Thermal response of structure and hydroxyl ion of phengite-2M1: an in situ, neutron diffraction and FTIR study

of 11

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
11 pages
0 downs
Thermal response of structure and hydroxyl ion of phengite-2M1: an in situ, neutron diffraction and FTIR study
  Introduction Micas are very common minerals occurring ina wide variety of geological environments.Dehydroxylation of micas (including the dehy-droxylation of illitic clays) in sedimentary basins,in subduction environments, and in metamorphicrocks is a very important natural process.Addressing the problem of what actually controlsthe dehydroxylation behaviour of mica on anatomic scale is of crucial importance in under-standing the global dynamics of water. Here weinvestigate the nature of the hydroxyl environmentin phengite using diffraction and spectroscopicmethods.The structure of dioctahedral 2:1 mica consistsof octahedral sheets sandwiched between tetrahe-dral (Si,Al)O 4 sheets, with two out of three octa-hedral sites occupied by divalent or trivalent metalcation (typically Al, Mg and Fe) (Fig. 1).In the trioctahedral micas all octahedral sheetsare occupied and the O(3)-H vector is perpendicu-lar to the (001) plane resulting in maximum K + -H + repulsion, whereas in dioctahedral mica the O(3)-H vector is inclined away from the (001) plane atan angle less than 90°. The angle between theO(3)-H vector and the (001) plane in such micas isobserved to range from 1.3° to 23.1° (Giese,1979). This leads to a strong interaction betweenthe apical O atoms of tetrahedral sites and theO(3)-H group. Hydrogen bonds occur between Hand the apical oxygen of tetrahedral sites as wellas to the bridging oxygens of the tetrahedral sitesmaking the O(3)-H bond within the hydroxylgroup weaker (Farmer, 1974; Langer et al ., 1981).The projection of the O(3)-H vector onto (001) Eur. J. Mineral.2001, 13 , 545-555 Thermal response of structure and hydroxyl ion of phengite-2  M  1 : an in situ neutron diffraction and FTIR study M AINAK MOOKHERJEE * , S IMON A.T. REDFERN AND M ING ZHANGDepartment of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, UK Abstract : The thermal dependence of the structure of natural phengite-2  M  1 with chemical formula(K 0.95 Na 0.05 )(Al 0.76 Fe 0.14 Mg 0.10 ) 2 (Si 3.25 Al 0.75 )O 10 (OH 1.96 F 0.04 )has been studied by in situ neutron diffraction. Theshort-range correlated behaviour of the hydroxyl group was probed by FTIR spectroscopy while the long-range cor-related hydroxyl structure was studied by neutron diffraction. Changes in long-range ordering of Si and Al on thetetrahedral sites were not observed from neutron diffraction. Structure refinement of the neutron diffraction data bythe Rietveld method suggested that the apparent average hydroxyl bond length decreases on heating. The infrareddata show a decrease in the stretching frequency of hydroxyl group, however. Possible explanations for these resultsare explored. It seems most likely that the apparent shortening of the hydroxyl bond length may be an artefact due toan increase in its vibrational amplitude. The anisotropic vibration of the hydroxyl bond as revealed by the anisotrop-ic displacement parameters of H, increases so much that the average length (shown by the neutron refinement)appears to decrease at high temperatures while the local length of the bond, as indicated by FTIR results, increases. Key-words : mica, phengite-2  M  1 , hydroxyl, neutron diffraction, FTIR, hydrogen bond. 0935-1221/01/0013-0545 $ 2.75 ã 2001 E. Schweizerbart’sche Verlagsbuchhandlung. D-70176 Stuttgart DOI: 10.1127/0935-1221/2001/0013-0545 * e-mail: paper was presented at the EMPG VIIImeeting in Bergamo, Italy (April 2000)  M. Mookherjee, S.A.T. Redfern, M. Zhang makes an angle with the b axis, which ranges from–30° to 32° (Rouxhet, 1970).Ahuge volume of work on the mica and mica-like minerals has been published and is reviewedto some extent in Bailey (1984). Early work on thedehydroxylation of mica was performed by Gaines& Vedder (1964). They demonstrated the loss of the hydroxyl group on heating, on the basis of theshift and disappearance of the O(3)-H peak in theIR spectrum. Most of the vibrational spectroscop-ic studies that have been carried out focus on thecharacterization of the environment of the hydrox-yl group in terms of the nature of the cationicneighbours (Vedder, 1964; Wilkins & Ito, 1967;Farmer, 1974; Robert & Kodama, 1988), the influ-ence of vacant octahedral site in the dioctahedralmicas (Vedder & McDonald, 1963; Farmer, 1974;Langer et al ., 1981), the influence of the tetrahe-dral layer through cationic substitution, e.g . Si-Alorder disorder relations and layer distortions(Farmer & Russell, 1964; Farmer & Velde, 1973)and the effect of the orientation of the O(3)-Hdipole ( Serratosa & Bradley, 1958a and b) and theeffect of the proton-interlayer cation repulsion(Kodama et al ., 1974). Other studies include hightemperature X-ray and neutron diffraction, inves-tigating tetrahedral ordering (Pavese et al ., 1999)and dehydroxylation mechanisms (Guggenheim et al ., 1987).Recent studies suggest that the stability of OH – groups in micas is intimately linked to the siteoccupancies of metal cations in the octahedralsheet (Drits et al ., 1995). The present work wasinitiated to study the relationship between thecation disorder (as a function of  T  ) and dehydrox-ylation of micas.The main aim of this study is to determine thetemperature-dependence of the long-range order-disorder of the octahedral and tetrahedral sites andthe behaviour of the proton using in situ high-tem-perature neutron powder diffraction. Infraredspectroscopy is sensitive to OH – structure andshort-range order, and it has been used as a com-plementary method to neutron diffraction. Experimental procedure The one phengite sample studied so far srci-nated from Livadi mafic-ultramafic complexes,N.E. Greece. The Si 4+ content of this phengite(Nance, 1976) varies from 3.20 to 3.50 per formu- 546 Fig. 1. a) Asection of single ditrigonal ring formed by the TO 4 polyhedra perpendicular to the  Z  * direction (at100°C). The interlayer potassium/sodium (not shown for clarity) sits at similar  x,y position coordinate as that of thehydroxyl group (O(3)-H(1)) in the figure. The octahedral cations are omitted for clarity. Apical oxygen atoms,labelled O(1) and O(2) are part of T(1) and T(2) respectively and those labelled O(4), O(5) and O(6) are basal oxy-gen atoms. O(3) is the part of the hydroxyl group and is roughly at the center of the ring and at the same  Z  coordi-nate as that of the O(1) and O(2) atoms. b) Two ditrigonal rings viewed perpendicular to (a). The major axis of thehydrogen thermal ellipsoid is at high angles to the hydroxyl bond. Structures are from our refinements of neutrondata collected at 100°C.  la unit and indicates a metamorphic condition of about 250°C to 500°C at 0.5 GPa. The sample wascharacterised by electron probe microanalysis andthe composition determined as (K 0.95 Na 0.05 )(Al 0.76 Fe 0.14 Mg 0.10 ) 2 (Si 3.25 Al 0.75 )O 10 (OH 1.96 F 0.04 ) . The srcinal sample was magnetically separatedfrom the other impurities and then the phase wasidentified to be phengite-2  M  1 polytype by powderX-ray diffraction, using a Philips X-ray diffrac-tometer.  In situ high-temperature neutron-diffractionstudies This sample was studied using in situ high-temperature neutron powder-diffraction at theD2B diffractometer ( l  = 1.5943 Å) at ILL,Grenoble, France. Approximately 3 cm 3 of phen-gite powder was loaded into a vanadium samplecan within a vanadium furnace. Diffraction pat-terns were collected over a time span of fourhours, a time interval which was chosen as theoptimal compromise between being that the count-ing statistics on low- d  -spacing peaks were small,while short enough that the data could be collect-ed to high temperature without the risk of sampleHigh-temperature structure of phengite-2  M  1 547 Fig. 2. Comparison of measured (crosses) and calculated (solid line) diffraction patterns (from Rietveld refinementsof the crystal structure) of phengite (at 100°C). The lower curve shows the differences, and the tick marks show thepeak positions.Table 1. Experimental and instrumental parameterspertaining to Rietveld refinements.  M. Mookherjee, S.A.T. Redfern, M. Zhang548 Table 2. Room-temperature to high-temperature data of unit-cell parameter, fractional position coordinates, thermaldisplacement parameters and  R p =   S ( ç I o –I c ç × ç I o –I b ç  / I o ) /  S ç I o –I b ç and w  R p = [ S w( ( I o –I c ) × ( I o –I b )/ I o ) 2  /  S ( I o –I b ) 2 ] 1/2 , where I and w are the intensities and weights of the profile points respectively, and I b is the backgroundcontribution to the profile.  break down. Data were obtained from 293 to 973K and on cooling at 773 K, 673 K, and 573 Kunder vacuum for four hours each (Table 1). Usingthe information we obtained from X-ray diffrac-tion, we were able to refine these patterns byRietveld analysis (Rietveld, 1969) using the GSASsoftware package (Larson & Von Dreele, 1986)(Fig. 2) and a 2  M  1 mica structural model(Guggenheim et al ., 1987). We tested for the pres-ence of phengite-3 T  , using the structural model of Pavese et al . (1999), which was incorporated as asecond phase within the refinement. Peaksattributable solely to phengite-3 T  ( i.e. ones that donot overlap with those of 2  M  1 ) are absent from ourpatterns. Most noticeably, there is no evidence of intensity at the position of the (1 0 5) and (1 0 7)peaks of the 3 T  structure between 25° and 30° 2 q .The peak shape profile was a pseudo-Voigt func-tion and the background was modelled with a ten-term shifted-Chebyshev function. The preferredorientations have been taken into account usingthe Dollase (1986) model. Aconstrained refine-ment scheme has been adopted because of thestructural complexity of the mineral. Accordinglythe following procedure was adopted:i) the thermal factors for all atoms except Hwere considered as isotropic;ii) cation ordering was constrained to the src-inal room-temperature value since no significantchange could be observed and no regular patternwas observed among the Al and Si occupanciesacross the tetrahedral sites;iii) the fractional co-ordinates of the atomswere refined unconstrained.All data were refined until convergence wasachieved. The  R -factors corresponding to theBragg contribution to the diffraction pattern ( i.e. not including the contributions from background)are listed in Table 2, where this value represents  R p =   S (|I o –I c | × |I o –I b | / I o ) /  S |I o –I b | . Infrared spectroscopy studies Silicon wafers have been successfully used asinfrared windows in other studies of minerals athigh T  (Bray, 1999; Bray & Redfern, 2000). Asmall amount (approximately 2.5 mg) of the pow-dered sample was applied to the wafer with a spat-ula, in order to obtain a usable area 13 mm indiameter. For high-temperature in situ infraredmeasurements, the sample was positioned within acylindrical platinum-wound furnace with a recy-cle-water-cooling system. The cooling system wasused to avoid the outside of the furnace to becometoo hot. Two thermocouples were equipped in thefurnace. The heating and cooling of the furnacewas controlled by using a Pt/PtRh thermocouplelocated close to the heating platinum wires in thefurnace and an Eurothermal 815 temperature con-troller, which allowed a temperature stability of less than ± 1 K. The temperature of specimen wasmeasured by using a NiCr/NiAl thermocouplewhich was pressed against the sample and coupledwith a Comark microprocessor thermometer. Thethermocouple was calibrated against the a - b phasetransition in quartz and cristobalite. The heatingrate was 15°C min –1 . The spectra were collectedunder vacuum with a Brucker 113v FTIR spec-trometer. Aliquid-nitrogen-cooled mercury-cad-mium-telluride detector coupled to a KBr beamsplitter was used to record near infrared region(NIR) spectra. The spectral resolution was 4 cm –1 .Atotal of 150 scans was accumulated for eachNIR spectrum. All spectra were recorded asabsorbance a , with a = –log 10 (  I  sample  /   I  reference ),where  I  is the single-beam transmission intensity.Spectra were recorded both in runs lowering andincreasing temperature and the spectra obtainedfrom heating and cooling are similar. Spectra areshown in Fig. 3. Results and discussion Lattice expansions The neutron diffraction pattern gives the cell-parameter evolution as a function of temperature.High-temperature structure of phengite-2  M  1 549 Fig. 3. Three-dimensional plot of the infrared spectracorresponding to hydroxyl streching region, with vary-ing temperature.
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!