D. LOUER.

Transcription

1 DIVOL06.txt Last change: P 7 Oct :43 pm ========================================================================== ========================================================================== DDDDDDD IIII VV VV OOOOOO LL DD DD II VV VV OO OO LL DD DD II VV VV OO OO LL DD DD II VV VV OO OO LL DD DD II VV VV OO OO LL DD DD II VVVV OO OO LL LL DDDDDDD IIII VV OOOOOO LLLLLLLL ========================================================================== ========================================================================== TWO OPTIONS ARE AVAILABLE WITH DIVOL >> OPTION 0 IS DIVOL04 (OPTIMISED SEARH TO BE USED WITH NORMAL QUALITY DATA) >> OPTION 1 OFFERS AN EXTENDED (EXHAUSTIVE) SEARH (FOR DIFFIULT ASES) [OMPARED TO DIVOL04 (2004) ONLY ONE ADDITIONAL INPUT PARAMETER IS ADDED IN DIVOL06, i.e. OPTION IN ARD 5 (SEE BELOW)] TO EXEUTE THE PROGRAM, ENTER : DIVOL06 THEN, ENTER THE INPUT FILE NAME and OUTPUT FILE NAME PROGRAM FOR THE AUTOMATI INDEXING OF POWDER DIFFRATION PATTERNS BY THE SUESSIVE DIHOTOMY METHOD >>>> THE REFERENE FOR THIS PROGRAM, TO REPORT IN PUBLIATIONS, IS: >>>> (1): A. Boultif & D. Louer, J. Appl. ryst. 37, , THE FOLLOWING RYSTAL SYSTEMS AN BE ANALYZED: UBI, TETRAGONAL, HEXAGONAL, ORTHORHOMBI, MONOLINI AND TRILINI. PREURSOR PROGRAMS: P1 & P2 (REF. 2), DIVOL (REF. 3), DIVOL91 (REF. 4), DIVOL04 (REF. 1) ======================= FOR FURTHER INFORMATION ONTAT D. LOUER Daniel.Louer@free.fr OR A. BOULTIF LABORATOIRE DE RISTALLOGRAPHIE DEPARTEMENT DE PHYSIQUE FAULTE DES SIENES UNIVERSITE MENTOURI - ONSTANTINE c ONSTANTINE - ALGERIE TEL./FAX (213) ABoultif@ifrance.com Page 1

2 DIVOL06.txt G E N E R A L I N S T R U T I O N S DIVOL PROPOSES SOLUTIONS; THE USER DISPOSES OF THEM. THE AUTHORS DISLAIM RESPONSABILITY FOR ALL USE OF THIS PROGRAM ===> BEFORE USING DIVOL06, PLEASE READ REFERENE 1, AS WELL AS THE ===> INSTRUTIONS AND REOMMENDATIONS BELOW. INPUT DATA FILE ARD 1 TITLE ARD 2 N,ITYPE,J,JT,JH,JO,JM,JTR N NUMBER OF LINES USED FOR SEARHING SOLUTIONS. (This number is, generally, lower than the number N_TOTAL of input lines). e.g. N = 20. ITYPE SPAING DATA TYPE. =1 THETA BRAGG ANGLE IN DEGREES. =2 2-THETA ANGLE IN DEGREES. =3 D-SPAING IN ANGSTROM UNIT. =4 Q SPEIFIED IN Q-UNITS AS E+04/D**2. J =0 UBI SYSTEM IS NOT TESTED. =1 UBI SYSTEM IS TESTED. JT =0 TETRAGONAL SYSTEM IS NOT TESTED. =1 TETRAGONAL SYSTEM IS TESTED. JH =0 HEXAGONAL SYSTEM IS NOT TESTED. =1 HEXAGONAL SYSTEM IS TESTED. JO =0 ORTHORHOMBI SYSTEM IS NOT TESTED. =1 ORTHORHOMBI SYSTEM IS TESTED. JM =0 MONOLINI SYSTEM IS NOT TESTED. =1 MONOLINI SYSTEM IS TESTED. JTR =0 TRILINI SYSTEM IS NOT TESTED. =1 TRILINI SYSTEM IS TESTED. ARD 3 AMAX,BMAX,MAX,VOLMIN,VOLMAX,BEMIN,BEMAX AMAX MAXIMUM VALUE OF UNIT ELL DIMENSION A, IN ANGSTROMS. (IF AMAX= 0.0 DEFAULT= 25. ANGSTROMS) BMAX MAXIMUM VALUE OF UNIT ELL DIMENSION B, IN ANGSTROMS. (IF BMAX= 0.0 DEFAULT= 25. ANGSTROMS) MAX MAXIMUM VALUE OF UNIT ELL DIMENSION, IN ANGSTROMS. (IF MAX= 0.0 DEFAULT= 25. ANGSTROMS) VOLMIN MINIMUM VOLUME FOR UNIT ELLS IN ANGSTROMS**3. VOLMAX MAXIMUM VOLUME FOR UNIT ELLS IN ANGSTROMS**3. (IF VOLMAX= 0.0 DEFAULT= ANGSTROMS**3) BEMIN MINIMUM BETA ANGLE FOR MONOLINI ELLS IN DEGREES (IF BEMIN= 0.0 DEFAULT= 90. DEGREES). BEMAX MAXIMUM BETA ANGLE FOR MONOLINI ELLS IN DEGREES (IF BEMAX= 0.0 DEFAULT= 125. DEGREES). ARD 4 WAVE,POIMOL,DENS,DELDEN WAVE WAVELENGTH IN ANGSTROMS (DEFAULT=0.0 IF U K ALPHA1). POIMOL MOLEULAR WEIGHT OF ONE FORMULA UNIT IN A.M.U. (DEFAULT =0.0 IF FORMULA WEIGHT NOT KNOWN). DENS MEASURED DENSITY IN G M(**-3) (DEFAULT =0.0 IF DENSITY NOT KNOWN). DELDEN ABSOLUTE ERROR IN MEASURED DENSITY. (DEFAULT =0.0 IF DENSITY NOT KNOWN). Page 2

3 DIVOL06.txt ARD 5 EPS,FOM,N_IMP,ZERO_S,ZERO_REF,OPTION EPS =0.0 THE ABSOLUTE ERROR ON EAH OBSERVED LINE IS TAKEN TO 0.03 DEG. 2THETA (DEFAULT VALUE) WHATEVER THE SPAING DATA TYPE (ITYPE IN ARD 2). =1.0 THE ABSOLUTE ERROR ON EAH OBSERVED LINE IS INPUT INDIVIDUALLY FROM ARD 6, AFTER THE OBSERVED 'D(I)' ON THE SAME LINE, AORDING TO THE SPAING DATA UNIT (e.g in deg. 2theta) EPS NE 0.0 AND 1.0 THE ABSOLUTE ERROR IS TAKEN AS A ONSTANT (= EPS),IN DEG. 2THETA, WHATEVER THE SPAING DATA TYPE (ITYPE IN ARD 2) (e.g. 0.02, which will apply to all input lines). FOM LOWER FIGURE OF MERIT M(N) REQUIRED FOR PRINTED SOLUTION(S) (DEFAULT=0.0 M(N)=10.0). N_IMP MAXIMUM NUMBER OF IMPURITY/SPURIOUS LINES AEPTED AMONG THE FIRST N LINES [N_IMP takes into account both impurity lines and peak positions out of the input absolute angular error EPS]. IF N_IMP <0 THE SEARH STARTS WITH ZERO IMPURITY LINES, THEN, IT ONTINUES WITH ONE IMPURITY LINE, AND SO ON UNTIL 'N_IMP' IMPURITY LINES IS REAHED. ZERO_S A PRIORI SEARH FOR A ZERO-POINT ERROR IN INPUT DATA. =0 NO SEARH =1 SEARH IF ZERO_S NE 0 OR 1, THEN ZERO_S REPRESENTS A KNOWN ZERO ORRETION (e.g ) IN DEG. 2THETA. ZERO_REF =0 NO 'ZERO-POINT' LEAST-SQUARES REFINEMENT. =1 'ZERO-POINT' LEAST-SQUARES REFINEMENT. OPTION =0 DIVOL04 OPTION (OPTIMIZED STRATEGY SEARH WITH DEREASING ELL VOLUMES). =1 OPTION WITH EXTENDED (EXHAUSTIVE) SEARH IN VOLUME DOMAINS ONTAINING MATHEMATIAL SOLUTION(S) (LONGER PU TIMES). IF OPTION IS OMITTED, DEFAULT IS DIVOL04. ARDS 6 to 6+N_TOTAL D(I) [,EPSIL(I)] (ONE FOR EAH INPUT LINE, FOR ALL N_TOTAL LINES) D(I) VALUE DESRIBING THE OBSERVED POSITION OF LINE 'I' AORDING TO SPAING UNIT 'ITYPE'. EPSIL(I) ONLY IF EPS=1.0 (ARD 5): ABSOLUTE ERROR FOR 'D(I)', AORDING TO 'ITYPE', [IF EPS= 0.0 OR EPS IS A ONSTANT NOT EQUAL TO 1.0 (e.g. 0.02)(SEE ARD 5), ONLY 'D(I)' ARE REQUIRED]. NOTE: IF ITYPE=1,2,4 THE VALUES OF 'D(I)' AND 'EPSIL(I)' MUST BE PREFERABLY (THOUGH IT IS NOT MANDATORY) LISTED IN INREASING ORDER. IF ITYPE=3 THEY MUST BE PREFERABLY LISTED IN DEREASING ORDER. OUPUT FILES DIVOL06 GENERATES TWO OUTPUT FILES: - A FILE ONTAINING THE SOLUTIONS AS FOUND BY THE PROGRAM (ALSO SEEN ON THE DISPLAY WHEN DIVOL06 IS RUNNING). - A SEOND FILE WITH THE SAME NAME AND THE EXTENSION '.ord', IN WHIH ALL (MATHEMATIAL) SOLUTIONS ARE RANKED AORDING TO THEIR de WOLFF FOM. Page 3

4 DIVOL06.txt ***************************************************** * * * THE USE OF DIVOL06 * * * ***************************************************** USE AS INPUT DATA ALL AVAILABLE DIFFRATION LINES (N_TOTAL). ONLY THE FIRST N (eg 20, SEE ARD 2) LINES WILL BE USED FOR THE SEARH OF SOLUTIONS. THESE SOLUTIONS ARE SUBSEQUENTLY USED AUTOMATIALLY FOR REVIEWING ALL N_TOTAL INPUT DATA. THE IMPURITY TOLERANE 'N_IMP' IS RELATED ONLY TO THE N LINES USED FOR SEARHING SOLUTION(S) (OF OURSE, OTHER UNINDEXED LINES AN BE FOUND AMONG THE EXTRA INPUT LINES IN THE REVIEWING PROESS). BE AREFUL IN USING THE IMPURITY TOLERANE: SPURIOUS LINES INREASES THE RISK TO MISS THE ORRET SOLUTION! ==> WITH GOOD QUALITY DATA, OPTION = 0 (i.e. DIVOL04) IS REOMMENDED. ==>THE OPTION 'EXTENDED (EXHAUSTIVE) SEARH' (OPTION =1 IN ARD 5), AN BE USEFUL IN ASES OF (i) UNSTABLE ASES, i.e. MANY SOLUTIONS ARE FOUND IN A SMALL RANGE OF ELL VOLUMES, (ii) EXPETED DIFFRATION LINES ARE MISSING AMONG THE FIRST INPUT LINES, (iii) LOW QUALITY DATA, (iv) SIGNIFIANT NUMBER OF IMPURITY LINES, ET. IF A SOLUTION IS FOUND IN A VOLUME SHELL, THE SEARH IS EXHAUSTIVE WITHIN THIS SHELL. THUS, THE SEARH IS NOT EXTENDED TO THE NEXT VOLUME SHELL (FOR THE SYMMETRY BEING ANALYSED). IF NEEDED, THE USER AN RUN AGAIN DIVOL06, FOR SELETED LATTIE SYSTEMS, TAKING THE HIGHER VOLUME VALUE OF THE ALREADY ANALYSED VOLUME SHELL AS 'VOLMIN' IN THE NEW INPUT PARAMETERS SET (ARD 3). GENERALLY, RUNS WITH OPTION =1 TAKE LONGER PU TIMES AND THE NUMBER OF MATHEMATIAL SOLUTIONS IS GREATER (IT HAS BEEN LIMITED TO THE FIRST 100 FOUND SOLUTIONS IN THE DIVOL06 OUTPUT FILE). BY ATING ON THE INPUT FOM (GREATER FOM IN ARD 5) SOLUTIONS WITH LOWEST FOMs AN BE REJETED). BE AREFUL: AS SOON AS 100 SOLUTIONS ARE FOUND THE ALULATION STOPS. DIVOL06 OFFERS THE POSSIBILITY TO ANALYSE YOUR INPUT DATA FOR THE PRESENE OF A ZERO-POINT ERROR. THE METHOD EMPLOYED AN SUGGEST MORE THAN ONE OR TWO SOLUTIONS (DUE TO UNERTAINTY IN SEARHING HARMONI DIFFRATION LINES). THIS OPTION IS TO BE USED PRIOR THE INDEXING PROESS. PLEASE, AS FAR AS POSSIBLE AVOID THE USE OF THIS FAILITY BY ENSURING THE QUALITY OF YOUR OLLETED DATA! USUAL 'ZERO-POINT' ERRORS (IF NOT TOO BIG) ARE GENERALLY ORRETED BY THE LEAST-SQUARES REFINEMENT OF THE 'ZERO-POINT' PARAMETER (ZERO_REF IN ARD 5), TOGETHER WITH THE REFINEMENT OF ELL PARAMETERS. ZERO_REF IN ARD 5 IS, IN FAT, AN EXTRA DEGREE OF FREEDOM IN THE REFINEMENT. IT INLUDES ALL SYSTEMATI MEASUREMENT ERRORS FROM DIFFRERENT ORIGIN, E.G. ZERO-POINT SHIFT, SAMPLE-SURFAE DISPLAEMENT, UMBRELLA EFFET, TRANSPARENY, APILLARY DISPLAEMENT,.. SUGGESTED STRATEGY: IT IS REOMMENDED TO USE A TWO- OR THREE-STAGES PROEDURE (i.e. TRILINI LATTIES SHOULD PREFERABLY BE STUDIED SEPARATELY), e.g.: 1- SEARH IN HIGH SYMMETRIES DOWN TO ORTHORHOMBI. ARD 2 : N,ITYPE,1,1,1,1,0,0 2- SEARH IN MONOLINI SYMMETRY. ARD 2 : N,ITYPE,0,0,0,0,1,0 3- IF NEESSARY, SEARH IN TRILINI SYMMETRY. ARD 2 : N,ITYPE,0,0,0,0,0,1 NOTE THAT FOR SOLUTIONS WITH MONOLINI AND TRILINI SYMMETRIES THE PROGRAM PROVIDES THE REDUED ELL. IF VARIOUS EQUIVALENT SOLUTIONS ARE FOUND, ONLY ONE OF THEM IS LISTED IN THE OUTPUT FILE. Page 4

5 DIVOL06.txt TRIGONAL SYMMETRY ASE WITH RHOMBOHEDRAL LATTIE: THE PATTERN IS INDEXED WITH AN HEXAGONAL LATTIE, HAVING A UNIT ELL VOLUME THREE TIMES GREATER. REOMMENDATIONS READ AREFULLY REFERENES 1 AND 6 (AND 9), AS WELL AS THIS USER'S GUIDE. PLEASE, SPEND TIME TO ENSURE THE QUALITY OF YOUR OLLETED DATA (SEE REF. 5). WITH AURATE DATA, THE SUESS RATE OF DIVOL06 IS VERY HIGH. PEAK POSITIONS SHOULD BE EXTRATED WITH A PROFILE FITTING SOFTWARE. AN INTERATIVE PROGRAM SHOULD BE PREFERRED, SINE AUTOMATI EXTRATIONS AN MISS LINES (LOW INTENSITY, SHOULDER,..). WITH BAD DATA, THE HANE TO OBTAIN THE ORRET SOLUTION IS SMALL AND THE ALULATION AN BE TIME-ONSUMING. WITH MODERN X-RAY POWDER DIFFRATOMETERS (THE USE OF MONOHROMATI RADIATION IS REOMMENDED), ABSOLUTE ERRORS ON PEAK POSITIONS LOWER THAN 0.02 DEGREES 2-THETA AN BE ROUTINELY OBTAINED. FOR INDEXING PURPOSES, ERRORS SHOULD NOT (IDEALLY) EXEED 0.03 DEG. 2-THETA. [IN EXEPTIONAL ASES, A FEW LINES WITH GREATER INDIVIDUAL ESTIMATED ERROR AN BE INTRODUED IN THE INPUT DATA. IN THIS ASE, USE THE PARAMETER EPS= 1. IN ARD 5, AND ENTER INDIVIDUAL ERRORS AFTER D(I) FOR EAH LINE]. WITH HIGH RESOLUTION POWDER DIFFRATION DATA (ONVENTIONAL OR, PARTIULARLY, SYNHROTRON X-RAY SOURES), THE ABSOLUTE ERROR IS USUALLY LESS THAN 0.02 (OR EVEN 0.01 WITH ULTRA-HIGH RESOLUTION) DEG. 2-THETA; ONSEQUENTLY, EPS=0.02 (OR EVEN EPS=0.01) IS REOMMENDED; THE ONVERGENE OF THE DIHOTOMY PROEDURE WILL BE IMPROVED. HOWEVER, BE SURE THAT THIS ONDITION IS TRUE FOR ALL LINES USED AS INPUT DATA. (REMEMBER THAT ALL MATHEMATIAL SOLUTIONS WITHIN THE INPUT LIMITS AND ERROR BOUNDS ARE FOUND, THE GREATER THEY ARE THE GREATER IS THE NUMBER OF MATHEMATIAL SOLUTIONS). ===> N_IMP PARAMETER: N_IMP AN BE USED IN ASE OF EXPETED SPURIOUS LINES (i.e. IMPURITY LINES, AS WELL AS OBSERVED LINES OUT OF THE INPUT ERROR). N_IMP ATS AT ALL SUESSIVE LEVELS OF THE DIHOTOMY ALGORITHM. AS SOON AS AN INDEXING SOLUTION IS RETAINED, A LEAST-SQUARES REFINEMENT OF LATTIE PARAMETERS IS ARRIED OUT. FOR THIS REFINEMENT A LARGER ERROR ON OBSERVED LINES IS ONSIDERED. THEN, A LINE REJETED AT THE LAST DIHOTOMY LEVEL AN, BY HANE, BE AEPTED WITH THE REFINED LATTIE PARAMETERS. [EXAMPLE OF POSSIBLE ASE: N_IMP = 0 NO SOLUTION; N_IMP = 1 ONE SOLUTION; HOWEVER, AFTER L-S REFINEMENT OF THE LATTIE PARAMETERS ALL N LINES USED FOR SEARHING SOLUTIONS ARE INDEXED WITH THE REFINED PARAMETERS.] NOTE THAT THE PROGRAM DIVOL06 IS EXEUTABLE FROM 7 LINES- 8 LINES IF THE 'ZERO-SHIFT' IS REFINED - (THOUGH IT IS NOT REOMMENDABLE SINE L-S REFINEMENT UNSTABILITIES AN BE EXPETED). LONG AND SHORT AXIS ASES (DOMINANT ZONE ASES): IF SUH ASES ARE EXPETED, THE NUMBER 'N' OF LINES (ARD 2) USED FOR SEARHING THE SOLUTION SHOULD, GENERALLY, BE GREATER THAN 20. THE MINIMUM VALUE FOR A LINEAR LATTIE PARAMETER HAS BEEN FIXED TO 2.5 ANGSTROMS. RELIABILITY OF INDEXING SOLUTIONS: READ PARAGRAPH 8 OF REF. 5 AND REFS 7 AND 8. NOTE THAT WITH THE OPTION DIVOL04 (OPTION =0), AS SOON AS A SOLUTION IS FOUND, ONLY SOLUTIONS WITH SMALLEST VOLUMES WILL BE SUBSEQUENTLY RETAINED. IF (FOR SOME REASONS!) YOU ARE NOT SATISFIED BY THE SOLUTION, YOU AN RUN AGAIN THE PROGRAM WITH AN INPUT LOWER VOLUME LIMIT SLIGHTLY GREATER THAN THAT OF THE FOUND SOLUTION (THE EXHAUSTIVE SEARH IS THEN EXTENDED TO A HIGHER VOLUME). Page 5

6 DIVOL06.txt NOTE THAT THE SEARH IS EXHAUSTIVE WITHIN THE LIMITS ON THE INPUT DATA. IN PARTIULAR, THE SEARH IS ONSTRAINED BY THE HIGHER AND SMALLER BOUNDS ON PARAMETERS, VOLUMES, SELETED FOM AND ABSOLUTE ERRORS ON PEAK POSITIONS. PLEASE AT ON THESE PARAMETERS WHEN USING DIVOL06. A LATTIE METRI SINGULARITY OURS WHEN UNIT ELLS DEFINING TWO LATTIES HAVE AN IDENTIAL SET OF ALULATED d-spaings. THIS AN BE OBSERVED WITH HIGH SYMMETRY LATTIES (READ REF.5, SET. 4.2, AND REFS THEREIN). SIMPLE RELATIONS EXIST BETWEEN THE PARAMETERS OF THE TWO ELLS, AS WELL AS PARTIULAR ELL-VOLUME RATIOS. A TYPIAL ASE IS: AN HEXAGONAL ELL [a, c, VOLUME V] AN BE INDEXED WITH AN ORTHORHOMBI ELL [PARAMETERS: a/2, a SQRT(3), c, VOLUME V/2]. DUE TO THE STRATEGY USED IN DIVOL, BASED ON AN ANALYSIS THROUGH DEREASING SYMMETRY, ALL ELLS SHOULD BE, IN PRINIPLE, DISPLAYED IN THE OUTPUT FILE (EXEPT IF A SOLUTION IS REJETED BY THE INPUT MAXIMUM VOLUME). HEK ON VALIDITY OF AN INDEXING RESULT: PLEASE READ REF. 5 (SET. 8) POSSIBLE SPAE GROUPS: LOOK AT THE HKL ONDITIONS IN THE OUTPUT LIST OF THE REVIEWING OF THE OMPLETE INPUT DATA PROVIDED AFTER A SOLUTION IS FOUND FROM THE FIRST N LINES. ADDITIONAL INFORMATION ON DIVOL06 SHOULD APPEARED IN REF. 9. REFERENES 1.- BOULTIF, A. AND LOUER, D., POWDER PATTERN INDEXING WITH THE SUESSIVE DIHOTOMY METHOD. J. APPL. RYST. 37, (2004). 2.- LOUER, D. AND LOUER, M., METHODE D'ESSAIS ET ERREURS POUR L'INDEXATION AUTOMATIQUE DES DIAGRAMMES DE POUDRE, J. APPL. RYST. 5, (1972). 3.- LOUER, D. AND VARGAS, R., INDEXATION AUTOMATIQUE DES DIAGRAMMES DE POUDRE PAR DIHOTOMIES SUESSIVES, J. APPL. RYST. 15, , (1982). 4.- BOULTIF, A. AND LOUER, D., INDEXING OF POWDER DIFFRATION PATTERNS FOR LOW SYMMETRY LATTIES BY THE SUESSIVE DIHOTOMY METHOD, J. APPL. RYST. 24, (1991). 5.- LOUER, D., AUTOMATI INDEXING: PROEDURES AND APPLIATIONS, IN 'AURAY IN POWDER DIFFRATION II', NIST, SPE. PUBL. No. 846, GAITHERSBURG, MD, USA, pp , LOUER, D. AND BOULTIF, A., INDEXING WITH THE SUESSIVE DIHOTOMY METHOD, DIVOL04, Z. KRISTALLOGR. 23, (2006). 7.- DE WOLFF, P.M., A SIMPLIFIED RITERION FOR THE RELIABILITY OF A POWDER PATTERN INDEXING, J. APPL. RYST. 5, (1968). 8.- SMITH, G. S. AND SNYDER, R. L., F(N): A RITERION FOR RATING POWDER DIFFRATION PATTERNS AND EVALUATING THE RELIABILITY OF POWDER-PATTERN INDEXING, J. APPL. RYST. 12, (1979). 9.- LOUER, D. AND BOULTIF, A., POWDER PATTERN INDEXING AND THE DIHOTOMY ALGORITHM, EPDI-10 ONFERENE (1-4 SEPT. 2006, GENEVA), TO BE PUBLISHED IN Z. KRISTALLOGR. Page 6