ASSESSMENT OF POSITRON ANNIHILATION AS A NON …

-- - ---- --------_ _---__-

,

NUREG/CR-ll29SAND 79-1217R7

y ,, ,. ..,,..,,._._y,

,--.,,n, n,.-,,,..,~.-,n-.,. n,.n.-..,~v..,- .-, y.m--i._.ms...m> < , _,... a w. _ . . - . - . w w ._e . ., ;, _.. . . u- - _:- .. -ml --w w-A - . a- A - a Ax,.a.s.a .b i- au m.&m ~.u .u3

.

e

n

Hou!dui:s

!.

!,lASSESSMENT OF POSITRON ANNIHILATION [

dAS A NON-DFSTRUCTIVE EXAMINATION TECHNIQUE " e/g ;

> ,

IECE!\:L:n i;*

g'

JAN 27199?% t"'l ': i

- - i.-

r:1.nq f,hD%c $$.

-

KM ;

v, c.. 3 -.

O,, ,. f'

J. A. Van Den Avyle, W. B. Jones, k' / Tyr"; ~

W. B. Gauster, W. R. Wampler '''-

-

Printed March 1980

.

<

# . 6 . .; . ; y .j.*'

*,,-'{ 'i' ' * y ? f't , J - s a <

,

3i-looPrepared for 5

U. S. NUCLEAR REGULATORY COMMISSION

.-

8~90'9R01% ;2. e

l_ _ _ - - - - - - - _ . - -J

. - -

.

!

:s

i1!

!,

l-4

1

1s

l *

)|

!

NOTICE

|| This report was prepared as an account of work sponsored by ani agency of the United States Government. Neither the United States: Government nor any ager .i thereof, or any of their employees,j makes any warranty, expressed or implied, or assumes any leoal' liability or responsibility for any third party's use, or +he

results of such use, of any information, apparatus, product or,'

process disclosed in this report, or represents that its use byj such third party would not infringe privately owned rights.;

| The views expressed in this report are not necessari . those ofj .he U. S. Nuclear Regulatory Commicsion.!j

!2

|1

|.

:

!

II

i

i1

:!

i

!1

I1. Available fromj National Technical Information Service} Springfield, VA 22161i;

}_. __ .__ -_ . _ . . . _ _ _ _ . _ _ __ _ _. - _ . _ _ _ _ , _ - _ _._

. _ _ . __ _ . _ _ _ _ _ _ _ _ _ _ _

NUREG/CR-1129cArm79-1217

R7

ACEE'C:E:JT OF PO3ITRON ANNIlfIIATIONA3 A ';3N-DEETRUCTIVE EXAMINATION TECIU1IQUE

J. A . Van Den Avyle an ! W. B. Jones"echanical idetallurry Division 5835

W. B. GausterPhysical Research Divisien 63!7

W. R. KnmplerIen-Colid Interactions Division 5111

Printed: March 1980 .

Candia LaboratoriesA lbuquerque , New Mexico 87185

operated byCantlin Corporation

for theU. S. Eepartment of Energy

Frepared forDivision of Reactor Cafety Research

Office of Nuclear Regulatory ResearchU. C. Nuclear Rogulatory Cecniscion

Wachingtcn, DC 20555Under Interagency Agreement DOE I40-550-75

NRC FIN No. A-ll72

1-2

- - - '

_ _ _ _ _ _ _ _ _ - _

ABGTRACT

Positron annihilation measurements respond sensitively to atomic-scale defectsin metals. Therefore, scoping studies were initiated to evaluate the potential of

the method as a non-destructive exnmination technig a In this program the Doppler.

broadening technique has been used to measure defect densities generated during

plastic deformation by cold work or fatigue at room Lnd at elevated temperatures.

The primary goals have been: 1) to assess the censitivity of the technique, 2) tocorrelate positron annihilation readings with o' served microstructural changes in

order to better understand the physical bases for these readings, and 31 to determine

correlations among positron ant.ihilation measurements and fraction of life or damage.

Positron annihilation measurements have been conducted on cold worked 316

stainless steel and pure nickel and on lcw cycle fatigued 316 stainless steel cycledat 293 K and at E66 K. The readings increase monotonically with damage and saturate

at approximately 20% strain in cold work and at 10% of life in low and elevsted

temperature fatigue. In 316 stainless steel much lower sensitivity is observed forfatigue at 866 K than at 293 K.

Annealing stuiles on deformed camples, combined with transmission electron

microscopy and microhardness results, indicate that positron annihilation is

sensitive to vacancies generated by cold rolling or fatigue at room temperature.

The same results indicate that positron annihilation is sensitive also to disloca-

tions in pure nickel, but is insensitive to dislocations in 316 stair.less steel.

For that reasen, the positron annihilation techniauo does not appear to be well suited

to ncn-destructive examination of elevated temperature deformation of 316 stainless

steel since excess vacancies generated by deformation above 600 K are mobile and annealduring deformation.

This is corroborated by the observed low sensitivity to deformation.

A ramber of engineering questions on the application of positron annihilation

for non-dest ructive examination of components and structures have been identified.

While nene rule cut the method, considerable development work would be necessary forits implementation.

3-4

._

.. _ - - ._. _ _ _ . ,

ij

f4

J

=|

i. I

i

1a

iia CONTENTS

Page

1.0 I!iTF0DUCTICN. 7. . . .. ... . . . . . . . . . . . . . . . . . . . .4,

2.0 EXPERD1NTS 9. .. ..... .. . . . . . . . . . . . . . . . . . . ..

T

| 3.0 FISULTS . . ... . . . .... . . . . . . . . . . . . . . . . . . . 15$ 3.1 COLD WOFED NICKEL AND 316 STAINLESS STEEL . 15) . . . . . . . . . .

J, 3.2 316 STAINLESS STEEL FATIGUE. . . . . . . . . . . . . . . . . . . 15i 33 ANNEALING FESPONSE OF DEFOR?ED NRE NICI":L AND 316 STAINLESS4

; a""m'"" L . 18.. ...... . . . . . . . . . . . . . . . . . . . .!,

-

3.3.1 Annealing of Pure N1. 18. . . . . . . . . . . . . . . . .

< 3.3.2 Annealing of Cold worked 316 Stainles:; Steel. 22; . . . . . .

I 3.h ANIEALING RESTC:ISE CF FATIGUED 316 STAINLESS STEEL , 2h1

. . . . . .

i 3.5 DESCRIPTION OF ADDITICNAL POSITRON AICiIHIIATJr'fi l'EASURE!ENTS . 29.

!.0 DISCUSSICH CF EXPERIMENTAL F2SULTS.4

31. . . . . . . . . . . . . . . . .

i

5.0 MICRCCTEUCTUPAL EVO!rIICN AS A '*0NITOR OF DA!! AGE. 331 . . . . . . . . . .

i

i6.0 ENGII.EERING APPLICATION OF POSITECN A!HIIHIIATION TO NON-DESTRUCTIVE

I .XAMINATICH CF STEUCTUPAL COMPO!!ENTS. 35. . . . . . . . . . . . . . . .

| 6.1 FOSITPCN AICTIHIIATION SUDFACE MEASUPZ:ENTS CF COMFONENTS 35f . . . .

4 6.2 ItSI"'ECN SOURCES AND S0t'RCE PIACE?EITI . 351 . . . . . . . . . . . . .

6.3 St?FACF PFIFARATICN ANL AREAS CF INSECTICU. 37*

. . . . . . . . . .

6A INITIAL STATE CF THE ALLOY 37,... . . . . . . . . . . . . . . .

7.0 CONC LUS ICNS . . .. ... . .. . . . . . . . . . . . . . . . . . . . 39

ACKNOWEDGE:ENT . 11........ . . . . . . . . . . . . . . . . . . 4

avFe ut v..w.e .nr ?h~o

~ . . . . .. ......... . . . . . . . . . . . . . .

b

5

/

. - - . - - - -- .- _~ - - - - - -.- - .- ---. - - . . - - - -

ILIUGTRATIONS '

Figure Page

1 Poaltron annihilation Doppler-brandening spectrometer arrange-ment and exnmple of dnta . ,

. . . . . . . . ........... 10 '

2 Photon energy spectra of well-annenled and defeeted samples withindiented channel width aren: uu d to calculnte L nnd R. 13. . ...

3 Positron annihilatten reopense of cold worked 316 stainicus steeland nickel 16. . . . . . . . . . . . . ............... >

!* h Positron annihilation rosponce of 316 stainlesc cteel ratigust at i

t

P93 K and 866 K. 17 f. . . . . . . . . . .... . . . . . . . ...i

L') Diclocation microstructures of 316 ctninless steel produced byroom temperatu re fat igue n t Se = 1.of . . . . . . . . . . . . . . . 19

6 Microntructure of 316 stainless steel specimenn fatigued tojcaturation of the positron annihilation linechnpr parameter;

n) 293 K, de a 0.60I, N = 471h3 cyclen; ~ (III), nenr [110]4 ='

uno ; b) 293 K, tv ' l.8T, Np = 3700 cycl e s , g = ( 002 ) , nea rfl00 cone; c ) 866 K , Ac = 0. '/$ . NNone,) not two beam diffrnetion conb = 10078 cycles, nonr [110]!

itians. . ........... 20

| 7 Icochrr.nnl annenling response of pur" nickel: a) 0-parameter! mensuremente of cold worked nickel with tvsintivity data forj quenched nickel by Wycick et al. (10); b ) calculated H-parameter j

versua nnnenling temperaturt ............... 21. . . .

! 8 Inochronal annenling recronce of '9% cold worked 316 atninlesssteel: a) G-parameter vercur nnnenling tempernture; b) R-pnrnmeter ,

;versus annenling temperature ............... 23. . . .

Micruhardneca changen produced 1 y etep-nnnenling colrl worked 316o

atnini": et"el. . . . . . . . . . . ........... 25

10 Microctructuren of 7A' cold worked 316 stninlens steel produced byisochrenal step nnnenling: n) na cold rolled; b) nr caled to 873 K;c) nnn m ica to lo73 K. . . . . . . . ............... 26 .

11 frochronal annealing r"sponce er fntigued 316 stainleca steel:n) fatigued 293 K, ac O.G$, N = 471h3 cyclen (prior to failure);b) fatir,ued P13 K, b J.Df, Ne = 3700 cyclec; c) fatirued 806 K,=

ar O.SI, Up = 10&l8 cyclet. tLineshape parameter difference , AC,define! in text.). . . ............... 27. . . . . . .

I

t

!I

| '

i

|

t.

. . . . . . - - - - - _ .-- - - _ _ _ _ _ ___ _J

___. .-

1.0 INTRODUCTION

This report describes progress in work perf ormed as part of an ongoing contract

with the Nuclear Fegulatory Commission entitled " Elevated Temperature Design Assess-

ment" (Contract DE-AC0h-%-DP(Xf/89). A primary activity in this program has been to

investigate new ncn-destructive examiration techniques which could serve to monitor

detmage accumul. + ion resultire from creep and fatigue during elevated temperature

se rvice .

Current ADE design rules for nuclear structural components operating at

temperatures in excess of 700 K use a "drimage accumulation" approach. This methal

identifies segments of the creep and fatigue history, assigns life fraction values

to each of there segments, and sums the lifa fractions to calculate a damage frac-

tien. In this approach, when the damnge fraction equals a critical value, a fracture

should occur. Implicit in this concept is the assumption that the material has ae

m"mory of its past history; this analysis does not censider flaws, crack initiation

and growth, or specific fracture mechanisms of creep and fatigue.

It is well knewn that tulk microstructural changes do occur during creep and

fatigue of metals. Dislocation density can increase or decrease depending on the

initial state of the metal, nn.1 these changes may alter the cyclic and creep flew

st re s s . Precipitates can form with time at elevated temperature, with the rate of

formation strongly affected by concurrent deformation. During creep deformation,

internal voids can form, principally at grain boundaries, and these ultimately link

to crente cracks and eventual failure.

The occurrence of microstructural evolution during elevated temperature deforma-

tien suggests that thir evolution mirht provide measurable variables which could be

correlittod with acemulatien of " damage" . If a non-destructive examination technique

we re doveloped to make these measurements, the technique, with the necessary correla-

tions , would be a monitor of damage and would fit in the framework of current ASME

design philosophy.

A new method for studying internal defect structures in metals is positron

annihilatien. This technique has been used successfully to measure dislocation

7

_ ._ . - _ _ _ _ _ _ _ _ _ _ __

. - . = . . _- _ _ - . . - _ . - . _ _ - . _ - _ _ . . _ , . _ . ~ . - - - . . . _ - ~ _ . . . . ,.

!!

i

i 18

I

i

!

dencity chnngen in cold worked metnic (1-5) and changes of vnenney concentrationsiP

nnd vnenney clustering in quenched met <tls (4). Ove r the pnet two years, a cerles of

neoping experiments has been carried out at Candin with the purpose of accercing the [i

use of pocit ron nnnihilation ac a non-destructive examinat ten tool for advanced ',

i

r" actors. Three gonic of the recenrch have been: 1) to access the censitivity of

the technique, 2) to correlate positron annihilation rendings with obrerved micro-

structural changes no ne to better un !crstan 1 the physical bnce: for there rendings,

nn,t 3) to detemin" correlatiene among positron nnnihilation measurements an d frac-

tion or life or damav". As part of item 2,.ictalled metallurgient examinntions haver

been perfomed en damnged samplen. fi

In a r" port published ontlier (6), ' me exper imental data and tentative corre- *

Intiens wer" presente l. The present re port include. ndditionn1 data taken on

faticim i snmplos of 316 ctninlesc steel eyele.1 nt 866 K (1100 F). Also described

reverni nnnenling experiments which, eteb!ned with trnnemission electron microccepyn re

; observat ione , provi> ldentifientien of the rpec ter of defects which enuce ehnngen in

pe :lt ren annihilat ion rendiner . There roruita allow conclucions to be mnde on thei

npplienbility of paritrcn annihilntirn nn n non-dect ructive exnmirntien technique fo ri3N s tn inle :e ct ne l at e levntml tempernture Finrtily, n number of potentini problems I.

n r" d iscu rre.i which deal with hypott et le"1 nppliention of the techniquo in n ren1

r"netor rt ructu rr .I

!

r

!

i

!!

lr

|

L

tb

|

,

B

,

8

.-.---. - . - -. . _ _ _ _ . . - _ -

_ _ _ _ __ _ _ _ - .

2.0 EXPERIMENTS

All exp;rimental data were generated using +he Doppler broadening method of*

positron annihilation measurements. This method, along with specimen preparation

technicues and mechanical testing were detailed in n previous report (6), and only

the major points will be outlined here.

Measurements were performed using a Dopplar broadening apparatus, shown sche-

matically in Figure 1, which was available from previous work. The configuration

required using small disk-shaped samples approximately 0.6 cm diameter by 0.13 cm

thick. These samples were sectioned from mechanical test specimens, so this

particular measurement scheme was not truly non-destructive. As described later

(Section 6), however, there are practical ways of making the measurements without

sectioning.

Positrons were obtained from sources containing either Na or Ge candwiched -

between the two samples. As shown in Table 1, the average depths of penetration of

the positrons are small, ccnsecuently only the near-surface regions of samples

centribute to the positron annihilation measurement. Fer the sectioned samples,

this still corresponds to the interior of mechanical test specinens, provided the

sectioned pieces were electropolished to remove damage caused by cutting (6). Surface

preparation for non-destructive examination of structures is discussed in Section 6.I positrons which enter a metal slow down rapidly and subseauently annihilate with

-10electrons within about 10 s. The positrons may annihilate with electron- in the

lattice, or they may become trapped at defects and annihilate with an electron asso-

ciated with the defects. Ideally, the annihilation event most often produces two y

photons, each of which would have an energy of 511 kev and be emitted at 180* to each,

other if both the positron and electron were at rest. In fact, however, the positron

is usually close to rest while the electron is not, and this causes the energies of

annihilation photons to be Doppler shifted thus conserving energy. In the Doppler

broadening method, an energy resolving detector is used to register photons, and the

result is an energy distributlen plot of counts versus energy (channel number in

Figure 1). The shape of the photon energy distribution curve can be described by a

!

9-. -_-. .-. - - - - - . - .- - - .- -- - - .-._-- _ -- -- - - _ - - - -

_ _ _ _ . _ . . _. . _ _ _ _ _ _ ..

|'

2;

POSITRON ANNIHILATIONDOPPLER BROADENING

;

SOURCE4

:M M

DETECT 0R :sr|| !| j 514 kev[ PREAMP

i LINE;

: :: :

a ; :Y i 1z

.

SAMPLE $ . ....... ' '

.......

cry 0 STAT umMm% .'.a i i. 511 kev >'MAIN oo : FE M Cu-

; AMPLIFIER .. ,.f .

.,

.

-.

* *..

BIASED ... ' " ...,*.; ADC + .

AMPLIFIERt +DIGITAL - 3800 3900

STABILIZER ICHANNEL NUMBER*

MEMORY

|

|

|

FIGURE 1. Positron annihilation Doppler-broadening spectrometer arrangementand exaznple of data.

;

= . _ _ _ . _ _ _ . _ _ . . . _ __. _ _ _ _ _ . . .- --

'|

(t

!,

!

TABIE 1. DEPTils OF IENETRATION OF POSITRONS IN COPEER,

t

| (Values for Steel will be ~ 10 to 20% Higher)|i i

! * '

4 Source e-Folding Distance Maximum Penetration|4

i 22j Na 23.1 pm = 0.0009 in 0.31 mm = 0.012 in!

I

Ge 165 m = 0.006 in 1.42 mnt = 0.056 ini

*

Depth in material at which positron flux is 1/e (37%) of itsvalue at the surface.

|4

4

<

1

A

i:,

, :J

|!

!A

4

I

j

ii4

h

1

>

|

.

! |t

111

- _ _ . _ - _ _ _ _ _ . . - . _ - _ . . . _ -- _ _.-__ _ . _ _ _ _ , . , , _ . _ . _ . - - - - - ,.i

,_ _ _ _ _ . _ . . _ _ _ . _ _ _ . - _ _ _ _ _ _ _ _ _ - . _ _ _ _ _ _ . . - , . . . . _ _ _ . . _-

4

!ii

l

I general line shape parameter, S, which in this work is defined as the sum of counts

in the center six energy channels divided by the counts in the center h8 channels

of the 511 kev distribution, multiplied by 100 (Figure 2).,

i

It has been determined experimentally that the Doppler broadened line shape

bec .nes more sharply peaked with increasing concentration of open volume defects

y (i.e., vacancies, dislocations, and vacancy clusters). Thus S increases as the

total mimber of these defects increases. The effect is measurable for atomic

concentratiens of defects between 10' and 10 , and saturation occurs for concen-

-3trations of approximately 10 to 10 These observations are consistent with the

picture of positrons being trapped at defects (vacancies, vacancy clusters, andi

dislocations) and then annihilating with electrons thet have different momenta in

| the defect volume than in the perfect lattice. In addition to the S parameter, the.t

defect-specific R parameter, defined and discussed by Triftshaliser (7) and Mentl anel

friftshaliser (8) was used for analysis. The R pnrameter as defined in Figure 2 is a

estic of the change in peak areas to wing areas for defected versus annealed material.!'

i

Values of R are postuleted to be independent of the number of defects for a givenf!dominnnt positron trapping site, i

Th" materials tested were 316 stainless steel (Republic steel heat No. 8092297,

cbtained from Oak Ridge National Laboratory) and pure nickel (nominally 99.995%,

obtains.1 from Materials Research Corporation).

,

i

|

!

!.

12 ;t\

| I

. _ - . - . - . . _ - . . . - _ . . - _ . _ . . . . _ - - - - _ - _ - - . . . . _ - . _ _ . . . . _ - . . - - . _ ~ - _ . . . - _ , _ - - . . . . _ -.~. - _. .. . ~ . ._

!

16j

1

!

DEFECTED!:

I12 |A-Al -

#

" 18 - B1 [) ANNEALED</ /,

,/ /, . . ,

I !' v i t

'O / /;; "x

,

/,/

-

g > ; S, <

f B' /'

,

O / '

/ !U

/,

//,

A.,

/ \

!'' '

q\_ ____________

/

hg\ /- - - - - - - - -,

O y

2012 2028 2044 2060 2076channel number

|

g FIGURE 2. Photon energy spectra of well-annealed and defected samples with,

4 indicated channel width areas uced to calculate S and R.v

. _ _ .

. _ , _. _ -. __ _ - - . _ _ _ . . . _ ._

'Y

1

1

Ia

.!1

!e

i! 3.0 PESULTS:)j 3.1 COID NORKED NICKEL AND 316 STAINIESS STEEL!( Initial scoping measurements were performed on cold rolled 316 stainless steeli2 to assess sensitivity to varying amounts of deformation. Sewples were cold worked

in stages up to 75% thickness reduction. In Figure 3 the measured positron anni-!

! hilation response is plotted as percentage change in lineshape parameter, relativeij to the well-annealed condition, versus reduction in thickness. Measurements werej |

j made on material with two starting conditions: mill-annealed "As-Received" and

re-annealed (1338 K for 1 hr in vacuum). The data show an increase in lineshape

) parameter, reaching a saturation value at approximately 25% reduction. Data fori

both as-received and re , nnealed 316 stainless lie along the same line. Counting

statistics yield calculated error limits of approximately 1 2% of the total measuredi

l e f fec t .l1 Also plotted en Figure 3 are lineshape parameter changes for pure nickel

generated by Dltebek et al. (9). Although the plotting scales have been adjusted to

provide a common relative saturation level, the important point is that the behaviors

ei t he two face centered cubic netals are nearly identical.

32 316 STAINLESS STEEL FATIGUE

A summary of data generated from low cycle fatigue of 316 stainless steel at

both room temperature and 866 K is precented in Figure 4 The major observations

|

which can be made are:

|

1) For a given test cenlition, saturation in linechape parameter occurswithin approximately 1C4 of life.

j 2) At 294 K, the apparent saturation value is slightly lower for the smalleri

strain range tests than at the higher strain range.;

3) The ulti2:nte caturation value for the tests at 866 K is much lower thanthe ?)4 K data at roughly equivalent strain ranges.,

1

31 h) Tests with several combinations of hold periods show positron annihila-

tien saturaticn values equivalent to non-hold period runs, even thoughfatigue life was reduced by a factor of four for the tensile hold c ..e.

<

!I

ij,

f 15

i-__.- - -.-- . _ . - - - - - - . _ _ . . . . - - . - - - . - . - . - . . . - - ~ . . . -

.-___-_ ._ _ __ - _ __

EI I I I I i j i I

. g.5.-,

.

,('g*** Ng - 12,; *

8' y *-

,

#.*- 10

6 -^

g ^

- .- 8 @o -

' o?lm E T

om mm$ ee 4 -

- 6

M oE

!O ANNEALED 316 SS

: e - 4

2 o AS RECEIVED 316 SS

NICKEL (Dlubek et. al.)(9) - 2;

*k

i f

Of ' ' ' ' ' ' ' ' ' OO 10 20 30 40 50 60 70 80 90

dd;REDUCTION IN THICKNESS ( 7,)

d o

FIGURE 3 Positron annihilation recponse of col.' worked 316 stainless steele.nd nickel.

,

,

__ _ _ . _ _ _ _ _ . _ _ _ __ __ _ __

_ _ _ _ _ _ _ _ _ _ _ - _ _ _

.

.

1

35.5 |i , , , , , , , ,

35.0 -. A + = FAILED

_

' o -a c = 0. 5%34.5 - 886 K < + -ac = 1.07a -

3.40 -A-ac = 0.5% + 10 MIN TENSILE HOLD

_

o-de = 0.5% + 10 MIN COMPRESSIVE HOLDm 3.35 - 0-de = 0.5% + 10 MIN TENSILE + 2 MIN -

5 3.30 - COMPRESSIVE HOLD _

b a-de = 0.5% + 10 MIN ZERO STRESS HOLD2 32 5 -

'

= c H-

g 32.0 -

294K D-

E' 31,5 - A c = 1. 8% C'

$m 31.0 - 294K -

z e-

30.5 - a t = 0. 6% 4 _

30.0 - -

j g $929.5 - c. ,

/ I29.0 - g g/28.5 - / -

/#28.0 /

-

a*

| | | | | | | | |

0.1 1 10 100 1000 10000

FATIGUE CYCLES

P h!IEE '4 Peaitron annihilation respenco of 316 ctninless eteelfat ic20 1 nt ")3 K an.1 F46 K.

||

17

,. . . . . . .

-- _ ____ _ _ _ _ _ _ _ _ _ _ _ _ ____ -.

Ex enirnt it nr of tb se faticuo specimens using trnnemission electrrn icrmecpyrhcw, fc each given t ect cen lition, n inerence in both tislocatien m r u i t ;, an!

lin"rhnp ; nrrete r with rude r cf cyc l"r. The microgrnphs in Ficure tnb fr aI

th .:1+ ilmona cycl" It +

'

1.M reveal inerencing dir lcent in r. .i. ncitienK nn *,

with cycles until nttninm s of a mt uratirn dens ity. The clnnge. clcerve, in th"

pt altrcn nnnihilntion mencur" "nte en thic ucple alro chewed utt u rn t i; nt > cut,

10 .cl.' Fic! +). Furth"r eyelire pro!uco+ - +

n rearrangement of t b !!rloen-

t i en :- (Firure S ) ruth"r thnn an, :icnifiennt chance in their dencit;, t r the lire rhn;

;nr w tor.

E x ren cuticn of "c: 1- fre tb rnturnt i n regi e of ench t at t en lit ich in

'F im r. c h ek " that tb linochnp p't rnv to r wa s net n unique' functicn af th. dirle-

intien 11 + . For exr ple, t h+ lin chn;- p rnmetere n*, nt uraticn fm t h- twe*

K+ :.t ' 'it i c n r* not4 -

r nt. : i f f e r"r.t , b u t. th. !iclocaticn '- ncitit ntu

s tu r iti: ilfi, r i lticantly Il icure t u n r. (t). ilco, th' u l t innt * ti irleent!cn

.t -it, af tl.e ' 1- t t em: "rn tu r+ fati n n;*'. "n: nt a n tu rn+ 1on ir +cuni to r

+,r thnn f; ' 'itb f t h- r: : n te .: c rnt ura contitlenc (Figur. Lej , ye t tbutu:nt. 1 in :%;- r a r v. L. r fm t h. elevat+ t em pv :n tu r. cnr. nye n 1:wer.E- ' nn 91 ir., :l~ rit . In the n"xt

*

ti n intient( that th m < t i ffe re ne.e

n r- t, lat t ic' 'i-r-,

3. . , 7, 7. . .,,.. . . . .i'. .rc , ; y. c.# r ..r 3 ,- . 7 ,. n.. ,u n o,., 2 . , , , , ,.r,sa <. -- ; p. < m m u; a. . . , . ,i...

o ,,

. 4. ! A n: 91ing f N r" li

< wne 1" t .. , - "I ~nte rinl fo r fl + n r.nc n l ing inv+ ct in t ico :ince<

t' 5: i. n w"11 c t, i , u s in, r' r i s t t v i t'. : 10,11 ) nn ' ca l o r ime t r;, (11 ) techni<:ner ,'

nr. ; t< nur it xhil itro a r" -n r. e - nrnble to thn t of 316 ctn inle: cte"I in cold *

k rk nc m~ n t .< ii: 2 c r e- in octirn 3.1. A ;'o ly e rys t a l l i n+ rnnpl. wn colt>

r.Ile 7. , < l.'ve 1 " r r* . -n 1 i r t. ;x citr' n annihilation nturntir-n, oni ir chrcnnilynnen b '

:o minut e: in K in . ~ nt: fr,m c 4 E to c0 K . I.in"chn;r | n rnm" t e rvi ,

luer r,nrn: 4 aft"r( nch unt e nlinc ricp nre i 1: t t".i a c cpe n c i rc le r in Figur" 7ao

" r; tb r uling t e mp" r, u r" TU i n rna te r i;: reen to decrence in two lictinct

et v :, curring ut n"arl. t h' c rn tv mpe rat ure; na t he nnnenlinc '.n re . of recidual

IF

_____ -

_ _ _ _ . _ _ _ _ _

, . . . ,

m e,s , ..

3;en- ... . . .

'* "

p,.s,

gt,,..g.,+. g ..

[N !,,

jRE-

z~ ~

E. ,

pyggypt.y ,-* - m .. . ...

,, . J vt -,s.-

,

.r .s

?

. l .: 1. ' *t t i n m i .c r. :- t r. e t ' . r. ' " . . f 4th at ninl - teci ; ro.lue. ! tyr " t 5; c ra t '. r<it i r , nt te 1, tsy ,

19

..

_ _ _ _ _ _ _ _ _

|

raC

gy w- y N \' .1

'

a .",

-,

(4,s

\ . Y '|h '

-[' ~^'

' hs. .

3 g' >; 2 .c :. .

g s_N-[Msp*gg' s }, )i - -^

#M* } v t!gn ' ; ''

Y, Q kg ' 'g[I 0.5 am1 3 r',.jd , , ,,

'

(a) (b)

7 4<w pr ; :%, m- '

*- - ,

f.g:: fy & , .

p.r qp~ -*

4:~:Q '', 4t ^'**;

$> 46,

-*

<'

FIGUFE 6. Microstructure of 316 stainless steel,

f!'j.3 ' , cpecimens fatigued to caturation cf

'

*the positron annihilation lineshape*'

parame te r: a) 293 K, ac = 0.60%,. ' , .N! / N = 47143 cycles; g = (111), near [110]'-, . .

W,' W- zcne; b) 293 K, Le = 1.61, N, = 3700.,

cyc le s ; 2 = (002 ), ne e r [100) zone ;* ' M" =

g., e

'

c ) E66 K, Le = 0.5%, Ny = 10078 cyclec;.,

'.vf' nent [110] zcne, not two team diffrac-

.gy''' *W i um tien ccnditicns. |,

.,

,Q' ' ? e% _,% .. M!

(C)

|

- - - - -- - - Ij ._.

..- - - . . . . . .

,

____ ____ - _

vacancy dislocationannealing annealing

} "40 - T T I I i ~T I I

S Ni 25To COLD WORKED

[- 1.0

e Or (* <, o39 r * g

o3p' ^ "

, o- h p,o

W =w o-

{ 38 - *0

, 000$ o' , o - 0.6w .c- * o b)$ 37" *

*w \ -w .. -. 0. a' ' , .= / *o" ,p Ni QUENCHED *L .

36(W cis k e t. al. ) 00)

-

y o - 0. 2*

00 eo oe

e35 I I I I I I I O

'

250 300 400 500 600 700 800 900 1000ANNEALING TEMPERATURE (K)

i i i ! I I i

1.6 -

-

61.4 - h6 4 f ->--

31.2 -

fI{? - (b)Te Y<4 1.0 -

X1I -e

0.8 -}{y -

0 -

1 I I I i 1 e i

250 300 400 500 600 700 800 900 1000ANNEALING TEMPERATURE (K)

F 11; 7. Ir c chr: n tl nr.n al ine re r pon e.e of' pu re nickel: a) 3-parametert' co'c! wcrkmi nick"1 with recirtivity .latri l'o rne rt r u r' r n t e :

que nd.e i n: , ' l to ' ycick et al. (10); b) enlculatei. F -pa rame te r..

v"rcus rtnnertling temperatur"

21

_ . . . .

resistivity (also Figure 7a) measured for quenched Ni by Wycick and Feller-Kniepmeier

(10). From this close correspondence with resistivity measuremente, we conclude that

the same mechanicms are operative: the first lower temperature annealing stage occure

when vacancies become mobile (recovery), and the second stnge is due to the disappear-

ance of lislocationa ( rec ryst allization ) . This interpretation was confirmed by

trancmission electron microccopy: micrographs of 2S% cold worked Ni samples showed

no change in the dislocaticn structure after annealing at 575 K (the end of the first

stage) but did chew a low dislocation density after annealing at 800 K. Thus the

dnta chow clearly that positrens are sensitive to dislocations in Ni and that the

drop in C near 700 K is due to the disappearance of the dislocations.

The R parameter, which is another lineshape parameter mentioned in Section 2.0,

in designed to have a value specific to the dominant trapping defect, but independent

of de fect concentration. It is calculated from the same data used to calculate S.

(See References 7 and 8 for detailed discussions concerning the definition of R nnd

relevant acrumptiens.) Values of R for annealing of 25% cold worked Ni are plotted

in Figure 7b. Consistent with the S-parameter annealing stages, the decrease in R

between 375 K and 550 K is acacciated with a change in the dominant type of trapping

::ite from vacancies to dislocatiens; as the dislocation dencity decreases during the

recend annealing stage (550 K to 725 K) no further change in R is observed.

Fram thece results we conclude that the positron annihilation response in cold

worked Ni is due to positron trapping at vacancien alone or at both vacancies and

diclocationc. The recpcnce after nr.nealing above 550 K is due to positren trapping

principally at dirlocationc.

3 3.2 Ar.nealing of Cold Worked 316 Stainless Steel

The resultc of cimilar annealing experiments on 316 ctainless steel cold worked

to ?% thicknera reducticn nre chown in Figure 8. The decrease in S begins at a

higher innmling tem;3erature thnn for Ni and proceeds more gradually without recolution

into two distin:t etnges. By 673 K, O has decreaced to the value of the undeformed

alloy. The R parameter remains nearly constant to 600 K and showc a slight decrease

at higher temperatures; errors in the i< calculat ion become large above 600 K since it

is fourd by taking the ratio of differences between two pairs of nearly equal numbers.

??

_ _ _ _ _ _ - _ - _ -

32 0 , , , , i i i i i i i

31.0 I***** 116 SS -

e ** ;OLD WORKED 25%,

5 *;

.g 30.0 -

-

a2 ..*

(a).

k29.0 - *-

,

5 .Ws

28 0 - * -

..

' ' ' ' ' ' ' ' ' ' '27.0300 400 600 800 1000 1200

I i i i i i i i i

1.6 --

b f '

f - (b)g 1.4 -

1

h 1.3 - oa n -

e1.2 -

-

! | | | | I I i i f I300 400 600 800 1000 1200

ANNEA LING TEMPERATURE (K)

FIGUFF f!. Iccchronnl nnnenling reopence of' 25% cold worked 316atninlecc steel: a ) 0-parameter versus annenlingtem & rature ; b ) it-parame ter vercun annenling temporn-t u re .

23gE| .. __

- - __ _ _ - _ -- - - - -

|1

To further investignte th- nnnenline rec pense of 31h ctninleen eteel, miero-

h n r r.e - -- 13 _ctrennl annenline cu rve n we re d e t e rmine d . lInc,:nesc for thic nlley Jet"ndsa

pri.nnrily en the dis icentien .:ensity, so a decrence in hardnera nfte r a given ennenling

*en would imply n decr.nce in numtvr cf dirleenticnc p re ce nt . Microhnroness vnluen

L"re ~um rure ! n t rnem temr"raturr . allowing ceouentin1 30 minute nnnenling steps

b e twe( n 4 ' nnd I t 'U Intn for '' nn! 7M cold worked esmple a re give n in

Ficu ro 1 The majcr ctnervntici tc ' ra n f rom thert me neu rementc ic that the hard-

nera ac not -!ec~-nre until n: enlin,; well in exo en cf 673 . This ir in centract

to th+. lit ren r v rihilaticn recults of Fimre S which rhow n coe ren s" to th- pre->

:e rc rmati n ' v a l u.- e: +ntin113 complete E 'u r Tre m.iccion electrcn microrcopy"

c b re rva t ? c nr. ei r 7' r cid work rn=ple nnne '. le d tc c73 4 ( F ig re ICL) indiente e

ii n l .,ce t i c . e n c i t .', unchnne ! rom th- nc-wcru ctate / Figure len ) - n rerult con-

sictent wi+h tht mic rcha r :ner s r"n :ingc . Enrplo nnr.en le d to 1073 e hei* n well-

nnnen1 t ru 7 tu r 'icure loc ) n "- low har in< n r .

F r r: thn e ri< cf ex; riment en col: relle 316 rtninl rc ct"el, w haveenclu thn+ th- ; itra < nnih ilat is techn :t" n , plie- hore ic in c e n c it iv> to

tir1c- tic - in thic 311~, Tho .e c re ve in ' c n nr.nonling to 873 K in primn rily due

te nnr+nlina cf nc v 1or c-r vnenney cluct"rr

$.s . . u l .~. , -. ,

~ r s T . , m. ,o a l,cd .-,i.7.., ~ a :. uo,.... ,

. ar a.. . a

Th- -: + 1in, w cr nce cf rnmplec * nkm f r m th< entu"ution pIntenu r"giene oft h. thr. faticuo eur < . "nm F irm r- are prer<nte in ficure lln-c. Linerhnpe- nm~+ r rhnn * nr ~xrr+: he r* n r .

rrc ** .e2o

. .

1 O

wr re tb ; n re.* '" r o f t e r n civ'n a nn .: n l in, rtep, ,g ir the nnnen le ! vn lue , nna

. i' *? tinl 'n l u th' .- 1. M , roc , tor rnture fnt ipp eurve1 .z1

a

'N .li c u r-/* ,<-n< r ** ! fo r tl.< t w:. 10 , t+ .y-ra tu re f a t icu+ 1 s pecimo nc*

4

.

( F i ,, .r" 11n nn! t > I n r. *"ry c imilur nn ! "xhihit th' raro behavior an the 2% col 1roll. , .: l e , the cu rv e: nten til; ecrt a r. s ta rt ing n t 500 K nn ! rench the well-

nne n le ! li .at C- W * Ir s nt ru t, nni"alinc of n fnticut reple cyc1"d at.

P!4

- _ _ - _ _ _ _ _ - _--

i| || ||

0 60 1- - - - - 4 3

1

d9.

O, ._

I 3 r0 e0 k

D o1E wKR 0 d

0 l, O I

2 oW c1

gD 0 n

0 ii LI

1 lO G ,,C ,

1 aen% 0 n

0 ai

5 I

0 -7 .- O- 1 p) eK t(

0 s,

' I 0 E-,

O O 9 R y,,

bD UE T d

i KI 0 c0 A e

. O R 8 R uO E d-

i

-

- u W P o- r0 M p-- O D I 0 EO L 7 T s

n- O e" g C g

0 ni .- o I 0 a

v % 6 h- .

5 clO 2 e* g seg 0 sti -

. I 0 eso 5 nc d srs

O ae.*

0 hli O I 0 on. o 4 ri.. g ca* it

Ms* 0i

- - O - - - I 03 .

0 0 0 0 0 0 0 0 95 0 5 0 5 0 53 3 2 2 1 1 ERU

]wZg<r c_ooZx GIF

_

| | L

r

.o v3 '.,,

. ., ol

< | '" Q?!z]! - o

kl. 5$,.g.-,h.

~

..

. -;,

s., '~ ' ti 0 C;> - cy' '

. it a o

I? I' *t$ .. ''*-

:.:.- ; v~

s - i. 9 j C ''

d E. [.*'

- ,,

. , 9"

:4 ?! t; Or71.c 4 ()

4 , *

2m .

T [ja.

,.j .{ {%:' ' "

*- t .

'Pt '*

EE,.

*1 ~

r e

e' %..s.- .,- -.,

* *

'.*. , _ _ ~ . . ,

. ,'sva

- -, 3

,

s .* g.

n;;5. .> .. .; ..

. .-.,- -

,

-'c [,;

,$, - 'i|,

a'

O *. . . > ~

< ' L~ lf,' & b y} %~.

g g:d;. ,.' x

,

i,,794,,.

g ..i.

.., Q i . ..'- -

..

h La $$ kk-

. . . .

. t?

m . . .. _

_.

.-

, . , , , I i , i

1.0 ~

** e 316SS

~

* * *e . ROOM**

TEMPERATUREFATIGUED !")

d't = 0. 6 Toe ,

% 0.5 -

e e -

wo e

e ezW em ew

ei e3 'o* e . ee o,o , , , , , i i i i i iwy : i : i i i i i i i i||E<cr 316SSs * ROOM' * * * ** e * TEMPERATUREyo. 1.0 -

FATIGUED -< eIen att = 1.8 %w eE"

e*e ,

e0.5 - ** -e

e

o ,

(b),

e oe

0.0 -

-

300 4b0 Sb0 6b0 7b0 8bO' 10b 1200 1400

ANNEALING TEMPERATURE (K)|

FI T: 11. Iacchrenal nnr.ealing r+ spcn m of fatigued 316 stainlecc '' eel:-

a ) f a*. ir'a <l L 93 K , as = 0.6%, II = L71?d cycloc (prior tc. failure );L '- fat'..me 13 K, as - 1.8% , il = 37CO cycles; c) fatigued F66 K,'e - 0, 'jf , Nr = 10078 cyclec. fLinechape parameter differr nce, C,defin+ 1 in text.)

27

, __ _ - -

!

. . . . . i i > > 4 icnm 316 STAINLESS STEELw FATIGUED AT 866 KO 1.0 d ( =0.5%-

-

eifsaCU 0.5 -

.

m2

UOy O 000O og o o< O O 000 O O OoOO1 O o

Ow O

@ 0.0 -_

IU)WEJ

i . . . . . . . . , ,

300 400 600 800 1000 1400

ANNEALING TEMPERATURE (K)

(c)

K

26

EI6 K dcor net produce any cirnificant dec reare in a0 (Figure lle). This ic cont.ic tent

with the view that C is primarily cero ttive +o vacnneies: in t.his alloy vacancieswculd b. mobile and would diffuse ;o cinks during fatigue cycling at E06 E. Thuc the

in c rea re in C alth cyclec we mi te rmall (Figure h) and there would be few residual

, r enneies to annen t (Ficure licl.

3.5 Ei SPIITIC CF ALF.TICNAL FOSITECN /di';IHI!ATION MEASLTEhNTS

In th< ccurro of thic recenrch, reveral other experiments were performed whichyiel.iv i nynth rorultc but which were informative.

I' opp 1"r b r::n :"ning roarurem"n ta we re ma i+ on 316 ctainless cteel creep cpecimens

torte- at EG up to 7I creep atrain. C did nct eher.ge c;gnifienntly after any of

the:+ tect een iitiona. "icroctructurally, the creep tects produce n cmnll increase

in dislocatirn dencity, nn i incron y c in vacancy ecncontration during long term testingnt E(6 K woul: h- nogligible I;o cther defects were generated in significant density

to trnp positrcr.c, no the Ir.d cf chnnge in 3 is not rurpricing.

A trial eri"c cf moncurement> wnc nnde ucing an alternate positrcn annihilatien

tochniqu", nngular corr"laticn, which moncures the deviation from a 160' path angle

tetw"en the twc nnnihilation rhetenc. This method uns deceribe:i in en earlier report

(6). The confia rnticn of th- nnglar correletien set-up is not suitable for in-o rv i c" incr<cticn, nnd its - here was inter.ded primarily en a recearch tool since:

1) it is ir t rinciple more necurate than the Doppler brcadening technique, and 2) withtb "xirting ret-up, encurem+nte enuld to mnde en the curface of specimenc without"ct ic ning. Thic woul! ellcw cequential canpling of individual mechanical testa.

T'r technicuo, hcw"ver, recuirec ecunting timos of the order of 36 hr, compared to

0.5 hr for Doppler trca "ninc counting, tcing the rather Icw activity pocitron cources

m il" :10 h" re . ' < u l t c of me n c u re ne n +, - en cevern] fatigue camplec chowed large

rentter and statictien1 orror tarc, caused by 1cng term electronice dr ift during theIrne ecunting perini In principle, thare are ceveral poccible improvements which

could la ma# , but it was deciled to drcp angula" correlation in favor of a redesigned

perpD r trendening apparatuc which will allcw curface meacurementc.

29

_ _ _ _ _ _ _

I

i

Preliminary Deppler brcariening measurements have been made using a modified

sample-source geometry with a armote positron scurce which allowr measurements to be

r. vide on the surface of samples without sectiening. With the new geometry, sequential

mencurements can be made on individual fatigue specimens. Results so far . indicate

that the sensitivity and time to accumulate data are comparable to the sandwich

source geometry.

.

9

*

30

. . . . . . . . . . .

_ _ . . _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ - -

.y,- a. .. , Lr ., n. .,. v . c_.a. P.,.,a. ,, . y. a mo n ._ r p Ls.,.-

Th- !nck ^f pc.:itrcn annihihtirn r"rpcnre to dialccat icn.: in 316 atninle . st"el

ir plier either that lislocatiene trap resitrcnr cnly wenkly, or elce the precence cfall cin,c elem<nte enuren the rnremoter for positrens annihilating nt diclcenticnr

'

to 1+ r r.ilnr to th. ; nramot + r fcr pcritr;ra annihilnting in the perfect Inttice.

iceriminnte lotw+"n tr. two ypoth'rer woulJ r"ouire more t aale re cenrch nn t --

' L' , re t i"a l int * : r" t u t ia. . In nr', enc", the ab i l i t ', cf pocitron annihilatitn to

n .it c r "le + + 'onterst "fc mntia (ntcv. 700 k:) in 316 ctainlerr stoel ir"

a : ni t fu l : . ~ cl. g. " rat"J n ico r inult anecualy nnnen 1 nyt t he re cu l' ant

tr"rv sienni ic w"nk.

Th< o rv i incr"nr ir for fatigue cf 316 etninless rteel nt 666 K (Figure h)'

te "xplnin" t ;, iner"are in lis1centien Jenrity with cycling. The positren<. *

+ rn t which cc:.t ritute to this incrense in 1 hnve- nct l'een definitely ectabliche f.

P ir ; r e it' le ttrt rcn" rori :unl vnennei"r or vnenney clunters enureq by deformaticn

r.mnin in *ho re plo the "ni of th+ fntino tent. Cince the crecimen coola dcwn*

relntiv"ly rapiil:. at t h* on: cf tb t o n +. (~ DOC K min), nome vnenneies mny be/

:ur nc he v in nn: r . r' . nr par it-cn t rnpc. T: r inta cf Figure lle, hcwever, chew no.vi- _. vr ancy nntenlinc. A roceni ptential rcritren trnp which has been

:1c.> e ! t; trnn n im icn *' trcn -icronec;;, f;2) are emnll (.0? pn) enrbidec 5.hich

cr" iritnte v oterce *" curly curinc faticu" of thic n11cy at 666 K. The trapping-

.n. t!'..""sn cf ther" cnrbi :er have not t en invectiented fu rt he r .

: ntn f::r ruro nickol :o 9 .cnetrnte censitivity t , dis locations , nn,1 McKee et al.b ve "rnnt re " ! ornitrcn t rapping nt iirloentions in pure copper (1). Elevntod+ m:"rnture i r.nt ic n c: ul t prot ably b< fellowed for these two pure metnis. It lan' knxi whethor meitrcn annihilatien could icnitor iirlocaticn chnnges in any other

v re in 11y nynilni:le hich temporeture alley, nr the unterstanding cf elloying effectr

r--it rcn t rn;: ping at dirlocaticns is nocrly developed.-

? ;ecitren onnihilatien r"cronce to 1cw tem;wrature fatigue nn3 monotonic

t"formatien rhcws roc 1 ennitivity in the early stage: cf deformation up to lQ% oflife d col t work. Hew" er, th- reniince n;prcnch n ecnntent value a f tm thin,

which wt 21.1 t r .iuo eitb r to caturaticn in the nu.mber cf defectc genorate-l or to n

31

_ . _ _ _ _ _

_

:nturnti.n " t e r~. i n Ly trappirm cf every pccitrcn at n efect. In th+ latter enr",'

t he en cu r' * "f t ect woul 1 1 " pe r2 m ; citron flux ani not cn lefect <encenerstien.

Th' r" r u l ' ir that the t"chni:u ic not c o n r i t ivt to th" inter tan rf " fo rr.n t i c

Thic v;ul ! r" t r i c t it - npplienticn to a menitor rf ently lifa or "r trancient evor-

Icn : It ecul :, Lc m a r, t , use < 90 ccreenine. woi to t"i mine whethe r a f1nw. -

"t eticn n - :+ c t ru c t i '. "xr.innt icn "ncurement rhoul i l e perform m . Alco, it in

;- rit1" th * "n r la. lifv curfs ren t inr m uld !+ ca rr" Int" i with <!"fc rmatien 1"n iln,

foticu' fruck init iati+ ,

TL. limit- e r i c- ;"rform,+ her. en "le s.r . i tempo rntu re crcql"c cycl 4-

witi nrirm. h; 1 : ri ( F icu r' L ) c h c4 that pr en nnnihilation n"r nat'

e e r r" I n t - wi'L ir lifo tu- to arm p'fatirue interaction. Tr etc withn. t 1* -

u ri ' et . inn t i: cf Lol: "r i- |c fall a lr % the a nn" cu rv" a c emnlec withcut

h 'l : TL. ir r * xr < t iven th- r~rultr of "le ct rrn microccop:/ an : pccitrcn. ,

E'

nnihil t!: t n ' ! ~." n r 4 f o rm* ty creen or fntigue alen" nnd with creep /fnticu+-

i t ' e r n c '. l > !' cr- n!" r. a t prcvi: Inr ' numh v of def"etc i v s c rtnc i- cr

.!i r !- *ir -)- -) t!. nly ni:; r rt ruc turn l d i ff" r"nc" not 4 by trnr.cmircicn el"et ron.

m 'r: '; i n: i n,_ or, in nudern of rrsl! e n rb i i< precipitnt" in the creepffnth ;e'. rte ~ n r- 'e tb 'ntic ''etc. T9 :ecreare in lifi dw te c reop/fnt igu" in te r-

'

n ti r >t < .m rapid frnettre pracocc, ruch ac onhance! crnck initiaticn

ef vnt i r n , -- mrit r- n ibilntit r. n . !!n.fr n r" rot lirect ly related to*

o .... .

A n u t i l t i c r.n l hich Lac ! * "n 1.:entifiel Lor <- ic for n d"fett-crecific line-*

r, ruch na the Ja rr.ete r, wh ich wcul i c"rve to identify the type of.n ; nrn .t.

' . i r/- n t f. e t . Tl 1 w 1 ! cc mplem"nt t h- unntitativ< pnreeter m cinco O mericu r";-

4,

n l:, t : t n l . b f" c t ernc<*trnti<n. I t w '?-u l l incure thrit rencuremento of ecncont rn tion

char * - !! v ! ' nt1 th" c r> ty p. of " f. > c t . Th- E prumeter values re p rte,i* v i vu r" 1 x:. ct< i for 9 : foc t-c ;"c i f ic p ram"te r, tut R

n1w n l cu l n t. :( - 410 ctninl"cr ct""1 r nr:p!"n are nmbiguoun (Figure b) . Further

um ri e n 1 e n ! '. n i r n- :' *atr perferm i en onnihilatio. pheten enorr;/ cpectra dntn

s lrm t, collect"! in th. prxrna to t "tte r refin" the "nergy chann"1 wirithe uced in

t he eniculntien of (Figur" ?).

U

_ _ -- . - _

_ -_____ - - - -

', . o '4 TPbrfTFirTUil.h iV0hilTIM AS A MMIT(a CF DAPNE

Th" re< "nt w o rk in t h- nre a of bulk micr( structu ral chnnm. cecurring during

c ree ;, nnt fot igue hoc provi:irl n rm >l un te rstan ling of the na tu re nni extent of th"cech nn /4 in terts < f ty p M an l 316 stainlere steel (13.14), with thir information,

the nil i ! i t;, c f using n Unrticular ~tho! ruch ac positron annihijntion to < !e t e rm i n<

t u l k nie rc r t ructu ral evolut ic-n t o predict f ract ure enn b+ dircuccel.

TE- !: . i r.a n t nierc s t ructu ra l ch tne r<rulting from cycling, cr ip lon' ting, ce

tb it a l ra t I t n ic t he ren rrnnge:" at of d icloentiens toward n eten-ty-state cub-

s t ruo tu ra churneterirtie of the impoced ctr cser an ! ctrainc. C hn rc"c in d isloention< r.r it; my nica ecur, wit h the dencit, inerencing siuring tecte of nnnen]el materia lor +h, e n r i t :, ce r nc ine curing tectr of initinlly cold worked materini. Ther. a retb 'nly varintic- cccurring during 1cv h mal a;ou c rm pe r a tu re deforr.nticn cf cingleo

;h- allny3 rm i n re c t i ll t h:?,

dominnnt changen du" to deformnLien of nucteniticrtninle :: t " I r atx ve 600 1, "ven th ough c om. en rbi.ie pr"cipitation occurs.

-

Thir rui c t ru c tu rn1 evolutio n 1+g ine w i th the firct fatirue cb le or the firstminute of cr . ' ct nn' ir urunlly complete in lecc thnn nbaut ?QL of the totalr p, f i ne r. l i ft ('lT. The c ten t';-s tate liclcention anncity an:1 lict rit'ution pluc the

pr cipitation ocarring turing the tert do influence the finnl fracturepr .r.-

er hcw "r, t he cut'c t r* cturnl changer occu rring en rly in life ar" not dir"ctly,

. e l n t, t t. o the vi e r of c:ic l t te failur' or the timo to f a i lu re . In gen"rrtl, th"

t u lk cut'rtructure ev >lve: nc n function of the numb"r of cycle: Oc timo from thect o rt o f t l.+ t. ct n: rmch- the eteniy-stnte ccn<titic.n well before nny clear in lien-

'

t ien of crack initintion. C rack init intion ic d irect ly trncenble back from the eni*

n test uring cr~'k propngntien bhnvior, nn i cubct ructure evolution in related tothe ci nrt of t h. tert; l'ut th"re is no discernible relntienchip tetween the attnin-

rr n t of n rten ly-rt n t" rubetruc ture nni crack initintion. This type of relntionchip,at 1. r t Indirect, wou]J 1+ poccib1" cnly by gaining a ccmplete un ierctan-ling of allt he pr m e: "c involved in cr'ep an! fatigue f rac tu re and their interactienc in enennlley cyctom.

33

W$ _ _ - .

_ _ _ _ _

||

t1.

Those ccnclucicns are teced primarily en results from transmiscicn electrcn

F.i""cac cr'/ v.h ich is a Eco:! tcol fer c'rrerving the impcrtant microstructural ch9nvos

whi^h offect f racture rrecessec. Is t b e r t , a ncn-ciest ructive examination te chnique

v.hich .encures these v9ririticnr wouli =cnitcr the s eco chrngen ctrervei in thes

el"c ',n micro.wcre. If the chang . ct rerve d by electrcn r i.croscepo cannot te

dir"ct ly relate- to fatice cr creer life results , then no tierectructurally rencitive

n^n-fortructiva examinntion technique exists which couli be used airectly for life

Irotieticn.

i

E

1e

Js

fe

.

>e

3a

1

--_

6.0 F ';GI EEFING APPLICATICN CF POSITRCN ANNIHIMTICN TO NON-DELTEUCTIVE EXAMINATICNLF :ThPCTURA L COTC n NT"

A nutter of potential engineering < aesticns have been posed on the applicaticn

of poritrcn annihilaticn er a ncn-destructive examination system. These are hardware

nn.1 cperaticns problems which are independent of the positrcn annihilation-damage

s orrelntion cuesticn which has been dircussed in Section 4.0 Here, several of there

items nre identified and discusred, but considerable engineering development would

to ..eeled to cbtain complete answers.

0.1 m0ITE AZlHIIATICN EGJACE EASGI.'ENTS CF CCMPCIENTS

The current test esti~ ate of a prototypic system would make Dcppler broadening

menrunemente using cceputer-controlled counting electrcnics and data processing. A

coincident counting techniaue cculd be used to minimize spuricus counts of background

v ttctcnr; thic wculd use two cetectors at 180" to each cther and have electronics

c cnficu re to triger n ecunt cnly when two "'s from single annihilation events

nrrive cicultanecurly. Usine suitable detectors and scurce (Secticn 6.2 below),

ccunting timer for a mearurement would te cn the order of one-half hour with reference

rtan ier !c use d to celibrate the syctem. The test unit could be reasonably pc: table

nna <cul.i fit in a dcuble electrcnics rack. A prcblem with the 180' coincident

ecunting technique ic that thc geccetry wculd nct allow sampling of recessed or flat

n rear or ccmrenentc with lare radii cf curvature.

o ' WITFf'L CCGCES AhT COUCCE FIACE'ENT

A variety of iretopic pccitrcn emitters are potentially available. The majorarinbler are the half-life of the isotcpe, which dett rMnes the usable life of the

nnd th'> end-point energy of the emitted positrons , which determines theirr c t. rc e .

depth of penetration into the rample. A listing of several candidates is given inTable D. V a trade-cff's between high and low end-point energy positrons and their

effect en positren annihilaticn sensitivity have not been established. For practicalnon- ie s t met ivo exa .ination applications, it is likely that long half-life isotopeswould be preferred; and high activity sources would minimize counting times. Source

placement on the exterior surfaces of components is expected to be straightforward.

35

__ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ - _

F__ _ _ _ _ _ _ . . _ _ - -

,

TA B LE 2. ICOTCES USED AS PCSITPC:; SC5CES, WI~H THEIF HAII, c.m,,. t. c , n um, ._ r - .. .s, C i Iv ..c.c..e, ...

-,- 7 . ~c , r .mu.. n.c ve.1 4 -o-.n , r - . . ;; , ~.a . . . a l.r e .tt .

PositronEnd-Point Energy Production uIsotere Half-Life E (MeV) Reactions

o~

^ Mg( p ,2 )2#:!a"5 ^

:!e c.cd yr 0.5h

' ''

~'Ti/"Ec L7 y- 1.h7 ~p~C a '2 ,2n ) "T i''

ssc e*, ( n

,.%55o' ' v' m 7Q.m .k. q /c.

n., a s,- um,

c- cOr e / p,ncn)J Co/ c

s

s 7'!i .

0.85 56' Fe ('He,2n)5730.0 h '!!i

cP' '2 0 'l.3 days 0.LE c6 cc

' Ili(n,p)''CO,

'':.'n (2 , n ) ' 8cs c

Cal

-:b ,4,u ,

,,;c. C . o p- c3,.u ' n , v) *, Cut .c.w c

A" n 6"'Z 6.' inyc O.33 Zn(n,v)65Zn

tm , .m,__ . c

,

-ae ue crp cays 1.uc- En(1,2nj3 ;Gea

..

U"R .

1.22 .'Ga(d,2n)60-

-Oh c0'Ge

R .m9C ,. / p ,n ,390 5,.; . 7 w., ,..n,n.- y

- m. s a-

GO'' Zr(,d,2n)GOib

|

36

, _ _ _ _ __ _ _ __ _ . _ _ _ _ - _ ._ _ - -

. _ _ . - - -

Fcr measurements en inside surfaces such as piping, h^ wever, the positron source must

te located cn the inside. This wculd be a difficult mechanical movement prcblem for

ec= plex structural geometries. Annihilation s's gene ated from inside surfaces could

be detected through several inches of steel.

6.3 SUFFACE F IFAEATICII A D ARF.AS CF II;SPECTICN

Since the maxime. penetration depth of positrcns is limited (of the order cf

0.3-1.k mm), any surface defomed regicn, oxidation, or coatings would need to be

removed in crder to take .easurements of .naterial representative of the bulk. The

preparaticn method used in this study, electropolishing, could be pc_cformed en

selected areas of interest. These regions would then have to te protected from

cutsequent curface dsmage luring the life of the plant.n rf the need for surface preparation, it is unlikely that the entireecause

surface area cf a ecmponent could be scanned. It would be most practical to select

critical 1ccaticns for inspecticn which analysis predicts as having high potentialfcr hence.

t5. L INI*IA L CTAC ' 'm A LLCY

It is certsin that differont areas of compcnents would have varying initial h

leve la cf vacancy cr iislocaticn defect concent rat icns . This would require baseline

~eacur tents for each locaticn to be monitored, and these could te compared to data

for well-annealec sa ples of the same alloy to assess the initial state. Any residual

ecid work perfcrmed en the ecmpenents or structurc cs a result of fitting er joiningc:eratiens wculd shew up. It is possible that regions with significant cold work

. ruld give high initial positren anni?.ilation readings close to saturation values

for C, thereby confusing interpretation of established da . age correlations.

37-38

_ _ _ - _ - _ - _ _ _ - _ - _ . _

___ _ _ _ _ _ - _ _ _ _ - _ _ _

. , ,m e . > I .n. ,s .' . -

7.1 In 316 c tsinlecc steel an : nickel cold worked er fatigued at room temperature,critron .nnihilaticn measuremants primarily shcw sensitivity to vacancies generateduring leformaticn. In nickel the vacancies anneal cut by 550 K, leaving dislocatiens

in the micrcctructure; for pure r ickel positren annihilatien is sensitive to disloca-tien density

'~ In 316 stainlers steel positrens are incensitive to dislocation density. Thiswas oncnctrat,; by ec= paring pcsitrcn annihilation readings, transmissien electren

micrcsecry, an; micrchardness measurements.

".3 or 316 risinlect st"el pcsitrcn annihilaticn lineshape parameter increasesr

crrtcnically with damage ani seturates at approximately 20% strain in cold work and

rt 10' cf life in Icw and e} evated temperature fatigur . Thuc positren annihilation

r-.ajince are tert suite-3 to mcnitoring the early steges of deformaticn. Pctentiallyt he early values of surface damnge could be correlated to fatigue crack initiaticn*

cri -ria.

In 316 ctainlecr teel r.uch Icwer rensitivity is crcerved for elevated terpera-

faticuo thsn :cr rccm tercerature fatigue. Thus positrcn annihilation dces not+

nr. r tr Ic .- 11 cuite d as a ncn-destructive monitor cf 316 stainless steel atelo r* t"nrerature. Thic results frcm the icw sensivity to dislocationc and thePrbiliti of ' acnncies generate t'; deformation at temperatures abcVe 600 e (620 F).

5 ''ince not all changee in tulk nierectructure are related to the fracture processes,ervntiene cf there cl.anges by positrcn annihilation, electrun microsecpy, or cther

t chniauec nsy cnly provide indirect ecrrelations to the extent of damage. These

corrolatienc wculd need to be established fcr each alloy cf interest.

7./ In r jiticn to the problem cf damage correlatienc , a number of engineeringav rticnr on the applicaticn cf recitrcn annihilatien as a non-destructive examination

technique for ecmpenents ani structures have teen identified. While none rule cut the

t^chr ciu ', ecnsiderable devalopment work would be necessary.

39-40

__ _ _ _ _ _ - _ _ - _ - _ _ _ -__ __ _

_ _ _ _ .-

ACIG0K EDGD2I.'T

We thank J. C. Banks for assistance with the positron annihilation measurements

and C. R. I! ills for performing the transmission electron microscopy.

/--

_ _

41

_ . . .

.s'.. f. ....'.'h'I. , .h

* * e '' ' ' , [ M l .':' O t f) , [s T. i t . se rt r t. , si' ; ''. .: + c t t , citit , ,1 , ; } g. 7 ,lo.). _.,.,,,

, ,

rk o n .1 . I . '' Frckin.* !b . .y, If . t t , p , p f ;1,,,7),,

>'8 8-,S 3 . O

' h a''

4 If | , f . ,ik k , ,j, 4. , , ,

. .

,

l' m *i t. ! n Eu,* ' :, in Nor m cl; T4chni.;u. In N n i. .:t ri e t iv. T. n t in,..

'. , .n r; o , ei., (aco 9 ,1c j9 ;, ; g y, ; pg 3n, ;, .g. , g . g 7 , , ,, ,1. <

'l' cin . 2 . .1 % . un' A. !: ' t.c.h : , Fi3

'

U n t"a ' I 1 p r t. A E iM 5 ' '_1s

.I

a ' t I;I' t t ? ' , . , [., e n m g } * = p 'f || J , ,, , ,

h,

f i t . . ,- . .- ,

1, . . . . . c. .-,, , , ,..

r . ,e-l, , , ,duelear s yu ] n t. :' ry ('- :m i c - { c n i .. pi > r t, -

:{ _ t )] } H \ |[O 1,'a <

i. i|* - h' 1 j !.- },.,, ,, . ; yn.,).. . t,

_ < t

I'i n t. i t!! ! 'i' r i + + . } ;'/ ;, , p , ;}; , ,tt _7 7, 7,,, ] , ;.g

% Q . , . t . S 'hy & ,

_' yy y ' , ''

l' ? l :' k 1 ] . . r - F~~t le "? r '.. ' ( . ! , .J . .. C.k* ' b< e.nt. . , tr) fd , t; } g, ( } -([g ) ,- i - is *

,

11. I.. e t r. 0: n . , . r- s'T, .mi H L m.++o, " N' m ' ~ k Mm1"

', , ,m

- ' t . , th r- '; e ry n n i . c r', c t n l l i r n t. i ., , , Eet ' :v o r:. nn1.

n, ..n,'cryn'nlli-- . . . .. , m. m,m. _m,1., , . , v n ., m , 1 c. , ,

l. 'nt O' !;. t" . . q r. h t_ {, ,g g p } j . t ,g g ,,,

}; l'. l.. e .e. e. , nn ir. i . . .n , E. pi,reka, Aetn M.: t. , y , J r.r/_ l iti (pcj'7),1i i

., .

" 'h 111snc+ * , nn 7 " tc+c.. ...t. -i r. n r,. , , 3,. .

i i. ,_...,3 ( 19,,,3 ) ,<f.

,9

w___.__ _ _ _ _ . _ _ . _ _ _ _ _ _ . _ _ _ _ _ . -- - --

_ _ _ _ _ _ _ _ _

D IST9 IEUTIGri:

U. ,. I;uclear Pegulatory Corraiccion(310 ccpiec for R7)

Livi_lon of Document ControlDictril'uticn Cervices Branch

.

Thi 0 I;crfolk Averu

Pe th"rd a , ?'D 20014

U. . I;uclear Eep ulatory Commincion (54)Pivision of Peacter Cafety FesearchOffice of I;uclear Perulatory Fe < earchR chingtcn, D. C. 20555Attn: C. *; . Kolber, Accistant Director,

Advance 1 Eeactor Cafety e ren rchE. T. Curtia, Chief

Analytical Adynneel Peactor Enfety Research, AESRM. Ciltert: erg, Chief

Experimental Fast Feactor Safety. i . Wright (50)

Exp"rimentr.1 Fact Peactor Eafety

D I erartment of Energyffice of fluelenr Cafety Cocr fination

. S nh ington , D . C . 20Sh5Attn: P . . . E a rt e r

U " . Fernrtnent of Ene rgy (2)A lt u uerque Creraticns Cffice

. i'ox SLDOA lbu que rque , I;M E7185Attn: J. Eceaer,I,irecter.

Cperational rafety DivisienD. k. :Iwlin, Directcr

Special Frcrra= DivicicnFor: C. P Guinn

D. -y.nle

Univers ity of Mich a nnI;ealear Engineerin,* DepartmentA nn 3. rl' c r , ''.I M102

lereral Electric :crroration310 Iw Gu ii:n*- DriveCunrcriale, CA A026Attn: ?'anager, Dynamics and Cafety

W. E .yer

P. G. Fox 1845I inho Fallc , ID 83kOl

Frejekt Cohneller lirueter (h)"rnfcrschungs rer.trum Karlsruhe GMPHh ;tr,ch 36LO.

L75 KarlcruheWer t G+-rmanyAttn: Dr. Keccler (2)

Dr. Heunener (2)

43

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ - _ _ _ - - __

D i s t ril ut ion ', Con t ' .i ) :

Institut ie Irrt"etien et <ie Curete i.u c le n ire (3)Ci Il F c n t"nny-nux-IO_. '" rF P f>

r^ bJ Pcnt"nny-nux-Ferer.

Frone.Attn: :1. Tnngu;,

M. Fr!=ittu.,. o,,et<..

Jnl'ety tu :1.' Lnb ernt cry (3)-

' :mi c s n rint 9 L'rs rgie Atominu.

n+. r" d ' E tt w ::u e l"ni re: d" Cninrnche'

B. D. 1, 1311S :'ni: t-Enul-len-::uraneFruch-n-Tu-H.F rn n e,

Att:: Milly'

', ' ,ye r He i n..

'. r"t

Ce n t r- t ' F. tu o s ' u c le n i r+ '!" G r"n cb leb5 - "ntre i" "' r iF. I .

3P 501 G rer ; b 1" , ce .xFrnns/>ttn: ". c. t n

. i,.o..~.., ,sn. . ,r,.

t ..:- ...n' n !'" t *, nn* " 1191 11 i t '. 'Ir"eternte

'.i i er h nw *a n., ,1 . n , n. .-

.inrrinct :, i 3 '.*.~cn,.* l n n

G. Fe l l am'.nctor Fu"1s Group

AFh H u n," 11

i x t'' sh ir", EX 11 < '-

.1 n: <

^, ,. . . ,,,...c ,.

r ac t :: r P , ," lc r="nt P ivin icne

1 n - tmic En"rg:. s t el- l id="nti n t' r i * h , T 2 rch" r t" r

: creet

r glnr. 4

' ' int :-eren rch Cent re (2)Isrrn stablichr"nt'l tN''' I n pra '.' n r+ s e )-

Italy;ttr F. ;1+ rc

H e lt b"e ke r.

&

- _ _ _ _ _ _ - _ - _ _ _ _

_ _ _ _ .

Distribution (Ccnt'd ):rcwt r Peact or & '?uclear Fuel

Development Corporation (P!iC ) (2)Fast Freeder heactor Eevelciment Project (FFR)ri-13, 1-Chome , A hr nka'dinate-Ku , TokyoJnpanAttn: Dr. ':ochizuki

Dr. Watancte

11CC C. D . P royler

Attn: G.t. !innech", 1120G. L. Cgle, 1125li. . Viney, 1130J. li. Pavis, 1136

1537 '.F. Ko lti erAttn: r. v. Actcn

T . Y . C h' :]W y, ; ,oilsen

Attn: C. J. =rc5.'.tt, 1552J. H liecke, 155?

?l5' T. L. h erb.an3?+ 2+ B. :i Ynter

C' A. ;nrnth.

'31 J . li Fenken

Attn: J A . linlb le it

. J. McEnnielJ. : 'Ma re l

'V .'. ?ny ier+

++1' L. J. McClerk"y''7? J . ', . Wa lke r+

'" F. L Ccatr++

W. J. - 3rcnacer'*

G. W. Mitchell+

., ' l' . A . Tcwerrt' ~

J. r. , p lyn r.i

A' D . '. Vnrela' ; J.F. Ecsell++

V3 G. L. Cano, ,

+3 a. c . c. .a t. r .,i n , ,u.. ~ .

'+'3 J . G . Kelly' ' '+. 3 n. A . McA rthur

'3 . O. Feila.

'3 II . L. 'ectt

'. '+ ? 3 . T. Ctalkor. 23 W . li Cullivan~' 3 ^ . A. bright.

.N P.~ Pickard!. O J . T. I!itchcock'

' ' . P . II . WorlNge'

JS X J. Camp'''5 R.W. Ortencen++

''+' O. C . Williams" ". F. Young+

W 5 R.J. Lipinski'?S W." Breitung

U. ?.1 M. L. Corradini!+ M. W. A. Vcn Riesemann (25)'460 J. A. Eeuscher%5CA L. D . Fecey

45

, . . . - - _ _ . - .. -

!: 1:' t.r i t u t.11 n 10 nt. ' I ) :

? '6 ' l T. i. :'ch:a l i t6

767. i D,it, 1, . t .

i11*i !< ! e } ,, .. t . 1'

>l1I T. i'lernux.

'1 I k 's%' ' t il; j '' } e ? 'r I' ,s 1-- ,1, n , .'

l ''' r rN.'i n r 1- i ,

' - (. li A, 1 s rt: t gj' '' * #. d .I, :"q st r. !y k

''N , i Inn;'

,<

'? l' 2: F. hioinv ,,

e I ?:% II.

k) *'a. i'tV.

I i>

'51 II . I ''if !;n :11

!4 :' (' !: L.i rn'

.

" B !i . i n. . ', )',

5'' v'*l!! ! * Ti [. ' .- } e ( .', I'

' 9 .I, ', ?, i :q::i

E, i . no ! ra-

.' ' , ' . ' . ' A, 'niIn-h'

*

*% 'i. '.

it . |9:1;; f , +.

'.'tf-. 1 A33

5 [ 's 1 I., ,et er :'I4

5 l'' l [,, i|q rgp ss ,

r: I< i' i . i n 11:n 1, ,. 3 e 1, q . y )1. ! ' 'a - ( 'r ;a .- 1 j ,s

l >r: N I t ril ut li n t *;T I . '.

,

'd!I

_ . _ _ _ . - _ . . _ .

. _. .-. _ _ _ _ _ _ _ _ _ _ _ - _ . __

,

1

9

0,, m.a n ., n.. ne- o,2 em o~ n~o-.

_ _ - . _ _ , . _ _ _ _ _ _ . . _ . _ _ _ _ _ _ . _ _ _ _ ___v- .-. - ~ _ .

1

---_-y.__- __. 1 ---f --- . - - , _ _ . - . - - - - -

| | !'

__u_ + - - . - _ _ _ _ - _ _ . _ . _ __ u _ . _ ____!.._2--_.-_____ ,|

jei

; i !

_________!... _ _ . . . _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ . _ . _ _ . _ ..}__ , _ . . _ . - - . . - - - - . _ . - - - -.-- L .

;-

iI ! , i

i}

'l ,

..--a_._.__+ . . . _ _ - _ . . _ ..r---- .- .__,__-_.-_,- -. _---_.---'

!ti |

J .

I

__________,r_ _ _ _ _ . _ _ _ _ . _ . _ _ . _ _ _ _ _ . _ _ _ . . _ _ - - _ _ _ _ _ _ _ _ . - _ _ .

6t

! >

__ _ .,.4____ . _ . . ._._ _ - . _ . ; __ - . _ _ . - ~ _ _ _ - - _ _ _ _.- ....______ _ . _ - - _

_ , _ __ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ . _ ._.._ _ _ .._ ___ . _ _ . _ _

7 ._ __ _ _. _ . . _ _ _ . . _ . . _ _ _ _ .

. _ ._ __ __ ._ _

- - - - -- - .- .-. - - . _ - . .

.

.[ __ . . . . __

,

Ii

j. __ - - . , _ _ _ .

>'

1

ie

i,

-* ~ - - ,-.p ..4-

i1 1.1

s

. ..