Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

X-ray diffraction studies on lithium intercalated MoS[2] Wainwright, David Stanley 1978

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1979_A6_7 W35.pdf [ 2.91MB ]
Metadata
JSON: 831-1.0085199.json
JSON-LD: 831-1.0085199-ld.json
RDF/XML (Pretty): 831-1.0085199-rdf.xml
RDF/JSON: 831-1.0085199-rdf.json
Turtle: 831-1.0085199-turtle.txt
N-Triples: 831-1.0085199-rdf-ntriples.txt
Original Record: 831-1.0085199-source.json
Full Text
831-1.0085199-fulltext.txt
Citation
831-1.0085199.ris

Full Text

X-RAY DIFFRACTION STUDIES ON LITHIUM INTERCALATED MoS2 by DAVID STANLEY WAINWRIGHT B.Sc. (Honours), Dalhousie University, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES the Department of Physics We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA . November, 1978 (c) David Stanley Wainwright, 1978 In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i l ab le for reference and study. I further agree that permission for extensive copying of th i s thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thesis for f inanc ia l gain sha l l not be allowed without my written permission. Department of P^ys»cs  The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date Oct . 31 y '37$ i i ABSTRACT L i MoS. samples were prepared by cathodically intercalating a MoS„ powder cathode with l i thium. X-ray di f fract ion patterns of these samples were taken using the Debye-Scherrer method. From these patterns, the crystal structure of the lithium intercalated M 0 S 2 was determined as a function of c e l l voltage. There is no apparent change in the powder patterns u n t i l a phase transit ion occurs in the sample at - 1.1 V. This transit ion appears to involve a slipping of la t t i ce planes in the M 0 S 2 crys ta l . Straining of the la t t ice is evident after this phase transit ion takes place. Non-uniform straining blurred the X-ray pictures with the result that the structure of the new phase could not be unambiguously determined. It does seem certain though, that an aligning of the Mo atoms takes place in this transit ion. The new phase could be f i t to a hexagonal unit c e l l . The la t t i ce parameters were measured over a charge-discharge cycle in the new phase. The unit c e l l parameters reversibly change throughout a cycle, implying that intercalation takes place. The c axis exhibits a maximum with increasing lithium content, while the a axis increases monotonically. The volume of the unit c e l l increases smoothly however, reaching a maximum value some 15% larger than the or ig ina l . Another phase transit ion may take place at =  0.55 V. Samples d i s - , charged below this voltage appear almost amorphous in the powder patterns, possibly due to extreme non-uniform straining of the l a t t i ce . Some decomposition products of the Li-MoS^ cel ls have also been discovered. i i i TABLE OF CONTENTS Page ABSTRACT .- i i TABLE OF CONTENTS i i i LIST OF FIGURES iv LIST OF TABLES v ACKNOWLEDGEMENTS v i 1. Introduction 1. 1.1 Layer compounds 1 1.2 The MoS2 battery 5 2. X-ray Diffraction Theory 11 2.1 Direction of the diffracted beam 11 2.2 Intensity of the diffracted beams 12 2.3 Debye-Scherrer method 15 3. Experimental Procedure 18 3.1 Cathode preparation 18 3.2 C e l l construction 21 3.3 X-ray specimen preparation 23 3.4 X-ray procedure 25 4. Results and Discussion 26 4.1 Phase 1 27 4.2 Phase 2 transition 28 4.3 Phase 2 30 4.4 Phase 2 3 transition and Phase 3 36 4.5 Decomposition products 36 5. Summary • 38 Bibliography 40 Appendix A Examples of X-ray spectra 41 Appendix B Line broadening 50 Appendix C Errors in Phase 2 measurements 52 Appendix D Intensity calculations 54 iv LIST OF FIGURES Figure Page 1. Layered transit ion metal dichalcogenides 2 a) General form b) Coordination units for MX^ layer structures 2. Common structures of the layered compounds 3 a) 3-dimensional drawing b) 110 sections 3. Processes involved in a discharge of the Li -MoS2 system 6 4. Voltage vs. charge transfer of a Li -MoS2 battery 8 5. Voltage vs. time profi les of ce l l s cycled in the three phases . . . . 9 6. Some la t t i ce planes and their Mi l l er indices in a hexagonal la t t i ce 13 7. Debye-Scherrer camera (side view) 16 8. Debye-Scherrer picture of 2H-M0S2 " " " " * ^ 9. Apparatus used for cathode preparation on aluminum substrates . . . . 20 10. Twin cathode c e l l assembly (side view of electrodes) 22 11. X-ray specimen preparation 24 12. Layer spacing and a axis plots vs. phase 2 voltage prof i le 32 13. Layer spacing and a axis plots vs. phase 2 c e l l voltage 33 14. a) Relative unit c e l l volume vs. phase 2 voltage prof i le 35 b) Layer spacing/a vs. phase 2 voltage prof i le 35 15. X-ray pictures of M0S2 cathodes at various stages of discharge . . 42 16. L i Mo0„ spectra 48 V LIST OF TABLES Table Page 1. Line spectrum of 2H-M0S2 • 43 2 . Line spectra of M0S2 cathode patterns in F i g . 15 44 3. Line spectra of L i x Mo02 patterns in F i g . 16 49 4 . Calculated intensities for various phase 2 structure candidates . . 56 5. Comparison.between observed and calculated l ine intensities in 2H-MoS2 57 ACKNOWLEDGEMENTS v i Many thanks go to my supervisor Rudi Haering who somehow always found the time to give me help on this project. A group at U.B.C. has been studying the Li-MoS2 battery since 1977, and much of this thesis is based upon work done by members of this group. I would then also wish to thank Dr. Jim Sti les Dr. Klaus Brandt Nelson Shen Ross McKinnon Ulr ich Sacken Chris Hodgson and Brendon Wong for contributions they have made. Peter Haas prepared the figures presented in this thesis, and his work is much appreciated. I am grateful to the National Research Council for their f inancial support in the form of a scholarship. 1 1. Introduction The desire to create a high energy density battery for use in e lectr ic vehicles has aroused interest in the group of transit ion metal dichalcogenides, many of which form layer structures. At the present time, several layer compounds are being tested as candidates for c e l l cathodes. Scientists at Exxon are working with H S 2 , while researchers at Be l l Labs are studying V ^ . (Results of these studies can be found in papers by M.S. Whittingham 1976 and D.W. Murphy, J . N . Carides, et a l 1977 respectively.) Our group is working with the mineral molybdenite, M0S2, the only transit ion metal dichalcogenide which occurs naturally in significant quantities. In many layer compounds, i t i s possible to insert , or intercalate, specific substances between the layers of the material. The process of intercalation i s not a newly discovered phenomenon, having been investigated in graphite as early as 1841 (C. Schafhautl). The fact that, in the afore-mentioned compounds, the intercalated material can be removed without destroying the host la t t ice makes these materials look l ike good candidates for secondary (rechargeable) battery cathodes. The state of discharge of our batteries depends on the extent to which the M0S2 i s intercalated. The purpose of this thesis was to determine, by X-ray di f fract ion techniques, the crystal structure of the M0S2 cathode as a function of the state of discharge. 1.1 Layer compounds The layered transition metal dichalcogenides have the general form shown in F i g . 1. The transition metal atoms, M, are sandwiched between the 2 (a) General form SANDWICH HEIGHT - x -GAP HEIGHT v -van der Waals gap (b) Coordination units for MX2 layer structures AbA trigonal pr ism AbC octahedron F i g . 1 - Layered transition metal dichalcogenides (after Wilson and Yoffe 1969) 2 H - M o S 2 / A b A B a B / -110 2H -NbS< - • r l l O / A b A C b C / 1 T - T I S 2 / A b C / A b C / 2 H - M o S 2 C A B C (a) 2 H - N b S 2 C A B C (b) • = Metal o = Chalcogenide I T - Ti S 2 A" 1> C A B C Fig, 2 - Common structures of the layered MX2 compounds a) 3-dimensional drawing b) 110 sections ( s o l i d l i n e outlines the unit c e l l ) (after Wilson and Yoffe 1969) 4 chalcogenide* atoms, X, i n planes with an underlying hexagonal l a t t i c e . The c r y s t a l consists of stacks of these X-M-X planes.(Fig. 1); There are two possible sandwich structures (Fig. l b ) , d i f f e r i n g i n how the c h a l -cogenide atoms are arranged around the metal atoms. In the sandwich planes, there are three positions i n the un i t c e l l where atoms can l i e , denoted A, B, and C. (X and M atom posit i o n s are represented by large and small l e t t e r s r espectively.) The t r i g o n a l prism coordination i s thus denoted AbA, while the octahedral coordination i s AbC. Complicated stacking sequences have been observed, but the most common structures are the simple ones i l l u s t r a t e d i n F i g . 2a . 2 H - M 0 S 2 and 2H-NbS2 type structures are s i m i l a r i n that they include two layers per unit c e l l and the chalcogenide coordination i s t r i g o n a l prismatic. (The 2 H - p r e f i x means that the structure has 2 layers per unit c e l l with point group symmetry i n the hexagonal d i v i s i o n of the hexagonal system.) Both structures belong to the space group ~B6^/mmc-I>^. The two structures d i f f e r i n t h e i r stacking sequence. 2 H - M 0 S 2 stacks with alternate Mo atoms unaligned, while the Nb atoms are aligned i n NbS2.(Fig. 2b) Note that by s l i d i n g the sandwich planes i n 2 H - M 0 S 2 so that the Mo atoms a l i g n , the r e s u l t i n g structure i s that of 2H -NbS2« The other common stacking arrangement i s that of I T - H S 2 , which has octahedral coordination of the s u l f u r atoms i n the sandwich planes. (The IT- p r e f i x stands f o r 1 layer per unit c e l l with point group symmetry i n the t r i g o n a l d i v i s i o n of .'the hexagonal system.) 3 I T - H S 2 belongs to the space group PSml-D^^. Like i n NbS2> the metal atoms i n T i S 2 are a l l aligned. Large anisotropics i n mechanical and e l e c t r i c a l properties e x i s t between d i r e c t i o n s perpendicular and p a r a l l e l to the sandwich lay e r s . * X can be either S,Se, or Te. 5 There i s only weak Van der Waals bonding between sandwich layers as compared with the strong molecular X-M bonding i n the sandwiches. It i s possible to i n s e r t other atoms or molecules i n the Van der Waals gaps between the molecular planes. This process i s known as i n t e r c a l a t i o n . There are two d i f f e r e n t types of s i t e s for the i n t e r c a l a t e i n the sandwich gaps. A tetrahedral s i t e l i e s at the centre of the tetrahedron formed by four surrounding chalcogenide atoms. S i m i l a r l y , an octahedral s i t e i s located at the centre of the octahedron formed by si x surrounding chalcogenide atoms. There are one octahedral and two tetrahedral s i t e s per MX^ molecular u n i t i n the common structures shown i n F i g . 2. In the MoS 2 structure, the octahedral s i t e s are located at the C positions i n F i g . 2b, while the tetrahedral s i t e s are found at the A and B p o s i t i o n s . 1.2 The MoS 2 battery Our b a t t e r i e s consist of an MoS 2 cathode, a l i t h i u m metal anode, and non-aqueous e l e c t r o l y t e . Lithium i s used because of i t s weight and because i t d i f f u s e s r a p i d l y i n many structures. An e l e c t r o l y t e s o l u t i o n that doesn't react with l i t h i u m i s required, hence the need for a non-aqueous solvent. The e l e c t r o l y t e used contained a l i t h i u m s a l t , either l i t h i u m perchlorate, LiClO^, or l i t h i u m bromide, L i B r , dissolved i n propylene carbonate l i q u i d (PC), 1,2 propanediol carbonate. The processes involved i n a c e l l discharge are i l l u s t r a t e d i n F i g . 3. Lithium ions d i f f u s e through the e l e c t r o l y t e to the MoS 2 cathode whereupon i n t e r c a l a t i o n occurs. An equal number of ions must enter the s o l u t i o n at the anode. Electrons from the l i t h i u m metal provide the c e l l current and flow to the cathode v i a the external c i r c u i t , thus maintaining i t s o v e r a l l 6 •VWWr MoS cathode o = 0 = e L« + cio; Lithium metal Electrolyte L i B r / P C or L i C I 0 4 / P C or Br" Fig. 3 - Processes involved i n a discharge of the Li-MoS» system 7 neutrality. On a recharge, these processes occur in reverse with the lithium plating back on the anode. The voltage of the Li -MoS2 c e l l depends on the state of charge. F i g . 4 shows the voltage prof i le of a c e l l prepared as described in Sect. 3 and cycled at constant current. I n i t i a l l y , upon continuous discharge, the c e l l voltage follows the path ABGEF. Three different paths (BA,ECD, and FG) may be followed upon recharging, depending upon the starting point. The c e l l i s not reversible along paths BC or EF. With these exceptions, subsequent cycling is permitted along any pathway. Thus the c e l l i s capable of cycling in the three phases represented by the paths AB,DE, and GF on the voltage prof i l e . These shal l henceforth be referred to as phases 1,2, and 3 respectively.(Typical voltage profi les of ce l l s cycling at constant current in these three phases appear in F i g . 5.) So a c e l l in phase 1 can be converted into phase 2, and a c e l l in phase 2 can be converted into phase 3. The reverse situation is not possible, at least at pract ical recharge rates (See Sect. 4 ) . The transition from phase 1 to phase 2 begins at point B and ends at C. A battery discharged to a state midway between B and C appears to consist of a phase 1 and a phase 2 c e l l run in para l l e l . Similarly, the transition from phase 2 to phase 3 begins at E and ends at F i And,batteries discharged to a state midway between E and F act l ike a phase 2 and a phase 3 c e l l run in p a r a l l e l . Unfortunately, at the present time, accurate determination of the cathode's lithium content versus c e l l voltage has not been obtained. In F ig . 4 , the values of x in the compound L i x MoS2 were determined from coulombic transfer, ( ie. The amount of lithium present in a M0S2 cathode of known weight was calculated on the assumption that one lithium atom CHARGE TRANSFER t t t t t X « 0.5 1.0 1.5 2 3 F i g . 4 - Voltage v s . charge t r a n s f e r of a L i - I f c ^ b a t t e r y (constant current) This p r o f i l e i s r e p r e s e n t a t i v e of c e l l s prepared and cy c l e d as described i n Se c t i o n 3. The values of x i n the compound Li xMoS2 were determined from coulombic t r a n s f e r . 0 J :  TIME F i g . 5 - Voltage vs. time p r o f i l e s of c e l l s cycled i n the three phases (constant current) These plots were obtained from c e l l s prepared and cycled as described i n Section 3. (The difference i n capacity between discharge and recharge i s believed to be caused by a s e l f -recharge process. See Section 3.) intercalates per electronic charge transferred.) The possible presence of a side reaction (Sect. 3) introduces an uncertainty i n these measure-ments . 11 2. X-ray Diffraction Theory The unit c e l l dimensions and structure of the MoS2 cathode in various stages of discharge were determined from powder X-ray di f fract ion patterns using the Debye-Scherrer method. The directions of the diffracted X-ray beams provide information on the shape and size of the unit c e l l , while their relat ive intensities depend on the arrangement of the atoms in the unit c e l l . In this section, basic theory pertinent to the experiment is presented. For further detai ls , two good references are Klug (1974) and Cul l i t y (1956). 2.1 Direction of.the diffracted beam The Bragg law i l lustrates how the angle of the diffracted X-ray beam depends on the distance between la t t i ce planes of the crys ta l . The relat ion is s in 0 = 2 | (2-1) where 0 = Bragg angle of scattering A = wavelength of the X-radiation d = distance between la t t i ce planes Lattice planes are specified by their Mi l l er indices (hkl). Crysta l -lographers define these as " a set of integers with no common factors*, inversely proportional to the intercepts of the crystal plane along the * Mi l ler indices with a common factor represent an imaginary la t t i ce plane. Higher order reflections from real crystal planes are thought of as being reflections from imaginary la t t i ce planes. crystal axes. " (Ashcroft and Mermin 1976) Some la t t i ce planes in a hexagonal la t t ice are shown in F i g . 6. Equations relating the distance d between adjacent la t t i ce planes and their Mi l l er indices can be derived from the la t t i ce geometry. For example, hexagonal latt ices obey where a and c are la t t i ce constants. Diffraction directions are usually specified by the d spacing of the la t t i ce planes, since 0 depends on 2.2 Intensity of the diffracted beams X-ray patterns obtained using the Debye-Scherrer method are composed of di f fract ion l ines . (See F i g . 8) Reflections from la t t i ce planes of the same d spacing superimpose to form one di f fract ion l ine . This set of equivalent planes is represented by {hkl} where (hkl) are the Mi l l er indices of any plane in the set. The number of planes in the set {hkl} equals the mult ip l ic i ty factor, p, for that set. Thus the intensity of a {hkl} di f fract ion l ine is equal to p times the intensity of a l ine produced from a {hkl} plane. The intensity of a (hkl) l ine is proportional to the square of the geometrical structure factor, F . F depends on the position and elemental type of the atoms in the unit c e l l . (See Ashcroft and Mermin 1976) Suppose the unit c e l l contains n atoms with fractional coordinates (u_.,v^,w_.) relat ive to the crystallographic axes a,b, and c. The geometrical structure factor, F, V 1 associated with the (hkl) plane (2-2) the wavelength of the X-radiation used. 13 (001 ) (101) F i g . 6 - Some lat t ice planes (shaded) and their Mi l l er indices in a hexagonal l a t t i ce . a,b, and c are the crystallographic axes 1 defining the unit c e l l . In a hexagonal l a t t i ce , a = b f c Angle between a and b is 120° c is perpendicular to the ab plane Thus two parameters, a and c, describe the unit c e l l of a hexagonal la t t i ce , (after Cul l i ty 1956) i s given by T-i v f ± 2TT (hu.+ kv + lw. ) F h k l ~ . E 1 f j 6 2 j 3 ( 2 _ 3 ) where f. i s the atomic scattering factor of the i th atom. Values of f. J J for the various elements were extrapolated from tables provided in Cul l i t y (1956). Calculations of the relat ive intensities of the di f fract ion lines were made according to the equation I = | F|2- p f 1 .Va° s 2 2A (2-4) 1 1 Vsin z9 cos0 / (For Debye-Scherrer method only) where I = relat ive integrated intensity of the {hkl} l ine F = geometrical structure factor of the (hkl) plane p = mul t ip l i c i ty factor 8 = Bragg angle The angular correction term in equation (2-4) is known as the Lorentz-polarization factor. An unpolarized beam introduces an angular dependence to the diffracted intensity proportional to (1 + cos 2 29), the polarization factor. Other trigonometric factors, specific to the Debye-Scherrer method, introduce the (sin 29 cos9) ^ dependence, the Lorentz factor. Some corrections are not included since they are d i f f i c u l t to calculate. Absorption in the sample i t s e l f and thermal vibrations of the crystal 's atoms w i l l alter the observed intensit ies . These effects are not negligible and must be considered when comparing calculated intensities to experimental data. Both effects are re lat ive ly constant over small ranges in 9, so agreement between theory and experiment is generally deemed satisfactory i f the observed relat ive intensities of 15 closely-neighboring lines f i t the calculation. The relat ive observed intensities of two lines of large angular separation is not expected to agree well with calculations using equation (2-4). 2.3 Debye-Scherrer method Powder X-ray patterns are obtained using a cy l indr ica l camera as shown in F i g . 7. The specimen is rod-shaped and is mounted on the axis of the camera. The diameter of the specimen is kept below 0.3 mm so that the di f fract ion lines w i l l not be too broad. The sample is usually rotated to increase the number of grain orientations presented to the X-ray beam. Film is mounted along the circumference of the camera. The diffracted beams come off the sample as a set of cones and create patterns as shown in F i g . 8. The Bragg angle of each l ine is measured and converted to d values. To determine a new sample structure, f i r s t the unit c e l l i s found by f i t t ing d spacings of t r i a l unit ce l l s to those observed. Atomic arrange-ments are then s imilarly determined by f i t t ing the calculated intensities of t r i a l structures to the rough visual estimates made of the fi lm l ine intensit ies . Similarit ies to known structures simplified the determination of the intercalated MoS^ structures. If their characteristic X-ray spectra are already known, other compounds can easily be identified by matching the patterns . Decomposition products present in the batteries were identified in this way. 16 Fi g . 7 - Debye-Scherrer camera (side view) Specimen i s rod-shaped, < 1 cm long, < 0.3 mm i n diameter. I t i s mounted and rotated on the axis of the camera. Camera diameter = 114.83 mm Entrance aperture i s 1.0 mm i n diameter. 18 3. Experimental Procedure The study of the L i x MoS2 intercalation compounds by X-ray di f fract ion techniques is difficult'because these materials are unstable in a ir and because a self-recharge mechanism is associated with their cathodic preparation. The ins tabi l i ty of the Li^MoS^ compounds in a ir required that a l l steps involved in the preparation of the samples used in the X-ray studies be carried out in an inert atmosphere, and that the actual sample be sealed in a thin wall glass capi l lary . A further d i f f i cu l ty arises due to the self recharge characteristics of the L i MoS„ cathodes. Self-recharge refers to the tendency (observed in a l l cathodes studied) to show a r ise in the open c ircu i t voltage upon standing for a period of time. This voltage r ise may result from an equi l ibrizat ion of the ( i n i t i a l l y non-uniformly intercalated) lithium or i t may arise from an unidentified side reaction at the cathode-electro-lyte interface which results in a slow depletion of the intercalated lithium. It has proved impossible to avoid this d i f f i cu l ty and hence X-ray measurements had to be made immediately after the cathodic prepara-tion of the sample. Many t r i a l runs were necessary before an acceptable technique was developed and in total over 100 X-ray exposures were taken in support of the data presented in this thesis. 3.1 Cathode, preparation M0S2 cathodes were prepared on aluminum or platinum f o i l substrates. Superfine MoS„ powder suspended in o i l * was baked onto the degreased * Supplied by Molybond Laboratories, Australia 19 substrate for roughly 20 minutes at 575°C in a nitrogen gas flow. The o i l comes off as vapor and is carried away by the nitrogen gas. It was found that baking at such high temperatures was necessary to remove the heaviest tars which otherwise coated the M0S2 grains and prevented intercalation. Upon slow cooling (=20 minutes), the M0S2 adhered to platinum but not to aluminum substrates, presumably because of the greater thermal expansion of aluminum. In order to prevent peeling of the M0S2 on aluminum substrates, the substrates were i n i t i a l l y annealed for 20 minutes at 575°C and a small amount of oxygen gas was added to the nitrogen gas flow (Fig. 9). MoC^ is formed (determined by X-ray analysis) by incomplete oxidation of the MoS2« An M0O2 coating of the M0S2 grains has been found to be sufficient for good mechanical and e lec tr ica l contact on aluminum substrates. Approximately 4 cm2 of heavy duty aluminum f o i l (6 mg/cm2) were coated per bake. Only about this much cathode material could be made at a time since MoO^ formation was not uniform along the length of the preparation tube. The presence of small amounts of M0O2 introduces complications in the analysis of X-ray data, but cathodes prepared on aluminum f o i l in this manner had a slower self-recharge rate than cathodes with platinum backing, especially at low c e l l potentials. Consequently, aluminum backed cathodes were used for a l l measurements taken at potentials < 2 vol ts . The rat io of M0O2 to M0S2 was generally = 10-20% by weight*, certainly not negligible. To reduce * Determined by rough visual comparisons of the relat ive intensities of X-ray l ine patterns of M0O2-M0S2 mixtures of known composition. The M0O2 in these standard mixtures was present as individual grains, not as a coating of the MoS„ grains. The grains are suff ic iently small (=1 micron) so that X-ray penetration is f a i r l y uniform. Absorption of the diffracted beam in the grain w i l l make the M0O2 content of a coated grain appear too high according to the standard. But, M0O2 also acts as a battery, and the weight of M0O2 present can be determined "from i t s capacity. M0O2 content as measured by this method approximately agrees with that measured by X-rays. Apparently, absorption of X-rays under the grain surface is not s ignif icant. Furnace Quartz tube Water Quartz plate M o S 2 + o i l f f 0, F i g . 9 - Apparatus used for cathode preparation on aluminum substrates. (Platinum backed cathodes were prepared in a similar fashion only no oxygen was used.-)- . Dimensions: glass sl ide - 3" x 1" x 1. mm quartz plate - 4 cm x 15 cm x 6 mm quartz tube - 115 cm long, 4.5 era O.D., 4.15 cm I.D, furnace — 80 cm long distance from mixing point of ^ + 0^ to quartz tube - 75 cm Flow rates: - 800 cc/min. 0^ - 0.2 cc/min. for the f i r s t 3 minutes of bake 21 inhomogeneity of the sample, thin layers of MoS2 were applied on the substrate (0.3-0.5 mg/cm2). Cathodes made i n this way are quite porous with a density of = 1.5 g/cm3. The density of MoS2 i s 4.8 g/cm3. 3.2 C e l l construction C e l l s were assembled i n a Vacuum Atmospheres glove box under an argon atmosphere. (< 1 ppm water and oxygen was obtained by recirculating the box atmosphere through a Hydrox p u r i f i e r . ) Every c e l l contained two cathodes which were discharged i n p a r a l l e l . One was to be used for the X-ray sample, while i t s twin was used as a dummy i n order to monitor the voltage r i s e indicative of the self-recharge. The substrate of each cathode was 2 x 1 cm2 in size, with 1 cm2 on one side coated with frfc^. Thus each cathode held = 0.3-0.5 mg of hfc^. The twin cathodes were mounted as shown i n Fig. 10, with the M0S2 coated sides facing the lithium anode. The anode consisted of lithium f o i l pressed onto a nickel screen. The el e c t r o l y t e used was either a 1 M LiClO^ i n PC solution or a 0.7 M LiBr i n PC solution. Both salts are hydroscopic so each was fuse dried (heated under vacuum u n t i l melted) to remove any water. The propylene carbonate (Eastman Kodak) was twice d i s t i l l e d under vacuum and was passed through several columns of molecular sieve and activated alumina to remove water and organic impurities. (Remaining amounts of impurity, notably g l y c o l , are under suspicion as the cause of a self-recharging side reaction.) Solder connections to the electrodes were made to wires above the l e v e l of the electrolyte since the solder decomposed i n the solution. The entire assembly i s enclosed in a glass beaker ( I V diameter, 2" high) 22 MoSo Lithium 0.7 M LiBr/PC or foil | M LiCI0 4/PC F i g . 10 - Twin cathode c e l l assembly (side view of electrodes) 23 w i t h connections running through the neoprene stopper. ( B l a c k rubber stoppers r e a c t w i t h PC and e v e n t u a l l y crumble.) 3.3 X-ray specimen p r e p a r a t i o n C e l l s were a u t o m a t i c a l l y c y c l e d t o the a p p r o p r i a t e p o i n t oh the d i s c h a r g e curve ( F i g . 4) u s i n g a PAR coulometry c e l l system as a constant c u r r e n t supply. Each cathode was c y c l e d a t a constant c u r r e n t of 100 uA, thus the c u r r e n t d e n s i t y was always 100 yA/cm 2. I n order to minimize any s e l f - r e c h a r g i n g , the X-ray a n a l y s i s was performed as soon as p o s s i b l e a f t e r the end of the d i s c h a r g e . F i r s t , the t w i n cathodes were r a i s e d above the e l e c t r o l y t e l e v e l to minimize e l e c t r o l y t e c o n t a c t w i t h the cathodes. The c e l l was then t r a n s f e r r e d to the g l o v e box. T h i s i n v o l v e s pumping down the c e l l c o n t e n t s , but s i n c e PC has a low vapor p r e s s u r e , the cathodes remain-ed wetted w i t h e l e c t r o l y t e . I n the argon atmosphere, the M 0 S 2 on the t e s t cathode was scraped o f f the s u b s t r a t e and onto a g l a s s s l i d e , g i v i n g one a puddle of i n t e r c a l a t e d M 0 S 2 and e l e c t r o l y t e . A t i n y open-ended Pyrex tube ( l e n g t h < 1 cm, w a l l t h i c k n e s s = 0.02 mm, O.D. = 0.2 mm) was used to s e a l the X-ray sample i n an argon atmosphere. ( F i g . 11) When dipped i n t o the puddle, the tube loads i t s e l f w i t h sample through c a p i l l a r y a c t i o n . The tube i s only p a r t i a l l y f i l l e d w i t h sample s o l u t i o n i n order to a l l o w the ends to be sea l e d w i t h hard p l a s t i c i n e . I t i s necessary to s t a r t w i t h a t h i c k s l u r r y of sample s o l u t i o n i n order to reasonably f i l l the tube. A f t e r removal from the glove box, the sample tube i s cleaned e x t e r n a l l y with, acetone to remove the adhesive l e f t behind from the s c o t c h tape. (The tube i s taped to a s p a t u l a f o r easy manipulation.) The s a l t L i B r or L i C l O ^ i s d i s s o l v e d i n PC and consequently does not c o n t r i b u t e any d i f f r a c t i o n 24 Spatula Sample tube Scotch tape Glass slide Thick slurry of L i x M o S 2 and electrolyte Fig. 11 - X-ray specimen preparation 25 l i n e s . Some s c a t t e r i n g of X-rays due to the g l a s s can be seen. ( F i g . 15) 3 . 4 X-ray procedure The sealed specimen was centered i n a 1 1 4 . 8 3 mm diameter P h i l l i p s Debye-Scherrer camera having a 1.0 mm entrance aperture. Copper Ka r a d i a t i o n (A = 1 .542 A) was used at s e t t i n g s of 35 KV and 25 mA. The process of preparing the X-ray specimen takes roughly hour mainly due to t r a n s i t time i n t o the glove box. The X-ray run l a s t s \\ hours, so s e l f - r e c h a r g i n g times are l i m i t e d to - 2 hours. However, as w i l l be seen, at the lowest v o l t a g e s on the discharge curve, cathodes s e l f - r e c h a r g e so q u i c k l y that s i g n i f i c a n t u n c e r t a i n t i e s are introduced i n the measure-ments. The amount of s e l f - r e c h a r g e i s determined by measuring the open c i r c u i t v o l t a g e of the dummy cathode when the X-ray run i s f i n i s h e d . The dummy remains suspended above the e l e c t r o l y t e u n t i l t h i s v o l t a g e i s measured. 26 4. Results and Discussion Diffraction patterns were taken at various points on the voltage prof i le in F i g . 4. Within experimental error, both phase 1 and 2 L^MoS^ can be indexed to hexagonal la t t ices . The layer spacing (c/2 in a 2 layer unit cel l ) and the a axis in these phases have been measured versus c e l l voltage. The major structural change associated with the phase 1 -> 2 transition has been identif ied. The phase 3 structure appears almost amorphous, probably due to non-uniform straining of the la t t i ce . Also, some decomposition products of the c e l l have been discovered. X-ray patterns representative of the various points on the voltage prof i le appear in Appendix A along with their measured l ine spectra. The X-rays should uniformly*'penetrate the small (-1 micron) cathode grains, implying that the data provided in these patterns applies to the bulk material and not just to the surface. Line broadening is apparent in phase 2 and especially in phase 3 patterns. The cause of this seems to be non-uniform straining of the la t t i ce . Further discussion on l ine broadening appears in Appendix B. The uncertainties in the data due to this and other effects are examined in Appendix C. * An example of X-ray penetration in a copper sample is given in Cul l i ty (1956). Using unfiltered Cu radiation, 95% of the recorded information applies to a depth of 25 microns, but 50% of that information originates in the f i r s t 5 microns. The linear absorption coefficients of Cu Ka radiation in Cu and Ko are = 470 and = 1670 cm - 1 respectively. Thus similar penetration is expected in - 1.5 microns of Mo as in - 5 microns of Cu. 27 4.1 Phase 1 (A - B in F i g . 4) o o 2H-M0S2 i s hexagonal with c and a axes of 2 x 6.15 A and 3.16 A 0 respectively. The sandwich height is 3.19 A and the Van der Waals gap 0 height is 2.96 A (See Figs. 1 and 2 ) . Cathodes of ce l l s cycled in phase 1 v isual ly appear no different than ordinary M0S2, remaining gray in colour. X-ray spectra of discharged phase 1 ce l l s (Fig. 15a) are identical to that of 2H-M0S2. However, the fact that ce l l s can be cycled in phase 1 implies that lithium must be going to the cathode. It is unlikely that the lithium is merely coating the M0S2 grains since this would result in a c e l l voltage of approximately zero vol ts . Moreover, a coating corre-sponding to x =0.5 worth of lithium would be quite thick and should be detectable even v isual ly . Lithium probably intercalates even in phase 1, perhaps just in a surface layer. It would seem unlikely that the X-ray spectrum would remain unchanged i f the bulk material were intercalated with so much lithium. For example, the spectra of VS2 and L i ^ 3 3 ^ 2 a r e quite different (D.W. Murphy, C. Cros, et a l 1977). However, the octahedral sites in the sandwich gaps should be roughly large enough to accomodate lithium ions*. It may be possible then to f i l l - % of the octahedral sites without s ignif icantly distorting the la t t i ce . Another poss ib i l i ty is that the lithium is concentrated just beneath the surface of the M0S2 grains. Such a thin shel l could go undetected in a di f fract ion pattern. * Assuming that the sulfur atoms in a sandwich are close-packed spheres, the octahedral sites in 2H-MoS„ are large enough to accomodate a sphere of radius 0.77 A. The radius or a lithium ion as given "in the Handbook of Chemistry and Physics (1969) is 0.68 A. A similar calculation implies that the tetrahedral sites are too small, being able to accomodate spheres of radius 0.46 A. 28 4.2 Phase 1 + 2 transition (B - C in F i g . 4) A colour and structure change are associated with this transit ion. The cathode turns from gray to black and from a phase 1 to a phase 2 type structure. Phase 2 L i MoS„ is believed to be either a 2H-NbS_ or a lT-TiS„ x 2 2 2 type structure. The phase 2 spectrum at point C in F i g . 4 f i t s that of a hexagonal o o la t t i ce with c = 6.33 A and a = 3.34 A. The poss ib i l i ty that c is some o greater multiple of 6.33 A cannot be ruled out from the present data, but no lines have been seen that would require a larger unit c e l l . Intensity calculations of a M0S2 layer compound with these parameters and the common stacking sequences appear in Appendix D. The observed l ine intensities f i t structures where the Mo atoms stack on top of each other. The NbS2 and T i S 2 structures, with stacking sequences of /AbA CbC/ and /AbC/ respec-t ive ly , are the most l ike ly candidates*. (The quality of the X-ray patterns is too poor to identify the sulfur atom coordination or the number of layers per unit c e l l . For example, i f the structure is similar to 2H-NbS2, the {101} di f fract ion l ine should be relat ively strong. No such l ine is seen. But, theory predicts i t should be fainter than i t s two close neigh-bours, {100} and {102}. A broad {101} l ine may be washed out by i t s stronger near neighbours. An idea of the agreement that can be expected with a sharp dif fract ion picture is i l lustrated with the observed and calculated l ine intensities of 2H-MoS2 in Appendix D.) The NbS„ structure can be obtained from the MoS„ structure by shifting * More complicated stacking sequences, with some layers octahedrally coordinated and some trigonally coordinated, have been reported in the l i terature (Wilson and Yoffe 1969), but are not as common. the sandwich layers by , so that the Mo atoms a l ign . A transition to the TiS2 structure would also require reordering the sandwich layers themselves. (Landuyt et a l (1976) postulate that in TaS 2 a sulfur plane can s l ip with respect to the host sandwich layer.) Intuit ively the NbS2 structure might seem more l ike ly for phase 2 since less rearrangement of the crystal la t t ice is required. However, transitions from 2H-MoS2 type -> l T - T i S 2 type structures have been reported in the l i terature (Landuyt et a l 1976) as have 2H-MoS2 type 2H-NbS2 type structural transit ions. The austenite to martensite transition in steel also involves shifts of la t t i ce planes, as the crystal structure goes from face-centered cubic to body-centered cubic (Robert E. Reed-Hill 1973). Line broadening is a common feature associated with this transit ion. Uniform and non-uniform stresses exist in the steel; the former causing shifts in the di f fract ion l ine positions, and the latter causing l ine broadening. X-ray spectra taken at points along BC show a mixture of the two phases, dominated by the sharp, strong lines of phase 1. Generally only o o the 6.33 A l ine of phase 2 shows up c lear ly , and i t overlaps the 6.15 A ( {002} ) l ine of phase 1 creating what appears to be a very broad l ine o centered at = 6.24 A (Fig. 15b). Close to point C in F i g . 4, traces of other broad phase 2 l ines become v i s i b l e . The phase 1 2 transition is reversible, but only at slow recharge rates. This may be due to poor lithium diffusion in the sandwich gaps near point D (Fig. 4) in phase 2. The unit c e l l here is much smaller than elsewhere in phase 2 ( See Sect. 4.3) presumably with a much smaller sandwich gap. Trickle charging converts a phase 2 c e l l from point D to a strained version of a phase 1 cathode. X-ray spectra taken of these 30 reconverted cathodes resemble s l ight ly strained 2H-M0S2 (Fig. 15i) . The l ine broadening seen in phase 2 is much reduced after the reverse trans i -t ion, but the pattern is s t i l l not as dist inct as that of the original 2H-M0S2. A S U D S e c l u e n t discharge of such a cathode reveals a plateau-like feature in the voltage prof i l e , similar to the BC plateau in F i g . 4, except that i t occurs at a higher voltage (- 1.4 V) . At the end of this plateau-l ike feature, the c e l l has converted back into phase 2. 4.3 Phase 2 (D - C - E in F i g . 4) Structural changes of L i MoS„ appear to be f a i r l y symmetrical over a cycle in phase 2. Changes occur in the size and apparently also in the positions of atoms in the unit c e l l of a cathode run in phase 2. Possibly even the shape of the unit c e l l i s altered. There are two kinks in the phase 2 voltage prof i le (Fig. 4), one at = 2 V and the other at 2.7 V. X-ray spectra of cathodes run between the kink at 2 V and point E (0.5 V) are s imilar, with differences arising only because of small changes in the c and a axes of the unit c e l l . Between the two voltage kinks (2 V to 2.7 V), an extra faint l ine appears in the X-ray patterns. Assuming a 2 layer unit c e l l , this l ine can be indexed as {105}. Between point D and the upper voltage kink (3 V to 2.7 V) , this l ine is no longer present. However, another faint l ine appears in the spectrum. Again assuming a 2 layer unit c e l l , this l ine can be indexed as {103}. Also in this pattern, the relat ive intensity of the {100} l ine to the {102} has increased to the point where they are visual ly equally strong. lT-TiS2 type structures have only one layer per unit c e l l . These extra lines cannot be f i tted to such a structure. 2H-NbS2 type structures with la t t ice parameters in the appropriate range are expected to have very faint {103} and {105} l ines . (See calculated pattern in Appendix D.) Also, the {100} l ine is expected to be s ignif icantly fainter-than the {102} l ine . However, strong {103} and {105} l ines are characteristic of structures with alternate Mo atoms aligned, sucn as 2H-MoS2> ^ n C* ^ n s u c ^ structures, the {100} l ine is expected to be more intense than the {102}. F ina l ly , a {106} l ine is expected to be seen in 2H-NbS2 t y p e structures but is not evident in the X-ray pictures. However, this l ine would be located so as o to overlap the phase 2 {110} l ine or the 1.71 A M0O2 l ine throughout most of phase 2. Near point D, though, the {106} ref lect ion should be suff ic iently clear of these other l ines , and i t is not seen. Faint {106} lines are characteristic of 2H-M0S2 t y P e structures. It seems l i k e l y then that as the c e l l . is recharged in phase 2, the cathode i s trying to revert from a NbS2 type structure to i t s or ig inal structure. A decrease in the amount of non-uniform strain is also evident upon recharging, as the sharpness and number of lines in the X-ray patterns increase. Within experimental error, phase 2 L i x MoS2 can be indexed to hexagona l a t t i c e s . I t is possible that the shape of the unit c e l l i s s l ight ly distorted. For example, is hexagonal and has a structure similar to l T - T i S 2 . D.W. Murphy, C. Cros, .et a l (1977) have discovered two different monoclinic phases of L i VS„, a and 3, occuring in ranges near x = 0.33 and x = 0.5 respectively. Both monoclinic phases are nearly hexagonal. Possibly our L i x MoS2 system is s imilar. The phase 2 spectra with the extra lines may be indicative of phases with different la t t i ce systems. On the assumption that the unit c e l l remains hexagonal throughout phase 2, the layer spacing (c/2 in a 2 layer unit cel l ) and the a axis have been plotted versus the phase 2 voltage prof i le in F i g . 12 and versus c e l l voltage in F i g . 13. (An explanation of the experimental o > < LU \— O CL LU O 0 6.35 o <t CL CO 6.25 OC LU > < 1 1 6.15 3.35 o < CO X 3.25 < o 3.15 PHASE 2 VOLTAGE PROFILE 32 RECHARGE I DISCHARGE (coulombs) i ^ » 2H-MoS. . < — ' 2H-MoS 2 Fig. 12 - Layer spacing and a axis plots vs. phase 2 voltage p r o f i l e ( a axis sometimes not available with corresponding layer spacing ) o o V e r t i c a l error : layer spacing ± 0.03 A, a axis ± 0.02 A 6.45 6.35 O o CO £ 6 . 2 5 < 6.15 33 o < 3.35 co X < 3.25 3.15 1 2 3 2 I PHASE 2 CELL VOLTAGE RECHARGE < 1 > DISCHARGE 2H-MoS 2 — < — \ 2H-MoS 2 2 3 2 1 PHASE 2 CELL VOLTAGE RECHARGE <- ^D ISCHARGE F i g . 13 - Layer spacing and a axis plots vs. phase 2 c e l l voltage (a axis sometimes not available with corresponding layer spacing) o o V e r t i c a l error : layer spacing + 0.03 A, a axis ± 0.02 A 34 u n c e r t a i n t i e s i s g i v e n i n Appendix C.) There i s a maximum i n the l a y e r o s p a c i n g , = 6.40 A, o c c u r i n g a t c e l l v o l t a g e s near the k i n k i n the v o l t a g e p r o f i l e a t 2 V. Coulombic t r a n s f e r i m p l i e s a c h e m i c a l c o m p o s i t i o n of Li.jMoS 2 a t t h i s p o i n t . T h i s agrees w i t h Whittingham (1975), who determined o t h a t the c a x i s i n L i j M o S 2 i s some m u l t i p l e of 6.40 A. There i s a sharp decrease i n the l a y e r spacing above 2 V. At the top of phase 2 ( p o i n t D i n F i g . 4 ) , the l a y e r spacing has a p p r o x i m a t e l y r e t u r n e d t o i t s o r i g i n a l phase 1 v a l u e . Below 2 V, the decrease i s not as d r a s t i c , r e a c h i n g a o minimum of - 6.30 A a t the low v o l t a g e end of phase 2. ( T h i s behaviour has been noted i n other systems. For example, i n T i S 2 , the c a x i s i s a l s o seen to decrease w i t h i n c r e a s i n g amounts of i n t e r c a l a t e . For the i n t e r -c a l a t e d compounds Rb T i S _ and Cs T i S _ , c decreases f o r x > 0.42 and r x 2 x 2' x > 0.56 r e s p e c t i v e l y (J.Bichon et a l 1973)). The a a x i s , on the other hand, i n c r e a s e s m o n o t o n i c a l l y w i t h the amount of i n t e r c a l a t e . I t d e f i n i t e l y does not r e t u r n to i t s o r i g i n a l phase 1 v a l u e a t the top of phase 2. (Confirmed by the d i s p l a c e d {110} l i n e . ) There may be d i s c o n t i n u i t i e s i n the v a r i a t i o n of the l a y e r spacing and a a x i s w i t h c e l l v o l t a g e . However, due t o the e x p e r i m e n t a l u n c e r t a i n t y , more s t a t i s t i c s or b e t t e r r e s o l u t i o n are necessary t o c o n f i r m t h i s . Abrupt changes i n the u n i t c e l l s i z e may be a s s o c i a t e d w i t h the k i n k s i n the v o l t a g e p r o f i l e . P l o t s of the l a y e r spacing x a 2 and l a y e r spacing/a versus the phase 2 v o l t a g e p r o f i l e appear i n F i g s . 14a and b r e s p e c t i v e l y . Assuming a 43 hexagonal u n i t c e l l , the volume of the u n i t c e l l i s g i v e n by -- x the l a y e r spacing x a 2 x number of l a y e r s per u n i t c e l l . I t would appear from F i g . 14a that the u n i t c e l l volume g r a d u a l l y i n c r e a s e s w i t h i n c r e a s i n g amount of i n t e r c a l a t e . At the bottom of phase 2 ( p o i n t E i n F i g . 4 ) , the 35 O > < I -z: LU H O CL LU O ro, . LU ° < 0 S .E 3 ^ 7 70 o L U _ , > LU i=<~> < LU CC I 65 Q. </> CU ^ 60 o \ CD o r5 CO c r LU 5 l.95r .90 1.85 PHASE 2 VOLTAGE PROFILE RECHARGE I DISCHARGE (coulombs) 2H-MoS 2 (a) 2H-MoS 2 (b) F i g . 14 - a) Relative unit c e l l volume and b) Layer spacing/a plots vs, phase 2 voltage prof i le o o Vert ica l error : layer spacing ± 0.03 A, a axis ± 0.02 A cathode volume has apparently increased by 15% over i t s or ig inal value. The minimum phase 2 volume remains a few per cent larger than that of the original 2H-M0S2. ^ & ^ - a ^ e r s P a c i - n g /a ratio in phase 2 i s never too far removed from the 2H-M0S2 value of 1.945. There is a peak in the curve occuring near where the layer spacing i t s e l f is a maximum. 4.4 Phase 2 + 3 transition (E - F in F i g . 4) and Phase 3 (C - F in F i g . 4) X-ray spectra of phase 3 cathodes appear almost amorphous (Fig. 15h). Only one very broad band at a low Bragg angle is ever v i s i b l e . This band 0 0' extends from a d spacing of = 6.3 A to ^ 12 A (the l imit of the f i lm) . The featureless pattern may be due to the presence of severe non-uniform strains in the crystal lattice.(See Appendix B).X-ray spectra taken at points along EF s t i l l exhibit the phase 2 spectrum of point E . So far , no reversal of the transition has been observed in the di f fract ion patterns. Cells may have reached their maximum capacity at point F . Here, enough lithium (x = 3) has been intercalated to f i l l a l l the sites in the sandwich gaps. 4.5 Decomposition products Discharged cathodes are unstable in a i r , presumably* forming hydrated complexes similar to those in the L i -T iS_ system studied by Whittingham * X-ray spectra of discharged cathodes exposed to a i r were s i m i l a r to that of a cathode washed i n water. (1974) . X-ray patterns of such cathodes upon exposure to a ir bore s imilar-i t i e s to those of hydrated L i x T i S 2 . Occasionally the compound L^MoO^ was also present, but only in those more deeply discharged samples, where presumably the lithium content was now high enough to allow I^MoO^ forma-t ion. There is also evidence of L ^ S formation in cathodes discharged into phase 3. In some X-ray patterns, two lines have been seen which f i t the two strongest l ines of the H 2 S spectrum. The formation of is a poss ib i l i ty according to free energy considerations. A l l cathodes in phase 2 or 3 smell of H S^ upon exposure to a i r . This, however, may be due to decomposition of the cathode to Li 9MoO, and not the decomposition of 38 5. Summary The M0S2 cathode undergoes two phase transitions upon discharging. In the f i r s t transit ion, the sandwich layers shift such that the Mo atoms stack above one another. A rearrangement of atoms in the layers themselves is also a poss ib i l i ty . After the transit ion, the crystal la t t i ce is subject to non-uniform stresses which diminish as lithium is removed. This f i r s t transition is reversible to some degree. The original la t t ice structure is regained, but with some non-uniform straining present. L i t t l e is known about the processes involved in the second transit ion, since the phase 3 structure appears almost amorphous. The transit ion may involve greater straining of the crystal l a t t i ce . In the powder photographs, phase 1 cathodes always resemble 2R-HoS^, while phase 3 cathodes seem almost amorphous. The unit c e l l in phase 2, however, is seen to vary throughout a cycle implying that intercalation takes place. The intercalation process appears completely reversible, since X-ray spectra taken at corresponding points on the discharge and recharge voltage curves are ident ical . The layer spacing reaches a maxi-mum in the middle of the phase whereas the a axis increases monotonically with discharge. Discontinuities, corresponding to the kinks in the voltage prof i l e , may exist in both the layer spacing and the a axis. The volume of the unit c e l l (proportional to the product - layer spacing x a 2) gradually increases with increasing lithium content to reach a maximum value that is 15% larger than the or ig ina l . The minimum phase 2 value is s t i l l a few per cent larger than the original M0S2, because although the layer spacing appears to return to i t s original value, the a axis remains 39 somewhat larger than before. Unless exposed to air., no decomposition of phase 2 cathodes i s evident i n the X-ray patterns. Phase 3 cathodes may decompose to form Li„S. 40 Bibliography Ashcroft, N.W., and Mermin, N.D. 1976. Solid State Physics, Holt, Rinehart, and Winston, New York. Bichon, J . , Danot, M . , et Rouxel, J . 1973. C R . Acad. Sc. Paris , 276, 1283. Chiane l l i , R.R. 1976. Journal of Crystal Growth, .34, 239. C u l l i t y , B.D. 1956. Elements of X-ray Diffract ion, Addison-Wesley, Reading. Klug, H . P . , and Alexander, L . 1974. X-ray Diffraction Procedures for  Polycrystall ine and Amorphous Materials, 2nd Edit ion, J . Wiley & Sons, New York. Landuyt, J . Van, Tendeloo, G. Van, and Amelinckx, S. 1976. Phys. Stat. S o l . , A 36, 757. McClune, W.F. 1977. Powder Diffraction F i l e , JCPDS International Centre for Diffraction Data, Swarthmore. Murphy, D.W., Carides, J . N . , DiSalvo, F . J . , Cros, C , and Waszczak, J . V . 1977 . Mat. Res. Bull.,-1_2, 825. Murphy, D.W., Cros, C , DiSalvo, F . J . , and Waszczak, J . V . 1977. Inorg. Chem., Vo l . 16, No. 12, 3027. Reed-Hil l , Robert E . 1973. Physical Mettalurgy Principles , 2nd Edit ion, D. Van Nostrand Co. , Princeton. Schafhautl, C. 1841. J . Prakt. Chem., 21_, 129. Weast, R.C. 1969. Handbook of Chemistry and Physics, 50th Edit ion, The Chemical Rubber Co. , Cleveland. Whittingham, M.S. 1974. Mat. Res. B u l l . , 9_, 1681. Whittingham, M.S . , and Gamble J r . , F.R. 1975. Mat. Res. B u l l . , LO, 363. Whittingham, M.S. 1976. J . Electrochem. S o c , 123, 315. Wilson, J . A . , and Yoffe, A.D. 1969. Advances in Physics, 1_8, 193. Appendix A Examples of X-ray spectra X-ray patterns representative of the various points on the c e l l voltage prof i le appear in F i g . 15. Common to a l l the pictures is a diffuse band in the forward scatter region. This is caused by X-rays scattering off the glass specimen container. Also present in the patterns are l ine spectra of various L^MoG^ compounds. (These spectra a l l exhibit sharp dif fract ion l ines , making them easily distinguishable from phase 2 and 3 patterns.) Those L i Mo0„ spectra showing up in the patterns in F i g . 15 are presented in F i g . 16. The measured l ine spectra from a l l these pictures are also presented in this Appendix. Ten X-ray patterns are shown in F i g . 15, and they are-a) Phase 1 (discharged to 1.1 V) b) Phase 1 + '2 transition (discharged to a point near C in F i g . 4 at 1 V) c) Phase 2 (recharged to 2.65 V) d) Phase 2 (recharged to 2.3 V) e) Phase 2 (discharged to 0.8 V) f) Phase 2 (discharged to 0.5 V) g) Phase 2 + 3 transit ion (cycled in phase 3; incomplete transition from phase 2 + 3 ) h) Phase 3 (cycled in phase 3) i) Phase 2 + 1 transit ion (by self-recharging) j) Phase 2 + ? (phase 2 c e l l washed in methanol) Fig. 15 - X-ray pictures of M0S2 cathodes at various stages of discharge (See Text) 43 The l ine spectrum of 2H-M0S2 ^ s e e F:i-S* 8) as determined by the National Bureau of Standards is given in Table 1. Table 1 Line spectrum of 2H-MoS2 0 1 d (A) {hkl} 100 6.15 002 4 3.075 004 16 2.737 100 9 2.674 101 8 2.501 102 45 2.277 103 14 2.049 006 25 1.830 105 4 1.641 106 11 1.581 110 12 1.538 008,112* 2 1.4784 107 2 1.3688 200 4 1.3401 108 5 1.2983 203 4 1.2513 116 2 1.2295 0010' 1 1.2224 109 4 1.1960 205 etc. * The {112} index was not printed on the PDF data card (McClune 1977). The following data was determined from the pictures in F i g . 15. The observed intensities given are visual estimates of the relat ive l ine strengths. In the-following tables, VS = Very Strong S = Strong M = Moderate F = Faint VF = Very Faint 0 A l l values of d are given in A. The calculated d spacings were determined assuming the crystal la t t i ce remained hexagonal. (See Sect. 2 for theory.) 44 Table 2 Line spectra of MoS0 cathode patterns in F i g . 15 a) Phase 1 ^obs. d , -obs. Identification VS 6.17 {002} F 3.56 LixMo02 (s S 2.74 {100} M 2.68 {101} M 2.51 {102} F 2.35 Aluminum S 2.28 {103} Remaining lines are those of 2H-MoS2» Some substrate can be scraped off while preparing the X-ray specimen. Aluminum lines were quite common in the spectra. b) Phase 1 -> 2 transition ''"obs. d u obs. Identification S broad 6.28 {002} Phase 1 and Phase 2 F F 3.44 3.14 superimposed LixMo02 (see F i g . 16a) {004}. Phase 1 and Phase 2 VF'. 2.86 superimposed {100} Phase 2 S 2.74 {100} Phase 1 M 2.68 {101} Phase 1 VF * 2.60 {102} Phase 2 M 2.51 {102} Phase 1 F S 2.43 2.28 ..LixMo02 {103} Phase 1 VF = 2.11 {104} Phase 2 Remaining lines are those of 2H-MoS 45 c) Phase 2 (c = 2 x 6.17 A, a = 3.24 A) I , obs. . d , obs. calc . Identification S 6.17 6.17 {002} M 3.43 LixMo02 (see Fig VF 2.99 L i C l M broad 2.78 2.81 {100} M broad 2.56 2.55 {102} M 2.43 LixMo02 VF ^2.28 2.32 {103} F V broad =• 2.07 2.08 {104} F broad 1.71 LixMo02 F broad 1.62 . 1.62 {110} F V broad = 1.55 1.57 {112} 16a) L i C l i s an insoluble impurity found in the LiBr/PC electrolyte. d) Phase 2 ( c ''"obs. -2.x 6.40 A,. a = 3.28 A) S F VF M S V broad F M V broad VF F F F broad obs. 6.40 3.43 3.20 2.84 2.60 2.44 2.13 1.89 1.72 1.64 1.59 calc, 6.40 3.20 2.84 2.60 12 90 71 64 59 Identification {002} LixMo02 (see F i g . 16a) {004} {100} {102} LixMo02 {104} {105} L i x M o 0 2 , {106} {110} {112} e) Phase 2 (c = 2 x 6.33 A, a = 3.34 A) "^obs. d V obs. d i calc . Identification S 6. .33 6. 33 {002} VF * 3. .50 LixMo02 (see Fig VF * 3. .17 3. 17 {004} M broad 2, ,89 2. 89 {100} S V broad * 2, .64 2. 64 {102} F * 2, .43 Li x Mo0 2 M V broad « 2, .14 2. 14 {104} M broad 1. .68 1.70, 1.67 {106},{110} VF - 1, .58 1. 62 {112} 46 f) Phase 2 (c = 2 x 6 o o ,31 A, a = 3.36 A) *obs. d u obs. d i calc . Identification S VF M broad S broad F M V broad F broad 6.31 3.65 2.92 2.64 2.44 = 2.09 1.69 6.31 2.91 2.64 2.14 1.70, 1.68 {002} LixMo02 (see F i g . 16d) {100} {102} LixMo02 {104} {106},{110} g) Phase 2 3 transit ion obs. obs. Identification F F VF F F VF 30 43 27 89 63 42 {002} Phase 2 (Table 2e) LixMo02 (see F i g . 16a) L i 2 S ? {100} {102} LixMo02 h) Phase 3 obs. F ring F obs Identification 6.3 -> upper l imit of film 3.44 {002} 11^002 (see F ig . 16a) i) Phase 2^-1 transit ion Spectrum is that of 2H-M0S2 except that the lines are broader than usual. This i s suspected to be due to some non-uniform strain s t i l l present 0 0 0 in the la t t i ce . L i x Mo02 l ines appear here at 3.43 A, 2.43 A, and 1.71 A. 47 j) Phase 2 cathode washed in methanol. o Spectrum is that of 2H-M0S2 with one extra l ine at - 10.3 A, possibly indicative of a superlattice. The lines in this pattern are a l l sharp, implying that the phase 2 l ine broadening is not due to small c rys ta l l i t e size. M 0 O 2 cathodes were prepared by part ia l ly reducing MoO^. Cells were then constructed in the same manner as the M 0 S 2 ce l ls (Sect. 4.2). Patterns of Li^MoG^ compounds were obtained following the same experimental pro-cedure as outlined in Sections 4.3 and 4.4. Mo metal is present as an impurity in most of the samples since some MoO^ is completely reduced in the cathode preparation process. The spectra of four different Li^MoC^ compounds are shown in F i g . 16. The M 0 O 2 ce l l s apparently also undergo a self-recharge, but possibly at a different rate than M 0 S 2 c e l l s . As can be seen, the voltages corresponding to the LixMoC>2 spectra in F i g . 16 do not always match those corresponding to spectra in F i g . 15. This may be due to inhomogeneity in the sample during the self-recharging process. The observed l ine spectra of those patterns in F i g . 16 are given in Table 3. 48 F i g . 16 - Li^MoO,, spectra a) Mo02 x = 0 b) Li x Mo0 2 (discharged to 1.6 V) c) Li x Mo0 2 (discharged to 1.5 V) d) L i Mo00 (discharged to 1 V) 49 Table 3 Line spectra of Li^MoCv, patterns in F i g . 16 obs. obs. b) obs. obs. VS S M F VF F M M broad F F } 3.43 2.44 2.42 2.23 2.17 1.84 overlap 1.71 1.53 1.40 Mo metal F S S F F VF M F 3.57 3.49 2.45 2.39 2.22 -<- Mo 1.78 1.75 1.71 metal + others + others obs. obs. obs. obs. S 3.57 VS 3.66 M 3.46 M 2.61 F 2.53 M 2.56 S 2.44 S 2.45 M broad = 2.22 -<- Mo VF 2.30 F 1.79 metal F 2.21 Mo S 1.76 S 1.83 metal S broad 1.78 + others + others Appendix B Line broadening Several X-ray patterns representative of intercalated M0S2 appear in F i g . 15. The dif fract ion lines in most samples are much broader than in unintercalated M0S2, and w i l l thus appear fainter than narrow lines of the same integrated intensity. This l ine broadening, at least in phase 2, is apparently caused by the. presence of non-uniform strain in the crystal l a t t i ce . Stresses arise in the la t t i ce when the material undergoes i t s phase transitions (Sections 4.2 and 4.4). Line broadening can also be due to small crys ta l l i t e size or inhomo-geneity of the sample. If the sample grains are less than = 10~5 cm in size, then there are not enough la t t i ce planes in a grain for complete destructive interference of off-angle diffracted beams. The result is l ine broadening, becoming worse as grain size decreases. An inhomogeneous sample, consisting of crystals of similar but not identical structures, w i l l produce many similar but not identical X-ray spectra that can overlap with the net result being a pattern with broad l ines . These effects have been ruled out as the major cause of l ine broadening in phase 2 from the reasoning to follow. On a macroscopic scale, considerable cracking of layer compounds has been observed upon intercalation (Chianell i 1976). At f i r s t , i t was thought that on a microscopic scale, similar cracking of the cathode material into small grains took place. However, the following experiment demonstrates that such an effect is not the cause of l ine broadening. After undergoing the phase 1 + 2 transit ion, samples exhibit l ine broaden-ing. Washing such a sample in methanol removes much of the intercalated l ithium, and the cathode material reverts from black to i t s original gray colour. The subsequent X-ray pattern is as sharp as that of unintercalated M0S2 (Fig. 15j).(The spectra is that of pure 2H-M0S2 with one extra l ine 0 at 10.3 A, perhaps indicative of a superlattice.) If the la t t i ce were stressed when lithium is intercalated, the removal of the lithium could remove the stress. But, i f the sample sp l i t into very small grains, broad X-ray l ines would be expected from any resulting product too. So far , i t has not been possible to demonstrate that l ine broadening in phase 3 is not due to small crys ta l l i t e size. Inhomogeneous distribution of the intercalated lithium is also un-l ike ly to be the main cause of l ine broadening. Samples from a l l phases have been X-rayed after being allowed to equilibrate for several days with no apparent reduction in l ine breadth. (Cells are capable of high discharge rates in phases 2 and 3. Over several days, the intercalated lithium is expected to have more than adequate time to diffuse evenly in the sandwich gaps.) 52 Appendix C Errors i n Phase 2 measurements Line broadening It i s d i f f i c u l t to determine the exact posi t i o n s of the l i n e s i n phase 2 spectra, because they are so broad. Generally, the broadening gets worse with increasing Bragg angle. The layer spacing, given i n F i g s . 12. and 13, was mainly determined from the {002} r e f l e c t i o n (assuming a 2 layer unit c e l l ) . Other l i n e s i n the pattern are not very s e n s i t i v e to small changes i n the layer spacing and to make matters worse, these l i n e s were quite broad. S i m i l a r l y , the a axis was determined from the {100} r e f l e c t i o n and/or the {110} r e f l e c t i o n , depending on the q u a l i t y of these l i n e s . (The {110} l i n e i s more s e n s i t i v e to changes i n a, but sometimes i t s p o s i t i o n could not be accurately determined. For example, the {106} l i n e i s expected to overlap i t i f the c e l l were discharged to the C - E region i n F i g . 4.) Several measurements of each l i n e were made. The maximum uncer t a i n t i e s expected i n the layer spacing and the o o a axis are = ± 0.03 A and - ± 0.02 A re s p e c t i v e l y . The other broader l i n e s i n the pattern roughly confirm the values determined by these measurements. Other errors i n the l a t t i c e parameter values are n e g l i g i b l e compared to those introduced by l i n e broadening. Self-recharge The self-recharging c h a r a c t e r i s t i c of the c e l l s introduces an uncertainty i n the discharge state as represented by the c e l l voltage. The l i m i t of the c e l l voltage d r i f t was determined from the dummy cathode. This assumes that the material scraped o f f the cathode ( i n a minimum volume of e l e c t r o l y t e ) does not self-recharge f a s t e r than the dummy (suspended above the e l e c t r o l y t e i n the c e l l , but s t i l l wet with a s i g n i f i c a n t amount of e l e c t r o l y t e ) . In fac t , subsequent X-ray patterns from samples run weeks e a r l i e r suggested that the self-recharge (as represented by actual structural change) may be slowed down i n the c a p i l l a r y configuration. The actual discharge state of the X-ray specimen, then, may be quite different from that of the dummy. For th i s reason, the specimen discharge state, as given by the c e l l voltage, was taken to be the i n i t i a l discharge state with an error bar to indicate the amount of d r i f t of the dummy cathode. The self-recharge phenomena causes the greatest uncertainties i n the phase 2 data. Appendix D Intensity calculations Attempts were made to f i t the X-ray spectrum of a phase 2 cathode discharged to point C on the discharge curve in F i g . 4. The calculated patterns for the three most common structures are presented here in Table 4. These are 2H-MoS2, 2H-NbS2> and l T - T i S 2 type structures. The calculated intensities depend mainly on the positions of the Mo atoms (being the strongest scatterers of the X-rays). Thus the calculated spectra of 2H-NbS2 and l T - T i S 2 type structures are quite s imilar, since the Mo atom positions in both are ident ical . Also presented in this appendix is an example of the agreement to be expected between the observed and the calculated l ine intensities (Table 5). In a l l these calculations, the lithium atoms were ignored as they are very weak scatterers. The sandwich height was assumed to be the o : 2H-MoS2 value of 3.19 A. (The gross features of the calculated spectrum are not changed signif icantly even i f the sandwich height is varied by 0. ± 0.15 A. In fact, the layer spacing in phase 2 has only increased - 0.2 from the 2H-MoS2 value.) The l ine intensities were then calculated according to the procedure outlined in Sect. 2.2. The fractional coordi-nates for the atomic positions in the unit c e l l are -2H-MoS? type structure : S - (0,0,0) - (0,0,3.19/c) - (2/3,1/3,1/2) - (2/3,1/3,1/2 + 3.19/c) Mo - (2/3,l/3,3.19/2c) - (0,0,1/2 + 3.19/2c) 2H-NbS„ type structure : S - (0,0,0,) - (0,0,3.19/c) - (1/3,2/3,1/2) - (1/3,2/3,1/2 + 3.19/c) Mo - (2/3,l/3,3.19/2c) - (2/3,1/3,1/2 + 3.19/2c) lT-TiS„ type structure : S - (0,0,0) - (1/3,2/3,3.19/c) Mo - (2/3,l/3,3.19/2c) 56 Table 4 Calculated i n t e n s i t i e s fo r var ious phase 2 s t ructure candidates (At point C on the discharge curve i n F i g . 4) Calcu lated i n t e n s i t i e s d o b s ^ ^"dbs. 2H-MoS2-type 2H-NbS2 type l T - T i S 2 type d c a l c ( A ) -Indie* 2H- type 0 0 12.7 {001} 6.33 S 100 100 100 6.33 {002} 0 0 4.22 {003} 3.17 VF 2 2 2 3.16 {004} 2.89 M 31 20 20 2.89 {100} 18 14 2.82 {101} 2.64 S 17 71 91 2.63 {102} 0 0 2.53 {005} 84 9 2.39 {103} 2.14 M 1 66 65 2.14 {104} 6 6 6 2.11 {006} 40 4 1.90 {105} 0 0 1.81 {007} 1.68 M J 5 17 23 1.70 {106} 22 ,22 22 1.67 {110} 0 0 1.66 {111} 1.62 VF 14 14 14 1.62 {112} 6 6 6 1.58 {008} 0 0 1.55 {113} 2 2 1.53 {107} 1 1 1 1.48 {114} 3 2 2 1.45 {200} 2 2 1.44 {201} 2 9 12 1.41 {202} 0 0 1.41 {009} 0 0 1.39 {115} 6 4 4 1.39 {108} 13 1 1.37 {203} 0 13 12 1.32 {204} 7 8 7 1.31 {116} 1 1 1 1.27 {0010} 1 1 1.26 {109} 10 1 1.26 {205} 0 • 0 1.23 {117} 1 5 6 1.19 {206} S' = Strong a = 3.34 A M = Moderate c = 2 x 6.33 A {001} {002} {100} /{101} \{T01} /{102} \{102} {003} 7{103} 1{I03} {110} {111} {004} {112} {200} ({201} ({201} ({104} l{-104} /{202} \{202} {113} {005} /{203} {203} VF = Very Fa int sandwich height = 3.19 A Table 5 Comparison between observed and calculated l ine intensities in 2H-MoS I* I , d , hkl 0 12.3 {001} 100 100 6.15 {002} 0 4.10 {003} 4 2 3.08 {004} 16 31 2.74 {100} 9 18 2.67 {101} 8 16 2.50 {102} 0 2.46 {005} 45 87 2.28 {103} 14 8 2.05 {006} 1 2.04 {104} 25 37 1.83 {105} 0 1.76 {007} 4 5 1.64 {106} 11 21 1.58 {110} 0 1.57 {111} 12 j ~ 6 1.54 {008} 13 1.53 {112} 2 2 1.48 {107} 0 1.47 {113} 1 1.41 {114} 2 3 1.37 {200} 0 1.37 {009} 2 1.36 {201} 4 6 1.34 {108} 2 1.34 {202} 0 1.33 {115} 5 13 1.30 {203} 4 9 1.25 {116} 0 1.25 {204} 2 1 1.23 {0010} 1 2 1.22 {109} 4 9 1.20 {205} 0 1.17 {117} 2 1.14 {206} etc. O O 0 a = 3.16 A, c = 2 x 6.15 A, sandwich height = 3.19 A National Bureau of Standards data (Powder Diffraction F i l e - McClune 1977) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0085199/manifest

Comment

Related Items