clhynabee lu aba - vtechworks

205
,AN X-RAY STUDY OF THE BEHAVIOR OF TITANIUM FILMS ON SILICON SUBSTRATES , by Frank E.,Dietrichy;, Thesis submitted to the Graduate Faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE in Metallurgical Engineering APPROVED: x A: CLhynabee Lu Aba C. R. Houska, Chairman W. R. Hibbard . _-4 J. T. P. Floridis L. Lytton” June, 1977 Blacksburg, Virginia

Upload: khangminh22

Post on 26-Feb-2023

5 views

Category:

Documents


0 download

TRANSCRIPT

,AN X-RAY STUDY OF THE BEHAVIOR OF

TITANIUM FILMS ON SILICON SUBSTRATES ,

by

Frank E.,Dietrichy;,

Thesis submitted to the Graduate Faculty of the

Virginia Polytechnic Institute and State University

in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

in

Metallurgical Engineering

APPROVED:

x A: CLhynabee Lu Aba C. R. Houska, Chairman W. R. Hibbard

. _-4

J. T. P. Floridis L. Lytton”

June, 1977

Blacksburg, Virginia

TY) SSS

VASS \At7I

DS¢+

pea

h ib

FP

ACKNOWLEDGMENTS

The author expresses his thanks to Dr. C. R. Houska who provided

the challenge and assistance required to complete this work. Thanks

also go to Drs. T. P. Floridis and J. L. Lytton for their interesting

and useful discussions.

Thanks are also due to Mr. Harry C. Dudley and Mr. Lawrence

McDonald, both technicians for the Department of Materials

Engineering. Appreciation is extended to Mrs. Teresa K. Belcher ©

for the typing of the manuscript and Mrs. Jeanne Warner for her

assistance. Thanks also go to Lawrence P. Johnson for his help in

computer graphics.

The author wishes to express his thanks to his family and

especially to Mr. James E. Kramer for his untiring encouragement

and financial assistance to both his daughter and son-in-law. Long

overdue thanks is offered to his wife, Robin, for her loving care

and encouragement.

The author is grateful to the National Science Foundation for

providing the funds that made this research possible.

Li

TABLE OF CONTENTS

ACKNOWLEDGMENTS . 6 6 we ee ee ee ee ee ee ee

LIST OF FIGURES. 2... ee ee ee ee ee iv

LIST OF TABLES... 1... ee ee ee eee ee ee VEE

I. INTRODUCTION. . 2. 2. 6 6 ew ew ew ew ew ww we we we ew ee eh el UL

II. REVIEW OF LITERATURE. . 2... 1 1 eee ee ee ee ee ee

III. THEORY. 2. ee ee ee ee 9

IV. EXPERIMENTAL PROCEDURES...) 1. ee eee eee ee ee 24

V. ANALYTICAL PROCEDURES . ~~... . ee ee ee ee ee ee 38

VI. RESULTS AND DISCUSSION... .....-----4-024. 43

REFERENCES. . ee ee eee ee ee eee ee 1168

APPENDIX Le ww ee eee ee ee eww LL

APPENDIX ID. we ee ee ee ee A LB

VITAL fc ee ee eee 193

ABSTRACT

iii

LIST OF FIGUR:S

Figure Page

1. Equilibrium phase diagram of the titaniun-silicon syst em e ° *. * ° e s * a . ct . a ° *. e e e e ° a . s * e e 6

2. Plot of linear expansion vs. temperature for the

components of a reacted thin film. ............ 10

3. Illustration of axis designation for a texture diffractometer . . . 1... 1 ee ew ew ew ww ew ew we ew we et 1

4, X-ray diffraction geometry ......... 4.4.44 .64. 14

5. High resolution Koy diffractomete« with line source. . .. 33

6. Schematic drawing of elimination 95f Ky, component. ... . 34 2

L. 7. Plot of heat treatment temperatu) 2 (°c) vs. (time)? to

fracture of cyclic anneals on thii films ......... 48

8. Scanning electron micrograph of titanium thick film

surface (2.2 um) after cyclic anr2al at 650 C, total

time of 2 hours. . . 1. 6 6 ee ee we et we tw ww ew ew 49

9. Scanning electron micrograph of titanium thick film

surface (2.2 um) after cyclic anreal at 650° C, total

time of 5 hours. . . 6. 1. 1. 1 6 we ew ee ww wee ew ee OD

10. Scanning electron micrograph of titanium thick film

surface (2.3 um) after total annealing time of 1 hour

at 625° CGC. ww ew we ww tet wt tt lt le le tle te lt lw ew OL

ll. Scanning electron micrograph of -:itanium thick film

surface (2.5 um) after cyclic anieal at 600 C, total time of 745 hours . . . 1 1 ee ew ee ee we ee ee ee ee «C2

12. Experimental intensity simulation for as-received

2. 2 pm titanium-silicon thin fila, 26 region from 35° to AS ee ee ee ee 60

13. Experimental intensity simulation for as-received 6

2.2 micron titanium thin film, 2¢ region from 70 to 80 . 61

14. Experimental intensity simulation for step. anneal for

2 hours at 350° C, 28 region from 35° to 45°... 2 wee 62

iv

Figure

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Experimental intensity simulation for step two hours at 350° C, 26 region 72 to 73°

anneal for

Experimental intensity simulation from oreo anneal for

2 hours at 450°C, 20 region from 35° to 45

Experimental intensity simulation for 9 Step

2 hours at 450° C, 26 region 72° to 79°

Experimental intensity simulation for step)

2 hours at 500° C, 26 region from 35° to 45°

Experimental intensity simulation for step

2 hours at 500° C, 28 region from 72° to 79°

Experimental intensity simulation for step.

2 hours at 550° C, 20 region from 35° to 46°

Experimental intensity simulation For step

2 hours at 550 C, 29 region from 72° to 79°

Experimental intensity simulation for step

2 hours at 575° C, 26 region from 35° to 45°

Experimental intensity simulation for step |

2 hours at 575° C, 28 region from 72° to 79°

Experimental intensity simulation for step.

2 hours at 600° C, 28 region from 35° to 46°

Experimental intensity simulation for step.

2 hours at 600° C, 26 region from 58° to 65°

Experimental intensity simulation for step 2 hours at 600°C, 20 region from 71° to 79°.

Experimental intensity simulation for step. 2 hours at 625° C, 28 region from 35° to 46°

Experimental intensity simulation for SreP o®

2 hours at 625° C, 28 region from 59 to 63

Experimental intensity simulation for step.

2 hours at 625° C, 26 region from 70° to 83°,

Experimental intensity simulation for. sted

2 hours at 650° C, 26 region from 32° to 40°

anneal for

anneal for

anneal for

anneal for

anneal for

anneal for

° . . e °

anneal for

anneal for

anneal for

anneal for

anneal for

anneal for

anneal for

* . . * e °

oanneal for

» a

Page

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

Figure

31.

32.

33.

34A.

34B.

35.

36.

37.

38.

39.

4Q.

41.

42.

43.

44.

Experimental intensity simulat:on for step, anneal for

2 hours at 650° C, 260 region from 60° to 63°... .

Experimental intensity simulation for step. anneal for 2 hours at 650° C, 28 region from 72° to 79°

Experimental intensity simulation for step anneal for

2 hours at 650° C, 26 region from 82° to 87°

Optical interference techniques measured the elastic

bending of the silicon substrate taken at right angles

Optical interference technique; measured the elastic

bending of the silicon substrate taken at right angles

Optical interference techniques measured the elastic

bending of the silicon substrate cu’ to within 3. 5° of

the 111 planes ........4.+4.-.64+24484

Experimental intensity simulation for as-received,

2.3 pm titanium thin film, 26 regiou from 34° to 42°

Experimental intensity simulation for as-received,

2.3 um titanium thin film, 2@ region from 52° to 54

Experimental intensity simulation for as- -received,

2.3 pm titanium thin film, 20 regior from 62° to 64°

Experimental intensity simulation for as-received,

2.3 um titanium thin film, 29 region from 70° to 84°

Experimental intensity simulation for 15 minute anneal

at 625° C, 28 region 34° to 42° 7. ke ee ee

Experimental intensity simulation for 15 minute anneal

at 625° C, 20 region 75° to 79°

Experimental intensity simulation for 15 minute anneal

at 625° C, 20 region from 80° to 84°... 2.

Experimental intensity simulation for 30 minute anneal

at 625°C, 28 region from 34° to 42°

Experimental intensity simulation for 30 minute anneal

at 625° C, 28 region from 59° to 62°. .....2..

vi

Page

79

80

. 81

82

83

84

85

. 86

87

88

. 89

90

91

92

93

Figure

45.

46.

47.

48.

49,

50.

Sl.

52.

53.

54.

35.

56.

57.

58.

Experimental intensity simulation for 30 minute anneal

Page

at 625°C, 26 region from 70° to 83°. oe - 94

Illustration of cross-section of thin film structure

for as-received and 350 C step anneals for 2 hours . 95

illustration of cross-section of thin film structure

for 450 C step anneal for 2 hours. . . woe - 96

Illustration of cross=section of thin film structure

for 500 C anneal for 2 hours .... oe 97

Tilustration of cross-section of thin film structure for 550 C step anneal for 2 hours... - 98

Illustration of cross-section of thin film structure

for 575 C step anneal for 2 hours. . . 99

Illustration of cross-section of thin film structure

for 600 C step anneal for 2 hours. . . L00

Illustration of cross~section of thin film structure

for 625 C step anneal for 2 hours. . . -LO1

Illustration of cross-section of thin film structure

for 650 C step anneal for 2 hours. - 102

Plot of "G' factor vs. disilicide thickness; in final step anneal treatment at 650 C for 2 hours ..... .103

Plot of strain vs. temperature increment above

deposition temperature perpendicular to the crystal-

lographic planes in step annealed thin films... - 104

Experimental intensity simulation for step anneal at

350 C for 2 hours, to illustrate the location of mono-

Silicide in this film. .......... 2... .42.. . .105

Terminal selection panel on the PDP 8/e rack .... .173

PDP 8/e computer... 1 4. 1 ee ew ee ee ee ew ew ww We LT4

vil

Table

II.

Ill.

IV.

VI.

VII.

VIII.

IX.

XI.

XII.

XIII.

XIV.

XV.

LIST OF TABLES

Completed Reflections of TiSi, (C54).

Atomic Scattering Factors .

Reflections for TiSi. .... . . . s . °

Intensities for 20 Region, As Received 2.2 jm

Titanium Thin Film. .....

Intensities for 29 Region, Step

at 350°C.

Intensities

at 450°C.

Intensities

at 500 C. .

Intensities

at 550 Cc. .

Intensities

at 575 C.

Intensi' ies

at 600 (. .

Intensities

at 625 °C.

Intensities

at 650 C.

Intensities

Thin Film.

for

for

for

for

for

for

for

for

20 Region, Step

° *. ° . °

26 Region, Stef

. * ° e °

26 Region, Stef

* ° » e e °

26 Region, Step

26 Region, Ste;

26 Region, Step

26 Region, Stef

. * e

° e .

Anneal for 2

Anneal for 2

Anneal for 2

Anneal for 2

Anneal for 2

Anneal for 2

Anneal for 2

* e * . e

Anneal for 2

Hours

Hours

Hours

Hours

Hours

Hours

Hours

Hours

28 Region, as Feceived 2.3 pm Ti

°

Intensi:y for 26 Region, 15 Mirute Anneal at 625°C.

Intensities for 26 Region, 30 Minute Anneal at 625°C.

28 Positions for Experimental Intensity Simulations .

viii

- 112

120

. 122

. 124

127

130

133

. 136

. 140

144

149

152

154

157

Table

XVII.

XVIII.

Page

28 Positions for Experimental Intensity Simulations . . . 163

Thin Film Specimen 2.5 Micron of Titanium Cut Within 3.5° of the 111 Planes Annealed at Various Temperatures

for 2 Hours, Measured at Room Temperature ....... . 166

I. INTRODUCTIO.J

Preliminary investigations of titanium-silicon thick films are

incapable of presenting quantitative models describing their behavior

at intermediate temperatures. The silicides associated with diffused

-thin film structures have received limited attention thus necessitating

conclusive determinations of structure and lattice parameters. The

confusion in the literature on the reaction between titanium films

and silicon possibly stems from the acceptance of previous incomplete

and inaccurate studies of the bulk silicides. The characterization

of this transition metal system is the purpose of this study.

The nature of X-ray diffraction makes it a particularly attractive

tool for analyzing diffusion on the order of several microns in

specimens. Refinements in the analysis of diffraction data have made

it possible to investigate uniform and non-uniform strains, grain

size, stacking faults and other defects as well as texture and crystal

structure. Grain boundary diffusion and volume diffusion in films

can be studied quantitatively in detail with an X-ray technique.

The titanium-silicon thin film system has been mainly examined with

MeV He backscattering along with some supporting K-ray diffraction.

A detailed study on thick film specimens which are free of oxides

at the interface was carried out. Titanium films of approximately

two microns thick, a factor of ten thicker than in previous studies,

were deposited on 111 oriented silicon single crystals. An additional

complication evolved from the highly deformed nature of the deposited

metal films. The diffraction line broadening found with these films

is probably considerably greater than that of cold worked titanium

filings at room temperature. A computer simulation data analysis

technique was developed to evaluate the diffused structure of the

films..

The build-up of strain about the various interfaces and its affect

on the long term mechanical stability of the reacted film is a major

concern. Examination of both uniform and ncn-uniform strains resulting

from various heat treatments and substrate orientations supplied some

insight as to the origin and subsequent reduction of the mechanical

instability.

Lattice parameters and structure factor determinations of both

titanium monosilicide and disilicide eliminated a major problem for

identification of these phases with X-ray diffraction. A range of

compositions along with the identification of TisSi, as a high

temperature phase will modify the most recent phase diagram‘ (see

Figure 1). Results of the bulk titanium si icides represent

significant differences from those integrat:d intensities and lattice

parameters previously reported.

Finally, it should be noted that the d:terminations presented

in the following sections have not completely charécterized the

titanium-silicon system. A continuing effort in tle examination

of diffused films and bulk standards will eventually complete the

study.

IIT. REVIEW OF LITERATURE

‘In the past ten years, a renewed inter:st in the silicides of

titanium evolved from their importance in iutegrated circuits and

energy conversion devices. These applications impose an acceptatle

period of operation at a given temperature ind consequently encourage

the development of new techniques to measur: composition and structural

changes on a fine scale. The thin film rea: tion between silicon and

titanium may differ from the ordinary diffu.:ion process in bulk

samples at elevated temperatures. Baluffi ::nd Blakely‘) have

attempted to identify and describe a number of special characteristics -

for diffusion in thin films. Although prel minary work has been done

on this system, it is not as clearly unders ood as many conventional

engineering materials.

A technique using the energy spectrum of MeV backscattered He

ions provides sensitive measurements of relative atomic composition

as a function of depth. This method along with glancing X-ray

diffraction furnish specific pictures of the transformations in thin

films after thermal treatment. Although there is some disagreement

as to whether ion backscattering is nondestructive ‘7??? (since it may

create some defects which could influence the diffusion data) it is

at worst categorized as "nearly nondestructive" , The subsequent

information, which has been collected for the thick film reaction

between silicon and titanium (less than one nicron) has made use of

this technique.

Silicide formation is most commonly obtained through direct

reaction of a vapor deposited film on an oriented silicon substrate.

Typically one, and in a few cases two, silicide phase(s) are formed

3) in thin film structures after thermal treatment ‘ . The minimum

reaction temperature and the silicide phase(s) present in reacted thin

film structures are dependent upon the extent to which the native

(253). "The thickness of native oxide

(3)

silicon oxide film is present

layer is markedly different for different substrate orientations.

Film cracking and peeling aftcr thermal treatment, which is often

enhanced for thicknesses greater than a few thousand Angstroms, has

(3) also been correlated with these silicon oxide layers

(2) Mayer and Bower have reported that titaaium disilicide forms

at approximately one-half its absolute melting point (600°C) in thin

film structures where the substrates are oxide free. Prior to the

formation of TiSi Tu 6) »? has suggested that the lower free energy

monosilicide of titanium will form. In both the zirconium and

hafnium-silicon systems, these transition monosilicides have formed

(3) during thermal treatment » but no direct evidence was found for

1 a

the analogous titanium-silicon system. A relationship of (time)?

(3 os was found for the growth of titanium disilicide ), which is character~

istic of a diffusion dominated process. Silicon was identified as

(6) the diffusing species in the titanium-silicon system by the use

of implanted nobel gas markers.

(6) It has been postulated that since silicon is the diffusing

species, it will leave vacancies near the silicon-silicide interface.

If these vacancies coalesce to form voids, the silicide could fracture

(1) easily under an applied stress. Balluffi and Blakely~~ have reviewed

several explanations as to the origin of re!atively large biaxial

(7,8) stresses often present in thin films - Commonly found stresses

, 9 10 ~2(7,8,9) , of the magnitude 10° -10 dynes cm are likely to affect the

thin film diffusion process and the stabilicy of the conesive inter-

(10) face forces. Dearnaley and Hartley emphasized that several

intermediate phases formed in the diffusion zone are brittle and

easily susceptible to fracture. These effects are recognized in oxide

films but little attention has been given to thin metal film processes.

Direct information on crystal structure, grain size, internal

strains, and the texture in thin film structures has not been obtained

with ion backscattering. It should be pointed out that there is

disagreement among authors (11> 12) as to the structure of TiSi,.

(2) Titanium disilicide, identified by Mayer and Bower » was reported

to be of the orthorhombic C49 type. Preliminary work by Kato and

(13) Nakamura using X-ray diffraction has reported both the C49 and C54

orthorhombic structures in the same thin film of titanium and silicon.

(11), Cotter, Kohn and Potter have claimed that TiSi

(14)

9 is dimorphous

Another author has inferred that the C49 orthorhombic

disilicide is a ternary phase containing aiuminum.

(15) An examination of the phase diagram for the titanium-silicon

system (Figure 1) indicates that a number of intecmediate compounds

are observed under equilibrium. At the resorted reaction temperature

for thin films, approximately one-half the absolve melting point,

TEMP

ERAT

URE

°C

2200

T J t 1 ' qT | tT 1

B3h8 Ti,Si., i Si

; 2130 f3-4 | 920°

1800; TiSi, 4

TiSi oO ,

1665 1570° ,

' 3X 1500 ,

softy P1420 1420° 1400} £340° \ :

5% 13%! 1330° . \ |

1170° “| U7”

1000+ /

1.1% 860°

600 —l 4 L 1 1 _4 t

Ti 20 40 60 80 am.%

si—

Figure 1. Equilip riu

system 15),

lum phase diagram of the titanium-silicon Anea in dotted circle is incorrect.

the solubility of silicon in titanium is less than one percent. The

solubility of titanium in silicon is negligible. The crystal

structures for the Ti-Si system are: orthorhombic Tisi, with a C54

structure 12) | a = 8.236, b = 4.773 andc 8.523A; a C49

structure‘)? also orthorhombic, a = 3.62 + 0.01, b = 13.76 + 0.01

and c = 3.60 + 0.01; TiSi is also orthorhombic ‘1° a = 6.544,

b = 3.638, and c = 4.997; Ti,Si, is hexagonal of the D8, type with

a = 7.429 andc = 5.13924. 2”)

Reacted transition metal films yield silicides which range from

(18) nearly random to textured or epitaxial , and consequently some

preferred orientation or texture may be anticipated in any sample.

The subsequent study is an effort to exemplify the use of X-ray

diffraction as a technique which can better characterize both texture

and strain effects in thick films. Emphasis is placed upon the strains

which are present in the plane of the film and their effect upon the

stability of the reacted structure.

Summary

1. Existing studies have been of a preliminary nature using

nearly nondestructive techniques, mainly MeV backscattering

with some supporting X-ray diffraction.

2. There have been no studies which kave measured the effects

of either stress or texture on the diffusion process in thin

films.

3. Grain boundary diffusion and grain size have aot been

properly addressed in silicide formation.

The effects of epitaxial growth on the stability of the

diffused film have not been addressed but have been shown

. . . . (19) to be important in the Pd-Si system by Hutchins and Shepla

A detailed kinetics study on thin film specimens which are

free of oxides at the interface is still to be carried out.

Conclusive determinations of structure and lattice parameter

of both TiSi and TiSi, in bulk samples are absent. 2

III. THEORY

An X-Ray Diffraction Approach to a Kinetic Model

Introduction

X-ray diffraction provides non-destructive measurements of both

(20, 21,22) in films. volume diffusion and grain boundary diffusion

High diffusivity paths, such as grain boundaries and dislocations,

are known to be important at lower and intermediate temperatures

and are of significance in determining the behavior of material

systems.

(18) Random to textured or epitaxial silicides form in reacted

transition metal films. The problem of texture must be considered

if accurate measurements of the thickness of growing phases are to

be made. The diffraction equations should be sufficiently general

to treat phases with various degrees of preferred orientation.

Thick metal films are capable of withstanding a surprising amount

of mechanical deformation during cooling from elevated temperatures.

The deformation results from large differences in thermal expansions

of the individual materials (see Figure: 2). Strain data can be

measured in three ways: (a) X-ray linc shifts (uniform strain),

(b) line broadening (non-uniform straii), and ‘c) the elastic bending

of the substrate.

An X-ray diffraction approach is « escribed along with experi-

mental results to illustrate methods wi ich enatle the determination

of a kinetic model. Only after understanding the texture and strain

10

1.0 T ¥ } ’

TiSig —

0.8 |

&

et oO

” < O.6F oa Titanium a . Gs (// a-axis ) a 44 © . Y a Titanium

37 .0C(C«OOW4 LL (// c-axis )

rt ow E o & E

0.2 -

0.0 , 200 800 1000. 1200

Temperature (°K)

. . . 28 Figure 2. Plot of linear expansion vs. temperature ) for the

components of a-.reacted thin film.

il

developed in thick films, along with the identity of reactant products

either deposited at the grain boundaries or at the interface, will

this system be characterized.

X-Ray Diffraction Theory

The diffraction equations below are for a polycrystalline film

deposited on a silicon single crystal substrate. Polycrystailine

materials require texture information to obtain quantitative results.

The use of parafocusing geometry results in simultaneous measurements

of an orientation function and the reactant phase thickness. At later

stages of this development the diffraction equations will reflect the

experimental limitations observed in this study.

Two mutually perpendicular angles, x and © (Figure 3), allow

various sample orientations. Phi is perpendicular to the specimen

surface. The plane containing the incident and diffracted beam also

parallel to the specimen surface contains the x axis. Specimen size

limitations restrict examination to 60° in x and a full rotation in

d. Chi should be taken as 0° when the specimen surface is perpen-

dicular to the plane defined by the incident and diffracted beams.

The area of the specimen sampled by tie X-riy beam is given by

A, = A,/sin 8 cos x (1)

where A, = cross sectional area of the incident beam, and @ = angle

of incidence. Sampling depths«range from the specimen surface to a

depth at which the absorption from the material <ebove will reduce

the diffracted beam intensity to that of statistical background.

L2

Figure 3. Illustration of axis designation for a

texture diffractometer.

13

The maximum path length of the X-ray beam (Figure 4) 1-2-5 depends

on the linear absorption coefficient as well as on 6. As the

sampling plane with cross section A. is lowered parallel to the

surface and the interface, regions of composition "m" are examined.

The fraction of irradiated area with composition "m"’ intersected by

the sampling plane, at a depth "Y" is defined by the function HY).

For one dimensional volume diffusion, the sampling plane will pass

through regions of uniform composition, Hs = ] in the depth Ya to ae +

AY and He = 0 elsewhere. Irregular composition surfaces associated

with high diffusivity pzths may also intersect the sampling plane.

In either case the function H@) will be useful in subsequent

calculations, since X-ray intensity is proportional to the volume

of diffracting material of composition "m".

The total volume of material of composition "m'' is given by

W

Vy =A ‘/ 4 (y)dY (2) m @o mM

where w is the sample thickness. The form of equation (2) neglects

absorption of the incident and diffracted beams, i.e. the term

exp [-K u(¥)Y¥] (3)

where K_ = 2/sin @ cos x m m

must be included. The angle oF is the Bra :g angle for an element of

composition 'm" and p(Y) is the average linear absorption coefficient

for all material from the surface to a depth Y. Equations (2) and

(3) combine into a single expression defined as the effective volume

SIZE —_—____—__-»

SURFACE

INTERFACE

hNTene

Fiber Axis

——-A 6 ly

8m

1° { % —TH

aa v ae 7 x Z — 7

C f= — f- / " pa X 7

X 7 X Z

Je mee mes ee eee ee a ee 7 x ZL

w x 7

é , N TELE EEE

a

La \| samen \ / _f

PLANES 7

1. =

t | s a Catlin

Figure 4. X-ray diffraction geometry.

15

<Ve ms sO OOA, i H_(¥)exp[-K yu (¥)¥]d¥> (4)

The integral has been extended to = Since the integrand vanishes for

thick substrates. The orientation factor, 00> represents the ratio

of the actual pole density for a specific composition "m" to that of

the ideally random case. Rotational symmetry may be assumed for the

present application of diffused planar films deposited onto oriented

single crystals.

The diffracted intensity is proportional to the effective volume,

Vie? and is given by the following equatior

Pa ~ ten me” . (>)

where Ow the reflectivity per unit length, 72°2” .

rn 1+ cos26 cos”2a 2

Ge a7 OF, exp[-20, ] (6) vn sin 26 (1 + cos’ 2a)

The constant terms are:

I_ = intensity of incident beam, o .

re classic radius of the electron (= e”/me*)

A = X-ray waveleng <h.

Terms dependent upon the composition, 11, of the diffracting element:

Via = volume of unit cell,

N |

F = structure factor = £ f_ exp[?a(hu_ + kv + dw_] (6a) m 1 2 n n n

16

fo = atomic scattering factor, /

us Yn? = coordinates of the atoms in the unit cell

N = number of atoms per unit cell

exp[-2M_] = Debye-Waller factor,

oF = Bragg angle for diffracting composition

angle for the monochromator. a

Equation (4) may be written in terms of an average over a hemisphere

for a specific oF

A n/2 “9

= —?. f -K u(Y) <V nee sin a 5 0&0 tan x S H(t exp l K CY ]d¥dx (7)

a/2

with the J 00 sin x dx = 1 (7') oO

The integral can be carried out if the transformation products are

planar and of uniform ecmposition to a depth Y. The general case has

several planar phases b:tween the suxface and the diffracting phase.

The intensity for an intermediate phase of composition "m" is

A, w/2

Pr. Tal au : 0x) sin x [1 - exp(-K u Y)] x

exp (-K_u(Z)¥) Idx (8)

where Y = distance to the surface,

Y thickness of planar composition"m", and the average linear m

Nt

absorption coefficient is given by:

17

Y I ucyv)dy (9) Q

KL

u(y) =

The integrated intensity of a single uniform phase with a texture

represents the limiting case of equation (8):

A, n/2 = —_—— f ; Pa TQ, 2h ; 0% sin x)dy (10)

letting Y > O and va +o, The last integral is unity by (7') and the

effective volume expression reduces to

<V >= co (11)

which is the effective volume for a uniform phase.

Existing literature on silicide formation in thin films indicates

that the planar model is adequate. Equations (4), (7'), and (8) (or

simplifications) can be used to determine phase thickness, we and

the distribution function, 0 6X) - In the case of epitaxial phases,

(20,2122). In this methods are available to give similar information

case, X-ray rocking curves and integrated intensities supply the

distribution of subgrains and phase thickness respectively.

Experimental Limitations

The long term mechanical stability of the reacted thick film is

limited by the build-up of strain about the various interfaces.

X-ray data obtained by systematic isothermal anneals provide

diffraction lines which overlap and are considerably broader than

those obtained from cold worked metal filings. Chi rotation, carried

out with an Eulerian cradle, would result in a complex analysis of the

X-ray data. A simplification of the development presented above is

given in the following section.

The effective volume, equation (4), reduces to the following when

conventional reflection geometry is assumed

Ve. = A.8 i HY) exp[-K_u(Y)¥]d¥ (12)

where g = 00x) and chi is restricted to + 3.4°, The intensity for

an intermediate phase of composition '"m" is given by

rd I TQ A.8ft - exp (-Ku_Y_) :exp[-K u(Y)Y] (13)

where Y = distance to the surface

Y thickness of planar composition "m'" and the average linear Pp Pp g m

absorption coefficient is given by equation (9). The orientation

factor, g (eqn (12)), designates the volume fraction of grains that

have {hk&} plane normals lying parallel to the fiber axis (see

Figure 4).

Bulk standards allow the X-ray data to be reduced to the

effective volume term of the various silicides (and pure materials)

given by equation (11). From an ideal powder standard of a, the

intensity is

A e)

P, = 1,9, a (14)

19

where "g"' = 1, a randomly oriented material. For a given hk&

reflection the ratio of a standard powder to that of the general case

of a thick film structure yields the following intensity relationship:

<P ,> QQ, #

a O a _—

P =o => X,8ll - exp(-K <u, >¥, Jexp[-K u@)¥] (15) a a a

. (25) Harris has shown that the values of Syke? averaged over all

orientations, is equal to unity:

zg hke _ TO ee)

where N is the total number of reflections. If only a few reflections

are available, equation (16) may no longer be valid.

Equation (15) or a modification of it can be used to consider

a homogeneous dispersion of "o" in a thick film or a columnar reaction

product at the grain boundaries. In this case the grain size is

assumed to be much smaller than 1/ny- Through the use of the

preceding equations, a kin2tic model for silicide formation in the

titanium-silicon system will be developed.

Simulation of X-ray Patterns

The complexity of X-ray diffraction patterns obtained from highly

deformed materials necessitate the use of computer simulation in

their analysis. Symmetrical diffraction peaks are often represented

by simple mathematical functions. Gaussian or Cauchy distributions

20

are often inadequate in representing the X-ray intensity distribution

near the peak position and at its extremities. A Pearson Type VI

(26) distribution can be varied from the Gaussian to a Cauchy by

appropriate selection of a parameter in the function. A modified

Lorentzian and other forms of this function can be used to effectively

approximate many X-ray diffraction peaks (27) |

The Pearson distribution has the following form:

y(X) = Yo[L + (K - %)°/ (ma), (17)

and the maximum value, Yo? at X. The integration of Pearson

distribution yields

Pm = 3) y (18) l= i Yf1+ &@ ~ X)7/ (ma) J Max = vim a T (m) oO

where I is equal to the total integrated intensity. This integration

contains the distribution maximum and solving

_ I tT (m)

Yo vim a F(m - 4) (19)

where "a'’ is related to the full width at the 1/pt® maximum.

W(Y/P) = 2avm(pl/™ - 1) (20)

The limiting forms of the Pearson distribution are

-1 Cauchy, m= 1, Y(X) =Y [1+ & - %)*/a-] (21)

YG = I/(at)

and Gaussian, m= ~, Y(X) = Y exp l[-(x ~ X)*/a7] (22)

Y, = I/(ayn).

The form of the Gaussian is derived by expanding the function as a

binomial and considering the limiting form of the general term of

the series. This study will make use of intermediate forms of the

Pearson Type VII distribution in simulating X-ray diffraction patterns.

Diffuse Intensity

The contributions to diffuse intensity from thermal vibrations,

Compton modified, and X-ray fluorescence is part of the measured

integrated intensity which is not included in equation (6). Thermal

vibrations reduce the intensity of crystalline reflections (Debye-

Waller factor, exp[-2M]) and also produce a diffuse intensity (TDS).

The-incident radiation will fluoresce the absorbing element, given the

required absorption edge and produce ¢n additional contribution to

the diffuse intensity. Compton modified scattering is present in any

diffraction pattern. Significant infccmation on the reacted

Specimen is available from these intensities.

The diffuse intensity from Comptcn modified and TDS can be

approximated by the form

2 D _ _ sin 6

Tcomp.,tos ~ *, 1 ~ exPI-K, 2 )) (23)

which gradually increases with 28. This is a convenient form for

22

fitting the experimental data, and both constants can be related to

theory at a later time if needed.

The X-ray fluorescence from titanium by the incident Cuky,

radiation takes the form

T (CuK, ) = KG - exp[-K,/sin 6]) (24)

Wi where K! > = Ing, (Ca) + Mpg HIT

Hp; (Cu) linear absorption coefficient of titanium with Cu

radiation

Ups gy (Cu) = linear absorption coefficient of silicide with Cu

x

radiation

Hp, (TL) = linear absorption coefficient of titanium with Ti

radiation

Uns ge (Ti) = linear absorption coefficient of silicide with Ti

° radiation

where Ung = FS? + “risa”

T = thickness of titanium-silicide films

X = Thickness of titanium

and Ki = I, C Afi (Cu) + Mpg (TL) I (24a)

I= intensity of incident beam

A, = cross sectional area of the incident beam

C = intensity of TiK /intensity of CuK = efficiency constant

which increases with decreasing. 26.

23

The effective volume of titanium after each diffusion treatment can

be cross-checked by the use of this flvorescence, which is independent

of orientation. The application of these two forms of diffuse

intensity will be described in detail in the Analytical Procedures

(Section IV).

IV. EXPERIMENTAL PROCEDURES

An examination of both thick films and bulk powder standards were

necessary to complete this study. The existing ASTM powder dif-

fraction card files for TiSi and TiSi, were discovered to be incom-

plete or inaccurate, and standards had to be prepared. These silicide

standards along with a Ti powder sample supplied the fundamental

information essential to pursue tre development of a kinetic model

of silicide formation in titanium-silicon thick films.

This chapter describes the procedures employed in the preparation

and reaction of both thick films and bulk powder standards. Certain

preliminary results from the preparatory stage of the powder standards

are also presented. A detailed description of the experimental

equipment is given for completeness.

A. Thick Film Preparation

This is an important stage of the research because one of the

purposes of this study is to obtain a thin film structure free of the

native silicon oxide layer. In order to clean this interface, the

silicon substrates were first placed in acetone in an ultrasonic

cleaner for twenty minutes. The samples were then sputter etched in

the same chamber where the metal films were later deposited. This

procedure did not allow th: Si to be exposed to oxygen prior to Ti

deposition. The etching ws done at 0.8 kV, 60 watts for 30 minutes.

The samples were subsequently heated to 340 and 360°C*, and the

Oo, . . . *Samples (b) were heated tc 275 C in order to examine the effects this

different substrate temperature would have on the fracture process.

24

25

titanium was then deposited at 2.1 kV and 300 watts. The sputtering

pressure of argon was 5 microns. The purity of the Ti target was

99.9%. Auger spectroscopy was used to determine the amount of sputter

etching required to remove the native oxide film on the silicon

substrate. Specimens were prepared at the Oak Ridge National

Laboratory by the Isotope Research Materials Laboratory.

Three different sets of thin film samples were received from the

Oak Ridge National Laboratory at different stages of the study:

1. Ti-Si specimens with a uniform deposit of 2.3 microns of

Ti within 0.95 to 1.00 of the theoretical density. The

substrates were cut to within 2.5° of the lll planes and

measured 0.5 x 0.5 x 0.01". |

2. Ti-Si specimens with a uniform deposit of 2.5 microns of Ti

within 0.95 to 1.00 of the theoretical density. The sub-

strates were cut to within 3.5° of the 111 planes and

measured 0.7 x 0.7 x 0.02".

3. Ti-Si specimens with a uniform deposit of 2.2 microns of Ti

within 0.95 to 1.00 of the theoretical density. The sub-

strates were cut to within 0.5° of the 111 planes and measured

to 0.875" in diameter, thickness of 0.013". These substrates

were purchased from Pensilco in Bradford, Pennsylvania with

a dislocation density of less than 500 em? resistivity, 1 ohm-

cm, and polished on both sides.

Samples for cyclic annealing studies measuring 0.1 x 0.1" were

cut from the as-received sampled using an ultrasonic milling machine.

26

The specimens were sandwiched between two glass slides using pyseal

cement. A 345 mesh boron carbide abrasive was used for cutting. All

samples were cleaned in hot (50°C) chloroform which was found to be

the solvent inert to both titanium and silicon, yet it readily removes

pyseal cement.

B. Bulk Standards Preparation

High purity silicon and titanium hydride powders, supplied by

(30) Ventron Corporation's Alpha Products » were used in preparing

the samples. Emission spectroscopy suppl:.ed by the manufacturer found

the silicon powder to be highly pure (iron the gieatest at 500 ppm).

Wet chemical analysis of the hydride found 94.5 4 O.1 weight percent

titanium, and from hydrogen evolution, 4 + (0.1-(.3) weight percent

hydrogen. All powders were mixed in methanol anc evacuated in a vacuum

desicator (1077 torr) and allowed to dry.

Titanium and silicon were mixed in the exact stoichiometric

ratio of:

18.8206 grams Til,

21.1794 grams Silicon

to prepare a 40 gram sample. The mixed powder was placed in a 3-inch

long, high alumina boat, and subsequently evacuated to i0°° torr in

an ion sorption vacuum system (see annealing furnaces, Section IV).

A total reaction time of 279 hours of annezling ensured complete

reaction of the sample in five steps at 1000°C. The first anneal

was 111 hours, of which 5 hours at 107" torr was due to hydrogen

. 3 ~6 . evolution. All subsequent anneals were at 10 torr during the

27

reaction. The weight loss could be attributed entirely to hydrogen

evolution in the initial anneal.

After each anneal (111, 24, 48, 48, 48 hours) the samples were

removed from the furnace and reduced to less than 37 um (-400 mesh)

particle size with a steel impact mortar. X-ray patterns were taken

without the use of a binder, and throughout the anneals the diffraction

lines remained sharp and no shifts were observed. Full reaction was

found after the fourth anneal with no evidence of residual titanium

or silicon. A fifth anneal was carried out to assure completeness

of the reaction. All X-ray patterns were made using a graphite

diffracted beam monochromator. This allowed very weak peaks to be

observable.

Titanium and silicon were mixed in the stoichiometric ratio of:

25.5973 grams of Til,

14.4027 grams of Silicon

for preparation of a 40 gram sample. All preliminary preparation

methods were identical to those employed wit the disilicide.

A total reaction time of 528 hours at 100°C in 11 steps assured

complete reaction of the standard. A111 anneals were 48 hours in

length of which 7 hours of the initial anneal was at 107 torr.

Weight loss again could be attributed entirely to hydrogen evolution

during the initial anneal. All sequential anneals were at 10°° torr

as in the disilicide treatment.

The X-ray diffraction patterns obtained with the same conditions

as before contained peak position shifts. Examination of the final.

sample found Ti,Si., and TiSi in the completely reacted state. A

10 atomic percent deficiency in silicon was determined and a corres-

ponding sample prepared.

All conditions remained the same as those of the previous samples

except the reaction temperature was increased to 1050°C. Seven treat-

ments of 48 hours in length produced a sample of only monosilicide.

The X-ray diffraction lines remained sharp and shifted to lower 26

positions during the treatment.

Each silicice powder of less than 400 mesh was filled into the

cavity of an aluminum sample holder after the final anneal. Acetone

and Duco cement (10 parts to 1) was used as a binder after the powder

was settled with acetone. This produced a sample with limited

surface roughness.

C. Annealing Procedure

Thin film anc. bulk sample heat treatments were carried out in

oil free, high vacuum furnaces which incorporate sorp*ion and ion

(20 liter/sec) pumping. Using both control and specinen thermocouples,

temperature measurements were within + 2°C. {interlocks in both

furnaces did not permit operation unless a vacuum of 107" torr or

better was maintained. Both furnaces and a general procedure is

described in the following section.

1. Bulk Standard Treatments

Powder samples were reacted at 1000 or 1050°C in a horizontal

tube furnace positioned around a quartz furnace tube which contained

29

the powders under constant vacuum. A West furnace controller provided

many different constant rise and fall rates, ranging from 20°F/hr to

300°F/min (0.001 to 1.0). Temperature was measured on a Brown Elec-—

tronic recorder with a chromel-alumel thermocouple which was in

contact with an alumina boat containing the powders. The furnace

control thermocouple was platinum-10% rhodium. A three-inch constant

temperature zone was measured at 1000° or 1050°C which was the reaction

temperature for all powder samples. Prior to heat treatments of all

powders, each alumina boat was air-baked at 1000°C for 8 hours in an

alundum tube inserted in the furnace.

A fused quartz, 3/4 inch I.D., tube was thoroughly cleaned in

acetone and methyl alcohol prior to any heat treatments. The alumina

boat located in the end of the tube was sealed to the vacuum system

by tightening two flanges against a copper gasket. The furnace was

positioned around the glass tube so that the constant temperature

zone enclosed the entire specimen.

If the following recommended start-up procedures are employed,

the inefficient pumping of samples to reaction vacuums will be

eliminated.

The ion pump is isolated from the remainder of the system during

an exposure to atmosphere. Dry nitrogen is bled to eliminate the

vacuum and a constant purge is continued during exposure to increase

pumping efficiency. A roughing vacuum is produced by filling the

sorption pump with liquid nitrogen. After a»proximately 15 minutes

the valve between the ion pump and the remai ider of the system may

be opened after valving off the sorption pum>. The vacuum should

stabilize at 5 x 10° torr and be allowed sufficient time to pump

well below this level.

The initial anneal produces a large amount of hydrogen at the

reaction temperature and is easily handled by the sorption pump.

During this anneal it is necessary to supply liquid nitrogen to the

sorption pump so that the roughing vacuum is not lost. This is

necessary for 5 to 7 hours at which time the ion pump may be bled

into the system. All subsequent anneals were done at 10° torr at

the ion pump. The rise rate of the furnace was set at 200°F per hour.

At the end of each anneal the samples were allowed to air cool by

retracting the furnace.

2. Thick Film Treatments

Thick film specimens were enclosed in a one-inch I.D. fused

quartz tube and evacuated to 5 x 10> torr. The vacuum was produced

with the same procedure described in the previous section. The

specimens were suspended in the tube on a horizontal high alumina

platform which eliminated contact of the specimen with the quartz.

A water cooled, open coil furnace was positioned around the sample

tube to produce a 3/4 inch constant temperature zone at 650°C (the

maximum heat treatment temperature).

Heat treatments of all thin films were carried out with the

"Thermac'' controller in the remote option. This option follows a

predetermined time-temperature profile ruled on a totating drum.

The limiter and gain controls were adjusted such that there was a

maximum of 2°¢ deviation between the chromel-alumel specimen and

command thermocouples. Limiter control enables the maximum voltage

applied to the heating coil to be any desired fraction of the full

line voltage. The width of the proportional band is dependent on the

gain setting; increasing the gain will narrow the proportional band.

For optimum control, the gain was 8 with the Limiter set on 4.

Cycled specimens (see Section VI and Figures 9-12) were heat-

treated with 50°C per minute heating and cooling rates at temperature

times of 5, 10, and 15 minutes. All stepped and isothermal anneals

had heating and cooling rates of 100°¢ per mi.iute.

D. X-Ray Measurements

The two X-ray diffractometers used to collect all data on bulk

and thin film specimens will be briefly described in this section.

Distinct advantages obtained through the use of this equipment are

emphasized.

1. Powder Diffractometer

(32) A Siemens diffractometer, allowing the specimen to be

rotated about the % axis (see Figure 3} independently of the detector

carriage, collected all data on the powder standards. The diffracto-

meter was used with a one-half degree cntrance, 0.2 mm receiver slit,

and a Soller slit in the incident beam. The data was collected with

a scintillation counter, a pulse height analyzer, and an E&S Industries

diffracted beam graphite monochromator. All final runs were made at

32

1/8° per minute. A silicon powder* standard was used to provide a

small correction to the bulk sample line positions. Integrated inten-

sities were measured using a planimeter after subtracting background.

The peak to background ratio was considerably improved with the

use of a diffracted beam sraphite monochromator. This diffractometer

set-up allowed relatively low intensity line measurements.

2. Single Crystal Diffractometer

The X-ray unit used to collect all data on the diffused thin

(32) films was basically a modified Siemens diffractometer, equipped

so that the Kyo component is eliminated. The instrument halfwidths

for a line source are 0.037° and 0.081° at 29 =~ 40° and 95°

respectively, which is only a factor of two _arger chan the natural

half-width due to the spectral distribution of the Ka, line using

Cu radiation.

The Kao component of the radiation from the hi:,h voltage fine

focus Cu tube was removed with an incident beam Jagodzinski singly

bent quartz crystal monochromator (see Figures 5 and 6). The total

elimination of the Kao component would result in an asymmetry of the

instrumental function. A ratio of 1:200 between Kay and Kay intensities

provided an insignificant contribution of K,, without causing excessive 2

high angle asymmetries. The monochromator was adjusted at this position

and remained there during the completion of the expcriments.

The data was collected with a one-half degree entrance slit,

0.05 degree receiver slit, a scintillation counter and pulse height

*Supplied by National Bureau of Standards.

BANOS

BUTT UFIM

ABJOWOROeATFIP

Toy uoTINTOSeI

YBTH ~°¢

aean3Tyq

WLSAYD

YOLYWOYHIONOW

>

QNINSdO

wwe

-11TS

JONVYLNE YOLYWOYHIONOW

SIX

Ss »

“WS wu}

ONINSdO JO

LHOISH --\

SNe

LINS VVOILYBA

\ %9

—_

\ —

VY 3ygnodls

i /

WLI

/ <

33S /

\

s

YaLNNOD /

\ :

! h

NOLLYTULNIOS \

\ \

LOds \I

~ —7F

yoo. ANIT

. SO

» oe

os

, /

of O

LTS eno“

F SS

.

: SIX

X _

_

YOLVWOYHDONOW

_—_ 3JGONY

H3dd09

@ 3eNnols

Beni AVH-X

SNd04 33S

3Nis ALISNSLNI

HOH

SIXVM™SBEG@e

34

Ka, REFLECTED BY

MONOCHROMATOR CRYSTAL

_ ENTRANCE SLIT (0.5° DIVERGENCE)

Ka, REFLECTED BY Se VERTICAL ENTRA MONOCHROMATOR \ . SLIT— HEIGHT OF CRYSTAL aN OPENING 2mm

_

Figure 6. Schematic drawing of elimination of Koo

component (Section B).

analyzer. The arrangement provided high resolution and refined line

shape making it well suited for defect studies.

Use of the Diffractometer: The high voltage generator was warmed

up at the operating voltage for several hours prior to data collection.

The pure silicon 26 positions were found to be in constant error over

the 26 region (30° to 150°) and corresponding corrections were made

to the experimental line positions. The detector panel settings were:

Detector Voltage: 1200

Pulse Height Analyzer: 2.0 (base and window)

Linear Amplifier: 4 x 64

on both diffractometers

3. Thin Film Thickness Measurements

The weight gain measurement supplied by Oak Ridge Laboratory

was varified using an X-ray diffraction method. Using the integrated

intensity of one or more orders of the substrate reflections, the

thickness can be calculated as follows:

gn (I/To)

. Ung (25)

t = film thickness

I = intensity measured in the plated condition

1, = intensity in the unplated condit’ on

Weg = absorption coefficient of film material

Ko= 2

sin Oo.

Gay on

The silicon single crystal substrate first order reflection will be

decreased by extinction. The use of the third order reflection will

minimize this effect and yield reliable thickness measurements.

4. Locating the Specimen

The mounting of thick film specimens in vaseline eliminated

additional contributions to the large amount of strain present in the

titanium. Destruction of the sample was inevitable if conventional

backings were used (i.e. mounting clay), resulting in a local fracture

of the titanium from the substrate. The sample was positioned by the

weight of a glass slide placed on the sanple which caused the vaseline

to creep. This located the sample surface parallel to the face of the

X-ray specimen holder.

The thick films were cleaned of vaseline in ios-octane after each

X-ray examination. Prior to heat treatment, all specimens were cleaned

in acetone to remove all traces of the iso-octane and then rinsed in

methanol.

E. Optical Interference Techniques

The elastic bending of thick film specimens has bcen measured using

an optical interference (Newton rings) technique. St-ains in the

plane of the metallic film distort the substrate into a saucer-like

shape, thus presenting difficulties in mounting the specimens (see

previous section). Information on the deflection of the specimen

resulting from stepped annealing upwards from the depcsition tempera-

ture will compliment the strain measurements obtained with X-rays.

A monochromatic mercury (5461 R) light along with an optical

flat were used to produce a pattern of contours for the film surface.

The factors that determine the nature of the interference are

differences in optical path length and phase changes on reflection.

Assuming normal incidence, the respective minimum and maximum inten-

sities are

2d = md (26a)

2d = (m¥s)A (26b)

where A = wavelength

m= 0,1,2,...

d = deflection from one rin‘, to the next.

Perfect flatness will form a pattern of uniform brightness. Each

fringe represents a charge in height of 18 millionths of an inch.

' The surface of the specimens were wiped clean and the optical

flat placed in contact with the metallic side of the thin film. It

is essential at all times to have the flat level with the surface.

The monochromatic light, at normal incidence, illuminated the surface

so that a reflection of it could be scen in the sample. The pattern

was viewed from an angle of 30° to the specinen normal through a

macroscope (5X). The photographs takcn of each fringe pattern were

enlarged to 8" x 10", and the distance between fringes was measured.

V. ANALYTICAL PROCEDURE

Calculations of the structure factors and relative intensities

for various crystalline reflections of a material are dependent upon

the atomic positions of the atoms in the unit cell. Substances of

the orthorhombic crystal structure yield X-ray patterns containing a

larger number of reflections than those of a cubic material and

accordingly less non-zero structure factors. The diffracted intensity

pattern for a structure composed of several phases will consist of

the individual intensities from each sei of crystalline planes parallel

to the surface. The difficulty in dete mining the individual contri-

butions from each phase to the total di fracted intensity is magnified

if uniform and non-uniform strain along with composition variations

are present in the material.

The computer analysis of both comp ex diffraction patterns and

the general forms for the structure fac: 2rs will be discussed in the

following section. The simulation of t .e entire diffraction pattern

determines the effective volume of each phase present in the reacted

structure. This is possible only after having determined the

experimental structure factors for the constituents.

Silicide Standards

The calculated relative integrated intensity for a powder can be

expressed by a combination of equations 5, 6, and 11. The diffracted

relative intensity, neglecting a thermal correction:

38

39

1+ cos-20 cos 2a 2

Tel ~ 2 PF, [Tax sin 26(1 + cos’ 2a)

a = 13.3°

can be calculated by the computer program listed in Appendix II.

Atomic scattering factors for titanium and silicon are presented

in Table II.

In general, F. is a complex number, and it expresses both the

amplitude and phase of the resultant scattered wave. The structure

can be expressed as:

Py = f [cos 20 (hu, + kv + gw) + i sin 2 (hu. + kv, + ww)

(28)

where fA = the atomic scattering factor.

The diffracted intensity is proportional to FI? which is obtained

by multiplying the expression above by its complex conjugate.

| If the space group and structure are known for a material, the u_>

va W, can be introduced into equation 28) and the structure factor

can be calculated for each crystalline reflection. Experimental

structure factors and those determined from equation 28 can confirm

the positions of each atom in the lattice (us v5 and wi) by the

following relationship:

rl|F | - |FI| R = ——?—__+_ (29)

z|F | oO

40

observed structure factors ll where F Oo

F e

calculated structure factors (see Eq. 6a)

when R is called the residual for a set of structure-factor magnitudes.

In principle, R < .50 suggests a correct structure, although a con-

siderably smaller R is necessary to lend credibility to the proposed

structure.

Lattice parameters are determined from agreement of the high

angle lines and a deviation of no more than 0.02° of those at low

angles. The initial approximation of each parameter is obtained from

hk& planes of the form; two indices zero and the other having a large

value, along with substantial line intensity. An iterative computer

program has been written which gives four decimal place accuracy in

each lattice parameter.

Thin Film Diffraction Pattern Simulation

An X-ray intensity simulation prog:-am was used to calculate the

total diffracted intensity from a diffused thick film structure.

Contributions from each reflection of a phase located as a dispersion

or a film were summed to represent the iotal measured X-ray intensity.

The diffuse intensity was considered as a combination of temperature

diffuse, Compton modified and X-ray flucrescence scattering. A

Pearson type VII distribution approximated the symmetrical diffraction

peaks, since the asymmetry of instrumental function can be neglected

due to the large line broadening found in all crystalline reflections.

41

Detail about a simulated diffraction peak is obtained by solving

equation (17), which determines a 26 position given an intensity:

2 Y¥(x) vima P (ins) ji/m ~ 1) 20 + ([ma I, Gn) 28 lt C17")

20 uN the peak position

where Y(x) is the measured intensity at any 29 position. The intensity,

Ty is related to that of the powder standards by the following:

ri ] I -_ a , _ 1 a ~ Ig¢q & ¥ 14 - exp(-K <u?) lexpl-K <u (Y)°¥] (15')

q K 2/sin 8 ™

for the general case.

The full halfwidth, By and full quarter width, By can determine

the "m'' of the Pearson function for a peak having a shape intermediate

between a Cauchy and Gaussian distribution by:

m= kn 2 , (30)

B - 1)

L

4 ng

and "a" is related by equation (20). Uniform strain is treated simply

by the increasing or decreasing X in equation (17').

The form for Compton modified and TDS scattering (equation 23)

is evaluated by fixed point iteration at two high angle points. Using

equation (24), the fluorescence from titanium is fit to two low angle

. . . . . - os (38) . points and substituting the mass absorption coefficients is

expressed by:

42

I= K,Q - exp[.1391T/sin 6]) (31)

I CA K =—-0 _

1 1391.06

where T is the thickness of the titanium film in microns. The two

diffuse intensity expressions are summed to represent the background

present in the recorded diffraction pattern.

The criterion imposed upon the simulations are those of con-

servation of material and that "g'' will sum to unity. Each

diffraction region of interest is fit with the diffuse intensity

from the predetermined functions to which is added the contributions

of each diffracting phase. Reflections of the diffracting phase are

considered initially random in orientation (g=1). Uniform and non-

uniform strain along with a preferred orientation are introduced as

the simulation progresses. As a planar reactant product grows, a

linear variation of "g"' from the substrate interface to the metallic

film is introduced. The simulation of the entire diffraction pattern

provides a model for the amount, distribution, and structural

perfection of the reacted sample.

A comparison of the experimental data, introduced as points, and

the simulated X-ray regions are viewed on a Tektronix graphics terminal

during the refinement procedure. An area difference routine of the

simulated and experimental intensities increase the ease at which an

operator can converge upon a solution. The computer program along with

details of operation will be presented in a College of Engineering

technical report.

VI. RESULTS AND DISCUSSION

Results of isothermal, cyclic, and step annealed titanium-silicon

thick films will be presented in this chapter. Silicide powder results

represent the fundamental link in the analysis of the reacted thick

film structure. Thick film results are related to the bulk standard

kinetics observed and the appropriate analogies will be made. All

computer simulated experimental data is included to emphasize its

central role in the present analysis.

Silicide Powder Results

Titanium disilicide was confirmed to be face-centered, C54

orthorhombic, a space group of Ds = rada 632) | The atomic positions

are:

8 Ti in 8a: (000, *50, 30's, 0's's) + 000, eax

16 Si in l6e: (000,!:0,%0%,0%!s) + x00,x00, (etx), Ce-x) des x=1/3

The structure factors take the following forms for various hk&

combinations:

2 _ _ . _ 0 F 32(fa4 fo4) ; h 3° + 2, k and & are odd

Fo = 32(£..42F..)°: h = 3", k and & are odd Ti “Si? | °

2 2 n FO = 64(£,,-f55) ; h = 2°, k and & are even

2 2. _ 3 FU = 64(£,,,+2f, 5) ; h =n --n, k and 2 are even

RF = 0; h, k and & are mixed

where n = 1,2,3,...

The atomic scattering factors required in these forms were obtained

from the "International Tables for X-Ray Crystallography" °” , The

43

44

correctness of the structure was determined from equation (29) (see

Section V). When titanium and silicon are considered in the neutral

state, "R'' was found to be 0.09. With silicon in the *4 valance state

and titanium in the neutral state, "R'" is 0.07, and the corresponding

integrated intensities are listed in Table I.

The peak to background ratio was considerably improved with the

use of a diffracted beam graphite monochromater. This diffraction

set-up allowed relatively lov. intensity line measurements. Results

of the bulk titanium disilicide represent significant differences

from those integrated intensities and lattice parameters previously

reported (12233) | Our study found a unit cell intermediate in size

which may indicate that contaminates were present in previous samples

because of the less perfect annealing conditions. The unit cell

dimensions of TiSi. are:

2

Laves & Duffin, Parthe Present

Wallbaum (12) & Norton (33) Study _

a (2) 8.236 8.27, 8.2668

b, @) 4.773 4.81, 4.7987

c (8) 8.523 8.56, 8.5503

The initial anneal produced a powder which consisted of mono-

silicide and silicon. This indicates that the lower free energy

monosilicide of titanium forms prior to the formation of disilicide

in bulk samples. X-ray patterns taken after each reaction displayed

the absence of any significant solid solubility of either constituent

45

in the disilicide, since no line broadening or line shifts were

observed.

(16) found to be pt® space Titanium monosilicide was previously oH

group, orthorhombic structure based on powder photographs. A study

(36) L by Ageev and Samsonov reported a Coy space group. Photographic

work, confirming the compound stoichiometric composition (36.95% by

weight Si = 50 atomic 4) and established that titanium and silicon

are insoluble in this coipound. The lattice pirameters of the

orthorhombic cell were:

Brukel, Nowotny, Schob Ageev and

and Benesovsky ) Samsonov (36

a, (R) 6.54. 3.61,

b (A) 3.63, 4.96,

ec. (A) 4.99, 6.47,

6. space group were: The atomic positions for the Dow

+ (X,%, Z5 (4-X), 3/4, (1/2 + Z)

UW Hl 0.127 where Ti: X 5 0.179, 2

Si: X= 0.0355 Z= 0.61,

and for a Co Space group:

Ti (000, %s*s, 00's, 1550)

Si (0.412%, 40.588, 0.350’, 0.350 3/4)

*The parameter Z is given, but a diagram makes it clear that it is

incorrect.

£6

(16) The monosilicide sample of Burkle was sintered at 1000°C in

an inert atmosphere from titanium hydride and silicon powders. Ageev

(36) and Samsonov used iodide derived titanium (99.7%) and silicon

(99.7%) synthesized by sintering the powders in an electric arc

furnace with a copper bottom and tungsten electrode with no protective

atmosphere specified. The above information is presented to evaluate

the validity of the proposed structure for titanium monosilicide by

these authors.

Titanium and silicon reacted at 1000°C in their corresponding

stoichiometric ratio produced a two phase equilibrium bulk sample.

(37) - The presence of Ti and TiSi in this sample identified Ti Si Si 5 3 5 4

as a high temperature compound and a modification of the equilibrium

phase diagram (see Figure 1) is warrented. The low temperature

existence of TiSi at its stoichiometric composition was found not

to exist but instead, one ten atomic percent richer in silicon.

The lattice parameters for both compositions are:

TiSi Ti, Sigg

a. (R) 6.5291 6.5358

bd, (A) 3.3643 3.6363

ec (R) 4.9915 4.9969 0

and the respective integrated intensities are presented in Table

III (a) and (b). Table III (c) displays a comparison of the Burkle (1°)

16 experimental intensities to those calculated for the reported Day

space group. The residual, "R", is equal to 0.26.

47

Previous authors have not identified a range of solubilities for

TiSi as in this study. A systematic shift of the crystalline

reflections to lower 26 positions in the X-ray patterns taken after

each anneal identified this solubility. The volume of the

orthorhombic cell for Ti is considerably larger than the 40°*60 stoichiometric composition, thus indicating that silicon will occupy

additional positions in the cell. An intensity calculation introducing

titanium vacancies into the cell did not correspond to those measured

experimentally. A structure identification of the monosilicide

standard is unnecessary for the completion o” the thick film study

. Since experimental structure factors can be ‘ised.

Cyclic Heat Treatments

Large differences in thermal expansions (see Figure 2) between

the silicon and titanium-silicides, and not a thermal fatigue process,

were responsible for the mechanical instability of the thick films.

Three separate times, 5, 10, and 15 minutes, at temperature were

examined over a temperature range of 550 to 650°C (see Figure 7).

Identical samples were cycled until fracture at a given temperature

for different lengths of time at temperature. The results of these

tests at 650, 625, and 600°C showed the fracture time to be independent

of the total number of cycles in all samples (see Figure 7).

Samples cut to within a half-degree of the 111 planes exhibited

a gradual fracture process (see Figures 8 and 9) initiating at the

outer perimeter of the specimen. This fracture characteristic was also

found in the larger samples diffused in step heat treatments described

48

703. 00 | FF ' ,

yoy ae eee eM cul 4 660. 00 oe va to 0.5

™~ ceo Sumy cul + ONG LO Lo 2.5°

AA NN +

- a a“ Ty 620). 00 me NA 5 um,

TS NOt Lo 3.508 aa we or SN +

.. ——_

NN SS

‘S90. 00 NS NS

NO

540.00 + T

T tT

500, 00 -——+ he —+

co L. 00 2. OO 3.00 4.00 5 00

6 TIME

Figure 7. Plot of heat treatment temperature (°C) vs.

(time)? to fracture of cyclic anneals on thin films. The effect of substrate deposition

temperature, film thickness, and cut of the

silicon single crystal to the 111 planes on

the mechanical stability of the reacted film are major concerns in these studies.

* Lower substrate deposition temperature

49

Figure 8, Scanning electron micrograph of titanium thick film

surface (2.2 um) after cyclic anneal at 650 Cy total time of 2 hours. Magnification 100X; Tilt = 20°.

50

Figure 9. Scanning electron micrograph of titanium thick

film surface (2.2 wm) after cyclic anneal at 650 C, total time of 5 hours. The entire film flaked off at this stage of annealing. Magnification 100X; Tilt = 20.

51

Figure 10. Scanning electron micrograph of titanium thick film

surface (2.3 pm) after total annealing time of 5

1 hour at 625 C. Magnification 100X; Tilt = 20°.

52

Figure 11. Scanning electron micrograph of titanium thick film

surface (2.5 pm*) after cyclic anneal at 600 C, total time of 7% hours. Magnification 500X; Tilt = 20°.

*Lower substrate deposition temperature.

53

in the latter part of this chapter. A‘l1 other samples (described in

Section IV) had a spontaneous fracture point (see Figures 10 and 11).

The data collected on Figure 7 served in planning anneals for the

larger X-ray samples.

Optimum deposition temperature appears to be at 350°C since

greater interface instability exists in films deposited at 275°C. It

appears, as in mechanical fatigue, that an endurance limit may exist

allowing samples to be cycled indefinitely below this temperature.

Complete information on the effect of substrate cut, film thickness,

and surface roughness of the silicon substrate on the endurance limit

is not available. An attempt to retain the reacted structures after

one hour at temperature with a cooling rate of 3°C per minute resulted

in the same characteristic fracture at approximately 300°C. In every

case, fracture occurs on cooling after annealing within a well defined

temperature range. Most often the fracture correlates with measurable

Silicide formation.

Step and Isothermal Anneals

A total of eight two-hour diffusicn treatments at temperatures

from 350 to 650°C were carried out on a 2.2 micron titanium thick

film, cut to within one-half degree of the 111 planes. The first step

anneal was chosen at 350°C to facilitate comparison with the as-

received sample deposited under similar conditions. Subsequent

temperature increments were determined from cyclic arneals and general

observations of conditions allowing the metallic films to remain

adherent. Since the mechanical instability results from differences

in thermal expansions of the individual materials, the temperature

increment was systematically decreased as higher reaction temperatures

were approached. Systematic step anneal upwards from the deposition

| temperature indicated that X-ray line shift, line broadening and

elastic bending of substrate are influenced by silicide formation.

Tables IV through XII are a list of integrated intensities at

different two-theta positions of all diffracted intensity above

statistical background for each of the eight anneals. It should be

emphasized that very little high angle diffracted intensity is present

in these thick film samples. The computer simulations of these high

angle regions are only semi-quantitative and all foundations for a

diffuse model are based on the low angle regions. The half-widths

of lines lying at 26 positions greater than 50° are the only simulated

parameters in which confidence can be placed. Figures 12 through 33

are the corresponding computer simulations where the points represent

the experimental intensity. The individual simulated reflections

along with the summed intensity is displayed in each figure for

comparison of the fit. Table XVI lists the Pearson function parameters

for each step anneal. A modified Lorentzian (m=2) best approximated

the shape of the X-ray diffraction peaks throughout the anneals with

the one exception being the titanium (011) at reaction temperatures

of 600 through 650°C, where "m' equals five.

In order to measure the response of the elastically bent substrate

to silicide formation, optical interference measurements of the

deflection in the sample were made and the contours plotted on Figures

55

35a and b. The maximum deflection of the substrate decrease with

increasing diffusion temperature or time corresponding to increased

silicide formation. In contrast, Figure 36 displays the profile of

a 2.5 micron thick film cut to within three and a half degrees of the

111 planes. The convex profile suggests that orientation of the

silicon cut affects the amount of strain present in the film and thus

the mechanical stability.

An isothermal anneal for times of 15 and 30 minutes at 625°C

reflect the effect of substrate orientation on silicide formation.

Samples cut to within two and a half degrees of the 111 planes

exhibited a lack of appreciable strain in the plane of the metallic

film. Tables XIII through XV are lists of integrated intensities at

different two-theta positions for each anneal. The X-ray simulations

needed to determine the preferred orientation and resolution of the

silicides formed are presented in Figures 36 to 45. Once again, a

modified Lorentzian distribution sufficiently approximated all

diffraction peaks. Table XVII contains the parameters of the Pearson

distribution for each individual reflection observed as a contribution

to the total experimental intensity.

It is evident from each set of simulations for the step and

isothermal annealed samples that two distinct orientations of the

titanium are present although a strong basal plane orientation is

common to both. The isothermal annealed specimen contained a tensile

strain on the (010) prism planes and a compressive one along the basal

plane (002) which decreased with increasing diffusion time. Titanium

56

monosilicide (0.1 um) present at the interface was strongly preferred

in the as-received sample. The silicide thickness increased to half

a micron after treatment for 30 minutes and fractured on cooling from

a subsequent anneal for an additional 30 minutes (see Figure 9).

The fracture corresponded to the cyclic heat treatment analysis of

these specimens (see Figure /) and substantiated a previous conclusion

for the fracture mechanism.

The titanium film remained adherent throughout all eight stepped

anneals; however, cracks initiating at the perimeter of the sample

were present after the final treatment. The existence of titanium

monosilicide as a dispersed phase (see Figure 46) in the initial thin

film structure exemplified the dominance of grain boundary diffusion

and nucleation at 350°C. A step anneal at 350°C produced an

approximate doubling of the volume fraction of monosilicide contained

in the titanium, from two to five percent (see Figures 12-15). The

monosilicide was random in orientation during all anneals. and was a

constant volume fraction of the remaining titanium film. If the

monosilicide were a planar deposit at the interface, a 0.2 ym decrease

of the plating would be required. The simulation was forced to fit

at low angles, relaxing on the condition of "g" factors summing to

one and conservation of material, yet the nigher angle intensity

region failed to yield a reasonable solution (see Figure 56).

Annealing upwards from the deposition temperature resulted in the

formation of a planar titanium disilicide deposit (see Figures 47-49)

at the original titanium-silicon interface. A linear decrease of "g"

57

from the silicon substrate with increasing silicide formation was

observed (see Figure 54) in each subsequent anneal (see Figures 50-52).

The final treatment at 650°C resulted in a maximum thickness of the

disilicide (3.9 um) and the formation of a planar deposit of mono-

silicide at the titanium-disilicide interface (see Figure 53).

Uniform tensile strains in the plene of the titanium film varied

insignificantly for the crystallographic planes displayed in Figure

56 in this series of heat treatments. The prism planes (010), whose

(39) when compared to the (002) modulus of elasticity is ten smaller

and (011),.showed the largest Poisson's contraction in the planes

parallel to the surface. Correspondingly, the smallest non-uniform

strain (line broadening) was associated with the (010) planes which are

(40) the slip planes in titanium The X-ray strain data does concur with

the mechanical constants presented in tiat a low modulus of elasticity

for a given plane will display a large ’oisson contraction and less

non-uniform strain. Line brodening inc -eased with silicide formation

for all reflections of titanium.

The disilicide reflection remained sharp after all stepped anneals,

thus agreeing with the bulk sample whici found titanium and silicon

insoluble in TiSi,- In the completely diffused sample, the monosilicide

present as a film displayed additional line broadening, probably due

to the range of compositions found for chis silicide in the powder

study. A weak preferred orientation of the TiSi in contrast to the

9 layer, reflected the influence of orientation of the initial TiSi

distance from the single crystal silicon substate.

58

Table XVIII is a list of X-ray data obtained for a step annealed

sample, 2.5 microns thick and cut to within three and a half degrees

of the 111 planes. The absence of any appreciable silicide formation

indicated the lack of adhesion of the titanium film due to the lower

a O substrate deposition temperature of 275 C.

Conclusions:

ol. TiSi, was confirmed a C54 orthorhombic structure having correspond-

ing structure factor forms and lattice parameters determined by

this study. The compound is stoichiometric in composition exhibit-

ing no solubility of titanium or silicon in it.

TiSi is not stoichiometric at low temperatures; rather a range of

compositions and lattice parameters exist. Ti,Si, is a high

. te Oo temperature compound, absent in equilibrium below 1050 C.

The thick film reaction between titaiium and silicon differed from

that of bulk samples, forming a large planar film of disilicide

at the original interface. The disi:icide film decreased the

transport of silicon and a second re: ction product of monosilicide

formed at the titanium/disilicide inierface.

The dominance of grain boundary diffision and nucleation in a film

diffused at a temperature of 350°C was observed, forming a dis-

persion of the lower free energy moncsilicide. Temperatures of

(2) silicide formation were higher in previous studies because of

the presence of silicon oxide at the interface. The absence of

degassing in the thick films verified that the method of

preparation did eliminate the oxide layer on the substrate.

Mechanical instability of the film results from large thermal

expansion differences of the components and silicide formation.

The highly deformed nature, dependent on cut of the silicon crystal,

warrented the use of computer simulation analysis of the X-ray

diffraction patterns from the diffused structure.

Suggestions for Future Work

1. More detailed studies of the low temperature phase diagram for

the titanium-silicon system is essential in characterizing the

kinetics of silicide formation in thick film structures.

The effect of substrate orientations on silicide formation

temperatures and phases characteristic to each should be explored

holding the substrate deposition temperattre at 350°C.

X-ray data obtained at temperature should provide sharper

diffraction lines because of the absence of that mechanical

deformation which is prodtced about each interface on cooling to

room temperature. The intercomparison between at temperature data

and room temperature data should provide considerable information

on the deformation mechanisms which are active on cooling.

INTENSITY

10. OO

8. OO

6. OO

4.00

Z. OD

60

7 t i: + + —t {- t —— t — t -t- + — t +t t —+ t +

, &@

1 1 m u

+ } T

mn . A Ti (002)

| _ Ci a T

F}

| Ti (010) y / . 4

Sg 7 Gf o 2 coo . m

T ul 0 ~~

0 Nn 0,0 Q 7 m

QF ) 0

1 oo 0 +

: Q ¢—

0 0 | Q wh

W a

TiSi (210) TiSi (102) x . 4 \ Ti (011) +

| St 21) ;

<— : . > Sn — —

35 00 35 83 36. 67 37. 50 38. 33 39. 17 40. oO 40. 83 4\. G7 42. 50 43. 33 44.17 45. 00

TWO-THETA Figure 12. Experimental intensity simulation for as-received 2.2 um titanium-silicon thin film, 28 region from 35° to 45°.

INTENSITY

61

=—_

15 00 t—— tt ot tt Ht Ht tt tt} 1— t—— —

+ al

1200 7 4

+ {

9.00 7 1

| +

6.00 Ff a ° o Te 8

| 1

3.00 | Ti (112) ai

Ti (103)

| TiSsi (004) +

.0O a t = . _- .

70. OO 70. 83 7L. 67 72 5O 73. 33 74. L7 75 00 79 83 76. 67 77. 50 78. 33 79. L7 80. 00

TWO-THETA . . . . O oO

Figure 13. Experimental intensity simulation for as-received 2.2 micron titanium thin film, 26 region from 70° to 80°.

INTENSITY

10. GO

8. OD

6. 00

4. 00

2. OO

62

Figure 14.

TWO-THETA

Experimental intensity simulation for step anneal for 2 hours at 350°C, 280 region from 35° to 45°

{——++ +——+ {+ t +- ae + + +——++

4 fu) +

o, f

vy

4 ih iP r

f \ 1 F |

i, iT}

T + rf iF) rr

1 oy 4 .

y ]

C ‘ w

1 wd AO 4

er O . @ 7 m 4 a QP UO iu _

(¢]

0 ba) t Ti (010) Soa © 6

Ti (011) - —_J

a TiSi (210)

oo fist (211) L

_ | \ a "|

35 00 34 83 36. 67 37. 50 38. 33 39. L7 40. GO 40, 83 41. 67 42 50 43.33 44.17 45 00

INTENSITY

1& 60

12 00

3. 06

6, 06

3. 06

63

{ +- t t t t t t t —t t t t— t— t i —t —t t t

I _

Ti (020)

: Ti (103) "|

. oo ao (112)

| \ a __--TiSi (004)

7Z GO 72 SO 73 00 73. 50 74. GO 74 50 75 0O 7& 50 78. GO 76. SO 77. CO 77. 50

Figure 15.

TWO-THETA Experimental intensity simulation for step anneal for 2 hours at 350°C, 28 region from 72° to 78°.

78 00

INTENSITY

145 00

12 00

3. OO

6. 00

3. OO

64

+ ~~

~~

—_

=e

a

-~

45 00

Are

Ti (002) ae y of

fy) :

- -

} i m

= Pf eo

C)

(yp i m

ae Ms o 4 LS —

a 0 a iT} ad

. ha]

f oF ) u slog y A o 8 ao e

4 ul Dnm8 ul a < \ fi Z ) oO u

> 2 ex é, Oo a

1 _ OF 2 +

i TiSi (102) . Tos ta Ti (010) Ti (O11) Wl TT

we On

+ , —— TiSi (211 ie TiSi (210) a )

\ fo TiSi, (004)

1 . a Tisi, (022) 4 . 7 2

35 00 35 83 36. 67 37. 50 38. 33 39. 17 40. OO 40. 83 41. 67 42. 50 43. 33 44.17

Figure 16. Experimental intensity simulation for step anneal for 2 hours at 450°C, 280 region from 35° to 45°,

INTENSITY

1§ DO

iz 00

3. OO

6. 00

3. OO

65

t —t t t t t t t t —t t t t t t t —t t 7

a ai

| - Ti (200) | Tisi (412) — TiSi (004) _

TiSi (501) - a ~ . TiSi, (333)

A _N y Lo “TiSi, (026)

72 OO 7Z 58 73.17 73.79 74. 33 74. 92 74 50 76. 08 76. 67 77. 295 77. 83 78. 42 72.

Figure l/.

TWO-THETA Experimental intensity simulation for step anneal for 2 hours at 450°C, 26 region from 72° to 79°,

INTENSITY

15 00

12 00

9. OO

6,00

3. OO

66

+ t t t t t t t t f —t t t t t t f f { t t

; Ti (022) Ha + oN .

\ a a

4 \. oy m 4.

\ 2 \ 0

n A “

- Tisi, (311)

Ti (016) —~_ TiSi (102). J oN -—T

‘\ .

oR Ti (011)

f v | 7

| | 7 fr} \

Q 2 RD na —~L

TiSi (210) 5 ) Ca ae “8p } am, nh a 2 TiSi (211)

otf us ' 7 ne

7 ye Le ~ Oe = p- Q) m

{ Cj CF wt u eS

\ fo oN TiSi, (004) \ a OOO aa |

a 0a rn" a © )

SE (022)

_ —_ ™ eee EE |

35. 00 35 83 36. &7 37. 50 38. 33 39. 17 40. 00 40. 83 41. 67 42. SO 43 43 44.17 45 00

TWO-THETA Figure 18. Experimental intensity simulation for step anneal for 2 hours at 500°C, 28 region from 35° to 45°.

INTENSITY

67

“ +

| ao

4 4

LS 00 t t t t — +— {———_t t t -t t t— t

4 +

1200 7 T

| al

goo - |

500 92 age ;

[ Tisi (412) —_ Ti (200) T

TiSi, (333) --~_ 7 3. 00 1 2 ” / os r ~~ ZL “nisi (004) TiSi (026)

Tisi (501) aa

| ao ~ | . a“

. oO { t T t ——f t t + t tT T { t —t T ——— t —t t— / —t t

72 CO 72 58 7317 73.75 74 33 74. 92 7a 50 76. 08 76. 67 77. 25 77. 83 78. 42 79. GO

TWO-THETA Oo . Oo O

Figure 19. Experimental intensity simulation for step anneal for 2 hours at 500 C, 26 region from 72° to 79 .

INTENSITY

14 OO

12 00

3. OD

6. OO

3. OD

68

- t r—t —t +——t t t + — t t — t t + —t t

+

frye

i)

4 a

\

| be ch

C" J

ui

+4 cy

rr ie

r

4 r , ‘Ti (002) Tisi, (311) —_ f v a

7m

5 Q

S oO "

| q L fr ©

an ty OTIQ Com Ti (O11) ate! © a T4 (010) A A CT of J om

- yo o Ne at a en aS (301)

t 2 AAC hh, ROS /. ; Qn

eal Ml tas Al San _———TiSi, (004) . Pr (yu / oO SAMEN. 2

TiSi > a ae a | TiSi (210 St , iSi (210) (102) :

36 00 38 92 36. 83 37. 75 38. 67 39. 58 40 50 41. 42 42 33 43 25 44. 17

Figure 20. Experimental intensity simulation for step anneal for 2 hours at 550°C, 26 region from 35° to 46°.

INTENSITY

69

— A + 7

1S oO t t t t t t i t t t —

12 00

3. OO

‘Ti (200)

6. 09 | oo T ?

TiSi (501) ~_

TiSi., (333)

| \ 4 3. OO TiSi (412) ‘. iSi , Tisi,

(026)

Wen t t t t t — ' —— ! i ———t —

72 GO 72 58 73 17 73 79 74. 43 74. 92 72 50 76. 08 78. 67 77.25 77,83 78 42 79. OO

TWO-THETA Figure 21. Experimental intensity simulation for step anneal for 2 hours at 550°C, 28 region from 72° to 79°

INTENSITY

14, 00

12 00

3. 0G

6. OG

3. OD

70

t { t t — —+ + + t t— { t t t t t -t t t t t t t

wae

: bi T

m7

A 7 me

Tisi, (311) 9 | ae . rua) |

NS

a u 0

| 0 | m pH

4102) ___-- Ti (002) 1 TiSi (102 | a al

ee a

1 Ti (010) TH (011) a

\ Lo ‘

J Tisi (210) NN , C TAS (211) +

me cry fy UN Gi OO .

(ane “ \ is 08> DO me Kone Md a

Re it Y u

4 oe x DOM m OQ” Lo Qo eo TiSi, (004) at

J / yD a

tpAnTbaae” | a ee / / “ Ou C iy TTR

TAT d 4 ’ uy () A (" A

<a. = pw TEST) (022)

354. G0 35. 83 38. 67 37. 50 38. 33 39. L7 40. 00 40. 83 41. 67 42. 50 43. 33 4417 45 00

TWO-THETA

Figure 22. Experimental intensity simulation for step anneal for 2 hours at 575°C, 29 region from 35° to 45°,

INTENSITY

1& 00

12 00

9. 05

6, OO

3. OD

71

t t t t t t t t t — t t t + t t t t t t t

©

art +

ys m

fo a

Oo D 7 c us a a aus >

* w

| a a oer SS a mang2e om :

4 cy rm WN

a “2. 0 m

a "oO ao oO oD woo

+

[i (200)

+ —

| TiSt (412)\ Tisi (501) | \

Ti (103) TSI, (333) -

oo _-— Ti (112) — f |

\ 7 TiSi (026)

\ aw —_

— t -—— t + -— i = t t t t t —<——} —— { >=

72 OO 72 8 7317 7375 74. 33 74. 92 75 50 76. 08 76. &7 7%. 25 77. 83 78 42 72. GO

Figure 23. Experimental intensity simulation for step anneal for 2 hours at 575°C, 28 region from 72° to 79°.

INTENSITY

19. 00

15,20

ll. 40

7.60

3. 8B

72

t + { t t t t t —t — t f t t t t t t t

si _ C 4

" a

") + ui a

m C]

~oTiSi, (311)

4 Ti (002) 4

A _Ti (011) 4

ee TiSi (211)

4 Lf oy _

t TiSi, (004) “2 2

fhe / BEM OMB—p—m |

ST @—_H <M TASi, (022)

35 00 35 92 36. 83 37.75 38. 67 32. $8 40. 50 41. 42 42 33 43.25 44. 17 45 06 46. G0

Figure 24. Experimental intensity simulation for step anneal for 2 hours at 600°C, 286 region from 35° to 46°.

INTENSITY

16. 8D

12 60

8. 40

4. 2B

73

/ ~TiSi (312) |

4 +

TiSi, (131). TiSi (410) - TiSi’ (511) 4

4 -4

—— t i + t t t i t t t —

68. 00 58. 98 So. 17 So. 7S 60. 33 60. 92 6L. SO 62 08 62 67 63 25 63 83 64. 42 6S OO

Figure 25. Experimental intensity simulation for step anneal for 2 hours at 600°C, 28 region from 58° to 65°.

INTENSITY

21. OO

15. 80

12 50

8. 40

4.20

74

t —+ t t + t — t t t — — t { t +- I t t t

a

oye fs

“i A

OE] 1 ° o SOL

ASO BOON A a OTT " Ne 2 Ove a

a BItt nm 2 , ND ligigugugamunu eee TL (200): SOR COT

TASd (412) — 7

TiSi (501) \ TiSi, (333) T 103

TiSi, (602) #03) :

. Ti (112)

7 TiSi, (026) T

71. OO 71. 67 72. 33 73 00 73. 67 74. 33 75 00 79 67 76. 33 77. 00 77.67 78 33 79. BO

TWO-THETA

Figure 26. Experimental intensity simulation for step anneal for 2 hours at 600°C, 20 region from 71° to 79°,

INTENSITY

19 00

14 20

Ll. 40

7,66

3. BD

75

46. 00

Tisi, (311) we

N ey

u

‘ CS

TiSi (102) fs oN

QO aa (002)

Ti (010) o

Tist _ Ti (011)

+ (210) ~~~ 4

TiSi (211) 4 an ~

eT

aes \ ‘SS

mene \ NSO, ———TiSiy (004) PRAgsTre \ BB aa | |

a SOON ae a a Q TiSi, (022)

35 00 35 92 36. 83 37.79 36. 67 39. 98 40. 50 41. 42 47. 33 43.25 44.17 45 08

Figure 27.

TWO-THETA

Experimental intensity simulation for step anneal for 2 hours at 625°C, 28 region from 35° to 46°.

INTENSITY

Zl. OO

16, 60

12 50

8. 40

4. 20

76

T

|

4 TiSi (312) +

| TiSi (410) |

59. 00 59. 33 59 67 6&0. OO 60. 33 60. 67 61. 00 61. 33 61.67 G2. 0O 62. 33 G62 67 63. 00

TWO-THE TA

O Figure 28. Experimental intensity simulation for step anneal for 2 hours at 625°C, 20 region from 59° to 63°.

INTENSITY

ZL. OO

16. 80

12 60

8. 40

4. 20

77

a

4 UI _|

A)

CN

a "

Ao UB OO O iF, 7 T

Oh On if . te m7 (4

LF | v Sop goo ln , v. Roy |

nooo Oo no She

oO Shy ty Gag “ag pooo® -—Tisi, (333) erm = “Qoaanammmd 7 TiSi, (602) +

Ti (112)

t So TiSi, (040) -TaSi (223) 4 TiSi (501) ____ a nS (421)

TiSi (412) Y .

Ti (103)___ /

TiSi (230)

\

= ' +

«KN

= SON,

70. GO 71.50 73. 00 74, 50 76. GO 77. 50 793. GO 80. 50 82. 00 83. 50 85 G0 86. 50 88. 00

TWO-THETA Oo . oO oO

Figure 29. Experimental intensity simulation for step anneal for 2 hours at 625°C, 26 region from 70 to 88.

INTENSITY

13. OO

1520

il. 40

7,60

3. 8D

78

t

1 2 . Ma

—_

TiSi, ere a

\ 7)

. +

—-—Ti (002)

| TiS (102) o 7

Ti (011)

T Tisi, (113)—_ J

‘TiSi (211) - Ti (010) a Lo 4

os = iSi 004 TiSi (210) N fre, aa iy (004)

- )

5 Vigan as 4 1) +

tae ry ph, ug ph

TiSi (111. TiSt (002) a +t Mean / Rm, OU TiSi, (022) we — fs \

Qa 7

A ptm 4 (rh ua Poor OOOO eg ‘On "|

oO \ \ ! i? (RSs eee

| ——~ Ss SES |

32 00 33. 17 34. 33 35 50 36. 67 37. 83 39. 00 40. 17 41. 33 42. 50 43 67 44. 83 46. GO

Figure 30. Experimental intensity simulation for step anneal for 2 hours at 650°C, 20 region from 32° to 46°.

INTENSITY

79

T

16.80 7 T

12 60

B40 7 +

TiSi (312) TiSi (410) 60. 00 60. 60 6L. 20 61. 80 62 40 63 60

TWO-THETA

Figure 31. Experimental intensity simulation for step anneal for 2 hours at 650°C, 28 region from 60° to 63.

80

TWO-THETA

Figure 32. Experimental intensity simulation for step anneal for 2 hours at 650°C, 260 region from 72° to 79°,

21. oO t t t t t t t t t + t t t t t t t + + t t t

+ -

AS a 16,80 —

u ub

" \ | T . 7

° © > 0 q [- + m © m u u > 4

4 1? 60 7 oD Th Uren nanmoe® a

YY) os ° ~ thee Lon v op |

= , ee SOUT rr tresadt

Lu boaae

- — < 68.40 4 a Tisi, (333)

: TiSi (412) .

TiSi (501) oa‘

4.20 | AN TiSi (004) Ti (103)

oo .Ti (112) |

<> Se _ oN. . co “ Tt t t i t u T t a tr —— — =_— T t + t r t t t t

72, OO 72. 56 73.17 73.75 74 33 74 92 79 50 76. 08 76. 67 77.25 77. 83 78 42 793. GO

INTENSITY

16. 80

12 60

8. 40

4. 20

81

t t t t 4 + + t + t — + t——_ + —t—— t t t t——t t-

| T

+ +

+ 4

TiSi (223)

| Ti (202) aS (230) |

82 00 82 42 82? 83 G3.25 83, 67 8+. 08 8&4. 50 84. 92 85 33 85 75 86. 17 86. 58 87. 00

TWO-THETA

Figure 33. Experimental intensity simulation for step anneal for 2 hours at 650°C, 28 region from 82° to 87°.

82

*poqtsodop

SCM WIT}

WNTURITI

94. YOEYM

UO sOeFJANS

VY} SBEOTPUT

MOITIe TBOTISA

*SaTsue VYSsTA

Je usyZeJ

azeAASQGNS UODTTTS

ay

FO

SUTpuseqd

IIASeTe

9Yy} paansesauw

sanbruyoeq

souaTaTAaqUT

Teotadg

‘yrE

ain8tTy

(°WO)

9F0ueIsSTG

T°t c°

7° €°

co" T*

0 Te

mn, v

4 4

T T

a -

T

°

9,059

L. z°

9,068 ——\-—-

L qe"

- |

3,00 —

i"

L _

(°TFW) OT X Your

83

*poejqtsodep

SCM WTF

WNTUeAT}

ey

YOTYM

UO BdeJANS

9Yy Sa2ReOTPUT

MOIIe

Te°TIPA

«“*SaTBue WUSTA

Je Useye

|}eAISGNS

UODTTIS

3syW jo

BUTpueg

DTISeTS

sy} peanseow

saenbruyds}

souaTazZAaquT

Teotjdg

(°WO) soueIqSsSTd

‘qve einsTy

TT see

(THM) OT * your

84

as received

as

received

Po,

Figure 35. Optical interference technigues measured the

elastic bending of a silicon substrate cut to

within 3.5° of the lil planes. This graph contains measurement: taken at right angles

to one another for tro step anneals. Vertical

arrow indicates the surface on which the

titanium film was deposited.

INTENSITY

85

15 oo t t t t t t +t t +t +— t t t t t +- t t t +

Ti (O11

12 00 | TE (002) ae - ° \ A T “ ,

oo \ my

m7 a

FI

9. OO T 7

Qo

}

Ti (010) 1

6. OO T 7

rj

ma

4 - —

th 7

3. OO T 7 ©

a Q © ~~

a Q +

re Oe ee eer erertr ty 6 Hm 2 OO O»yrrrerrenernta

. OO t t t { + t f t —t- t | t { t f —r —t

34. 00 34. 57 35 33 36. 00 36. 67 37. 33 38. 0O 38 67 33. 33 40. 00 40. 67 41.33 42. 00

TWO-THETA

Figure 36. Experimental intensity simulation for as-received, 2.3 um titanium thin film, 26 region from 34° to 42°,

INTENSITY

86

soo 7 — { t t t t

/ +

12 ao 7 T

9. OO T L

6. OO T 7

q

ao o )

3. OO T a Ti (012) +

| |

. OO t “+ ' T t t ' t q

52 00 52. 40 52. 80 53. 20 53. 60 54 00

TWO-THETA

Figure 37. Experimental intensity simulation for as-received, 2.3 um titanium thin filn,

28 region from 52° to 54.

INTENSITY

87

15 oo t t t t t t t t t

iz00 7 T

3. OO | T

6. OO

(110)

3. OO 7 oo a

. OU T t t t t t t t a

62. 0O 62 40 62. 80 63. 20 63. 60 64. OO

TWO-THETA

Figure 38. Experimental intensity simulation for as-received, 2.3 um titanium thin film, 28 region from 62 to 64.

INTENSITY

88

15 00 -——- t t t { t + t t + f — t t t t + t ——

4 4

1200 T +

|

4 ASR _ + 9. OG af Un

TARR ) i) A

m A FF ny 7 c ty 7 m P pF} 7 N -

: oF eg rc a po) CI HoOMoMAwp- Mi Kon yy

0 OOCARSASSSSGR Sea DamoAa aaa ome ~ TABOO ORAMAAmAMAMewBer ©

DOVAAMM

6. OO ; T

| Ci (201) |

: Ti (004) L

. Ti (112)

3. 00 r |

| Ti (103) :

_ en 4 3 2 i 1 1 , 1 4 4 1 4 4 1. 1

. oO q qT 1 v 1 i t t | ' T 4 qt ' Oo t { of 1 T t

70. 00 71.17 72 33 73.50 74. 67 74 83 77.00 7817 79. 33 80. SO 81. 67 82. 83 84, 00

TWO-THETA . . . O O

Figure 39. Experimental intensity simulation for as-received, 2.3 wm titanium thin filn, 28 region from 70° to 84.

INTENSITY

30. 00

24. OD

18. 00

1z 00

6. OO

89

4 t t t t + t —t t t t t ——- t t t +t t t —t

u

Ti (002) —- ae ;

m

Ti (011) -

| 4 oo

ff

n 4 ; _

1 t - Ti (010)

Oo W a u am

Ua ua

T ua + DR .

/ 0 Fr . ra Q Q m

6-B--2r-F- [ OUTST Fe he hee)]nod © 8 8a pou U 8 Q Q OVW rae Re +

34. 34. 67 35. 33 36. 00 36. &7 37. 33 38. 0D 38. 67 33. 33 40. OO 40. 67 41. 33 42. 00

Figure 40. Experimental intensity simulation for 15 minute anneal at 625°C, 26 region 34° to 42°,

TWO-THETA

INTENSITY

15 00

12 0

a. OB

6. OO

3. 0D

90

ao (201) a

i. 1 4 4. 4 4.

v 4 q { t t '

75 00 75 GO 76. 60 77. 40

Figure 41.

TWO-THETA Experimental intensity simulation for 15 minute anneal at 625°C 26 region 75° to 79°,

78. 20 79. 06

INTENSITY

145, 00

1Z BO

9. OD

6, 06 Ty

3. OD

91

TiSi (421) ee eel (004)

80. 00 80. 80 81. 60 82. 40 83. 20 84. 60

Figure 42.

TWO-THETA

Experimental intensity simulation for 15 minute anneal at 625°C, 28 region

from 80° to 84°.

INTENSITY

25, 0D

20. OD

1& 0D

10, OO

3, OO

92

+

_~-

+— + t f t —+- +—- t t — t t +- t f t t am i t—

by)

4 2

Ti (002) ———__ 7

m

r Ti (011)

+

iy

{ : +

uw

4

| | af

" u

4 J

. Ti (010)

5 a ° J r a . TiSi (002) eH p O .

a 7 0 f

{ a owes & 82 Owe sess

34. CO 34. 67 35 33 36. OD 3b. &7 37. 33 38. 0D 38. 67 39. 33 40. 00 40, 67 41. 33 42. 00

Figure 43.

TWO-THETA

Experimental intensity simulation for 30 minute anneal at 625°C, 28 region from 34° to 42°,

93

: | : 7 20. GO

16. 00

12 00

6. OO

INTENSITY

4. 00 59. 60 59. 60 60. 20 60. 80 BL. 40 62 00

TWO-THETA

Figure 44, Experimental intensity simulation for 30 minute anneal at 625°C, 26 region from 59° to 62°,

INTENSITY

10. OO

8. OD

6, OO

4, OO

2. OO

94

f t— +— f t t— t —t+ —+- + t +— t —+- t — —+ t ti———_r AO

4 m |

u

yp ©

m Nc)

5 AC rw 3 T

MW HS ry n 7 ai ©

fe @ Gh mat f - DQ OD OCP Se ee eee MMR MnMAMOMOMMOMNMMD?In Q

NO OOM PPSEseBoeonMAADonoOaeBan

Q t oT

+4 -

+ 4

TE (201) Ti (004) ~

TiSi (421) ~.

i a Ti (103) Ti (112) mo, al

4 4+

70. OO 7L. 08 7Z L7 73.25 74 33 74 42 76. 50 77. 58 78. 67 79. 74 80. 83 Bl. 92 83. 00

TWO-THETA

Figure 45. Experimental intensity simulation for 30 minute anneal at 625°C, 26 region from 70° to 83°.

95

——=~Monosilicide

Titanium

! \ — - Ay i poccd) = 2.2—uM

| '

Silicon

bo

Figure 46. Illustration of cross-section of thin

film structure for as-received and 350°C

step anneals for 2 hours.

Figure 47,

96

a Titanium

Monosilicide

Silicon

oC

2.1uM

Tilustration of cross-section cf thin

film structure for 450°C step anneal for 2 hours.

-5uUM |

97

Titanium

Monosilicide

|

NN

4 ' ' | ' ' 1 if U '

Lut LL pet pa Reel = - 1 ' \ { ' \ \ aie ioc roo4 --- <1 2.0uM

i i —. ! ‘ | | r /| i

' | | | | L | 1 i

Disilicide op

\

Silicon

Figure 48.

Mo

Tllustration of cross-section of thin

film structure for 500°C anneal for

2 hours.

1.0uM

98

Titanium

fo a Momost Hicide

\ fi! i ; d 1 Le ee ---- ee r--

; J ‘ i 1.8uM

| poste te Hep Le i ) | | \ LJ

Disilicide nN

)

|

j = Silicon G=4.5

i

Figure 49. Tllustration of cross- section of thin

film structure for 550°C step anneal

for 2 hours.

99

Titanium

Monosilicide

i j j i Ha-—) ff L--- i- -¥ - 4 oped

| rool I~ 4 oT} ro 1.6uM

a: L a re Pr |

1. 5uM Disilicide ce 3.6 tH ~

geass

Silicon

Figure 50.

——__

Tllustration of cross-section of thin

film structure for 5750C step anneal for 2 hours.

100

Titanium

Monosilicide

=|

dim be — —— — SO «a Disilicide

Silicon

es Figure 51. Illustration of cross-section of thin

film structure for 600°C step anneal

for 2 hours.

101

Titanium

an

Monosilicide

1.0 iM

T 1 j i i t { ' Lm my —~--4 i- 4 ---1 rnd

’ ‘ \ ' ' \ t a . ‘ . a

| <— —|—ce2.5

G=3.2 Bebe Fp

| Disilicide « |———G=4.5

Silicon

ae _—

Figure 52. Illustration of cross-section of thin film structure for 625°C step anneal

for 2 hours.

102

Titanium

Monosilicide

-4M so r ] és FF Gf fot

é H

; G=1.2 |

G=2.5

G=3.2 _ ee — —} 3.9uM

G=3.6.— —_ — —. _—— — —e — awed

Disilicide G=4.5

Silicon

Figure 53. Tllustration of cross-section of thin

film structure for 650°C step anneal

for 2 hours.

G FACTOR

3, 00

4, 00

3. OO

2. 00

1.00

103

CO 8 1.60 Z2. 43 3. 20 4.00

MIIRONS

Figure 54. Plot of "G" factor vs. disilicide thickness in

final step anneal treatment at 650°C for 2

hours. These points represent an average "G" factor for each thickness.

OTRAIN

104

CO 6000 120.00 180.00

_070 { { i t f t t t =f

Ti (O10) fo tL (

: 8 o o . O58 m Go 8

. 045 : T

a Ti (011) +

aoe Ok Ti (002) x t

fs A T rm KO

A A “A 4 x

. 022 7 T

-olo + —t i—— i !

240.00 300.

TEMPEF ATURE

Figure 55. Plct of strain vs. tc aperature increment abc re deposition temp :rature perpendicular

to the crystallographic planes in step

ann2aled thin films.

O06

INTENSITY

10. 08

8. OO

6. OO

4. 00

2. OO

105

TiSi (004) g = ae

Ti (200) g=1.1 .

\

\

1 \ T \

TiSi (501) g = 2.0 \

| visi (412) g = 1.4 |

| \

Se v f f f t —t f == t t i t t > —t + t T —— t t

72 OO 72 50 73. 00 73. 50 74 OO 74 50 74 00 7% 50 76. CO 76. SO 77. 00 77. 50

TWO-THETA O, . . arr .

Figure 56. Experimental intensity simulation for step anneal at 350 C for 2 hours, to illustrate the location of monosilicide in

this film. Titanium thickness = 1.9 um.

7& GO

106

O°vL

O°T?

O° T?

G°T6

0°06

O°OL

Q°*OOT

bb LT

“a’s SOAP T

8°T

2°T

7°€ TT

7?

L°el

U'Ty

O°OOT |

T°?

G'2

Y°TT

Z 1° “aes

“9TBD

L*0

8° 8T

3°0

€°?

7°?

8°0

G° 4S

6°S

cel

83°C?

0°O0T

877

Q°6

9°OT

4 1°T

as

*dxq

6t°T

67°T

7O°T

c3°t

80°C

el?

6c°C

%P

SOAR]

GGLE*T

LO6E°T

VOLT T

8S8r7'°T 6887

°T

LLEG*T

9797S °T

TT8S°t

CLES *T

L£990°7 TSLO°?

€260°2

9LET*?

VIO?

767E °C

9TL6°C

9EEL*E

¥P

"oTeD

PLE TT

O6€ *T

OL7°T

687 °T

9€9°T C7G°T

O8S°T

TEs T

CLO"?

60°?

SET"?

OOE*Z

B7E°?

016°C

6cL°E

P *dxy

O€*L9

Gé°C9

c0°09

80°0S

Lye”

07°C

TE*6E

4% OC

SOAP]

(769) CrstL

JO SUOFIDETJeY

peeTduog

I ATavi

cIT "89

L9C°L9I

c8T°e9

TS7°79 LOe* 79

TcT°09 G78 °6S

60€°8S

OSL*6

POL EY

SLS°CH

TO? “€7

CVC CV

LOT" 6€

9LE°8E

970 °0€

cl8°€?

¥OC

*“OTeBD

(7°89)

67°19

(67° €9)

(6£°Z9)

(7°09) (6°6S)

GE°S8s

9L°6¥

(79 °€4)

CC EY

LO°C%

el oe

O° SE

90°0€

78°CTC

O¢ *dxq

cet

GTE

CCV

70%

7CC

TTS

TeT

STT

ETE

007 OC?

Zc0

700

TTe

etl

c0Z

TIT

"TAH

107

O°TT

O°TT

¢°96

0°79

O°SL

0°S6

O° 72

ZIT

“aes SOAR]

L°%7

L°Y

7°0

7°0

9° LT

Oo

9°?T

€°eT

L£°cT

7°O

T°St

ZH L°T

“as

“9TPO

G°¢c?

c°”

8°0

7°0

7° eT

7°?cT

G*?er

T°St

S*0

T°8st

VT'T

9T°T

6t'T

cel

7e°T

Teer

9E°T

as Tv

SOAP]

GLEO'T 6c70°T 6970

°T

6£90°T T990°T 8890'T

SSLO'T

9/80°T

HZIT'T

99ET*T BECT

YT

owe

-

StstT

LOST'T

S76. T

Lo6T T

cSée TL

SVVc°T

VITE°T

cLYeT

9S9E°T

a

T

"OTRO

€70°T 970°T

L90°T

GLO°T

L80°T

elt’

9ET*T OCT

YT

CST’

€V6T°T G66T*T

Gcc 1

270 °T

OTE *T

SverT

G9C°T

P *dxq

(penutjuo))

I ATAVL

TO °S8

Té°€8

L£9°08

O€°S8Z

O8°9L

€0°CL

66°89

x OZ

SOAR]

028°S6

7TC°S6 8c8°

76

71L°C6 LTS

°* 76 CCC

CO

8L7°T6

SLT "06

T79° 28

8ce SB

020°S8

7OL*ES8 GOS

’*€8

O8¢ °08 068°62

C06 *LL

EL7°9OL

076°TL

CVL°69

QL9°89

%O7

*“oTeO

€o°S6 88°76

(07°76)

OS*T6

07°06

69°L8

(7€°S8) (7T°S8)

(78°€8)

cf °08 06°62

COLL

OS‘*9L

L£6°TL

(S8°69)

02°89

C *dxq

O77 709 770

€TZ Ces

800

Gee

LTEe

CVC

TTL TES

GTS Set

0c9 070

9720

Cee

cO9

902

TEE

‘TH

108

L°C?

8° 2

G°L

S*0

c°0

h° I'l

“u's SOAPT

€°¢eT

0° TT

T*8

0°8

On

ooo

mor

oo

~ 1°T

“a's ‘OTe

7° IT

S°OT

9°9

8°9

€°€?

7°8

UL

9°0

9°0

£°0

7°0

O°Z

ZI!

%*P “U's

SOAR]

*dxq

8008 "0

T098°O

LE98°0

8c88°0 S88 °0

6S68°0 €L68°0

€T06°0 T706°0

9ST6'0

1976 °0

4EE6'0

6176°0 7646 °0 Z0S6°0

$296 °0

6796°0

S066 °0

¥P

“OTBO

8008°0

7S08°0

0098°0

£€98°O

8238 °0

0S88°0

TL68°0

€T06°O

9ST6°0

89260

TEE6*O

S066°0

P *dxq

(penut uo)

I WIGVL

¥O7 SOAR]

OV °S4rT

CTO0°94FT

LGT° LCT

L6T*9cT

667° TCT

T96°O?¢T

OLS°STT

SOE*STT

PER LTT

OS8°9TT

OVS FLT

GcSs’*?erl

Tee ttt

ELL*S8OT

9S47°S80T OcE*

80T

VTE *90T

£70°90T

¢80°cOT

¥OZ

“OTPBO

EC °S8vT

TO°9#T

LT° LCT

Te °9CT

cS Tel

66°OcT

(ZE‘ SIT)

(77° LTT)

9G°YTT

(19°ZTT)

(82° TIT)

(7°80T)

(€°90T)

60° OT

OC *dxq

Se

OTZO

€T6

€Se

6TE o79

97¢ TT6

TSE est

LEE

6TT

77Y

TSt 8047 Bcc

LTS LET

909

‘THH

109

4° II

“a’s SOAP]

8° TS

ee €°6

8° LY

Aa Ra HW

me Ww

%P

SOABT

TE6L°0

TE6L°O

696L°0

996L°0

. 0261°0

996L°0

616£°O

*P P

"OTB °dxq

(penutjuo))

I aATaVL

%O7C

SOABT

Pyndy

ecyv est

OL2 OST

cS? OST

T89°67T

VEL

OUT

¥O8Z

“OTeBD

cv CST

S7° OST (2°64)

82 *dxq

TE6

ESS 89

870 GT6

THH

110

TctT

LéE

ets

FOL

Cod

Li

Ee

661

el2

C¥e

Tt?

Les

ve cm

us

GS5

126

T64

c66

COG 7+ "3

GCO"R TIN*R

“3 y1E°2

Tec? "e

cout Logte

"e TLS

"9 Zeb

"8 SEL

52 agets

3 Tears

YOO" 6 -

sya Pyree

"6 ESS °6

FLO%E 6

LEL°G BTS

“6 ones

poens “6

eH TTCT SAETCT

“6 E9E°CT

G25 °C)

"6 COGENT

SOLAN "6

Zoe° ot ——

"6 CETTIt

ZRETU "6

C1e* TT

LoS*tt "5

cOnttt 6YBtTT

"6 CLOT TT

ete zy

*6 coe

LTy*21 "6

Gestalt G69

°2ZT "6

CLeteT T2st2t

"6 ELT

ET 60Z°E1

"6 Ise

tet SER TET

“6 CES*ET

STOTT "5

CULTET LOL‘ ET

"6 bog

ted YOS*ET

“6 ELG*ET

9L6°ET “Ol

CCOtHI COOtHT

TPA TS

TS TS

sio}0ey

Surzez7eoS

DTMOIV

ceo sy

SOLE T

COpeT

CLE

S47]

OCS

PST

bh SOT

SHC t9T

T6587 ceo

ent

OF 6TOT bOc*lt

Cette

™.

OX

y f

ice:

A

ii?

4

rod

‘tf oO

yp

my

» a

OR a OE OL AL OAL AL AD pat are ete -

=) 1 t

*

OL aH om AR md Oot [OL Lt ed sd ee COI eet Be

oad eet end £5

wo *:

CES "1S

4SO°T?

GOC*

22

TL

“IIT WIV

Yee SAY

oCce*e

rover

LETS CGE

FO

CHe

tO

nave

COE

8

CET

TEC cosy

er

CETTO

OET*O Cette

i

MA Mm 0°

DANO. dd met ot

e

2

ro en

ar ue

nm cr

G &

fe .

QO

kA NR DY Am Pe

Mm De YQ

MAA CT D-AMDMMO- e

a

aa NN

¥/@ UTS

TABLE

II

(Continued)

+4

val

712853

72794

)

Sin

6/ir

12929

ce

Ay Cy 2h

SO oe

ag wean

WOOLY

on

wh oH

Lil.717

11.271

10.0592

il.

7.2519

ra 3Q5

1 nh

4

7.21

mp oy om

oy oN

a mr o

” a fod

1.9093

5.884

2ols

Oe 7id

2/79

“I oF

ay

TS.

4 _?

in 2

o.437

&

r) Pts

"pry

ony

ry

we

Sho

o uy

es)

Cy oO ivy

*

ro

“~ 24

ry

0 Sa

yo

2f&%S

4 eo

Le

4.876

5

os

wy NO

oT

lll

~

NJ

LY

ny ™ ny

LEN

0 5

we

orm “ny

Ta ‘

ry

ry

Ls

wh

4

at” ™I

4647

9,

TIL

~+

3

af

mM

wy

rr

it

321c

3.999

673

Ze 326

cn

3 Cc?

AY

m Le

Pm ay a

4#.u

72

ms,

ST

co fete!

2? 1.37,

1.712

~~

ef Pam

ay

a

1.42)

1.5!

i

Ln

3e 63

a

~ ore

1.439

1.257

ir ¢ ny ke.

ed

4 aN

oy

Le

sy,

i

a

of

1.313

367

Ll,

1.313

1.254

wy

a7

le VD

st re

oO)

sf 1.26

7) on

Cx)

_s

os oe

7

oO!

iSi

Reflections

for

T TABLE

III.

TiSi

(50%

Silicon)

Previous

Rel.

Exp.

Rel.

Previous

Calc.

Exp.

I.1.

I.f.

20

20

20

hkl

wt

BO.

tO

OL!

~

29290

A

80.5

hm OY

NS

my

bie

fol

on

3236

41

ill

332

Kw

204953

De IT

7209

3

L390.

cr

st

™ im

on)

59

~o*es

Loe

a

cn

os Mae

S32

2el

414°:

\ 1

. 2

2 4 $

— ol oy

“& 1

a 3

Ye.

5029

om

ce)

sw “~

uy le

)

112

1.9629

+ AG

sb

cc my

ave!

m™ |

el

59

3294

9 5172

1.7483

1.5514

l.

pe

lH59

C at

elo

>

7 oN

7.8

ond

*

iN LY

xr ae

tN!

my LN

~r

as -

uy, Dre L.

+ Jed

ue

us

LY

nn

La

~ oy

ts \

ws

2139

i.

LW

NI

ol.

nl.el7

ond

MT NJ

cr wy

wy

Ww

ped

cy ed

wy

Les+Col

Lu

1.4

}

“~Y

r X SOY

1 °

4 74R

41)

1).3

Je

x Oo

A

L113

e OLS

> *

a) ay

soo

“ « woos

™!

bo

oad

or SroNP

» « ret a

ft -

»

- | Rg a

ey nay

Goo

wd

tm sr * a

are Ly A

a ef yor

ma

1.3725

1.3661

. 19

xy,

an ve

ey

od

OCe

54

>

)

wy

= c. a e

coal N

an ec e eo

Ol oo a

Ow oy CO ead

SP onyn my eae

a e ®

rel end oot

- rd oo Od

a cd

ie oo

9 h-

Cy ey Ue

Peo a

TN eet Td

a e e

LY ed

Sh hs

ey m

oO OS a s

op ret

“OD h

mat TP PS)

OA OL

SN ON of

en

Ll.27o7

iN

<1.9

|

i

foeld

7501

wt

uN

1.2479

~y

Ot as

© aD

TABLE

IIIa

(Continued)

Previous

Rel,

Exp.

Re

l.

Calc.

Previous

Exp.

I.1.

I.I.

26

20

26

hk1

A

1.24

a43

sO

TG.

5S

313

OO

.

vV-)

ad

oon

«

NS pal

CR rm Gy hm in fe Bh mod et

* @ @

eed pe reed

me

Oo

a

ot

x)

on. Tr OY Sy he Ph

e* ee 8

ret med ON]

92° ©

-~

icy a

rod

co

one

rod date

SY OO

FON,

LJ.d

1)

G

ny

> won

F

9

mo

went

gy

au

}

A“

oy rx)

114

i3t

4 4 »

l ° L457

oy nN)

ce)

a ae)

CN ey

Co ce

4}

52-9

1.21099

1,11

3a7.7a

-

aa

on)

21

113

Dd

i ots

127

413

512

Mm OM >? *, e « ?

TAL om OX O) MN OJ

ay m4 e |. @ @®

h mM FP os

12.1025

38.33

ES

oy SB)

0)

cn!

A m

cst

Jell®

JL

* 44

2

on™

a OT)

ao

on

os ws

1.094275

2

13. 5

OLA

un oo,

or oS

cL GY

Doe

Lb

CN OP eed

SELLY

oT “p>

eviaw

. 5

a

re Le

cot

12

1.9794

oN ~

5 2

f.15

“sy

a

DD

Lo

37.9

)

LOo2.377

il

Ka

51

l t)o

wy

un

0.9811

Je97T451

0 +4

L 4.

24

nm

ONS

1J4.

.

im cy

Om!

“ iN

2

un

~m

ry

Lj7.2°

(107.99)

0.9321

1 1).

34

C8)

TABLE

IIIa

(Continued)

Exp.

Re

l.

Calc.

Exp.

I.I.

20

26

hkl

sO

cy

cr

{1l11.19)

2)

OD

Li3.93

uN

ILO7

t *

9.8

i

ra

ics

oy

rd ed

on i

eh)

rd

wy

cy

LQ.

é G5

~

120.5594

4 ~*

oe

nn .

uy

tO

aA

4. % ey

=

ae

SRR

ODO

0.

6

uf

1

wo

NI ~d

A)

oy .

oy oa

SO

a w L33.11

tie

490

8 Qo.

114

een

ON co

oD

poe

Lif.

242

-

tm -

oa AM

r ‘ 4

fy

in

mt

I.I.

$0.9

4

Previous

Rel.

ot

Exp.

Rel.

I.I.

1.1

L4

4 ca

ws

>

2.7359

“”

L

@

J e

4 Q

“+1

20

2,

TABLE

III

(Continued)

Previous

Calc.

26

60

Exp.

20

33230

*40

2 2

hkl

JL

201

7)

115

oS ON ENS ODN SO 2 cy o> oO On a e * * ® > a e e . e « . a e

OM DP QIN MAA oO o™ > um cy Qe Co GOR OS Ap oe ™ wot iy rl “Tr AEM wot

RD LD et OVE STO A Ram re e a a e e a * ® ¢ n a * ® e a eo

wth sb a oh a SO un Ch peed Mh Lo wm oo

SO LO em MS A rod a weed ol rod

Eh a Od mA OD et MN Om OY AM Cry Mm APG ee a wr De Shot Ph SOs at ANON eet SL re ed OL A Be DN RO SCT me Re SIN LO ttm a ar ay

e e & a e . a * a a = * 2 a a& a a * ° ° *

COOKE CNT md me ed ded et md eg md et ed ed ed ed en

Geop cg. CP am Mh Le va ny g Cl fo oc co rts

CTU Mey NE DP Ra * “y — 2 Ap om es * " ” ® * . * a * * * . * e °

DO me ND TN greed od tay Nt ” en ret iN

oy af Srp ao oe O oO a Pp

ped th RN | Dee Be fee EN Be pa eT et om IN + 0 fw ee ot A CLIN wd by meek AO OM Oo om em ch ot

CRO wed Pe ee AD et DCR ed ad ™ e . # s ° ® a ® eo . e » * a » a ® o *® * 2

SV ee EN OT emt at ee RE NOU RD ea ST SOA EIN DA OTD ADD a eS

UN PD OTE 0 -4aoa3 aa an] crt ETON EL OO nt met do nel ie moNUN “Ay om myo)

e a * * e * e ° * * ® e e . * a

Teh EQS pet es tof > ™ met

TA A EY LA 3 » 2 a ny he fh

CTE ON eh A OT mE INT ON met cE A ot et

eed TT met AY ed met ed dD ee TUS ONE TTD CJ SU Wy eet OM SRO TD tp PLN

13.9

22

ke

15.6

{5.135

(7

2» toy)

TABL

E II

Ib

(Continued)

Previous

Exp.

Rel.

Calc.

Previous

Exp.

I.1.

Rel.

I.f.

20

20

20

hk1

m™

313

Lim

1 s

1 3 2%

a

fo

SO,LE

e1l.3

rnd

o

oN

2 32

aN

“es a

il

Je

“F

car

D4

- 7

3 {

ott:

uy ot

L.L77s

1.1664

1.673

33

wd

19.09

tel

57)

3 OL e

> (.

A

2

Ouse

LN rot ey

we

Aly

116

l.

o

a rod

1.1497

Te

LO

3

, >

33.273

Stel

oO

> Ni

Gi es]

wa Ww

-™ nt

ny

ny

A

4

a

waa

1.11907

LeLLil

xh

Pe

fe

roy

>

uy

sy oa

ow

OJ

OO ee « «4

UN LO

M™ md

mL a a

ri AS

we rm fr hee

woot rd e e

wed el ed

CY OL 004

mo Ne

e ¢ ®

CO penned

Cow

TIMI NOT Oo

* « @

TD ed

coo

ere, en

ae) > uw

rm m7 wh

eo * a

OO pent

ay om OD nil fae

NS TON ad TY A

uy od

a)

De

LO.)

NE

Gy

™i

Ww

om .

ry

mt

m™

* oa

uy

rm,

1,5332

ob

ra

Doe (45

5314

bo

_

OS

11.4

cy

roan

oe

50

ran)

je

1,928

3 a

7.0

303

oll

uy 4 “sy

m

Ff.37

Loe.

uy

cn

md

ry

iN

ae

UN

ot

193.32

2

134.03

a] C fy

<a

io

ion

oN

ne

ey

nvy =?

74

ID

0 4 4

LI7.13.

f.5))

a) CL

+1%

TABLE

IIIb

(Continued)

Exp.

Re

l.

Calc.

Exp.

I.I1.

26

26

hk1

029344

oO

LideeT]

LL1.935

elt}

{LlG.93)

\ 3 2

(11.

3

a

Fy

e 1243

om:

ee

112,388

527

Oe9117

oO SI —O

LLS.

my

coy

~

uN

129.736

ny

(121.

~f

(Thos

uN

123.3%

531

uy

oN a a

al

Nem

tO oo

lan) o

wed

ot

iP

f12

L23.%e2

117

ta re

e e

SO ey road ood

MD LY

Oe apPocias OO oo

¢ 6"

Oc 2

ry OY

me Pam

Bheoooosr * ®

aper Pp.

co nm

reed ie ed

a atti,

Cth eet C9 Me OS

« a e

ost é 3

mor wy

wed ct pod

bateteli eal

wh) ON

mye Mp 3 ON

ry

7.6

190

9.23145

CN |

os

we

Wy

cm

my

a

ua ww

17%

ee.

t 142

cn

118

* ™

6°62.

C° Ge

T*O

T° 472

C°o2d

T°?

Ae OE

C eee

f° 27

CeO]

E°4

Leh

Cee?

& °C

c“S

C°CT

T° t?

GPoty

¥°G

ort

>

G*T

G*T

>

Get

C*Ce

bene

cece

G*aay

cece

E°LG

C8 OG

E°8E

C°CY

e*go cece

T*15

G°aey

t* cd

C°CE

T°?

O°62

L°Lt

G°*GZd

G*O

C*Tt >

“T°! “I°!I

“Tay °oTe9

"Tey snoptaelig

SOTITSUSJUL

SUTT

DATRIETSY

snotTasig

jo uostieduoy)

(penutzuop)

III ATAVL

Cee Tee cCh amar TTt

~J

—4 ee ed med ow al

NTE NT werd r NS 4

Dis TOE AEN St md ce

™~ Po "oe ret OP wed OY o™ _

mee ee OP et dN Ty ot

) pond DAO tre

AS

we

Tad

a)

Rel.

TABLE

IIIc

(Continued)

Previous

Rel.

Calc.

hkl

119

SO IN TO LA SR LO Be OS LO OO et a PO I OTT HOO

eo ¢ # 8 @ 8» 08 8 #8 8 8 & 8 9 * #8 8 &@ %8© #@# @ G8 #8 &

AINA OO POD Sat mor rm er OOD ND ~ ed | Ldn ot rt

Cco™ > om ry ™ Om 7 TM OY DVAO oe) o . e e e * e e * © ® e a a a e s ®

CY rt TD m2 my on wm EYL moe SBI a) uy rd Cy Ol nd | Gobet mr} A ONG es “Yd

WV

ON met SE eh et med OR RE ON POO OO OOO CV TE OD mE TD Tmt Oe a TD OA AOS RHO OW

VEY ON OO ER AD SON TD BIN OH TM AO

TABLE

IV

Film

in

Th

Lum

A.

Intensities

for

26

Region

35°

to

45°,

As

Received

2.2

um

Titani

Int.

26

Int.

20

Int.

20

Int.

20

ony) fi?

urs

eS

G

OM

mt!

e

ny

ve

oy

Ny:

4.?

ty 36.

“ha

4.5

T 37.37

37

: Ye

“fh

as?

cel?

8.1

23

aad

aN

4.8

osl

in

“!

” MA

mm

120

38,59

fi.

ay

SD

wey

a

a?

de

36

a wt

‘be

39.290

3 ~ +

5.7?

~ al

or

l.7

el

) o

Ino At

en @

PU Ln

OO OD TS LA LIN

* ® e

abo oa

PN fs AD

*. *

QOL IN

Phe Be. 1m

"QU TA MS - mm ©

CN ead

my xb oot

Iya ® a e

Uo UIN

Wey ey a fe mo

. e oo

(Py ed

CV wPoer

“ye e oo» «

Ay: in

ee Le

e *

eS ~y we

oe LF

. °

LA

o- mf

a *

OL oe; aT At

wt C

a

ion i

cay ms he

« a

rd 5

wht

a *

-_ ie)

fhe em

nema

° e

a4 OL

wt

Wa 7) s a

iN SP

ON LS rr os

e *

cot | eo “te

309

44.90

St:

ny

5 42

“+ De

ae

end

121

“OL "Qh

“2b °il "OL "ol "Sl "Sh aan "Oh "tl "ek "él

"el "tL

* a

0 0 0 0 0 0 0 40 0 Pe 0 0

i DOV Oem CO RS CO ty cr Pe

>

MLN OM POM NN DIN GY OO °

WOON DO wonwnowe nn;

a & n

>

tO ory a

u es H

Oc

ee © = @ © 9 ® a * e « 2

CAN STH ORTON BEN Ca o

9oOOOW ODOM HOV OO

-4uy

CEo*ed

CS*Rh

Cicie

COT LE

cl °*sol

ceecL

Gi°GL

Ce*Sh

GLA

G2° 7h

Gite

ce*

el

GLeek

e?g° dl

COtTE

e e * 2 ® e »

APDAXHMOOMOP DAN DM AO e

oO.o Op O ON 0 CEM 1.6 0°09 9

« qUI

GCTARL S2°Sk

corse

ertid

£9"*OL

CL Sol

CO*Gh

eT*cL

Cot ye

CTT¢d

georee Etech

c9o°2t ET*2l

OG*O)

OC

ON OR LOD ®

>

WO AM Ce BOA wD a

x

. qu

CETEe

COPEL

corel

OG tle

CC%LL

CE*GL

COSSeL OGc*tc) CO°G?L

ear

eee y

COtH)

CGY?

OOP

Ed

CS°?l

CCe% ed

CoO*oaL

UTFa UTYL

wnzueITL wl

7° peatesey

sy ‘0g

02 (OL

uoTSey eZ

AO} seTATsueqWT

“g

122

CR OS eed SEO a

LAA On OL

wy oe

s * ¥

an Oo t-

"9

AVP he BOW OOD eo 8 @ # 8

wor sh oO fh

FON «¢ * 6

he CO LA

Mine. «

+

‘qu

~ 3

a CG

OL

O€

OF

GC

O& US oe)

OO ON TM ANI SRO ee LN TL Bee OW MINN OO MIN HL TOW Ln

oh

SL

as.

ot. 44

eh

"ty

"cv

"co?

*é%

“17

"TY

"1

“OF

"CH

“Us

"AS

"6E

“6¢

"BE

"SE

"ee

"Ee

"Je

"LE

"le

"Le

"LE

"OE

"GE

"SE

vy - 8 @

Mo DAA e 2 8 * 9 * ~ o

DIU In DOW IAS AMA 2 ® a e a s « e

Oar wr Ah OO OP

WO OG OR PR ASU et I a

Om PP IN

PO MmcaN a

. qUI

Corer Ore

OL Ev

cto

CGT EY

L2°*e CL

*?Y¥ e°4

CE SCY

O°S

CE° TY

C*°9 Cle

Ty E°G

CG °

Tt aS

OL' Ti?

E°%S GL*O¥

£°S Cb

° Ob

%*9

Ce

C*s Ol

°6E

&*6

GLSG6E G°6

CE TSE

T°9 CT

6E 3°9

CE*EE

*S CY

OE eet

CORE

Qfy Ce

ele Be

OG FFE

L°9 Gee

Ze q°o

CELE

c°S

GE°GE

Crs CL

TSE pet

CL*9€

TT" 06

°GE e°e

QZ "qu

gout}

emer et

wee oo¢

1G*° Ev

05°

24% CSY

2H

Ce CY

C6°

TT?

° TY

ere,

iy

“T+

"OF

"Ch

- ane

Lae €

"BE

ry

~

~

ae OO + DLA MIAN OD

® on

ey

UM OPA pm ® % 4 =

cn

aa

oe

ey.

*

Lm he co

*

hm fh PD SEA oy oN

e

Med ¢ e

or -- yoy

SETS PALA LO LA UN LEY M2

s

5 a

OC

TON M™ Aa e

NIN 13 0

65 mF

9

Gee

G*6

f. 7

fF

L°>

T°s

i‘

34 9

C YG

or)

aa

6*¢

Q°%sG

Ge

O° 4

RFE

“Aur

red

~

a 1

QO o a

WOT SDT A Mey Ot se

OQ ry

n

OM Te RN

D,0S€ #8

SsanoH Z

Joz [vouuy

dais “Oo”

0} SE

uoOTSey 9Z

AOF SoTIIsuequl

‘vy

A WTdvi

123

ra A6%)2

OSstiL Cr

sed Gi°gl

Un* OL

06°SEL OS*Sl

O1°SL

OL Te

Oct el

C6 %EL

OsSs* el O°

ed OL

ed

Gee dh

* * @ @ a ew #@ @ we wmvowuo COO OOOH & OO

Mo et A SP OLA OD OO OO rs Bh LAS e u

a 4

8

whe

eae #® @

NO

9 s s e

mary OOOH OL * ”

oon -> * 6 ®

MO Os OW WO DMP OO ©

qu

CE SDE

O7els

cCo*cled

cg*ol

Cer

se

O8 *SL

CH °St

COSSL

O09

° 47

Og

* HL

CE EL

CY el

COPEL

C9%CcL

Oct

dl

8d

* a e ° e 2

CMN MIN IA SO com OO! OD or » e« # #8» oe @

Oo .0 00 0 0 OO 0 10M OO OO

“3UT

OL LL

tee

vw dd

06°

OR

OS °9OL

CT"

ol

O2°SL

OF *SL

C6°%HL

OG*

he)

CT°ol

CL*Ed

mce

ot

C6%2L

CS°*?2L

2l

OC

*

“O40 « * @ © @

WoO OO 090 0 OO Dh O00 48 ° e a ® e

DOaAAFINO RT OMN OD Ole st + « ‘

* UL

CO*SL

C9)

Cero kd

ag*

Oo)

Cy? ok

COToE

o9°cL

OC“

GE

CET HE

Ov *

OOS FL

ogreL

Ce

* EL

Ce ch

Oy ed

COc?cl

0% O,0SE

328 sanoH

Z 10}

Teauuy

daqs

* 584 O73

oct uoTB2eyY

QZ AOF

Set ATsuejuy

°g

124

mop TF PO om et OL SP GY OO * .

Shook OD ME INNO * * ° e ®

*qur

OC SH

OF ey

Ub°

¢%

OS°?4

OL°

2%

Gl°

1%

U7

°

OT" Ty

OL TOY

UE CY

G5S*6E

GL YGEe

GSG°EE

Gere

Gt

G6

°

Po mo

Da Qw man

ao @ @

me mm OO me eo

=

a

aa Qc un

e

C)

oO a a

a

fh Pe rye ee

C4°9€ OS°9E OT°IE GL°*GE

9Z

% a e * e e a ®

PO am HM FOL ALM OO RAN UNA ed etm OLN 2 e * e a . ti] e e a

weet pet reed

a * *

PR Re LOLA IA UN OO DOOM OF 8 ON xr OO

>

u

c°s

gS °

€°%

G*%

“QUI

chtehy Lt

CZ°Eh Tt

C&etty ory

Ch°Zh #8

COt ey

Z*9 CL° TY

9°9 Getty

G*9 COtTy

19 CIO

Ete cette

te CO"6E

£°6 OL

TEE cl

CS*6E 1°t

CETSE —

LG CT*6c

Gt Getee

8% CL

“RE Q'S

G4 '8E

2°s CoTke

bt CG*LE

bt Sb

LE Gy

GeLe CFL

SUTLE Bw

1g "FE 1°S

CFE

ott CO

"9€E Zh

CErse oy

02 "Ul

red

uw Oo

« Clo)

Oo OQ IN a °*

.

e

~ GS ©

e« 6

UV LY SO Sa om hein ih

Co h

*

> wy

*

AM AM AMaAM mec

°

PSO WO DN SS MS woe

OTA a

Mm FS e *

Nb et UN Le e

cD re

“jul

~GG*LE

J,0S7 38

Ssinoy Z

ioz Teouuy

days “O97

07 {SE

uoTSeyY QZ

AOF soeTATsusqutT

cy

IA ATIVE

125

‘UI

GL°HY

c°e G

A * UT

(penutjuo))

(V) IA

WIdVL

tre

“UL

126

G*9 O93"

SEL v9

cee ©

GPL OE

PLL

er O¥P

LL 9°L

OOTLL

BL

CS*9L

1°sS Oc

OL

0°8 Oe

Sl

Z°l Or°sl

S*2 CL°GE

G°l CL

°oL ofl

Cet hl

L*t 06°

EL SPL

OS* EL

ye GTtel

Tek OL

el

ToL Oe*

dL

“Uy

A

e * e t f * ° s e ° e e e e

Oe EEE EE Bee eM OO

mt SOM OIN OO DW OH A OOM OWM °

“quT

C6*El

OG*

ar

Ci Ge

OL LL

NE %s)

06°91

OS's

CT*SL

OL GL

oe tcl

OG SL

O9°oL

O¢* +L

OS cet

Ch el

COS EL

OS*dl

Oe" 2k

82

>. °# ee & @

Q 0

AMO DMDOMWCOWVWO OA Om MOooOoNM ° e

SO Pe fe Be fee Pee fe Be Pe 00 CO fe fe Be OD

en «

e UT

OS"

Sl

Or sh

COS

CST*Le

Cereke

GeeoL

OF

° OL

CO*OL

O9°sSt

Od *eh

C6*?)

GG Soy

OT St)

OL El

Cetel

C& %?2L

0G*dL

O1T*Zl

82

e @ Oo NO

a

7

e e e

Be Pee Be rr

*

wo

a eo

cn e a * ° e

No a

Poet GE LAM SO LALA IN DD Rt OM I ed OO LN *

e qUL

CL

* Sl Oe

fed Boia

CG*2d

CETLE

OL S92

Oc

° ol C6°S)

OS*S2d -OTPSL

O8 °o)

OF* aL

COSH)

Go ek

CZ°eLk Cece

Cy°?al

00O°?7L

82

0,0S7 38

Sanoy Z

Joy [veuuy

deqs * 64

07 2L

uoTsay

QZ TOF

SaetTATsuejzut “*g

127

G*] Cat

ey S°T

CL EY

S*T Co*ery

o°T CG°CY

i*t Ov

Cy

6°F Ce

* CY 2°

2 Gere

C8? GL’

eh ye?

OO*% Ee”

Be? C5°eF

6° OS

8 e+ O°E

OL cy

C°E COT?

T° €e

CS° Cy

e°e Coe

ey EE

CET CY

GE

Co* eh

2° €

C1l* 2s

6°%e COt?Y

oO" %

GC6* T+

oc’ Ghee

Ly eet

CLS TY

7h

Co* hy

G° 7

CG°TY

9° 4

OU" 1%

G* GOST?

£°% O2°T%

ce #

OT* Te

T° CO*

TY &°E

06°C

E FE OCS

Ne ere

OF oO’

Ble Uo

Cy Cee

CG" Ct

Tt. woe

cee CORY

yey O¢

°OF Gee

Og *G4

€ ° 4%

GT ° OF

Gre OT*

OF e*s

GGS Oe

G°G COP

Oe ES?

66 °EE

c*? C6°HE

BES St

Woe wy

eeooLe ars

crere

pee OP

TEE C8

GO*GE

€°€ COOGEE

6°82 SE°SE

G"&

OG

* GE &°

6 SYP

Oe 2°6

CH EE

9°6 GETEE

76

CEC AE

ye *

Cet Ge

US

Ce PGE

ore oA

ES S&L

TPE?

e°k SO*E&E

G*9 COE

6°C GG“SE

O° COP

Se G*s

GEt ee

L*4 CET

SE G*

> GLYSE

a*y OLTEES

G*E GOTEE

See CF

eee ore

37

GE we

CS TRE

E° Ee OPT

CE rte

Oe TS

Tee C2°RE

T°e OTS

ee

6°? CC*SE

6°?

COo° LE

Cre GSE°LE

GP? SLeLle

oF E GGT*LE

Te

CG SEC

eee Saeko

cee roere

&*€ Gee

le c*%

CvTlE

Ewe Ge

ThE 1°S

CemLle c°s

SC°LE

T*s OF LE

f° GT°LE

Oe D

OT°sLE oe

4 GUT

LE c°

COPLE

T° €

O05 °9E

S°E NOT

GE oe

OL os

ne eee

per cote

aa CH

OT

O°E

0% S9E

a°? Qc

PSE 9°

OT°3e

G°2 COPE?

G*2 GE°GE

S°2 CO*SE

Se

GZ°GE

G°? CO

"GE

*3ut OZ

“UT

QC “UT

QC “WUT

8c

9,006 372

Sinoy Z

103 Teouuy

days “Oo?

03 SE

UOTSIY

OZ AOF

SOTITSUeqULT -“°y

ITA WTEVI

128

e*T o07SY

Gotey oT

OG tee

T4 "qU]

Oz

(penutqzuog) (V¥)

LIA WIAVL

129

T°g OL°SL

e°9 O°

Sd Be"

9 enone

a4 O°L

OG*LL

ek

OlfsL

0°8 OL°OL

3°83 Jo° Ll

o°S 06*Sl

C°s OS*SL

Aee OT°SL

S*L OL°RL

S*2 Ge

eel yok

Ob

% EL SPk

US ek

O°h OT*el

L°9 OL

% ck T*9

Oe*ck

* ULI 8¢

®

CCP

ed

co* el

CE°R?

Cress

CyeLb

COtle

Cg"

oL

Ce °9Lk

Ce*sk

OV" Gk

Co ece

Cott

Od

° HL 08°

Cl C#

eh

GG*El

Co*eL

él

Om eNO - e oe *&® #8 @ a e @ * * ° a

DORE EEE MOO OHO DD DO

AOR FOE EH ONO M +O a

*3UurT . 8¢

"qUI

OS*Soh

OT*el

CL Shh

o€ fil

O6°9L

CSo"SL

ToL

OL°StL

cCE*soL ar

Of 9%

Cot OL Olt

OLtEL cere 06°22 CS*2L O1°zL

82

a @ @ @ «4 ®

uw? hh Mh D0 40

*

Com

J e e

hh

"Lh

o°L 8°9 Z*9 org

eA EIN ON MLD POD DIN ©

*quy

cer°sl Creed

Cot el

Corl

Coole

Oe* ok

ObP OL

Oe od

O0So°sL O¢c*Sh Oe

theL CHL

OO° FL

OP el

O2* Eb

CBr ed

CH°SL

00O*%?¢L

OZ

9,005

378 SAnoH

Z AOF

Teoauuy deas

* 562 0}

ocd uoTsey

9Z OJ

seTyTsuequyT

-g

TABLE

VIII

for

28

Region

35°

to

46°,

Step

Anneal

for

2 Hours

at

550°C

Intensities

A.

Int.

20

Int.

20

Int.

28

Int.

26

35.75

2 el

2.20

205

LN

LA

my

2.4

Ol

sa)

2

2

35290

N

209

; 4, 36

e

uN i? ‘

~m

LAN mx

un

209

36.25

35245

7

3e1

NS

3.1

—t

sO

rey

uD

coy

230

36

35,59

323

c ad

36.75

5 366

302

No

ood

cy

ov

ms

oe)

cy

LY

» “+

my

2

sO on

we

oN

Sta

od

2Ge

sb

37.225

4.22

37239

36

3.4

37245

4.4

37215 in

ey 26

cS

37. 4

3.7

37.35

37.70

329

LN ¢ M+

cy

°

my

ALY q

oy

iC, im:

oy.

130

i

co

fen)

oo

mi

a

Biec

oy e

33.30

3el

320

3.7

ro

© mt

mW

eee

vO

me

38.55

> In

tO ‘

aa)

49

3

st

LS

549

126

a

39290

33.95

5.3

o.9

Ged

39.279

343 40

9.1)

wr

ra)

oo

Li

ef

uN

ch

ny

uw

5D»

2 12.4

Ya

fel

mM

A

Jaw

49259

40.09

X ye

~

a

eo

aT

a

146

h-

en)

A

39,655

39.285

629

4.9

39.95

ay

Des

39.99

“od

41

40220

fy

OQ »

4 0

iT

49,45

iy

(Ge

LO

40230

Ze

3

fy me

40

& GD

40.225

uy re,

ww

4.20

~

Ny Fa)

4.0

3.20

49.50

Gel

4.9

Ly

©) -

329

49.39

wm ~

on;

) 40.7

49.55

i

42%

> ms

wr

°)

cy

41.

50

41 eh

Let

4.6

41. 4d

4l.

4.3

4

A

at

oO

uy

41.79

4.5

41.60

131

G*T GL*SY

S*t o°

T GLY oy

O° 7]

L°T

Cot

ey

i*tT

GY?

UG*

EY O°d

T°2

OT*

cy £°?

g°2¢ OLY

Sy Gk

G°€

Oe*

2% T°

*3UuyT 6Z

*qUrL

OG *G¥Y

CG* Hy

8 ee

CYTE

CC*

ER

COP ey

Od

* S47

ON * 8 a

te af

e

or on

®

POA NH ated

Oy es ®

*quq 9¢

(penutquog)

(Vv) IIIA

WIGVL

FLO Mo Mh OLN *

SPOT ND ND 4 ed eet

W S eI

132

e

0 TO AO! co co ®

et OO OD ioe eed

9°*cT T*et

o° tt

€°tt

S* OT

Z°OT

O°TT

O°tt

S*OT

G°6& T°6

° JUL

O9° se

C6*LL

CS*kib O1°LL

OL °9L

Oe° OL

GS°scl OS*sGe OT°SL

OLS oh

Cer el

06°el

OS* ee

OT*EL

OL*é2

Oc*él

8

8 °6 8°6

e°*cl G°tt

6° 1t

e*ctT O*ET

9° TT

c°TtT c*Cl

i°ct

O°TT

Ott

c°ct

L£°6 0*°6

“UI

GP

el O8*id

CHP Ld

COPLEL 09°9t

O2* OL

C8 °Gk

CH*SL

Core?

OG *hs

Corel

cer*el Ov°cd

OC TEL

CF

* ck Céo*cl

QZ

&*6

e°6

G°6

7°CT

c* tt

c° dT

O° eT

“eT

“TT

“OT

"OT

“OT

“tl

"OT

°oT

*6

*8

PN ae Oe DW DINO

*quy

CCT)

Oc

* 8h

O2°LL

Certil

GE *o)

OT*9OL

OL SL

CETGL

CEPbL

OG*HL

Cl

aL

OL cl

Qf *El

C6 cL

OG°*écl CT*

2d

OC

8°G O°OT

G°ct

y* TT yee

c° ey

C*ct

¥° TT

ns

mm

eed

9° OT

6°CT

C*TT

$°OT

mM ey oF a

Dan

-quUI

CEP)

CO*%ek O9°2L

COc°Lle O08

OL

02° OL

COPoL

O9°SGL C2°SL

Ge toyL

OF HL

CO° ey,

Cote?

Og "ck

OR TCL

CH Yeh

CC Zh

QC

D,0SS

38 sainoy

Z Joz

[eeuuy

dais * 62

03 QL

uoTsay

gz AOF

SeTATSsueIU,

“*g

133

Bre Cc

ey T°4

usr TY

eee One

14 G*°e

OOP Te

Gre OF

° OF G°E

SERCY

Be? SO°%CO%

O*°9 S8°6t

Cc*e SO°6E

L°*OT SVYSe

Yee Sc*

OF 3°6

CC°6E

e° GE

PE

O° tr CGP

OE e*t

OF FEE

B°C? OOF

EE c*€

CELE

Crs G2°LE

GP €e

06° SE

Ere GO"9E

ee

Gyro’

Cee GU*

oe or?

OG “SE

c°e OS°SE

T*? CESGE

er? OL

° VE T°?

Cer ve

“UL

OC

e s * . e

TUN Oo CIN MMe Ss | * @© » © © 2 @ «© @ ©

MNO

TOO OASF RH ™ MH We W MO ODO WLM AD «

“yr oY NI

on °

AEN TS ae @

AL MONE AL OA

an e

-quI

rao

dc Ate

ChE

Te T°

Oe * TY

ere

CE°O"

G°€

CGS °C

Z°E€

Cc°

ay €

°F

CO ea

ore

CE TEE

0

*h

Coté

C*H

CL"

G €°Ee

CET

Ed cre

OC *° GE

ye?

CEE

E&Y

peree

cee

GCETEE

O%e

CE°LE

O°E

GG FE

L°€

CES aek

C &

CEYSE

GE

OG *GE

Gre

CE SOE

ePE

CTY

SE ge?

GETCE

c°?é

OF PGE

ara

CO FOS

L*¢

OF *

Ee E°?

GET 7E

6*T

Od *4Uurt

pont peed

GO"?

Oo

* Te

C2°Ts

C&S OH

GH°*O#

cT°Ch

Coe

Oe

GL TGEe

GC*6E

GE*5E

Gite CE*GE

Gi eee

CETOE Od

SRE

Oe tLe

CY?

2de

OL se

CL SGE

GG*9O

SC'9E

GC °SE

CL YGE

oc *GE

OAS OE

GG PSE

CLI°Ve

02

Cr+

Ge

o°¢ Lee G°% Oem

Gtk 8°6 eth tai

Lee oh

TAN el AL A OMe GDN et e

mn

“Ur

CO. TH

CG* ET

CL ty

CL °Ov

GE° OF

CT*O¥

Ppaoer ce

wwe

tre

CLLEGE

CG*HBE

CES 6E

NT

Re

ot

woo

G6

° BE

CL TEE

chene

GT

Ss

CL ELE

CereLe

COLE

CL Y9E

QG*gEe

OZ

° OE

CCSSE

OG° GE

C?°SE

Of PEC

OF TE

COceEe

DooLS 7

Sainoy Z

1OzZ Teouuy

dais “597

07 ve

UOTSeY

9Z AOF

set AITsuseqUT

‘*y

XI FIAVE

TABL

E IX

(A)

(Continued)

20

Int.

20

Int.

20

Int.

Int.

20

42.50

3.0

202

42.60

42.00

323

3.7

420%

3.5

42,30

42.70

2.5

2.8

42499

on

pod

ct} wN

wT

cn

st

1.5

rn

pod

wy

yy v

tH

uN

a

css

a 2>

44.90

Ls

7~ 4

ew

5

134

135

E76 o9teL

S°6 O6°LL

6°6 CStEL

S°01 OTTLL

E°1T OLTYL

E°CT CEtOL

S*ET G6°SL

O°1T OS°S2

O°lT O1TSL

L°01 OLt HL

S*OT Ofte

&TCT O6°C2

8° OT OS*EL

6°OT O1°EL

16 GL*2L

O°6 OeteLl

"Quy OZ

e e * e *

OO DAHAN HODaA GD god pet ed med ped eed eed mel

a m=

aed

Cano oF IND Dt anm ®

QO —

e

o> T S°6 6°E

eo qUuy

Cy OL

CecrLle

CETL

cowed

o9*oL

Cé°

ol

Ce °sh

CHS

Sh

Geo)

C9° ee

Coco!

Cec EL

Cy*TEL

CO°

El

Co*7l

Odc° eb

€°6 U°6 E°6 T° CT J

* iT

Z°tt

G°et

e°etl O°

TT EP

OT S*OT

L°OT

R*OT

L°OT

S*°OT £°E 6° 8

“jul

CC*%6L

Oc

° aL

GCE eLed

GCetld

CHOP

cco °*od

OT*

SL

OL SL

CE*GL

OC6°HL

Ge ft

CTP ee

OL TEL

Oe *el

C6

%* 2h

OG*2L

CT ed

8c ° UT

CS °eL

GO*el

OS oka

Oc

LL

C8°9L

C8

OF

0O°

9)

G9°SGk

C?2°sl

O8 SUL

Ob

eet COOL

Co EL

Cc*el

C8e cl

Cee cd

COP¢l

I5S2S

38 sanoy

Z 1o¥

Tesuuy

doajs “64

03 QcL

uoTSay

QZ AOF

SaeTATsuequtT

“gq

136

c* 7

CT"

¢¥v c°4y

C*+

CL° TY

Hee

OC °T#?

£°E

C°¢

Ov" OF

ere

Swe

OUTCY

Lee

9° 4%

00° OF

O°S

T*9

GQE*6E

9°S

y°8

OS°6OE

O°S

46° TT

Ov *6€

yee

6° T

Cot ot

fe

1*6

OC *EE

O°8

c°S

OR8* FE

it

c°'e CG*he

Cft

£°eé CL°Se

or

6°Z

CL hE

cre

Yee

GerLeE e*€

9° ¢

GEi*dte wre

Gfe

SE°9E

Gre

L°€

Gitoe

2°E

ote

GG*OE

GTe

€°¢

GETSE

T*€

A*2

GI°Se

e°2¢

C°2

Ca*Ge

e°?

Cr?

OG °GE

Cf ¢

ie <¢

OLS SE

Ged

E°?

CL*¥E

cw?

3°17 OT

VE e*t

* Uy

8Z * Uy

reed peed

CO'C

CoS

Ty

O1°

T#

OL"O%

Ce tot

G6°EE

GLT6E

*£f o

"BE

e fe

ke’

GE

"EE

moo uN LAA ES LL

ooh fh oy.

e

a “

~ slg

eS

"LE “LE

CEPGSE

CLE

CGSOE

Co

ene

'

OL* cg: om ec US oct

om Or! wd OO

TA LALA

°

hw PAT MA OD wl

OZ

vty ErE Zee EcE 6°E E°G OTL 9°54 GthT €*

27 669 C *4-

Q°2

i°¢ Ete rE

nte QtF Gt

G°Z Z°2 0°2 at? Z*z 9°T

-4uy

C6 °T+ Ch

ly OGtTY 09°04 OZ

O04 CA%6E CLOSE OG*KE ares CT

COTE OL? G >

© fy i

vy

wo. PD

mm een ey rn

LA En ALN LN LN Dye Oo DAN:

e

9 roa

A

®

COL he BM OY A Od a

EW DA Tm UF OO OO

C1 et

e

_e te

*

an A

°

oe LN a °

TP OM IN md a s 4 e ¢ e

mA ON NOY NIM a oT MT ers

2 OO AO OU a

“qUT

ed et

one

OR

* Et

Ce STP

GES

OF

in

a 8

= a

e-

CQ ed

OG

7 wd

e

m~ o oy

a ony

MO Oy Al e

co

wr e

2 OD

DDININ MIAN INO Oo ™

e ay 7

7

3 @

™ “l

a

LO) fe fh Ph

~

OS

oge

olf

co*

cg

Od*

Go *

Cv"

COTE sroury ity <3 em a ”

MASA OV AO AMO ACA A OY OS MD CO OO OM NE

8d

09009 38

sanoy Z

Joj3 Teeuuy

days “697

07 ve

uotsey

QZ AOJ

SaTyTsuequtT ‘ty

X FTaVL

137

°

rm oF wa

a * my Oo

ee ®

MYL OS ed ed ed edt

quy

G2 °S4

G2° UY

Thee

OE*

ey?

CE CF

OS° CY

Og

sO 0 Fm e *

CON ed ee ed ed oe

wy SG OF .

-quT

COY%9Y¥

COSY

1O

* ty

wr

“C49

Cd

° eh

C&T CY”

Ob "24

67

(penut iwo))

(Vv) X

FIVL

- OM OO Fo e a a 2

CN AD ON ed wt ed et

gf

. qUL

° ¢

CA Pe eet OR Pe zoe * 8

NNT ET eT SON Popp sro

OM Ca OU OZ

e Ln

PTOI n a aetie

o

ee CON TN *

. 3UL

38

e* Tt

G* TT

€°ctl tat

S°oT

2°02

s*et

9°cT

L°utl

“qu

CO* Cs

C2Z%+9

G7 °e9

C9*

29

Ob *T9

COSTS

O¢* GS

O° 6S

O9°SS

82

e*ttT G*tt

9°}

c*T

O° et

O°) T

9*et

c°él

2°TT

‘uy

CE

° 29 COS

HS Cc

°E9 C7"

29 CS*T9

Coetce 00°09 02° 6S C4°8S

OC

e°ttT 6°

UT ere?

S*yT

S°6T

R°oT

e*el

T*¢etl e°tt

“url

Co Pn

C& YEE

QG°E9

O¢*

29

CH? Tg

Of *Ce

CB YES

O00

° 6S

O¢°8S

7° TT

Te?ci c° et e*st

e°Ce

Pet

6°cT

6° TT

6° TT

“qu

C7 8ue

C9*ES

CE°

CS ©

COr?Y

C2 °TS

Eo

99° 6S

ceress

OG" es

OC

0,009

38 Ssinoy

Z 103

Teeuuy

dais * 599

03 8S

uoTSay

QZ AOF

SAaTAISsusIUyT “g

139

s

rN Met et OO DR * @

oud praed

te et ed ee tt od rst eet

9° TT O°

TT 8° ot S*OT

O° OT

S*6 S°6

POA AN oc Tin Des Wma om e

wnt

poent

e UT

C6" BL

OS°&L

OTS EL

OL LL

Oer° ld

Or

Bret

wu

ad

oo*gL

OT*°9L OL

GL

OErsl

06° HL

OS HL

OT* tl

OL Eh

cece

O06

° cL Gt

él

OL* él

CL tL

OE PTL

82

* * &@ 2

SOO et ORE a

No set et OO OND wed wed eed ed eed red med

e

ord

ome

T*ttT €°tt

y°TT

G* IT

S° ET

8° OT

T*OT

O°OT

S°6 S°6

-3UT

Cs° el

OF

° el COGtel OS

Sid OewldL Cee

OL aa

COP9L

GS" oh)

Co°scl Cer

oL C2?

bl

CO PHL

Co El

Cee

Cee dL

OF*cl

OC* cL

Co* Tl

Oc °TL

a *

eA ALIN OO mM DH rood

et TOON O MIN FO Ww * e ® e« eo *

wood

rod ‘ed emf ped net md

pot rod

° qUI

OL Bl

ce eee

CE%LL

OG*/L

OLTAL

CL°9OL CE

eof

OS*sed OT*

GL CL

EOL Ceo

fee CE%EL

CS*ek

eirel

OL °?L

Cee eh

O6*TL

CS*TtL OT*tL

QC

G*6

S°6

O° OT

v°CT

G* OT

O° tT

i°*tt

Of ?eT

c°€

Cet

e* TT

"TT

“TT

"TT

"TT er7

tt "CT "6 "6 "6

“6

tA IN DH BD kt ODI at

qu

CC°6L

og°eL

Cc tel

OB cil

CHTLE

Corld

Co *SL

Cero)

Of °G?e

Cy*SL

COPSt

COP HL

Od PHL

O8°¢eh

Ch eed

GCotcL

OS°¢L

O20

* 2k

Ce ets

Ce Te

OO°

TL

9,009

72 sainoy

Z Joy

Teauuy

dajs

* 64

03 old

uotsay

9% AOS

satjtsuequy

“9

140

e » ° a e ® * 6©@ a 2 4 e e e e * e ® e s ® e * s o

AEAL NEN NCO CINE N NN OS DON DON SOM IAMS SNS

AAPM AL ODO Am DM IND TN DIM ODMMDAIN OD CIN ©

Og° CY

GS*% eV

Cy coy

COSY

O9*

1%

CoP

TY

G8 °O9

OP Ose

00°C

OB *6E

CY*° 6E

CY FSS

O¢* 6E&

uC*&t

CS°RE

OF SE

OE“ EE

CG%LE

OG°2LE OTTLE

OL Se

CEEGE

PO °GE

CS*GE

CT* Se

OLS v7e

CEP ve

8d

ee ¢© # » # @ @ &®& @ @© @ » @ #® &§ ® @ @® #& @ & @® @ #® 8 @

CEA AL AY OE NE OS SEN ALN Rr om ME SE MM Im Se NN rand

MAR OUN Dt OMOMmM MO OMIIN TN Me a tO WY

OT TEy

CL ech

Cot ey

CEST

Ger

Ty

CT*° ts?

UL SO

OE

° OR

CE*° EE

GSl°6E

SG°GEE

T°6E

€ SEES

2°8E

G*RkE

c° Se

&°lEe

Geese

COPLE

COor9Ee

CERGE

CECE

C4

° GE

COPSE

CG PVE

CC

RE

OC

IX aATdvi

* * ® ° ry ® e ry e

eC mM OA RA ON

4

ph

Ih NAO DOW test PUI PDN OH AHO OOM *o» *® «4 e# ®8® 2 @& 8 @o@ & @ @ @

mA ALN ALS NM) NANI ND 6 OS OO

OX CO wt eed

OO%EY

OG

* 2b Cote?

CeTTy

Of

FP Th OC

* T#

OC °O%

CB°6E

G2 Ge

4 i.

CS*HE

GEerae CLT6€

CHPRE

OLS?

CG “OE

CT"8e

OL°LE

Cerere

C6 "GE

05 °9€

C1LYSe GOLYGE CE

RGE CHT

VE CG*9E

CI°ve

TT IAN OR AR DO MR mR ng et OM 9 oO DAD at ed AG UY ore

ar

*

Cy

OO AG Cé

Oe

CVT HE

GG 8¢

0,629 #8

SainoH 7

toy Teouuy

deas “Oo”

07 Qvé

uoTZ9Y

OZ AOF

SeTATSsusqUy

141

Cotoy COS

SY

00 “7b

O5°

EH

OC

co SOO LY *

red red pt med

quy

GL° Se

SLA’?

06° EF

CS*° EY

OC ul

(panutquo)) (VV)

IX wIaVL

142

yO

e°OT

O°TT

L°*tt C*etl L°et

C*dl

G*tt

T° TT

6° OT

e JUL

U6°S9

0S°¢9

O° ¢?

OL

° 19 CCT

1g

06°OY

OS° O69

CT° Ce?

QL %6¢

CE* 6S

82

TP OT

L°OT

O° TT

T*2éT vt

Eel

€°el

L°Ttt e°

tt T*

tt 6°CT

e UT

C8 wC9

Cy" e9

00° 9

Co

* Ts

Cee tgs

C8 °O?

Cv*SO9 CO*OS

ECE

Ne

is

Cé*°6S

GS *OT

&°CT

1°TT

E°dT

Rect

i*dl

O° TT

T° TT

O° TT

6°CT

“quy

OL T?9

Cee?

O6°%T9

GG "TS

CITSTS

OL *O9

O&

° OG

06 °6S

Cetge

OT°AS

OC

S°OT

S°OT

C* TT

e° tt

G*Zl

R°ET

ye ?eTl

o°TtT

TTT

gS*AT

6°CT

"Jul

GO*Ees

09° 29

CE° 29

C8

° Ts

Ov"

T9

COP

TS

09°O9

02° OS

OR TES

PY

GG

CO°6S

0c

049¢9 728

SAnoH Z

10F¥ Teeuuy

dais * £9

03 46S

uoTSay

9Z AOFJ

SatjtTsuequy -g

143

T°tt

c* tt

L°TtT E€*

ZT 6°ET

Y°TT

O° TT

o° TT

O° TT

€*cl

G*éT

Z°ét

G*eT

T°ST

C°é6T e°St

G°st

H°GtT L°*7T c*

ct L°

tt £°tt

“QUI

047° 18

Oo9* 9€

08 °S8

0O"SS

02° 78

OF*ES

09°

ds

O&°

TS

OO*TE

O¢*

08

Ov°6L

09°8L

O8*

lL

OO°Ls

Od°

SL

Gl

O9°HL

Gere

Oct

EL

Ge ch

Oo*

TL

CS* OL

82

T° Ut

T°It

€*ttT

6° TT

G*eTl

8°¢T

c* tt

told

9° TT

2° Tt

G*¢cT

€* dl

e*ctT

5° ET

8°ST

0°02

G° sot

T°StT

O° GT

O° 7T

of lt

L*tt

&°TT

“qu

co*se o2°18 C7999 cee

C8 tHE

CC° HE

ez*ee 04°29 Certs og *0R coto8 CZ°6L OF

"SL COTLE Og

ToL 00°92 C2°SL ObtyL 09°

EL Ce

2L 00

*2L O°

TL OF "OL

OZ

T° ttl

T*Tt

bert

T'?eT QO°StT T*et

T° tt

e°TT

“Tt “ct °2l1 ‘ct “oT "oT “St “GT “St “ST “el ‘11 “TT

e* TT

APO MN AEN NU IN Or OLS

-qUr

CE*le

CO*°L8 C2

* 98 0F°S8

CS

* be Cle

CE

OUT EE

O¢ *Z8

Ce" TS

09°08

oe *6L

OC*6L

O2°8L

CY¥* LL

oor ol

OS °St

mere

Ne sd

OC 7b

O87 *%El

OS*2éL ce

ttl COTTL

Oc

° OL

° 37

* e e a 9

reat ACY IS UN ur, Pe ph

meted et eed met pe ed ed

NF OM ADS RE MLO MIN OQ EIN EN OD *

-qUI

OS*%1&

08°98

CC*GE

G2°S2

Oye be

UGS

ee

CO*

de

GCC°TEe

Gv* Og

O9°6L

Og*eL

GO* ek

Cech

-O7°Sl

COSGL

O8 SFL

60° Ll

O2*el

One eel

cCoOrTtL ceeck

oo°aL

OC

I4Se9

ye sanoy

Z 10ojy

[eouuy

daqs

* 588

07 904

uoTSaey

QZ AOF

SoTALSueqUyT

°*49

144

E°€ OZ*14

4° €

GOH

et Co°

0% G°G

OC *04 va

Of GE

50

O9°6E G*9T

OF" 6E Of bT

CZ°6E el

GO°6E L°4

OG°BE bE

Cytee g°2

Cette Ete

OGTLE TE

CS*LE HE

CTLE O°%

CLT9E S*€

OC°9E 7°72

G6"SE 5°

OS GE

C2 OLSE

2°z OL

HE BT

Sot ye

ett ObtEE

ect CG* EE

ary CT

CE gtT

OL° ze ict

cE

* ze

"Uy QZ

£rE

GTE

O° ¥

Gs

Beh c*

tt

G*at Cte1 L°s a

Z°€

Gz cre

G*¢

O° F%

Zre c°?¢

5°T T*2¢

c*2

He

8° T

g°1 QT

Z°T

itt

“UT

CO1L° TY

CATCH

CEPCY

GE*6E

cys

GG¢

LA ry

s

oo

in owed

°

Da HYD Vee aA TA FO

e

ns Ad

SOLA LA LA

SOLA Pm CN

® e y

™ 3 OD rm

a

Pe Pe cA rey

—~ at

* aed

0

*4

s CE

Sve CECE anne GOTEE 09°2E CZ°2e

QZ

* a e a . e

mod eed

TDL me er LN CA eH OS a € & b 9 eg Z L C

GE

cre

Ore Zt¢ are 6°€ gz

T*2 A°T L°z Lz Ort S*] g*T ect LT

L°t

qu]

oe mo LO he ca

* ® s

2 wy gS EP gs A eA eA EM

Vane

oO

[~~

e

ft

ce

Oe *

fi

rm

~

th,

°

0? 7

OGS*9E

0%

Nm 1 Pe Ap °*

eM r o

oO e

rt

pry “OM DA yn DB ee #e@ # e# @® go *® eo ©

LA Bh ®

ION DO s * e s ° e q

sont tmnt met med et INO OA NE PO TON OY OE OS pomed

Ph ™ 70 tO 9 ®

qu

06° OF

COS

° Ot OT*O%”

SB8°6E

OgFLe

OP

LE COP

Oe CH

POE COrSE

O9* CE

OCP EE

C&EN HE

CHE PE

COPE

Co* CE

OS "CE

OG 7E

CHEE

OCO°?E

A

D069

78 sanoH

7 Joy

[Peuuy

daqs * 59%

02 ,z€

uoTSay

QZ AOJ

Sat}Tsuaqurl

‘cy

TIX WIAVL

145

“ «@ ®© » & & @

SPOON NI et rt a es e n

we

Phe PR Um tO OS

rae

S1°SY

Cor

oY

Co*s

OL

CL YEH

Oe

* CY

Og

° ey

Cr "cy

GU

Et

O9*

1

9~

e +» ® » © @ »@ @ # @

CA CAN A eet mt et end

LV OD oy SPUN Bh EN ca

OS °S%

CE

% 74%

CY 4H

COSHF

CGT

EH

CL ey

Ol

% dé

Cee

cr

COSTLY

CG*

Ty

82

(penutquoj) (Vv)

IIX WIGVL

a e * * » a ®

Qh OA SS

SPONP OO) PO od et ed ret at

~ Ne

s ra ry

oS

moe

CCP Ou

CO*EY

OG

8 tty OCS

BY GS*

cr CE*EY

03° 2%

OG

° Cy CT

Cyr OL

PTY Oc

rly

146

9° OT 6*OT

G*tt

G°daT L°sctl 6°?T

c° tT

‘qu

CL °?9

O£° ¢9

Ca* 19

GoS*T9 OT’

19 OL°O9

Cc °OS

OC

9° OT

9° OT O°Ttt e°

tt S°Et

T°St

S*eT

O°TTt

qu]

oot E9 oo*z9 oz*z9 cBeTy CH°TS cette 09°C oz*09

©

° Oy]

9° OT

e*tTtT

c*ct

T*#t

yyy

C*dtT

O°TTt

“QUI

OE

* 2S 0S

°¢9 CTY

cS CLET9

Oc

* 19 CE°09

CS*O9

OT*O9

82 -qUqT

Cereg

07° 29

GO" <9

C8

1s O2°

TS O&°69

C#P C9 00°

O69

QC

0,069 #8

SsanoH Z

Joy Teouuy

dears £9

03 09

uOTSay

OZ AOF

Sat Tsusqzuy

-g

147

O°GT

T°OT

o' Ctl

C°TtT 9°TtT £°

eT 6°UT

O*9T

c*2t

6°TT

6° TT

6°Tl

tet

E*cl

G° tt

S*ot

9°6

“qUT

OL

° Se

Oe *SL

06°21

OS*ki

OTTLEL

OL° SE

OC*OL

C6°SL

C38

Sb

O1LTSE

OLS 4L

Geto

O6 EL

OG* el

OTted

OL Ch

Oe*dh

OC

Let

6°TT

6°tT

7° et

E*¢c!l e*

tt e°cT

G°6

uy

CO°6L

O9*% sk

° 8h

CGS

LS

Ove

le

OGL

CS* gL

Od

° 9k

CE YG)

CY Gd

CO°sl

09° OL

Oc

° vl

Gets

Cy.

EL

O0O°EL

09%

el

Cé°cl

QC

ALD aA oO 3 e °

ADO OOD rot nd red ed

0 e »

oN mt

o°9t

c°hT

L°ct

6° TT

6° TT

G°TT

G*ét

T*2l

O°TTt

T*OT

7°6

. UT

Cé*Se

CG°SL

OT *°el

OL TLL

Cech

Cot

OL

Coto

OT*9L

OLY

Se

Cue

ck

C6°% RL

CG eoL

OTe o2

CL YE

Cce*el

OE°CL

OS*2L

GI°ded

A

COT

C*cT

%* OT

O° TT e*

TT L° él ye

ot O°lt

c*et

C° eT

6°TT

6°TT

O° ct

vel

é° Tt

6° OT

8°6 7°6

“4UI

oeteh

Cy

* Sh

00° SL

Cgel?d O?*2L

cet os

CH os

O0°SE

COSCe

mF

& Ce

fGL

C8 coh

O”F HL

00% FL

09° EL

Cc °ek

C8°?L

dk 00*¢L

OC

0,089

ye sainoy

Z az0jy

Teouuy

deqs

“04

02 ott

uoTsay

9Z AOJ

SaTATsueqwyf

i:y*4

148

0°6 OL

98 0°6

6€°98 E°6

06°SE L°6

0S°S8 G*OT

ose

O°21 OL

48 e°tt

OEtee Lot

Co°Es L*6

os*ee T°6

O1° Es 16

On*ze 1°6

oe*z8

“QUT QZ

weetedt SOM OM cs OO} * @ © @* «

Peer aDdIan am PE ® ee ed pet od

. UI

cc"le

Cg*OE

Oc*9e

0& °S®@

uv?

Ss

00°SE

C9 *7E

02° Ue

CE ES

Cv" C8

Oot Ee

Co °?R

Cé°dcs

OC

ead AP OTM OH OLA A) DO a e 6 e os

RAPNADMINAMAAAN

QUT

od et med ed reed

G6° 98

OS *O8g

OT° 98

OL*S®

CiLYGE 06°

QG

°° CT"

2S OL°€s

GE *E8

06° 8

Co

* eS CT

*ds

Oc

O°6 08°98

C*6 C7*

9S £°6

00°98

G°6 OS°

Se ‘°OT

CE St

e°* Tt

O8° 98

e°ctT “07°

UR e*OT

OO eS

6°6G GOS°

ES 1°6

Oe

° Ch T*6

C9" ce

T°S OU"

2S T°6

OG* 8

* Ul

QC

9,069

78 sinoy

Z iojz

[eauuy

dag

* 548 03

28

uoTsoy

QZ JOJ

SaT}ITsuequy

<‘q

Film.

in

34°

to

42°,

52°

to

54°,

62°

to

64°,

75°

te ion

» as

Received

2.3

um

Ti

Th

* ties

for

26

Reg

Intensi

84

TABLE

XIII.

Intensity

28

Intensity

20

Intensity

26

Intensity

20

34.2

70

1.4

020

34479

ad

1.

1.2

1.2

34219

le

34209

oY ma

wt en

mt

342 4D

3%.

209

of

36

A

om

st ry

ON

34.9

9 2el

a)

mn a

L.3

4

fn.

oy

+ 72

@

cC

uw ce

w

on

un

ey

LA cy

1.5

us

uN

35.59

1.6

1.9

wt

un

Gn

uN a

1.3

1.2

1.2

~

wm

af

2/0

36019

Ln

om Snel

ns

NI

L.2

35.2

0%)

1.2

mr Ww

366976

0 5)

fe)

x rs

£

c of

am cs

1.2

37.219

3 i.2

mi

ond

36.3)

149

Ly oS 2 a

om ot

mo a Le

» 6

he fh cy

ren Ut,

a *

ret Cr}

oy

ar *

fh

my

We °

pal Of

CT Om fh

* 2

Ph my ry

MS! ® *

road

= > ALY

* «@

ef

my ety

37,39

oe

as, wf

my

rh

ey

cS

12.i

38215

33.35

11.2

2) 33.1

4)

: Oe

3;

9.5

co rt

ica

A

oO

oD Ww

em

uy

ny

ca en

~ eS

ES

~

a

Se

mI

a

ny

343

739

x re

ww

®

L.4

ue

ory

on

te

ea

ww

om

4

on

ay

ose

oe: e

wo

ao

ar ah ne

39019

ny

c™

aa «

&

39.95

Psy

39.99

+

11.5

Lli.

e 2215

Ae

40,19

li.

Jae

Ll.)

a Cc

fF

At

«OD

39

wr

IN

™]

t

Ww

#9.60

a

iy

7,

a a

a Y

Fa) Tr

St

1.7

49.75

41.9)

1.9

40,79

4Oe

22

49.255

4).

41.190

1.2

1.2

1.3

a 3

won

8

TABL

E XI

II

(Continued)

Intensity

20

Intensity

20

Intensity

28

Intensity

26

MS

1.2

41.259

1.

1.2

41249

41.39

1.2

41.39

1.2

1.2

41

s 29

a

41.279)

a) oO

OA! tf

ce

GS)

on

a

CS

on

Ln

S201)

»

on: un -

c ur

wt

ts 4

Ly

om un

pmo

qt m

~~ 8

{ 525

De

150

53259

5 3.

8 53.49

532259

36

3.4

3.290

53.239

3-4

54.09

529

62.30

a)

Ped

~™E

na

20

Sa

CJ

C> .

aan

0

DP

70

ey

i)

? Pa a.

& e.

a? rs

de

7.29

J Woe

t

12%

62439)

9g ?

» fy

J

ar

aD

7 ont

29

O32

©939

0

ry ~

0

2?

151

* a 9

CR COM *

Ia |

a Ld a *

Pre foe Pee Be,

e &

e e@ “CT ai

e

es

onal

- «8 @

ao mo *

Oo

OF AEM AR eM LTR et RI LO LN IS ea OOOO LOL

ATSUa UL

OL

° Eb

GE EE

CE ?E

UG*ée

GT

ée

CL Te

Oe*

Te

O6°OR

UG "OE

CT* Og

Ci*6L

Oe °6L

Core:

CS*EL

OT*SL

OL id

Oeil

CE*OL

Sa Gk

CTS9L

OLTGes CowGe

OZ

S°b

G*t

GL

Be

ck

°G

CCT

c°sé

6° L

Qk

G*L

S°2

e i

we

T"é

3° 8

2° TT

G*CT

G°&

yea

c*6

oe

O°§

AjTsus Uy

cc*ve

G*i

C6*Ce

GO *ce

G*)

GS* ER

Cé* ES

GP)

OT*ee

Cre

2s Te

Cit Ze

CH c&

Gee

ux cs

OO" Ze

BSG CE*TS

cece Te

CPt

Co*Te

Cé*1s

ifs

G1*Ls

€&°Ce

é°b

Ci*es

ote ce

cr) Me

foe

CO%Ca

Gtk

C6 *sk

CO GEL

c*Z

CG°%S

oe TEL

AS)

GES

as

CRYEL

T°S

Cl °sl

Cy oe

° Cee

Sh

CO TEL

°6

Ce*LL

2O°%22h c*?et

C°eb

CZtLL T*2T

CTtLL Ca

® Gh

4° G

Ce toe

CH SL

b*G con

od CO*%OE

ers

CG TGL

Cortes Gee

Oe"G] O€

* GE

Cre

GT eos

02 AjTsue

uy OZ

(penutzu0D)

LTIIX alavi

mL: a

he ft

eet sh

af

OT ed * &

aA

Spo LO OO mE UN I I AD ow *

i)

e

Pe Pe oe Be OT Oe

*

9 ®

a)

s uF

ran

“TT ef

"6 €a uw

Ge

O° 3

A}TSusqU

OM O M D> wr OD * oe 8 @

cn

wy on mM oO DQ mn eD ao

cD An *

PO ON et eed eed OI OAL try

2 a [b=

Cee

Ge°ol

OCT

6E

CSE O2°o)

Ce LD

Cetid

COtlL

OOo.

CE°ok

aa’ esd

CH Ges

Coe cy

152

u°9

g°9

g°9

GL

L*k

4; *

&

Bee

Lft

O°GZ?

O°L

L°€

E°E

Gt?

G*9

cre

G°e

G*E

Gre

6 °S

£°%

G*¢

A] Tsuajquy

Ae°eéed Lerrh

Ze*tl

LB°CL

2e*Ok

Z£5°T%

ee

Tt

[eauUuy eINUTW

CT

Nn e

Go e

i OT e a

Ph & 39 30 0

ped:

e

Bre

yee

Tey

T*e?

G* +

Let

° 4

"led “4

*€

"¢ ‘fo <

"€ “L maa

PO ee LOU LN ir

ARTSUa UT

Ger el

GL*TL

Gerd

Geol

GS¢°UL

ELLY

Cec te

SL° Cv

Ge "GY

CLSGE

Ce Gk te *

LS LY ~

* » 6 *

Sb OE ID IN DOM Mm om oD

a °

AeA AM TM eA A ere

La LA LLL CPt AT Be Ot he at

a

82

« <

of8 OF

(OL 6,27

03

e

[3 60 OG 4G ‘

f= ea on oO » eo

So

+ *

sw m > a 9

CM OLN @

mmm es tm em ak Re PR OO DID OAS

s e * *

LA Ph TUM MOIA IN DM OM °

AVTsusdjUy

ct°2@e e°g

Oot?¢el CO°

TL GS

OGTTL

cle te

f°S CC*TtL

CCfGh

Tee cere.

ct*OL

*O Catal

yee

CO EY

cd

T# ey

CST te

cl* ty?

Bre CatTs

CF" OF

O°S

OG*CY

Clay

G*cet CC

CF C9

° BE

Lee

CS 6L

e1°6¢C

L°e

OO*° EE

cS "Se

Cte]

CG*Ee

cT° Re

c°TtT COPE

CST Lhe

eee CGTLE

cTTLe

G°S COLE

CO" FE

G*E CS*9OE

CLS ee

C*e CCSSE

co *GE

1°y CE*GE

CT *GE

G°S CO°GE

ao fe

GSE

CEPSPE CL°VE

gre CO

8HE

QZ Aj} sue

quy QZ

"059¢9 38

“ATX

ATAVL

ove uot

say 9Z

aoz AQTSuUaqUT

153

« wo

. @

co ® e

0 0

e 9

oe © @ @ @ @ @ ©

sof Bh Be CAM 059 0

*o

Y; Cn CO OO De OM CIN IN Em Ir LAIN GS am A *

OO

°g 3

*

cr

Aj Tsusa

uy

c LA CU CLL

—~ oco coo

"te

e ~ > Cl OT ed AL a pe Co ro OO ee: 3 i

* ®

cA ®

he

LA, «

o

em

LEY LLNS °

1 * e

wo C3 SO co a

om e

SOs 60 SO 4D 0 6 GM SO om 69 42 SO 048

oO

A}TSUa UT

LESZE

o°9 SL%2L

2€£°22 Gey

C2°ZS

iste

C8

CL 1S

LETTE

Gee Sette

Le*Ge2 G*9

G1 *OS

£etcs

G°S Ce"cs

LE°6L

G*S cLeey

LE*6L

G*9 Gets

bet id

st9 GLteL

€°s G°¢

GZ°Rh

iethd

9°S Gl

LL

LETLE

Eth CetdL

igtel

Lty GL'SE

2€°9OL Qty

CEtoL

GL°SL

gen CeegL

GE*sl

e*g 21°S2

Githl

e*¢ ES

* a) G2tuL

g°9 CUtes

CL°EL

E°9 COCEL

Sorel

e*9 ZTE

Ce Zh

g°S Eat?)

02 A} TsueqUy

02

(penut 30D)

AIX FIVL

MO CG COAT LAIN LA LT EO te *

2D

oe * ®

Ph p

g°9 *O

= ny Moo Mma ™

a

“GO

AUTsua4uy

oO om x

xt

Intensity

300

26

34.37

f

Be.

Intensity

3

26

34°

to 42

°,

59°

to 62

°,

70°

to 84

°,

34.22%

ion C.

Intensity

3.9

Intensities

for

20

Reg

30 Minute

Anneal

at

625

26

34.12

Intensity

TABLE

XV.

20

154

Os AN A OL mma comm 4 = OD * e¢ @ 9 © &# @ & * @ #& © #@ 8 8 » 6

soso eT Oo Sm mm oS 9 OD sored rod wont

Od Pe ph hehe Ph Em he he fr Po A OMY ny men ee ey mm mn cs om. rs OG — CS

* 8 * 9» @ © @©@ *@® ® #© & 8 @ @ *

SPM ry Tm Oo Om PR OD OS et mn Pee CT A OT epee SS 8 Ln un

BP ON OS CI Py et OP Om & * « #¢ *e® @& &#® a 68 # @ & # @ 8 © * ¢@

MUD ETA Oe ED oem Om en wo OAD

Om rot

LOLOL LE LOR Em ETA EEUU LS ‘A As hm SIP am Pm AT Re GE p.m Be Gu me ag qo:

a @ @ ©» *#& @ © @ & @ @ @ © 8 © e

SPM LN 2 mh h coos Go: oe oO OD et me cn

my ee PA Ae Oe Le ND na Sh OSE ta LA

eH OCOD AL APR ee so mt OS Nm m2 O eo oo 4 o.mUmtmlCUHUmUCUCMOmUmCUMM eS eae es

CD OV MY OO met rt PMT CNS 6Q050 Os Cs!

QE AL Ca A OY SY BE OY AL mE AY GE oT Ag oS Atos Sed Dm me NT ret ea son

ex *# 08 &® # #®# @®& @® ® #& @ @® ® @ © «

SPUN iA Go De M00 oO OCR OD et od aA OD TH OTM mr py Se St i LY

DLN ALO AL AL AL mM mW Ooo mmm mo OD ee 8 *#® *® &® an *# & & *# @ #*§ @ am @ & a 6h!

COLA PEO oF oT oN ar Doo HV om % ond o~l

IO AOD Or oanrrwnannr aI IN DME OU DIA DME Mw Din WIA M man

> «© @® 8 @&@ 0 j8& @ ®& © @& *§ %S @& & f » 6f

SOL SD Be OD TT et et ot Mon PTA POR PA ty ho Et TES Uy

640

Ls o)

60.62

Lo.

5.9

529

61.900

O2-295

60.

59

f920

Liel

60.90

60.37

o2

O »

Q)

o

60.225

80.75

TABLE

XV

(Continued)

Intensity

26

Intensity

26

Intensity

Intensity

26

61

2&2?

6.0

6.0

9 61.37

6.9

ol.

£5

. )

61,25

5.9

62.09

529

79212

T)200

10.59

124

un

a]

©

&.5

Te)

T1212

Oo.

3 71

.2?

71.00

wn

sad

ws

Se)

uN a

read

fh

iA

so

OQ i,

155

Oe

Pe bY

? 7

ra)

wo

sp MN]

pod

mI ih

237

73.37

iO

Ol]

Cs) O

OMI

is

tt

72.930

wy

PS

on

73.2

i)

73.1:

13.00

3

we

Va

625

be

ph

Cus

Me

ve)

MI

“2

me

wn

M8

220

7%.37

14.37

5

o 7!

7 t~2 2°

7 2?

1%»

L

6.5

5 é

T/

~

Le ed

bel

ra Ae

uw CK

tm,

75.75

7 f Woe

oe

i

a x

Da

OD

tN

iy

UN

oO

<e) h-

h-

or

Me

O1

Vo

i 9

Te 7%

s 7%

Tq

T7.75

Om!

sO

ee

UN

920

wy

>

77

le

»>

q 77

.99

y T7.

a2

‘ 4 ide

52

t~

tim oO!

cn

ON

? 136

Del

29

~ at ¢

i ow4 @®

7

645

Oe

73.75

el

ce

Lo

a

fh

er

Ly

s

An

Se)

6.5

19.90

wy

79.57

nN! 2D

ga

156

a

Oo 0.9 00 a

ct e

wD wO se)

a

LV in OF OO MAIN *

AYTsus uy

LE °ECS

LETC

LS*Ze Le°ea L3*

1g LE*

LE 18°Os Leroe

QC

a xa SD

* we tes

a s

sO 0

e

IN NE NO ,

3 fp

Aj Tsuequy

EL TES

GEtER GLtze fz°78 cLt1e GETS Gi

*C8 GEG

OZ

, 8 e 2) 6D se

IN cl amin a A s 68

hm oc fe

°g

Gee

A} Tsua zy

20° ER

cL Pee

eg*eg et

ze CO*TE Z1°18 Zo

°OR Z1°Ck

OC

(penutquod)

AX AIAVL

Oy A) «

No

<r *

IN MIA TP aH

AjTsuajuy

COSEE

CS* Ee

OC fC

C&* 2&

ca*7ze OG6°TS

occ

Te

CS* Ce

COTO

972

157

i

&

un

LA Cy

e

- a8)

a°oh

rb

cru

Oot 19

CTS

Cet T+

CESEE

29° GE

PSOE 2"

EE

S € 0 ro as ™~ TT

(C€ ysnoszyq

ZI sean3tq)

suorqje[nuts

AjLsusqUy

[ejUseUtsedxg

AOF suotAIsog

97

is LA LS *

peo

Mapes uy

*

LA Pu LN OD CD Sj osr ch et am cp

*

fe TA I AD

009

ate Bee pe ™ om of @D N

Cy

oO e

co iis

mm my ot Lf LEN

*

3 CY ere!

cae xr

GLtoL CETEL OT°EL GETTY Leree Gerke

SLSy7

OLS.

Cc) *t]

G5°¢)

pee

ds

69 °EE

Bator

Cers?s

Eee

Tt

GY¥TG

92°LE

005 ,.

Ce

® eh

Getesy

cc] *¢e?}

Te

EE *Ge

cAS

Ie

Cots Of°GE GELE

05757

CztoL cet H! CL°EL

CW ef

mn ®

Px -

a.

ef

*

nad f

re)

Cech

Got

O9°6E QErLe

OSE 4.

Cec

£2?¢ Tey OO

TCS oly Cy

ete Tic

fOT OT? cce THT

SeuryT FSTL

GG *ol

Cera

C2Tel

CG

* Tt c9°6E

Lele

‘201 SB

8Z

bCO cll LO2

ECT Td

€60 O10

Tad

Sour] FLO

“PIAX ATAVL

158

ENG)

yt eb

cn

*

oe 4

ry

BN

NS Am ae

CT° ol

corded

co > pet ed OD

a a

ire ety ty ak OS

7 ‘ S294,

CociL O2°SL CG*EL CSCS

USSG

ZZ°E4 ager

CETEE NN

009,

Wm

sO 9 hf

C7 TY

wh ON met ON ON wy

2

alan

cm we NF (penutquoj)

PIAX AIaVL

*“O0902 SP

SOUTT

C#O °20 EEE eC

LTS Tel cc

5CO

TTe clt

TAY

CrSTE

159

Gert

GOST

Oar tT

CE eT

Cet

oct

un

eo 4

id

om, a

3

LA LS LA Ln

ac) m4 4 a 6a

oOnoa nar

“Ne tet SES

*

OT

C1?

GH°O

G1°T

0S9

YApTA

sTey

Oc°T

ee

Oo NS

° |

6 fe * * e

ed pod mend read

at

*

~

re

C

«

oO

QUA Mm IN LA cm

“’

So at

*

m}

t

i

ostt

ott Gite GEO ST*t Cz9

yapT” 3Tey

a

»O 0% * a

CO emt ed ped rend

oo ae

*

on CO e a

om

ad

e om

DLA IMS OR

CSP

Cet

oo wn

e

od e

t

oN

£3

MN

06°C 68°C

009 yapTa yrTey

OS* 1

Cé°T

OG*T

CO%?

GE YO

CE*O

CLS yIpT! FTeY

G2" CL*T

Ott

OO°T cert Cott

OT

Cott GE*O 94 °O

OSs yIPTA FTeY

mm Om

TU o

md met met

OTttT

ctl

c8tec

Cy ec

00S

YIpTa

JT PY

* e

{in oe LG

e

paved omg pan

CHT CHT Ot

cst COT Ef*O

OS¥ yIPFm sTeY

CceT

cy" tT

Ge? T

Get] CHT

CHT

GOTT

eg

cOtT EEO

OSE q3p FA 3Tey

“(ee Usnory.

z_ SeAN3Tq)

suoTje[nwtTs AATsuaequUl

TeJuewTsedxg

AoF YAPTASTeH

*

oO DO sFoxf ost

®

sand pect md

CEed ECC Ter CO

TCs CTY CITY

cle Tt? ect OT? c00 TTT

SsoutyT TSTL

GL°T COtT C12 cor] ce" CHO

‘d91 se

qIPpFA FTeYy

¥CO

ett CO? COT TIC 200 OTO

TAY

SOUTT TL

“qTAX ATAVL

160

Orman UN

Woah * * ® a

OO cd a

YIPpEA 3 Tey

Cac T

CUT

a2*o

SLO GL*G G7*O

679 Y2PTA JTey

009 yapTA FTeY

a ON

°

CL°C elre Ly CLS

qapra Frey

cett octT

coc CErO 240 —

OSS YIpTs FTey

00¢ qapTa FTey

(PenutqwUOD) qIAX

ATAVL

Cztt OztT

Cott Cort ocetT

OSy yap Te Frey

OSE

- CHO

- 920

~ ECE

- Zc9

- TTS

~ TET

- Zec

- »CO

~ LTE

- Ell

‘O01 se

TAY

QIPFaA YIPTA

FTey JTePY

SourTry CISTL

161

0O°T

OCc'T

Cee

GotG

OL*a

OS °O

Cott

Cet?

C9*G

CacT

oc*

aattT

Cet

Cort

On %T

Oo°T

Ge*T

oc *t

COST

co fy]

ZI‘ gic OOtT Te*0 262

oce cot are

Ww

ces 00° cot Get

wad pod end eel vod rud sod

s @# @

ITD eed OM med

*

f= Aled T™ CO A]

CO EPR ET OO md °

co oO

a7

ae)

e

nny

me * *

mm Te

*Q ce?

L£e%O

00S

aa

> *

~ 4 sme

C (

°

cCc*

a

“VT

a 2 8 ord md wd

COtT ect cett

CHC —

Leé°O

HE?

EC*O

OSE,

CE2

C2?

Tc

vA

TCs

cl?

OT

ele

TT?

eT

G12

cOG

Tt

SOUT TSTL

ec*T

cF*O

ett

TE%C

One? c9°C °Oo01

SP 9

LOO

ett (G2 eOT TTC cCC G1O

TAY

SeuTT TL0

(€€ yYsnoiy?

ZI seanstyq)

suotjertnuts Aq

Tsuequy [Tequeutaedxy

azoxy si1070ePT

N *OIAX

ATIAVL

TABLE

XVIc

(Continued)

Tisi,

Lines

CA 50

F500

Co

50

C575

S600

Ce

o5

S600

C359

hkl

as

rec.

1.99

"y

~ 3

LL

311

3220

2 wy

4.59

“} ny

a

ay

ae

O.

O79

9215

i on

Wn

O)

00)

ml

JO"

0.65

005

1 x

i

0.22

f L3l1

SLL

un oO

C 2 3

3 9

2»?

, &

9.75

Ww wm

S15

1.99

m 926

162

949

163

YL TE

CL TE

CL TR

Tc"

G6°6S -

~ TG

GO°SE -

- Zee

“uTW

- *UTW

"D901

SP

rm OF gz

rm St oe

QZ TAY

SoUuT] TSTL

- -

- Cot

cet CFG

- -

~- Leesatt

cle a

- -

ST°?

tT IT

- -

- GC*ECT

TTt¢2 -

- -

ee etl

Ev? -

- -

Ta ° 26

ETC

- —

- TL

* 98

CO? Ce*es

tee

T2° és

Set 7e

FEE CG*iL

cyeit

CLtil

c€*ilL Tbe

6c°OL

CEPSL

E5°SL

6c°9L

ett -

_ -

Ce erh

CCe2 -

- CO8

CL Go

ok COT

- -

coree

C&* 29

CT

- -

TO% ES

TO° SG

eT

GL°O?

ec * CY”

SE°C?

STTGY

TEC GE°sSe

SC°EE

ce °ee

CH SE

eG

TT*se

GT°sSe ST°SE

SO °GE

CTC

uTw 0.

UT ST,

val Ste

SPAed |

a

SOUT] TL

(Sy Ysnorzyi

9€ SeAN3Tq)

suoTAe[NuUTS

Aj[suaquy

TeJuewTIedxyY

AOF suoTATsoOg

eZ “eTIAX ATAVL

164

CE*G CZ*G GHG

suotzepTnutrs

AjTSsusqUy

[ejUuswuTAsdxy

10F syApraAs[eyn

ENC

ecter Té2F

TC

cCO

SourT

TSTL

aS °C

C6%C

Gite

CefC

age*¢ OG°O

- 08°O

- To

*e

- 64

°O

O+ °C

Ses

CHPee CG*°O

047° O

Ce °C

‘UT

CT "201

se

FOC Ce

clt EGT CTT 21

ETC cO0 CS

TAY

SeuryT TL

(cy yYsnorzyy

9¢€ SeansT4)

“GQTTAX WIdViL

165

CT we’

Ov" +

OG*e

Om AMO AQIS SL et

*

DO Oa A

Oey

e

NO @

OO Mm aw In

e

* ‘

o 2

oo cc's

T2b -

~ TC

~ ~

Z0C

SOUTT

TSTEL

OF te

O°?

FOC

on

C1" T 7

TCS 16°C

eL°O elt

- G2°O

col -

ctet CTT

- GHTC

éTe care

cato TTC

OT°%? LO%?

cCO 9€°O

ecta CTO

‘UTW CT

‘291 se

TAY

SeuTT TL0

(s7 Y8norzyq

9€ Ssean3TJZ)

suoTqe[nuts

AjTsuequy

[Tequewtiedxy

AOZ sz0

ey

9,

*“VLLAX ATAVL

166

TABLE XVIII. Thin Film Specimen 2.5 Micron of Titanium Cut Within 3.5° of the 111 Planes Annealed at Various Temperatures for 2

Hours, Measured at Room Temperature

28 26 26 26 20 28

Identification Cards As Rec. 350 450 500 550

a Ti (300 129.80 - - 129.7% 129.7% 129.6%

TiSi (610) 94.97 - - - 94.94 94.97

a Ti (202) 86.71 86.5* . 86.5% 86.6% 86.6% -

a Ti (004) 82.28 - - - 82.2% 81.9%

a Ti (201) 77.32 77.24 77.25 77.29 77.25 77.30

a Ti (112) 76.29 76.16 76.05 76.16 76.11 76.12

a Ti (200) 74.26 74.11 74,01 74.04 74.05 74.10

a Ti (110) 62.96 - - 62.88 - 62.78

oa Ti (011) 40.15 39.99 40.01 40.01 40.01 40.03

a Ti (002) 38.40 38.23 38.17 38.16 38.16 38.11

TiSi (210) 36.88 - - - - 36.87

ao Ti (010) — 35.06 34.92 34.97 34.97 34.97 35.00

Rel. Rel. Rel. Rel. - Rel. Rel.

Inten. Inten. Inten. Inten. Inten. Inten.

Identification Cards As Rec. 350 450 500 550

a Ti (300) 4 - - 16 16 11

TiSi (610) 10 ~ - - 10 15

a Ti (202) 2 1 1 4 1 1

a Ti (004) 2 ~ - - 4 5 a Ti (201) 14 8 16 19 18 16

o Ti (112) 17 2 7 10 8 8

a Ti €200) 2 1 7 9 8 9

a Ti (110) 18 - - 2 1 1

a Ti (011) 100 88 2/ 30 32 29

a Ti (002) 26 100 100 100 92 93

TiSi (201) 20 - - - - 20

a Ti (010) 30 21 91 92 100 100

*Determination of peak position better than one decimal place was not

possible.

Identification

Ti R

TiSi (610) Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti eerkreekk eee

TiSi (210) Ti 2

*Determination of peak position better than one decimal

(300)

(202) (004) (201) (112) (200) (110) (011) (002)

(010)

possible.

As

Ooo

Rec.

167

Oo

- 350

*

002 004 003

005 007

.003

oooooo

TABLE XVIII (Continued)

€ 450

*

*

002 002 -003 . 002 005 .007

-003

-003

005 .007

oooooc0oc0 ©}

003

place was

550

0.0

002 003 .003 002 005 . 009 -00 002

not

REFERENCES

1. Balluffi, R. W. and Blakely, J. M., "Special Aspects of Diffusion in Thin Films", Thin Solid Films, 25, (1975), pp. 363-392.

2. Bower, R. W. and Mayer, J. W., Appl. Phys. Lett., 20, (1972) p. 359.

3. Mayer, J. W. and Tu, K. N., J. Vac. Sci. Technol., 11, (1974), p. 86.

4. Picraux, S. T., 'Mass Transport Between Two Metal Layers as

Studied by Ion Scattering", 6th International Vacuum Congress, Kyoto, Japan (1974).

5. Tu, K. N., Electrochemical Society, Extended Abstracts, 1, Vol.

116, (1975), p. 267.

6. Chu, W. K., Krautle, H., Mayer, J. W., Muller, H., and Nicolet,

M. A., Appl. Phys. Lett., 25, Vol. 8 (1974), p. 454.

7. Chopra, K. L., Thin Film Phenomenon, McGraw-Hill, New York,

(1969), p. 266.

8. Campbell, D. S., in L. I. Maissel and R. Glang (eds.), Handbook

of Thin Film Technology, McGraw-Hill, New York, (1970), p. 12-13.

9. Gangulee, A., Acta. Met., 22, (1974), p. 177.

10. Dearnaley, G. and Hartley, N. E. W., Phys. Lett., 46a, (1974) p- 345.

ll. Cotter, P. G., Kohn, J. A., and Potter, R. A., J. Amer. Ceram.

Soc., 39, (1959), p. 11-12.

12. Laves, V. F. and Wallbaum, H. J., Z. Krist., 101 (1939), p. 78-92.

13. Kato, H. and Nakamura, Y., Thin Solid Films, 34, (1976), p. 135- 138.

14. Schob, 0., Nowotny, H., and Benesovsky, F., Planseeber. Pulvermet.,

10, (1962), p. 65-71.

15. Svechnikov, V. N., A. Kad. Nauk, SSSR, 193, No. 2, (1970) p. 393.

16. Burkl et al. Monatsh. Chem., 92, (1961), pp. 781-88.

17. Nat. Bureau of Std. Circular, 539, 8, 64, (1958).

168

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

169

S. S. Lau, W. K. Chu and J. W. Mayer, "Evaluation of Glancing Angle X-Ray Diffraction and MeV,He Backscattering Analyses of Silicide Formation", (submitted to Thin Solid Films, March, 1974).

Hutchins, G. A., Shepala, A., Thin Solid Films (to be published).

C. R. Houska, J. Appl. Phys. 41, 69 (1970).

C. R. Houska, High Temperature - High Press, 4, 417 (1972).

Houska, C. R., Thin Solid Films, 25 (1975) pp. 451-464.

James, R. W., The Optical Principles of the Diffraction of X-Rays,

Vol. II, The Crystalline State, Cornell University Press, Ithaca,

New York (1965).

Guinier, A., X-Ray Diffraction in Crystals, Imperfect Crystals

and Amorphous Bodies, W. H. Freeman and Company, San Francisco

(1963).

Harris, G. B., Phil. Mag., Ser 7, Vol. 43, No. 336, Jan. 1952,

pp. 113-123.

Elderton, W. D., and Johnson, N. L., "Systems of Frequency Curves",

Cambridge Press, New York (1969) pp. 77-78.

Hall, M. M., Veeraraghavan, V. G., Herman, Rubin, and Winchell,

P. G., J. Appl. Cryst., 10, (1977) pp. 66-68.

"Thermophysical Properties of Matter", IFI/Plenum Data Company, Division of Plenum Publishing Gorporation, New York, New York

(1975).

Schultz, L. G., J. Appl. Phys., 20, 1030 (1949).

Ventron Corporation, Alfa Products, Danvers, MA.

Dietrich, F., Smith, T., and Houska, C. R., J. Appl. Crystallo-

graphy, to be published.

Siemens America, Incorporated, New York, New York.

Diffin, W. J., Parthe, E. and Norton, J. T., Acta. Cryst. 17 (1964) p. 450-451.

"International Tables for X-Ray Crystallography", Kynoch Press, Birmingham, England, Vol. IV (1959) p. 74,76.

35.

36.

37.

38.

39.

40.

170

"0S/8 Handbook", Digital Equipment Corporation, (1974) Section 1, p. 93-97.

Ageev, N. and Samsonov, V., J. Neorg. Chim. 4, (1950) p. 950.

Hans-Ulrich Pheifer, Schubert, K., Zs. Metallkunde, 57 (1966) p. 884.

"Tnternational Tables for X-Ray Crystallography", Kynoch Press, Birmingham, England, Vol. III (1959).

"Handbook of the Phyicochemical Properties of the Elements", edited by Samsonon, G. V., IFI/Plenum, New York, New York (1968) p. 388.

Anderson, E. A., Jillson, D. C., and Dunbar, S. R., Trans. AIME,

197 (1953) p. 191.

APPENDIX I

Automated Texture Diffractometer

(29) Computer controls allow the automated texture diffractometer

to operate continuously without the presence of an operator. The data

collection technique is continuous, measuring diffracted intensity over

an increment of angular region rather than conventional step scanning.

The computer incorporates relative angular positions rather than abso-

lute values in execution of any movement. The purpose of this appendix

is to describe the operation of the following data collection programs:

1. CH.RL — operated with a chart recorder in the usual collection

of x-ray data.

2. "Search" - quickly collects data to identify angular regions

of interest.

3. RSCAN.RL ~— will provide integrated intensity, mean and

variance with scanning speed from 1° to 1/100° per minute.

Bringing the System UPY

The data collection system consists of two Sykes 7100 disks and

a PDP 8/e minicomputer which are interfaced into a timer-scalar and

stepping motors on the Siemens diffractometer. The following prepares

the system for operation:

"Refer to PDP 8/e small computer handbook pg. 3-1 to 3-5.

171

172

1. Insert the system programs disk which is placed with slot to

the back in the upper Sykes unit, push in, and lower cover

to the vertical position.

2. Insert the data disk which is placed in the lower Sykes unit

as above.

3. Turn the key switch on: the PDP 8/e (left side) to the power

position (center) (see Figure 58).

4. Select the appropriate terminai, either teletype or teleray

by switches directly above the PDP unit (see Figure 57).

5. Activate terminal

a. Teleray - turn switch on left to desired brightness,

position middle two switches down and all others to the

middle on box above terminal.

b. Teletype - turn switch on right to "Line".

In the event of a line power interupt to the computer, a boot strap

load (see next section) will be necessary if the following results in

no response from the computer. Initiating the system (first five keys

up, O through 4, refer Figure 58) is accomplished by:

1. Place HALT button in "UP" position.

2. Pressing ADDRESS LOAD, CLEAR, CONTINUE, on the right of PDP 3/e.

which results in a dot display on the active terminal.

Boot Strap Load

If a 7600 start has failed, the following load will bring the

system up without damaging any system programs. This must be loaded

on the switch register of the PDP 8/e in the following order:

"yoezr 9/8

ddd 9y}

uo Toued

uoTIoOeTes TeuTwAze,

‘fg eAinsTy

173

NAY

;LYOs7; NO

|SLIOA|SLIOA|SLION

TIVANVAAYANVAIYOLOW)

SI+ G+

E+

1H O71

qgAaw

ONINGNSE

3dVv SLHOIT LYOddNs

YAMOd

SSA INN

3SN

LON

OG Q33dS YOLOW

AAO 440

UOT IIeTAS

QO Teuluarsy]

AesoyaL ALL

174

GON _|

193738

‘zaynduod

3/8 ddd

* gg oan8Ty

| |

d31S

A300

SKIS omens

4NO0D Vyeana

tj

Luvs

|

| ovo]

ayvay

*! yagy

fotsfelz[s]}s[ele HTS

YSLSIOSY

HOLIAS

fe aul

r—

Sng

OW

aw

oY SNaivis

SLVisS

OOO

COO

000

000

O00

000

000

000

poo

Now |

| |

| L[/

ww |

SS3Y¥adgY AYONSW

/

a/g dpd

0/1 6}

1)}p

"NO Si

LHOIT NSHM

WVYdH9Odd

ONILNDSAXA

S! YSLNdWOD

LZ

LHOIT AYOWSAW

QSQN35LX4

//

175

1. Pull switch "4" up, press "ADDR LOAD", key down (all others

are down).

2. Enter the following numbers by pulling up the appropriate

keys and pulling up the "DEP" key after every set is

entered:

1127 6731 6731 7600 6732 5204 6734 7106 7006 3232 6734 7012 7012 0226 1232 3631 2231 2230 5204 1203 6731 5014 0017 0401 7400 0000

So

NR

NM

NN

we

©}!

Oo he

|

BW

h

wo CS

we w

WwoMmMororUuUUNPsAI

OU

Ov

v wow

‘ w

Av v7

iv)

w w

w w

1 ,5,7,8,11

oO Pe

PPE FAMP

AMOR EPLERRE

PCE PLELD

wow

a

HH

FOwWOONOrFRRrFEREN

FOC

OCrFOOCOCOCOCOCO CO

bh vue

ee

ee

lll

lhl lll

rPmoONnNrFPRNME

HN PNP

RP RP

NMRFPRP ENP PRE >

fea

v w

w we

« w

vt

we

\ iv

vy

w vw

\ wT

wy

At this point, go back and check the locations and corresponding

entries. This procedure is listed below:

1. Pull switch "4" up (all others down). Press "ADDR LOAD", key

down.

Location

400 401 402 403 404 405 406 407 410 411 412 413 414 415 416 417 420 421 422 423 424 425 426 427 430 431

176

Push "EXAM" key down (to sequentially examine each location).

The lower lights on the front of the PDP 8/e will indicate the

numbers which were entered.

The upper set of lights will indicate the location.

Light Numbers

bY ba

OOH

we

we “

ww

w

w

we

we

w

we

WKWWOREE

OOF

» oOOF

HbHRe

fs PHP

QOrF

OOF

. bs

. bs

= ja

bs

bs

w

ww

w

Wo lo

GH WH

WO

WD WH

W& DW Ww

WH wD HW

&

©

WD WD Ww www

MEN MINN NNN NS

WO OWDWDMWMAMMDAWO

WOO OH RH

w

ww

w

w

w

we

Ne

w

w

ww

we

w

we

w

w

w

w

w

we

w

we

we

w

we

DOW wowowoworh

Fb

4

HA

. w

we

w.

w

w

“se

we

When load is checked:

1.

2.

Enter 400 (Key 3 UP)

Previous Entry

1227 6731 6731 7600 6732 5204 6734 7106 7006 3232 6734 7012 7012 0226 1232 3631 2231 2230 5204 1203 6731 5014 0017 0401 7400 0000

Light Numbers

we “

WMmMeoOorMOrRPUU

Ps OW

Q-v

ww

we

w e

vy

“ iv)

v7

“ tv

) ‘e

»7,8,11

oO Pe

Wb

BPP

REFINE DERRY ERRYWOWN WEN

we

we

w pa fu

ww

ww

w ee

ee

eee

we

we we

ww

we

we el

OmlUlUlUOUlUlUlOlUlUlUOUlUlUlUOUlUlUUlUlUU

PONFENPE

NEAR

P NFP PRP NHR PHP

Bs

Rv

w w

« w

“ .

w vy

7 a

tv Y

‘ w

w “w

Ov vw

w w

FOODOONOFEFEHENFOOCOOFrCOCO

COCO CN

Press down "ADDR LOAD", "CLEAR", and "CONT".

Listing and Typing Files

A list of the file names contained on the upper disk are printed

by typing "DIR" on the active terminal. The information in each file

177

is typed by entering "TYPE File Name File Type”. Substituting "FLP:"

after each command, Dir or Type, will accomplish the same as that

above except the information will be generated form the lower disk.

Changing to Another Terminal

Either one of the two terminals (teletype or teleray’) can be

used with the PDP 8/e, at the discretion of the operator. During

the operation of the system, it may become necessary to change from

one terminal to another (i.e., to receive a hard copy - teletype) which

is the simple procedure listed below:

Type the following on operating terminal after dot appears:

R TLS (press return)

1. When computer halts, (PDP 8/e, "RUN" light goes off) change

terminal switches (see Fig. 57) for desired terminal.

2. Press continue.

3. Computer should print a dot on now active terminal.

PDP 8/e Programs

A. Position Program — POS.RL

This program enables the operator to conveniently position

the diffractometer at any 20, XxX, ®, or W setting.

Type on operating terminal after viewing dot:

R POS (press return)

The response from the computer will be a series of questions

which, when answered correctly, can position the instrument to any

+

Teleray is preferred because of its speed.

178

set point. The smallest movement in this program is 0.0025 degrees

which is one step of a motor. An example:

THE PROGRAM WILL NOW SLEW A MOTOR, INPUT MOTOR SELECT NUMBER: 1=2-THETA, 2=OMEGA, 3=CHL, 4-FI

1 IN WHICH DIRECTION DO YOU WISH TO GO:0=UP,1=DOWN. (UP-HIGHER ANGLE DOWN-LOWER ANGLE)

1 HOW MANY DEGREES DO YOU WISH TO MOVE? (F9.4)

100.0 DO YOU WANT TO SLEW AGAIN?: 1=YES,0=NO.

0 «

THE PROGRAM WILL NOT TAKE OUT BAKLASH 1=2-THETA, 2=OMEGA, 3=CHL, 4-FI

1 IN WHICH DIRECTION DO YOU WISH TO GO: O=UP,1=DOWN.

0 HOW MANY DEGREES DO YOU WISH TO MOVE? (F9.4)

1. DO YOU WANT TO TAKE OUT BAKLASH AGAIN: 1=YES, 0=NO

0

In this example, two theta was positioned 99° below its original

value prior to execution of data collection. It should be emphasized

that no data is collected with POS.RL.

B. Data Collection Programs

1. As in any of the following programs, POS.RL is available

as a separate program as given in "A" to the operator for positioning

the diffractometer. The program CH.RL is used with the usual chart

recorder operation while scanning either 20, xy, %, orw The program

is self-explanatory and is operated by typing the following on the

active terminal after viewing a dot:

R CH

The response from the computer will be a series of questions which

179

must be answered to initiate the scan. If a scanning speed of 1/32°/min

is desired the following should be typed on the active terminal instead

of that above:

R_CH32

2. A combination of a loading, running, and scaling program,

"Search" will quickly collect data over as many as nine different

regions. Scanning speeds of 5 or 1.25 degrees per min. along with

interrupt times of .1, .2, .3, and .4 min. yield a variable range of

detail which is then plotted by the teletype in the form of an

intensity bar graph. The optimum choice of speed and interrupt time

depends upon the c/m and is determined by experience. After viewing a

dot type:

R_Gl

The computer will now ask the appropriate question on the teleray

Screen so that the scanning program will be loaded. The load including

the data will be stored in a file called "KMAND2.DA". An example of

a load is given for a Chi scan:

.THIS PROGRAM SCANS SPECIFIC REGIONS FOR AVERAGE COUNTS, AND OUTPUTS THEM IN GRAPHIC FORM

FOR IDENTIFICATION, FILL IN THE INFORMATION BELOW THE * &

SPECIMEN MONTH DAY YEAR HRAKAKRKEAESH Ks ws RK

SAMPLE

TARGET VOLTAGE (KV) CURRENT (MA) MONOCHROMATOR wk* wk kk k4*

CU 50 20 C ENT. SLIT (DEG) REC. SLIT (DEG) KKK ReK

1.0 1.0

SCANNING SPEED (DEG/MIN) ARK

1.0 ON THE NEXT THREE LINES GIVE OTHER INFORMATION (20 CHARS./LINE)

180

DO YOU WISH TO SLEW A MOTOR?:1=YES,0=NO

0 DO YOU WISH TO TAKE OUT THE BACKLASH? :1=YES,0=NO

8)

IT IS NOW TIME TO COLLECT DATA ON THE CURVE 1=2-~THETA, 2=OMEGA, 3=CHI1, 4=FIL

3

INPUT SCAN SPEED:1=5,2=1.25 (DEG/MIN) 1 INPUT INTERRUPT RATE: 1=0.1, 2=0.2, 3=0.4 (MIN)

1 DO YOU WANT TO SPIN CHI: 1=YES,0=NO

0 HOW MANY PEAKS ARE TO BE SCANNED?

1 IN RESPONSE TO EACH QUESTION, INPUT LO, THEN HI ANGLE FOR EACH PEAK IN F9.4 FORMET. ? LO HI

00.00 90.00 0.0000 90.0000

INPUT PRESENT POSITION, FORMAT F9.4 0.00

To start data collection:

R G2

The run light on the front of the PDP 8/e will go out. The program

will start execution when the "continue" button is pressed down.

The program "GRAPH" will output the intensity bar graph on the

teletype and is run by typing the following;

R_GRAPH

The input needed for the graph program is the data file name and a

file name in which the graph is to be stored. If a second copy of the

output from the scan program is desired, type on the teleray/teletype

the following after viewing a dot:

R_READ

181

The input data file used for "READ" is "GRAPH1.DA".

3. Twenty different motor operations, each executing scans over

as many as nine different regions, can be accomplished by the following

pair of programs. Included in the output from RSCAN will be the

Integrated Intensity, Mean, and Variance without a correction for

background, see program "IMV" for the correction. Background points are

determined by the operator prior to the execution of IMV. To load this

program:

Type on operating terminal:

R_LSCAN

The response from the computer will be a series of questions which

will operate the 'RSCAN" program, for example:

THIS PROGRAM SCANS SPECIFIC REGIONS FOR AVERAGE COUNTS,

INTEGRATED INTENSITY, MEAN, AND AVERAGE VALUES.

DO YOU WISH TO SLEW A MOTOR?: 1=YES, O=NO 0 .

DO YOU WISH TO TAKE OUT THE BACKLASH?: 1=YES, O=NO

0

IT IS NOW TIME TO COLLECT DATA ON THE CURVE*

1=2-THETA, 2=2-OMEGA, 3=CHI, 4=PHI

3

INPUT SCAN SPEED: 1=1/5,2=1/10,3=1/20, 4=1/50,5=1/100,8=1 (DEG/MIN)

8 DO YOU WANT TO SPIN CHI: 1=YES,0=NO

0

HOW MANY PEAKS ARE TO BE SCANNED?

1

IN RESPONSE TO EACH QUESTION, INPUT LO, THEN HI ANGLE FOR

EACH PEAK IN F9.4 FORMAT.

LO HI

00.00 90.00

0.0000 90.0000

INPUT PRESENT POSITION, FORMAT F9.4

0.00

*the motor desired to operate

182

FOR IDENTIFICATION, FILL IN THE INFORMATION BELOW THE * 8

SPECIMEN MONTH DAY YEAR

Ke keRARKAS wAK *K* *w*

SAMPLE

TARGET VOLTAGE (KV) CURRENT (MA) MONOCHROMATOR

ws k* wx kx

CU 50 20 C

ENT. SLIT (DEG) REC. SLIT (DEG) KK* hike

1.0 1.0

SCANNING SPEED (DEGREE/MIN)

#AAKK

1.0

ON THE NEXT THREE LINES GIVE OTHER INFORMATION (20 CHARS. /LINE)

END OF IDENTIFICATION FIELD ANOTHER SET OF DATA?: 1=YES, O=NO

0

If an additional operation is to be made, a "1" should be typed instead

of the "0" in the last statement, which circles the program back to

the beginning. This load will collect data in the same manner as the

example Chi scan from the previous section. The data collection is

activated by typing on the operating terminal:

R_RSCAN

(Symbols between dotted lines will be typed by the computer).

1. Run light will go out on PDP 8/e

2. Press "Continue"

NOTE: Termination of a program at any point can be accomplished by

pressing down the halt switch on the front of PDP 8/e.

After the data collection is completed, the computer will return

a dot to the active terminal. A background correction can be made on

183

the integrated intensity, mean:and variance through the use of "IMV"

which again is a self-explanatory program. Type on the active

terminal

R_IMV

An input and output file along with the background points determined

by the operator are the only information required in this program.

Data Analysis

The IBM 370 has many distinctive advantages in analyzing the

data obtained from the automated diffractometer. A plot of experi-

mental data along with much faster printing rates are easily accessed

over the phone line. Connection to the IBM 370 is made by:

Part 1

1. Position button box on top of teleray as follows (left to

right)

A. I1st up position

B. 2nd middle position

C. 3rd down position

D. 4th middle position

2. Turn "VADIC" on (located in rear)

3. Dial "9" for an outside line

4. Dial 951-7020, a tone will be given when switching is complete

5. Lift "up" on white button on top, left of phone

6. IBM 370 will print on terminal "370 on line"

User I.D. Password

8.

9.

184

Press "Return"

When computer stops printing, press ''Return" again.

At this point the log-on has been completed. The data transfer to

the IBM 370 from the PDP 8/e requires:

Part 2

1.

2.

Place the data transfer disk in upper Sykes unit.

Re-positioning the switch box on top of teleray to the

operating position for PDP (middle two down, all others in

middle position)

Type on active terminal R IBM which requires an input and

output file name. This input will be used for "GAPHIT"

(see below in plotting).

Type on terminal:

| R TO 370

The computer will print "Enter name of file to be transferred"

Enter the file name and press return

Re-position switches on box at top of teleray as given in

Part 1, statement 1.

Press "Continue" on PDP 8/e

Computer's run light will go out after data transfer is

complete.

This procedure will copy all data in the given file to the IBM 370

permanently. Graphing of data is accomplished by:

A. Typing on terminal (switches in same position as for transfer)

"GRAPHIT file name file type A

185

NOTE: File name is name of file, file type is kind of information

(i.e. Data).

and the printing of a file:

B. Type on terminal (switches in same position as for transfer)

TYPOUT file name file type A

Editing Files

Information for editing a file on either the PDP 8/e or IBM 370

(35) or the VPI CMS User's Guide, can be found in the 0S/8 Handbook

respectively. Some useful commands are listed below:

1. PDP 8/e Editor

Command Format Meaning

R #R Read file

L #L List file

#NL List line N of file

#M, NL List lines M through N

C #NC Change line |

#M, NC Delete and insert following lines

D #ND Delete line

#M,ND | Delete lines M through N

S #S Search for character entered

E #E Close the output file

186

2. IBM 370 Editor

Command Meaning

TYP List line

TYP N List N Lines

N List next line

DO Down to next line

UP Up to line

C/QQ/ZZ Change characters QQ to ZZ

FILE Close output file

XG*E

TS XGSS

ef¥TT Sade OXI

(S°CASSXOTSS *S4C*XTTSTIEfXOT)

ives

SI (G8

E44 G8

84 4Gt

C4) AvRNOA

¥ (TIESTICSTIEf{TIESTIC*TIctTictties*tictitic!

TEE ATIC

STIC STILE

STILE STE

feet ef Te

fTief§tict# icf

Tic sf

XE TICfITLe

Tic {Tileftic*TicitileftTie

fTIieslicsTicftie*

Tle icf

tie fTie

fie

tice

Prete Tee ers

/PMOPTLE* TIE

Tie

Tie Ties

Ties ie

Tet

Lie sh

TICfTT

eft beftre {Lief

ef Pict

icf Tle STE fics Tle

Tic {Tie

{Tiley ivweod

€ (Zi€)ivwuos

T

RVYOOdd

FHL NOS

SAVevO4d JHL

Sav SSSHI

dS* Za

TAS * ZA

FY ETE

YAODEN

ATT

d TSGN WSt

(Osos

COs ddd (oe sos!

(CE S259!

(Ce lsaaf ice! éyLSI* (ONT 16€Ge}9?

VALE (CAEP

LT ECOPPSS

TE COOP

TEES

PIS ECOG

FENS (OSPSS ECCS

ITE(ORIAT

ECE DX ET (Codaf

(Cede fC

ssa hGb

ys ah

Os

y a

COS Fe

ye ECCT

OAS

(OZF 240

NOTSNSRTD

(Cefesrso*

(68

2st (CU TSS ds !X I

do

Gx XAIVdWOD

*3SV5 STHL

WE GFIDNVHD

38 ish:

LIDTIVLL

ING FHL

SChIM

ANY GL

CSLAICOL:

ar NYO

It

DiaAioH ohio

ST S3uMloONyls

ShL *Sy¥Cidvd

ONTYILLVOS

DiweDL\y Skil

GNY SedLy

WOOT

TCNT SHEL

40 SNGTL1SOe

DJIYOLY SHL

NOUS ChAIGeOoD

Aevile

Wood

ALTS IHL

I G3iVsSO4iNT

SSLVIGDTVO

bY OGUd

SIRE

é

GA

lk Xs ee

ESF XG

sGe

erga

*xe E° ea! xG

‘XH4,

abana XS VCShVIZ/NI

Ss

(+°944*xXe*7*

6G"

GAPXL

PS8 OB EXG SE

S°XU CIE

“Eyes

XS Sh ad tf Xk fade

S XOT Se

*¥OGF AVESHES

sf XL Sa

TV

(S°8a*xe*o*8a)

LVREd

ALISNSiNT

Fi

II XIGNdddVv

KEV LVDS

89]

exxiT

S¥G)IVeYOH

SOT

ly Yt

00 OOOO VOUNOO

187

188

Chase C (PRT) DC

TP STIP

NACHT CECSTICVAFICT

Hac CIC ETI

CUNT INISH+4

4S 39 4S CTs

COPD ED

CEC ETIPY

SCE Ee

CEO ETT

PDAS (PL

He (IC ELIS IND

SOOFEAS=ESS NN‘T=IP

L GG

O°C=¥4S C'C=€45S

dO=HNNCETO3°TIF)

Al FNFENN

GE TT OS*

FIC) al

c*T=1If

& CG

RON

* T=F1

9 OG

CoP ESIEECSEEZIAL

OUP

FA SINGCESTTEG)

ALTUM

CUP PCERSOOE

AS SCIAL (NE OSIM

(4%°G) GVay

dP*t=ur

Co ac

(CETIMECPETIAL

CF ETING OSTT

ESO) ALTUM

(FETIBRACEITIATOCR

ATIC (#Fo)ayv3ay

AVeaw

RY F159

LINA

HL

NI SBPOLV

GHL 30

SNCITLISOd

JHIL/

CNf ATH

SG Oa

CT XNEF-N=d0 (TLATHATS

CLV TSCT ISS ELH)

(ers)

CVS PNEANARONCT

ASG) Gyad

AvdadayY NY

*Ca3i¥

TA 9199

ae OL

SNCIIVNOETSSOQ

ANV Ids

ONIX

iSt1Td SHEL

JO SwOiV

40 YSGhON/

W223

LINN

dad SROLV

JO t3aahAN

TWLOLZ

(22 eh kViy)

G¥Vau 439

OL VYH

40 siGhoN

WioL/

CSLESFELLLLSITLLETL

EELS LTTE

I STILL EXO)

LN WSCS

LOEGRTES?C*S=lId

(s°e@4a'xoforea*

xo’ Gres

* xe dl ywucd

(G*84e) LVwdOd

(///e8SMOT104

SV ST

CISTL

MGA

SIEVE

FHL sf

XG) IVREUOS

wy

MAN

THH

TNF

WW Ob.

REe

66TT

ra A

TLT

Wo O OVQaUOO

189

V3W39 LINA

SHE

dO eslakvVuvd

@ 2HL/

Vso LI

ZHL

sO SALGNVveyd

Vo RHIS

JON

TINGS

(Ii) 2d=(04T)e45

CTS} T4=€4HT)TS

(INJV=OH1)7

(INI Y=OKITIDA

(CINJH=C€4I)X

T4H¥T=H1

SANTINC2?

NOT

OL GOCOSES*G*GNV EC*69°9)

Al

SPR TINGD

Gi Qi

69

Z2 OL

O5(C°GS°SG°ONRV TC*ES"V

Gl

(CIN

ZAIIVWTV=0

(CIN)

22) 1V3=9

CCIN} TAPOVETV=9

(CIN)

Ta) Wadev

WANS T=IN

OCT

OG

C=yHI

SANTINO D

(FIYZSS(FTITSAS

ECE) TECF TINS OF

LTIHOSTSS)SLIUM

SANT LNOD

(CHU

ST ayy

KX TdWO=(F

TTS

PANT LCD

§ OL

09

(CSe*TAN

XX TdwWo=(fT 24

C2 OL

G9 (T°OS°TIC)

Atl V7AS=2454

€iS=1T34

SANT INOS

aaa 9

VVV 9

OCT

6L

LL

3

190

GTTt O01

O09 YET

GL Po

COLSD

DORSAL CNT

DD TS*GN

VY CINTI On

LO"

(NTID

IS) 3]

T+iNT=1NI

SG*l=INI

GIT Cad

T=1N1

MT4T=N1IT

€1Tt aa

T4+dNIT=dN 1

(RI JASV

SEND ISS

(oh fe)

Gvau

TIT 3

“ESel4d LN3RI

Ia LSSTA

BL

a%d SAOIVA

SHI SCVde

eV uUSONd

Shi °LNabNST4

3 BOVG

SsOa VGEUY I/O

Lab ibis

G@%d OL

OF0 bOsd

SaNnIvA

GSS 40

TV¥LDE YV

2 Alidy

SOG TVIVESAGD

AVU-K

dad SlaVi

GOA YSLOVA

ON TeSLAVoOS

DNCLY/

4SV 9

VOARVTACVELAHLINISZ

ON 9

CVS 4

SAN¥V SBOLODVG ONT UGLivos

STW OLV

SHI 3 3

O¢*T=NI

TTT CG

SANTINCD

¢tt C=daNXT

T-NPENI

SNNELKNOD

€€€ SANTLINGD

OCT TEQeG*

TA(INP J4AS= ENE I

TS

Meee GONE

lig

“Lox CCN)

45 SuV=(NPF)

444 UNE)

a7 0" 271604 e

T)=ONF

194

CONP

IS) LdOS=(NF)

44 in

C8éxe DIDI SOC®

Cue (NG

VZ) DP 4ACCOP

Cee ad)

OT

C8 Prox

(NTA DEES See VVV

ISL CGT See

ENF IX

DD eC

TU SI

C® CexT GOS SLIP HCNP DS

M1‘T=Nf

O1T OG

D900 fead*vvVi?27's)

AVA

2 VW7S9

LINN

SHE JO

walakvevd

D SHL/

393 9

NUE

; CUNT

116

YD)

CSe

cil

y—-hQET?

iT

IN

i {

{ t

‘Gh

F(INT)-ASFCINI))/(>

8 =ASFCINT) +004

ATCINP,LIN)

QCINT)))

LN)-N

CONTINUE

115

113

CONTINUE

GO

TO

lle

NPLEQ.L)

IF

(I

D9

L190

=L,Ik

NIJ

r 1

L.-

EXPRESS

AULTIPLICITY

IS

THE

wae

ate

ote

ala

y yoUS a

a

ase

vr

ule

Tal

223

wie

se

ye

als

Ate ale

Fe

at

‘ sf

xc

S35

5;

< ? os

le

ube

de

Pe

Te

BL

BL

ar

At

ae

4, oe

oute

abe

we

ale

ale

ate

wt ie

ate

a ye

whe

ste

ate

ate

ate

FASS BAR

SS

AS

aS

PS SS kT

AS

oy bp

al ste

yy

TN OF

RA

wie

te

whe

+, at,

as

yemi

xk

Re

TFACINI

JS) =4

IF

(X(NITJ).4

5 \ qi

QV TEAC

{AT

I)

~Z{*

TS)

eet,

NO

03

i #

=O.

2.2024

C

4

191

tod IN|

ar La atined

TJ

1 i QeJVTFAC(S

by wh

e mts

ate

ale

ale

ate

we

ale

ale

RAL

AS

FR

tent nig

ie .

eae ety,

~ > Dent

O 7

“~~ >

«

i 4 Xx }

<L a

« O ~~ @

™ C7 wet LE

— *

wet ate

x 7 weet tot

bide an Pent ener

omed

a wot

i wh

a ade

tle,

MAS

RL

Se

a a 8

he + 3 Je a

Nie

Flos

Togs

we

ete

ate

we

QL

we

ste

SAY

AS IRS

Re

2s

2s

TS

aS

as

FD

BS

é 3, ’ tear

ate

we

Pee

- af,

Fo gh

gle

aie

ae

be

wl

te

ate

07 SR

I ES

OR FES

x -

oats

ote

we

sz

whe

ate

als

ale al

e alo

cole

slo aie

sic

ake

ie ate

fe

ofc o

le 3K

a C

3%

"yy,

UT

OYE

aN

OO

mR

o ty

as

IM A

T!

CHa

+ MO

° 5

(26

S$

ITS

)/a7e24955)

io Nd

x NDJd=

014+

(C

t LLP {

L/(it (Cost Frat

‘2))

wer

- 2,

a) ” ny i,

et

WY “>

—_—

~—s

sevontelh

Zz

one a . * “ 3s

t 3 ay ral ae

Ww . x

fl

3

=(1+

LP(NI

J)

L8))* (L

LP

Tl { i

yo

Pas

ale

Wo

vty

ale

ote hone

2 Teo

3

oe

stows

wee

v:

Lr

BS BLS

as

PA

RE

RS Ns

Be

le oe

ta

gh

at

ye et

ate

we

Pe

Pe URS

FAAS

TA tS

a.

F whe

Ale at

e i.

WW

whe

te

ste

fo

ie

* de

ve te

We ek

ale

= a

FS SAF.

78

at

ghee

ate

wy

ae Xs 3)

C t

TINU

a id a

L100

CON

=l1,I1

vy t a

D3

141

JJ

T))

4 rw

KFLQCJGD

V0 t04T 02,500) er

2s

“r }

(AT(1L,JJI

IG(CRICLyJIT))

Ce

FJCL,JJIT)

CFI(1L, JS

UT)

}M CFI(2,I55T)

CFF(JJST1)

FJC2,5ST)

rm Ve

ve we

C

UE

NY J

cy

oO)

L41

192

CPR VTO4

ECP RESP

LAD EER

EN ATP

LAD SOP

RNI OTE

CPR

DOWST

OC PRN)

4404 (PENT

SP yiv

(Pai

h TILE OGRE

TS oO

*, iat Bee

GNa eCi$

NISad

6 OF

ANIZ

SCP RNIAS

OPEN

IX OSSTS9)

SLIUM

WIAT=Feh

£OT OG

(SeTf'S)

SLIUM

(TZT*S)

BLUM

(ses*9o)

JLIYM

"OTe CLNO/ (BER

STE LADD HONE

fo?) 44d

ML‘T=NEF

COT ac

(RAN

ST) LAS=LUO CLUBS

CNN fT 149)

Sl

YES THANKN

29T

OG

C*0=1u49

SANT EROD

(NINVOV4AL&

(NIN) d Te (NIN SS9=ENINAT) 149

MIST=NIN

TOL GG

¥ST

EST

cot

Tot

193

VITA

The author was born on November 27, 1953 in York, Pennsylvania.

He received his elementary and secondary school education in York,

along with two years at York College of Pennsylvania majoring in

engineering.

In December of 1975 he finished the requirements for a Bachelor

of Science in Metallurgical Engineering from Virginia Polytechnic

Institute and State University. He worked on his master's degree

from the Fall of 1975 as a graduate research assistant for

Dr. Charles R. Houska. In July of 1977 he joined Bethlehem Steel

Corporation's Steelton Plant in the Steel Operations Department.

The author is a member of the American Society for Metals,

Alpha Sigma Mu, and Sigma Xi.

He married the former Robin Lou Kramer on June 28, 1975.

Frank Ew Dietrich

An X-Ray Study of the Behavior of

Titanium Films on Silicon Substrates

by

Frank E. Dietrich

(ABSTRACT)

The thin film reaction between silicon and titanium did differ

from the ordinary diffusion process in bulk samples at elevated

temperatures. The lower free energy monosilicide formed prior to

the disilicide in bulk standards and was also the dominant reaction

product at lower temperatures in films. Titanium thin films, 2.3

microns thick, sustained considerable mechanical deformation during

cooling from elevated temperatures. Line broadening, greater than

that obtained from cold worked filings of titanium at room temperature,

necessitated the use of computer simulation techniques to analyze the

diffracted intensity from the reacted structure.

The structure and space group of titanium disilicide was confirmed

along with a determination of both structure factor forms and accurate

lattice parameters. A range of compositions for the monosilicide of

titanium was observed along with the identification of visi, 0”? as

a high temperature compound. The identification of the inaccuracies

and omissions present in the ASTM powder diffraction card files should

resolve the confusion in the literature on the reaction between

titanium films and silicon.

The temperatures for formation of silicides in thin films was

found to be substantially lower than previously reported? , indicative

of the oxide free interface of the silicon substrates. The identifi-

cation of the reactant silicides, deposited at the grain boundaries

and as planar film at interfaces, establishes this study as the first

attempt to develop a quantitative model for silicide formation in

this transition metal system.