J Moradian-Oldak and S Weiner, Connect Tissue Res, 1991,25,181 W J Landis and M J Glimacher, J Ultrastruct Res, 1978, 63, 188 M I Kay, R A Young and A S Posner, Nature, 1964,204,1050 H C W Skinner, H T Hunt and J Griswold, J Phys E, Sci Instrum, 1980, 13,74 L M Siperko and W J Landis, Appl Phys Lett, 1992,61,2610 H Nishikawa, S Ikeda and H Monma, Bull Chem SOC Jpn, 1993,66,2570 E Hayek and H Newesely, Inorg Synth ? 1963,7,63 and the amount of chlorapatite formed during the oxidation of methane are apparent 33 34 F W Lytle, D E Sayer and E A Stern, Phys Rev B, 1975, 11,4825 D W Marquardt, J SOC Ind Appl Math, 1963,11,431 This work was partially funded by the Natural Sciences and Engineering Research Council of Canada to J B M ,to which 35 36 B K Teo, EXAF Basic Principles and Data Analysis, Spnnger-Verlag, Berlin, 1986, p 71 S Sugiyama and J B Moffat, Catal Lett, 1992,13,143 our thanks are due 37 S Sugiyama, K Satomi, H Hayashi, N Shigemoto and J B Moffat, Appl Catal A Gen, 1993,103,55 References 38 39 S Sugiyama and J B Moffat, Energy Fuels, 1993,7,279 T Suzuki, in Ion-Koukan (in Japanese, Ion Exchange), ed M Seno, M Abe and T Suzuki, Kodansha, Tokyo, 1991, p 141 1 Y Matsumura and J B Moffat, Catal Lett, 1993,17, 197 40 T Kanazawa, in Muki-Rin-Kagaku (in Japanese, Chemistry of 2 Y Matsumura and J B Moffat, J Catal, 1994,148,323 Inorganic Phosphates), Kodansha, Tokyo, 1985, pp 61-63 3 Y Matsumura, J B Moffat, S Sugryama, H Hayashi, N Shigemoto and K Saitoh, J Chem SOC,Furaday Trans, 1994, 90,2133 41 42 Y Amenomiya, V I Birss, M Goledzinowski, J Galuszka and A R Sanger, Catal Rev Sci Eng ,1990,32,163 J H Lunsford, Catal Today, 1990,6,235 4 Y Matsumura, S Sugiyama, H Hayashi, N Shigemoto, K Saitoh and J B Moffat, J Mol Catal, 1994,92,81 Paper 5/07380D, Recezued 9th Nouember, 1995 464 J Muter Chem , 1996,6(3), 459-464