UNCLASSIFIED

AD287174 Kap/iteuioacL

luf. Ute

ARMED SERVICES TECHNICAL INFORMATION AGENCY / ' I ARLINGTON HALL STATION f * ARLINGTON 12, VIRGINIA

UNCLASSIFIED J

f ~ -- .. I

I I

i

Best Available

Copy

NOTICE: When government or other drawings, speci- fications or other data are used for any purpose other than in connection with a definitely related goverrunent procurement operation, the U. £5. Government thereby incurs no responsibility, nor anv obligation whatsoever; and the fact that the Govern ment may have formulated, furnished, or in any way supplied the said drawings, specifications, or other data is not to be regarded by implication or other- wise as in any manner licensing the holder or any other person or corporation, or conveying any rights or permission to manufacture, use or sell any patented invention that may in any way be related thereto.

o

INORGANIC DIELECTRICS RESEARCH

PROGRESS REPORT NO. 1

Contract No. DA-36-039~3c-39lUi

November I, 1961 to February 1, 1962

Investigators (Full Time)

E. J. Smoke C. J. Phillips E. L, Kastenbein D- R. Ulrich R. C. Pitetti J. R. Smyth C. Breen

Signal Corps Research Program File No. 01007-PM-62-93-93

N. J. CERAMIC RESEARCH STATION Rutgers, The State University John H. Koenig, Director

Assistants (Part Time)

J. Thomson W. Hoagland J. Rubin J. Wasylyk J. Buckmelter

JISIA

«•*

Your comments, suggestions and criticisms on

the work reported would be appreciated. Longhand

notations on this sheet will suffice. Please

det^Sh and send to Dr. J. H. Kocnig, 6chool of

Ceramics, Rutgers, The State University, New

BrunswicP?, N.J.

No. o Copie

DISTRIBUTION LIST

CONTRACT NO. DA-36-039-SC-891U1 RUTGERS, THE STATE UNIVERSITY

Destination

, Spencer Laboratory, , Lord Trenton 1, N. J.

; Mr. Cedrick D.

J., ATTN• Mr. D.

Mr. George

Bell Telephone Laboratories, Murray Hill, N. J., ATTN: Dr, M. D. Rigterink *.J»IH U I ATTN

B. G. Corporation, 321 Broad Ave., Ridgefield, N. J., AT IN

General0Apptied Science Laboratories, 60 Hempstead Ave., ^mpsÄ, Long Island, New York ATTN: ^Iv Bartone American Lava Corp., Chattanooga, Tennessee, ATTN. M. J-

Mayfield, R. D. Dillender Raytheon Co., Microwave & Power Tube Div

Burlington, Mass., ATTN: Mr. Parker M Cook Ceramic Mfg. Co., 500 Prospect St., Coors Porcelain Co., Golden, Colo., ATTN

Brynner Frenchtown Porcelain Co., Frenchtown, N.

1 ulc s General Ceramics Corp., Keasbey, N. J., ATTN:

Ge^rafETIctric Co., Magnetic & Dielectric Materials, Room 11+, Building 3, Electronics Laboratory, Syracuse, N. Y. , ATTN: Dr. E. C. Henry ^ „,i

General Electric Co., Research Laboratory, The Knolls, Schenectadv 5, N. Y., ATTN: Dr. Louis Navias

Glad^ngfSSeaA I Co/, 2901 Lo^ Feliz Blvd.. Los Angeles 26,

aiä^JnroÜ'l^So^E^Kerre^fnue, Milwaukee. Wis.,

, Du^ToNLramlcPsacrSBox 587, Butler, Pa.. ATTN: Mr. John J- Duke „ . ...

. Illinois Electric Porcelain Co., Macomb, 111., Benedict C. Giltner

. Lapp Insulator Co., Inc.. LeRoy, N. Y., ATTN. Stettinius _ „,»,„„

. General Electric Co., P. O. Box 57, Baltimore,

Massichuseit^li". of Technology, Laboratory for Insulation ■ ^R'earch Cambridge 39, Mass.. ATTN: Dr. A; Von Hippel - Massachusetts Inst. of Technology, Ceramics Div., Cambridge

•jo Ma«« ATTN: Dr. W. D. Kingery . McSAef RJf'racIIry Porcelain Co., Beaver ^SL/*™^™1

* . North Carolina State College Dept of Engrg. Research

Raleigh, N. C., ATTN: Dr. W. C Bell & Dr. W. W. Kriegei

ATTN: Mr.

Mr. K. E.

Md . , ATTN:

I

-2-

DISTRIBUTION LIST (CONT.)

No. of Copies

Columbus,

Linden Lane, Wentvorth New Brunswick, N.

Destination

National Ceramic Co., Trenton, N. J. The Ohio State University, Dept. of Ceramic Engrg.,

Ohio, ATTN: Dr. Ralston Russell, Jr. National Advisory Committee for Aeronautics, Div. of Re-

search Information, 1312 H St., N.W., Washington 25, D. C, ATTN: Mr. Bertram A. Mulcahy

Porcelain Products, Inc., Findlay, Ohio Radio Corp. of America Laboratories, U2

Princeton, N. J-, ATTN: Mr. Chandler Library, Rutgers, The State University, ATTN: Mrs. Elizabeth Boyd

Mr. Frank J. Stevens, 23 Oakley Ave., Summit, New Jersey D. M. Steward Mfg. Co., Chattanooga 1, Tenn., ATTN: Mr.

Charles M. Grether Thor Ceramics, Inc., 225 Belleville Ave., Bloomfield, N. J. Univ. of Utah, Dept. of Ceramics, Salt Lake City, Utah Amphenol-Borg Electronics Corp., 1830 S- 5Uth Ave., Chicago,

111., ATTN: Mr. N. R. Kurtz Westinghouse Elec. Corp., 7325 Penn Ave., Pittsburgh Ö, Pa.,

ATTN: Mr. W. E. Blodgett Commanding Officer, U. S. Army Signal Research and Develop-

ment Laboratory, l^ort Monmouth, N. J., ATTN: SIGFM/EL-PEM Headquarters, W. A. D. C, Wright-Patterson Air Force Base,

Ohio, ATTN: WCLTLC (Lt. J. M. Kacicy, Jr.) Department of Defense, Office of the Assistant Secretary of

Defense, The Pentagon, Washington 25, D. C., ATTN: Dr. F. P. Hudd1e

Champion Spark Plug Company, Toledo 1, Ohio, ATTN: Robert J. Rea „ _,. _

AC Spark Plug Division, General Motors Corp., Flint d, Michigan, ATTN: Karl Schwartzwalder

Raytheon Manufacturing Co., Microwave & Power Tube Div., Spencer Laboratory, Burlington, Mass., ATTN: Parker M. Lord, Library

Oak Ridge National Laboratory, P. 0. Box Y, Oak Ridge, Tennessee, ATTN: L. M. Doney „..,*

Mycalex Corporation of America, 60 Clifton Boulevard, Clifton, N. J., ATTN: Dr. A. J. Monack

Electronic Mechanics New Jersey

Inc., 101 Clifton Boulevard, Clifton,

DISTRIBUTION LIST (CONT.)

No. of Copies

Street, White

Dr. W.

Bell,

D. C. ATTN:

Dr.

Pest inat ion

Eitel-McCullough, Incorporated, San Bruno, Cal Nuclear Development Corp. of America, 3 New

Plains, N. Y. , ATTN: Frank D. McDonough Carborundum Co., Niagara Falls, N. Y., ATTN:

Lambertson Philco Corporation, Research Division, Blue

Pennsylvania, ATTN: Mr. Robert B. Murphy National Science Foundation, Washington 25,

Dr. W. Grüner Speer Carbon Company, Niagara Falls, New York, ATTN:

F. Collins Corning Glass Works, Bradford, Pennsylvania, ATTN: ur.

Shaw , _ Electric Auto-Lite Company, Spark Plug & Ceramic Group,

Box 231, Fostoria, Ohio Star Porcelain Company, Box 1329, Trenton 7, New Jersey Allen Bradley Corporation, 136 W. Greenfield Avenue,

Milwaukee U, Wisconsin, ATTN: Dr. G. Eccnomos Lockheed Aircraft Corporation, 3251 Hanover Street, Palo

Alto, California, ATTN: Mr. W. A. Kozumplik E. I. duPont de Nemours & Company, Inc., Eastern Laboratory,

Gibbstown, New Jersey State University of New York College

University, Alfred, New York, ATTN; The Glidden Company, Pemco Division,

Baltimore, Maryland, ATTN: Dr. E.

of Ceramics, Alfred Dr. W. B. Crandall

5601 Eastern Ave., E. Mueller

TABLE OF CONTENTS

ABSTRACTS.....................................

Part I - Low Loss Microwave Ceramic Dielectrics, Part II - Transparent Polycrystalline Ceramics.. Part III - Hot Extrusion........................

PART I - LOW LOSS MICROWAVE CERAMIC DIELECTRICS...

Page

i

I i

ii

1

Introduction. 1 I. Hot-Pressing. ........!!!..'!! I! 2 II. Investigation of Crystalline Phases Present...., 9

A. Introduction. 9 B. Literature Research. 10 C. Experimental Work.......................... ll D. Results... .................. . . 16

III. Summary. l8 Future Work , iQ

PART Ix - TRANSPARENT POLYCRYSTALL INE CERAMICS,,. 20

Introduct ion 20 Experimental Work s................ , '.„'. 21 Results». .,,,.,, 23 Future Work ?A

PART III HOT EXTRUSION 28

Introduct ion , Equ i pment . , Experimental Work Discussion of Results. Conclusions Future Work

23 29 31 3? 33 33

Part I

LOW LOSS MICROWAVE CERAMIC DIELECTRICS

Abstract

Three compositions in the La203-Al203-Si02 system were hot-

pressed and the electrical properties measured at several

frequencies and temperatures. A study was made of the crystalline

phases appearing in a number of compositions within the system.

Two lanthanum silicate compounds, lanthanum aluminate and mull He

were found in the various compositions. Attempts to form each

lanthanum silicate compound by itself are being made in order to

evaluate their electrical properties. This information will be

used in attempting to tailor bodies to meet the specific require-

ments of this project.

Part II

TRANSPARENT POLYCRYSTALLINE CERAMICS

Abstract

Specimens of pure alumina were pressed and sintered using a

two-fire procedure. For the initial firing, one set of specimens

was fired in air and one set was fired in argon. The second

firing of both sets was performed in argon. Bulk densities of

the two sets of specimens were measured and compared. Specimens

which were initially fired in air were less dense than those

fired wholly in argon.

Part III

HOT EXTRUSION

Abstract

The hot extrusion apparatus is described and its operation

explained. The results of three extrusion attempts are

described.

-1-

Progress Report - Part. I

LOW LOSS MICROWAVE CERAMIC DIELECTRICS

Introduct ion

The object of this problem is to develop low loss insulation

with a dielectric constant of 15 and a power factor not to exceed

O.OOOlj. at microwave frequencies and at temperatures between -SO

and +300CF. The dielectric constant is above that of the

recognized range for insulation.

In earlier work, two lanthanum alumino-si 1icate compositions

exhibited these desired values as determined at 1 megacycle and

room temperature. The compositions were LAS-2 (1.96 weight

percent of AI2O3) and LAS-5 (4-76 weight percent of AI2O3)1. Both

of these compositions were fired conventionaL/.after an involved

preprocessing procedure. The processing of these compositions

consisted of drying the LagOz for one day prior to batching, wet

ball milling the compositions for one day, calcining the

compositions three times, screening the material between each

calcining operation, pressing the desired pieces, and then slow

firing the pieces to vitrification. This processing of the

compositions took about one week to complete.

1. Inorganic Dielectrics Researfch, Progress Report No. 1, March 1, 1960

o Rutgers, The State University Signal Corps Contract No. DA-36-039-sc-89111.1

Efforts were made to reduce the processing time by two

methods: by a high temperature prefiring, and by a single cal-

cination in place of the three calcinations used previously. Both

of these attempts resulted in materials exhibiting critically

short firing ranges, porous bodies and, consequently, unrepro-

ducible and unreliable results. For this reason, hot-pressing

the compositions was tried and resulted in the production of

dense bodies of approximately the desired properties within at

least a portion of the frequency and temperature range.

This report deals with the results obtained by hot-pressing

and with analysis of the crystalline phases formed in fired

pieces of a number of compositions in the ternary system.

I. Hot-Pressinq

Since firing of these bodies by conventional means was not

successful without calcining three times prior to the final

firing, it was decided to try fabrication by hot-pressing. This

technique produced bodies that had 0.002 moisture absorption and

electrical properties that were near the desired values.

Three bodies were hot-pressed: K15, K15+1 and KI5+3. The

compositions of these bodies in weight and mol percentage are

given in Table I.

Table I

K15

K15+1

K15+3

LaaOa wt. % MoJ_. %

61.56 26.lt

61.32 26.3

60.67 25.i|

SiOz Wt. % Mol. _#

33«1U 65-0

32.93 6U.5

32.68 61).. 1

Al203 wt. % Mol. %

5.30

5-75

6.65

8.6

9.2

10.5

The LazOs is extremely hygroscopic and was treated for 20

hours at l[|.00oF to drive out the water. The materials were then

weighed, dry-mixed for two hours, and wet ball-milled for twenty

hours. This material was then dried and passed through a 100

mesh screen.

Hot-pressing of the screened material was performed in

1" ID x 3" OD x 3" high graphite dies which had been lined with

0,005" sheet molybdenum to minimize carbon migration. The

dielectric constant of composition K15 decreased the greatest

amount with an increase in frequency between 8.52 and 9.6^ KMC.

Graphic presentations of the change of the dielectric constant

with frequency and temperature are given in Figures 1 and 2 on

pages [(. and 5 respectively. The graphs indicate that the diel-

ectric constant is not very temperature sensitive at these fre-

quencies, but that there are large changes in the dielectric con-

stant with changes in the frequency.

Table II

Body

K15

K15 + 1

K15+3

Bulk Density (g/cm3) Frequency (KMC) _°F

U.13

U.20

i+.U

8.52 9.6^ 8.52 9.61|

8.52 9.62+ 8.52 9.6i+

8.52 9.6U 8.52 9.6lj.

Temp. Diel sctric oF Constant

72 13 9 72 10 .15

300 li+ 0 300 10 35

7P- 12 2 72 11 1

300 12 US 300 11 3

72 10 8 72 9.U8

300 10 8^ 300 9 8)4

•»H

u.

5

H

8 u 1—1

o w u

i

u

u c

cr CD I-,

o o OC\J

o o OOJ OI>-

vr\ t*>

U, tt,

o r-

U

(-1 iS\

00

-r-

^UB^SUOQ otj-^oatata

CO

u CO I

'T- r^ O

i vO r^

-< a t- • CD O > 3

C *-> D o CD (-■

-!-> -kJ cd a +-> o

(D en

H (- o

■>U <n

(D (0 O) c

oc; to

•5-

U u y y y

+ + lAUMA

Vof

V

u as

M a,

5

H l/i

O a h 3 cn

OS H cj U

o It] u

u\

CO

0

J- CM ^t -^ vD UA ^D O

0s OD O ('■

0 • x a

o o

1 1 ( ^ ^1 -H

a-

bk o ""-' 0)

o u o p r\j

(0 t- ^H

(U -ct CL ^-H

B O- CJ CD (- I

U

I

m o

>, 1 -►-> < •tH a to (a • cv a > 2

■»H

g ■i-> s o OJ h O -tJ *J ^-^ CO C

*J o CO o

c^ H

0 D. u O « O

U1 u r~*

W 03 OTC

4-J cn 3 •r*

OS lO

• After hot-pressing the specimens were machined to fit the

waveguide in the dielectrometer (0.250" x O.396" x 0.890"). When

this size specimen was placed in the waveguide of the dielectro-

meter, it was found that the length dimension (.250") was too

small to determine tan accurately. Tan Is given by:

Tan S = f TCK^A)

where F(K, d. A) depends on dielectric constant of the specimen,

specimen length, and wavelength in the waveguide. However, the

effect of d, the sample length, on F(K, d.^ ) is minor compared

to its effect on £* . The 4 X, then, produced by the specimen,

is approximately proportional to d and tan $ . Therefore, for

low loss specimens, d must be made large enough to produce a X

which is much larger than the 4 X caused by waveguide losses.

The dielectrometer specimens were also used to obtain

approximate measurements of the dielectric constant and power

factor at 1 megacycle and room temperature using the Q-meter. The

size and shape of the samples did not lend themselves to direct

measurement by the Q-meter so the data obtained had to be

corrected for edge effects. There were no calculated corrections

for the size sample used so that the values given in Table III

are only close approximations. The plus sign given next to the

dielectric constant indicates the given value is somewhat smalle'r

than the actual value would be if the sample was a size where

-7-

there would be no "edge effects." The minus sign, next to the

power factor indicates that the actual value is smaller than the

value given.

Table III

P. F. Body

K15

K15+.1

K15+3

K

+16.0

+23.8

+15.1|

-.00108

-.00058

-. 00081+

BHiA-Den s i_ty (g/cma)

k'18

k-20

k- 11

If the values for the dielectric constant calculated at 1

megacycle and room temperature are considered with the higher

frequency values at 720F, then much more Information can be ob-

tained. With body K15, there is a sharp decrease of the

dielectric constant in the higher frequency regions. With body

K15+3, there is also a greater decrease at the higher frequency

but not quite as great as body K15. Composition K15+1 appears

to have a more linear decrease in the dielectric constant over

the entire frequency range investigated. A graph of this infor-

mation appears in Figure 3 on page 8. New specimens are being

hot-pressed for more precise electrical measurements.

It was apparent that a small amount of carbon or molybdenum

had minratpH intn -ill +1-,^ v,„i. _, . »taveu »ntg nn tne not-pressed specimens. in all probabil.

ity, this migration increased the power factor of these bodies.

An attempt will be made to hot-press these bodies in ceramic dies

to eliminate this contamination.

Figure 3

VARIANCE OF DIELECTRIC CONSTANT WITH FREQUENCY AT ROOM TEMPERATURE

Frequency (KMC)

X KIS o K15+1 • K15+3

3 8.52 9.61| 10

Rutgers, The State University- Signal Corps Contract No, DA-36-039-sc-391,'1

The application of the hot-pressing technique has produced

results which approach the requirements of the project. Further-

more, if the results of the dielectric properties at 1 megacycle

can be held reliable, then we can see that there is an increase

in the dielectric constant with increasing alumina percentage up

to 9.2 mol percent and then there is a decrease In the dielectric

constant with greater percentages. With the power factor the

reverse is true. The power factor is high at 8.6 mol percent of

alumina, low at 9.2 mol percent and then a slight increase at

10.5 mol percent. These results can be made reliable only through

further experimentation with these compositions.

Ii. Investigation of Crystalline Phases Present

A. Introduction

Since compositions within the LaaOs-AlzC^-SIO2 system

exhibit promise of producing bodies satisfying the technical

requirements desired in this phase of work, a study of the

crystalline phases occurring was deemed desirable. If these

crystalline phases can be identified, attempts can be made to

produce bodies containing a maximum quantity of each given phase.

Evaluation of the properties of these phases should be extremely

fui »r, tallorina HOHIRS to meet the technical requirements of

the contract.

-10-

B= Literature Research

The preliminary research done for this investigation was the

accumulation of information on the various binary compounds. The

binary compound of alumina and silica is mullite (3Al203°2Si02).

This is a well explored binary and thus there is sufficient in-

formation dealing with the identification of the compound.

Two compounds exist in the lanthanum oxide-alumina system:

LazOs-AlzOs and LaaOa-12AI2O3. The 1:1 compound has a perovskite-

type structure and is stable from room temperature to its melting

point at 2130oC. The X-ray powder diffraction pattern has been

given in earlier reports. The 1:12 compound is similar in

structure to CaO-6Al203 and is stable to 1920oC, where it melts

p congruently .

A recent thesis by Warshaw-^ has shown that there are two

compounds in the lanthana silicate binary: La203-Si02 and

La203°2Si02• The compound La203=2Si02 is typical of the pyro-

silicates of the larger rare earths, comprising La, Ce, Pr, Nd,

Sm and Eu. There is a high and low temperature form of La203o2Si02

{ß and oc. respectively). The orthos i 1 icatc (La203'Si02) melts

congruently at 1980oC while the 1:2 compounds melts incongruently

at 1760oC forming the 1:1 compound and a liquid. As in the case

~.p T_ r\ .OCJA- i^^+K^^^im ^r^ + Vi^ir-ili/~'3+o oloo Vipc ti-fn no! ■vrmn r nh ^

2. Warshaw, Israel, Phase Equilibrium and Crystal Chemical Rela- tions in Rare Earth Systems, Department of Chemistry, Penn State U.

3« ibid.

n

■ 11.

with the transition temperature being 12750C.

The phase equilibrium diagram given by Warshaw, Figure I4. on

page 12, bears little resemblence to that published by Toropov

and Bondar^ which was given in earlier reports. Instead of

finding two compounds, they found only a 2:3 compound. This

difference would completely alter the phase equilibrium relation-

ships in the remainder of the system. Only the compounds reported

by Warshaw have been identified in the present study.

C. Experimental Work

A number of compositions were prepared to investigate the

compound formation occurring in the ternary system. The lanthanum

oxide was dried at 1^00oF for 20 hours before the oatches were

weighed. The compositions were then dry bail-milled for seven

hours. One inch discs were pressed and fired with no soak.

Vitrification and compound formation was obtained using draw

trials method. The fired specimens were crushed to pass through

a 200 mesh screen and then X-rayed.

The compositions used in this study are given in Table IV on

page 13 and their locations are shown in the two ternary diagrams.

Figures 5 and 6 on pages li^ and 15. Indicated also in these

i^.. Warshaw, Israel.

Q

0) la

Y \

\

\ s.

v O P

(0 "H

\ V

\ S,

X \

O CO

>

/

/

/

/ ^ Of

Cr

C.i

m

—i

+

\ r.

0J

,-J

O

C\J

—p- o o CO

-I—' o o

CM

I

-t-J I

Si o

ir

CM

TJ

U H

rj

•T-l

H

o N td

S

. o OO

o

- o ■^3

o

to o

o . o -^

CM

«0

(0 O HJ

o CM

o o o

o

o U0

{0o) aan^BJadmai

CO c

u 01 I

O-

O I

vO CO

>»Ji *» 3 -■ Q (/) u • Q> O > s C *->

^^ co t- -P *J cd C *-> O (/) O

JC ft

O •vU

a) to en c +J O) 3 -<

DC LO

-13-

Tabl e IV

Compound wt. La203 % Mol. % Wt.

Si 02 % Mol. % Wt.

Al. %

'.03 Mol. %

1 70 35.6 2k 56.0 6 8.U 2 y. 2i|.2 28 55.6 17 20.2

3 hb 17.8 30 5k-9 2 3 27.3

h 33. 3 11.3 33- 3 55-U 33. 3 33-3

5 75 U6.2 10 28.0 13 25.8

6 60 29.8 15 35.1 25 35.1

7 k5 18.5 23 k7.k 30 314.2

3 50 18.U U5 76.5 5 5.1

8A 50 16.1 ho 69.9 10 ll+.O

9 U5 16.5 ko 68.1 13 15.U

10 40 11|.6 35 59.5 23 25.8

.Ik-

Figure 5

LOCATIONS OF COMPOSITIONS IN WEIGHT PERCENTAGES

SiOz

IB.ZO3 AI2O3

W.eight Percent

Rutgers, The State University Signal Corps Contract No, DA-36-039-sc-89lUl

n

-15-

Figure 6

LOCATIONS OF COMPOSITIONS IN MOL PERCENTAGES

SiOz

/ \

Laz03°2Si02

LazOs-SiOz 50

LazOs LazOs'AlzOs

Mol Percent

LazOa«12AI2O3 AI2O3

Rutgers, The State University Signal Corps Contract No. DA-36-039-sc-891l4.1

n

ü

-16-

diagrams are the locations of the compositions that had been hot-

pressed. Comparing the two diagrams (one in weight percent and

the other in mol percent), it can be seen that the compositions

fall in the high lanthana area when using a weight percent diagram

but they fall in the low lanthana part of the ternary when using

mol percent» It has been postulated that if a ternary compound

would exist it would be in the low mole percent lanthana region .

The results of the X-ray analysis of this work indicated that

there were no ternary compounds in the area evaluated. Two

lanthanum silicate compounds, lanthanum aluminate, mullite and

the three starting oxides in various forms were found in different

compositions. The compounds found in the different compositions

are shown in Table V on page 17-

D. Results

As can be seen from Table V, there appears to be no ternary

compound. Furthermore, La203»2Si02 appears to have formed in

most of the compositions, but LazO^'SiOz appeared in only a small

number of compositions.

It will be noted that all three of the compositions (Kl5,

Kl5+lj and K15+3) on which electrical properties have been

obtained contain mullite, La203°2Si02 and l^Oa'A^Os. The

electrical properties of all of these phases except La2U3"2Si>-'2

are known. Therefore, it is important to attempt to produce

5. Warshaw, Israel, ibid.

-17-

Table V

Crystalline Phases

LazOs" ILazOa- ILazOa" ILazOa- LazOs' Body MuH i_te AlzO* 12AI2O3 2Si02 2SIO2

1 X

2 X

3 X

k X

5

6

7 X

8 X

8A X

9

10

K15 X

K15 + 1 X

K15+3 X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

Si02 La2 Ol AlaOj

X X

X X

X X

x(?) X

X

X

X X

Crist.

X

X

X

s

X

K

X = indicates crystalline phase present

-18-

the latter compound and determine its electrical properties. If

the properties of both the 1:1 and 1:2 lanthana-si1ica compounds

can be determined, this information should be most helpful in

preparing bodies to meet the desired technical requirements.

HI. Summary

The dielectric properties of Kl5> K15+l> and K15+3 prepared

by hot-pressing at least approach the desired values. It appears

that these compositions are only slightly temperature sensitive

at the higher frequencies but are quite frequency sensitive. The

highest dielectric constant recorded was found using 9.2 mol

percent of AI2O3 and the lowest tan was also recorded at this

percentage. Any increase in the alumina content decreased the

dielectric constant and increased the tan

The compositional study of the ternary system revealed that

La203°2Si02 was the dominant compound formed in the areas studied.

Also formed within the ternary were Laz^s'SiOz, mullite, LaAlOs.

All of the compositions evaluated exhibit short firing ranges

which indicated that vitrified bodies will be extremely difficult

to form using the conventional firing techniques.

Future Work

1. Hot-press new specimens K15, K15+1 and K15+3 to evaluate

tan at microwave frequency.

2. Hot-press LazOs-SiOa and LazOs'ZSiOz to determine the

dielectric properties.

-19-

3- Hot-press ternary compositions and evaluate their

dielectric properties.

k- Use information obtained in 2" above to prepare compo-

sitions believed to have the most promise of fulfilling the

technical requirements of this project.

o

"•

-20-

Progress Report - Part II

TRANSPARENT POLYCRYSTALLINE CERAMICS

Introduction

The objective of this program is to produce a transparent

or highly translucent, polycrystal1ine5 single-oxide dielectric

of very high density and essentially void free. In addition to

transparency, the material should also possess the following

properties: a low dielectric constant, low power factor,

dielectric strength of 1+00 volts/mil for l/k" specimens, and high

mechanical strength.

As has been stated in previous reports, the preliminary work

in this investigation has been on alumina. Alumina was chosen for

this work since transparent, polycrystal1ine specimens of it have

been produced by other investigators1. Furthermore, a considerable

amount of information on the sintering of alumina is found in

the literature. Therefore, in starting with a study of alumina,

a material is being used which is not only known to be capable of

producing transparent, polycrystal1ine specimens, but about which

a maximum number of the conditions required for the production of

. . „„^ vr^.m rin^e examination of the transparent specimens ai^ nnuv—

I. Aus tralian Patent, "Improvements in or Relating to Transparent Alumina."

Rutgers, The State University Signal Corps Contract No. DA-36-039-sc-891^1

(>

-21-

parameters involved in production of transparent, polycrystal1ine

alumina should be extremely valuable in future work on other

oxides. The use of Isotropie crystalline materials should allow

production of specimens having a higher degree of transparency

than alumina since alumina is anisotropic.

Two methods of maturation arc planned: sintering and hot-

pressing. However, as of this date, only the former technique

has been studied. The transparent alumina previously mentioned

was fabricated by sintering. The aluminum oxide used was of high

purity, having only 0.3Ä of magnesia. Alpha alumina was the only

crystalline phase present. The crystal size after sintering ranged

from 15 to 65 microns.

In the present investigation, the procedure outlined in the

Australian patent was followed closely, with certain modifications.

One of these was that a different gaseous environment was used.

Experimental Work

The apparatus used for sintering has been described in the

previous report. A diagram of the sintering furnace appeared in

Figure 1 on page 38 of that report. It is essentially a water-

cooled, induction furnace, containing an induction coil, and a

molybdenum susceptor, surrounded by a quartz tube and Insulated

with zirconia sand as loose fill. The sample rests on a small

piece of molybdenum on a zirconia setter.

■ 22-

r The procedure as described in the above-mentioned patent

calls for two separate firings: First, a firing at a temperature

from 1650° to 1750oC for a period of from 50 to 300 minutes, to

eliminate most of the voids and gas containing pores; secondly

a firing at a temperature of from 1850° to 1950oC, for a period

of not less than fifteen minutes. The first firing may be done

in any atmosphere, while the body is always subjected to an

atmosphere of hydrogen in the second firing.

In the present program, argon was substituted for the hydro-

gen. There are several reasons why this was done. First, it was

felt that hydrogen could not be safely used with the present

set-up. Further, a literature review failed to reveal any

previous work utilizing an argon atmosphere. Hence, much new

information could be obtained in this area. Finally, the argon

serves to protect the molybdenum susceptor from oxidation.

Samples of pure alumina (99.9+%) in a colloidal form having

a maximum particle size of 0.3 microns, were pressed into compacts

l/2 inch in diameter and l/8 inch thick. Half were fired in a

gas-fired kiln in a slightly oxidizing atmosphere for one hour at

1700oC. The other half were fired in argon at the same temperature

and for the same length of time. Ail samples were then subjected

to a second firing to 19000C in an atmosphere of argon and held

for periods of 15, 60, and 300 minutes respectively.

o

-23-

Results

Densities, obtained with the samples of fired alumina are

listed in Table I on page 2^. The results are shown graphically

in Figure 1= As can be seen, the densities obtained were quite

low, and the specimens were not transparent.

The samples were all coated with a thin silvery metallic

layer, probably from condensation of vaporized molybdenum. How-

ever, the films were very thin and were easily removed by

grinding. The coatings did not appear to have diffused into the

specimens. Hence, the lack of transparency was undoubtedly due,

not to the contamination of the metal, but to the large percentage

of voids present in the fired piece. An examination of the table

reveals that the most dense sample obtained still contained voids

to the extent of nearly $%, In order to obtain transparency, less

than 0,$% of voids is believed to be necessary.

It is postulated that the presence of these vodds is due to

uninhibited grain growth. If grain growth is too rapid, gaseous

inclusions become entrapped within the crystals rather than at

crystal boundaries. Pores within individual crystals have been

found to very difficult to remove. Grain growth must be slowed

down enough to permit diffusion of interstitial pores from the

body. In order to inhibit this growth and to allow it to proceed

at a rate slow enough to permit elimination of the voids, a small

amount of some other material must be added. The next series of

samples is therefore being made up with 0,7% of magnesia added

to the alumina.

A.

-2h-

Tablc I

DENSITIES AND MOISTURE ABSORPTIONS OF SINTERED ALUMINA

Samples initially fired in air (1700oC)

Add itional Fire, 1900oC

Bulk Densi Lz

% True Densi ty

Moi sture Absorption

0 min. 3-67 92.0 0.88 15 " 3-71 93.0 0.11 60 " 3.75 9U.0 0,02

300 " 3-76 9U.5 0,02

B. Samples initially fired In argon (1700oC)

Additional Fire, 1900oC

Bulk Densi ty

% True Density

Moisture Absorption

0 min. 3.77 94-5 0,39 15 " 3.82 95.5 0,06 60 " 3.82 95.5 0,02

300 " 3.82 95.5 0.02

(Theoretical density of oC-alumina = 3.987)

n

-25-

o

l-

Cu

C o en

<

u

3 r~* a

H ta 2 u i—i 3 1,0 O)

■ ^( d. k. O

SO

3

o o

c

<l

CT- CO

' 00

u i c

o^

i II

o C sO

>»Ji

CJ -— MH

E P

! O

H O .—i n)

I

CD O > z

1^ C - 0) 8

0) u

■

4-i o

</> o u ■J (0

>>

(0 C +J o in o

CO

C o

C

d) (fi i NH n Ä a +J

o ITN X5 cd o f—t

^ U »H a V n) t^ OJ c | o *-> O) w pa co

..o 0> CM

(T- O

A^TSU3Q s^-1! ^

-26-

It is also planned to fire in an atmosphere containing

aluminum vapor. The purpose of the aluminum is to create

structural defects. The increase of vacancies in the lattice

should increase the sintering rate by enhancing the diffusion

of pores through the structure. It is hoped to thus be able to

eliminate the voids and produce theoretically dense bodies.

Table I reveals that those samples fired initially in argon,

have higher densities than those which were given an initial

fire in air. The significance of this is open to question at this

time since only a small number of specimens have been tested.

Polished sections of the specimens are now being made for micro-

scopic examination. This will permit determination of grain size

and amount, size, and location of pores which will be necessary

in evaluating the effects of different sintering conditions.

{

Future Work

1. Polished sections will be made of sintered specimens.

2. Specimens will be pressed with alumina containing a

small amount of MgO to inhibit grain growth, and will be fired in

the manner described above.

3- The furnace set-up will be altered slightly to permit

firing in an atmosphere containing aluminum metal vapor.

1|. A furnace is currently being built which will permit

firing in a vacuum or in various atmospheres. As soon as it is

finished, alumina will be subjected to firing in an atmosphere

of hydrogen.

o

-27-

5- Samples have previously been prepared by hot-pressing.

However, they were contaminated from the graphite die. Future

samples will be prepared by hot-pressing in ceramic dies.

6. When work is completed on alumina, investigation of

cubic phases, particularly magnesia, will begin. The literature

reveals that MgO can be fabricated to theoretical density by

hot pressing. However, no mention is made of transparency.

-28-

Progrcss Report - Part III

HOT EXTRUSION

Introduction

Hot extrusion of certain ceramic materials is foreseen as a

method of producing materials having the desirable properties of

hot pressed materials but being a more economical production

method. For instance, earlier work has shovm that barium

titanate can be hot extruded and that the extruded material

possesses qualities similar to that produced by hot pressing. The

previous work was only preliminary and, although it did provide

encouragement that hot extrusion of ceramic materials is feasible,

it also demonstrated that there are a number of problems which

will have to be solved before truly successful hot extrusion of

these materials can be accomplished. Briefly the problems are:

1. Lubrication of the material being extruded so that it

will flow readily through the die.

2. Maintaining the material being extruded at the proper

temperature during the time necessary for extrusion.

3. Controlling the cooling rate of the extruded material

to eliminate failure due to thermal shock.

The extrusion of materials such as glasses is relatively

simple since these materials exhibit a gradual softening over

Rutgers, The State University Signal Corps Contract No. DA-36-039-sc-891l4.1

-29-

a reasonably large temperature range and essentially true viscous

flow. However, crystalline ceramic materials have distinct

melting points, going from a solid material exhibiting only a

small degree of flow to rather fluid melts, over a very narrow

temperature range. Although it is known to be possible to slowly

extrude crystalline ceramics at some temperature below their

melting point, rapid extrusion at such temperatures is not

believed to be possible without lubrication. One means of pro-

viding lubrication without dilution of the desired crystalline

phase Is by coextrusion of a metal or glass cladding around the

crystalline material. The coextrusion of a metal was used in

the previous work on barium titanate and has also been success-

fully used by Nuclear Metals, Inc. for the hot extrusion of UO2 .

The coextrusion of metal cladding around the ceramic is the

method of lubrication being used in the present study.

Equipment

An exploded view of the extrusion assembly being used pre-

sently is shown in Figure 1. The press used is an electrically

operated, hydraulic, 150 ton, constant pressur« Dake Press, Model

No. 21-150. The ram speed of this press Is approximately 1?

Inches per minute.

J, G. Hunt and P. Loewenstein, "The Fabrication of Clad Massive UO2 Fuel Elements by Coextrusion," Fourth Quarterly Report, Contract No, AT(30-1)-1565, Nuclear Metals, Inc., Concord, Mass., June 6, I960.

-30-

Figure 1 Exploded Cross Section of Hot Extrusion Assembly

m p —h— i ^ ̂

////. ^ ̂ ^ 2 ̂

V///, '///A ^\\ s\\\< 1 1

Extrusion Ram Heat treated l+l^O steel , Rockwell Cl+O-li^

Coextrusion Can O.D. approximately 1.70". depends on expansion of metal and extrusion temperature

Extrusion Die

1-3/11" O.D, Heat treated 'l-lijO steel 5 Rockwell Chfl-hS

Extrusion Barrel 6-1/2" O.D., 1-3A" I.D. Heat treated Ij-lq-O steel , Rockwell Cl).0-I|.5

Base Plate 2k" x 18"

Rutgers, The State UAiversity Q„II i Siqnal Corps Contract No. DA-36-039-sc-89 Ul 1

-31-

In use, the extrusion barrel with the die in place rests

upon the base plate. It is positioned on the base plate directly

under the ram and over a hole to allow the extruded rod to pass

through the plate. The positioning of the barrel is accomplished

simply by pushing the barrel firmly against three pins set

tangentially along the circumference of a circle having the same

diameter as the barrel. The angular distance between each pin

being approximately 50°'

The extrusion ram is placed on top of the extrusion barrel

and off to the side enough so that it is clear of the inside

diameter of the barrel. The press ram is then run down until it

is only a fraction of an inch above the top of the extrusion ram.

The coextrusion can and the material to be extruded are

heated, either separately or together depending upon temperature

requirements, and arc then quickly placed in the extrusion barrel.

Experimental Work

The work performed during this period was concerned mainly

with testing the operation of the hot extrusion apparatus. The

proper operation is somewhat complex and depends largely upon

timing and teamwork of a group of three or four men. This is

necessary to accomplish the extrusion as rapidly as possible

before too much heat is lost to the cold extrusion barrel.

For this preliminary work, it was decided to use a material

that had relatively well known flow properties and which would

extrude at a lower temperature than barium titanate.

■32-

During this period three extrusion attempts were made. The

ceramic material used was a proprietary, glass containing compo-

sition used in producing transfer molded parts. This material

was chosen since the temperatures at which it can be transfer

molded are known.

The mix was hand tamped into a coextrusion can of 1018 steel

and placed in an electric furnace to heat. The die and barrel of

the extruder were lubricated with a mix of graphite and stearic

acid in CCI4. After soaking for about 10 minutes at the desired

temperature^ the coextrusion can was removed from the furnace and

quickly placed in the extrusion barrel. The extrusion ram was

placed in the barrel and the full extrusion pressure applied as

rapidly as possible. The time between removal of the can from the

furnace and application of full pressure was approximately 5-7

seconds. Neither the die nor the extrusion barrel were preheated.

The temperatures to which the material and coextrusion can were

heated were ii+OO, 1^00 and 1700oF.

Discussion of Resu1ts

Extrusion did not occur on any of the three extrusion

attempts. In fact there was very little difference in the appear-

ance of the three slugs after the attempted extrusion. In each

case, the coextrusion can did doform sufficiently to take the

shape of the die and barrel and each had a slight "nose" where

the can had been foreed a short distance into the straight portion

-33-

of the die. The ceramic itself did not appear to have been

heated sufficiently prior to the attempted extrusion since all

pieces were porous although the porosity of the piece preheated

to 1700oF was quite low» It appears that a ten minute soak at

peak temperature is not sufficient for thoroughly heating the

ceramic powder. This is probably due to its low thermal

conduct ivi ty.

Conclusions

The preliminary hot extrusion attempts have shown, at least

tentatively, that:

1. Using the present design of equipment, assembly of the

necessary parts and application of full extrusion pressure can

be obtained within 5 to 7 seconds after removal of the pieces

from the furnace.

2. A soaking time of ten minutes is apparently not long

enough to allow the ceramic extrusion blank to reach equilibrium.

3» The use of a coextrttslon can having a sof^ning

temperature than 1018 steel is desirable for use in extruding

the composition employed in the present study.

Future Work

A number of extrusion experiments will be performed using

1100-F aluminum. Trial runs will be made to determine the

temperature and pressures needed for extrusion of solid slugs of

■3h-

this aluminum. This will help in determining the parameters

necessary for coextrusion of the aluminum with the ceramic

material. According to the literature the aluminum should ex-

trude between 750 and 900oF..2 It Is felt that the present

extrusion apparatus will probably work most favorably if the co-

extrusion can is made of a metal which will extrude at a lower

temperature than the ceramic since the coextrusion can is in

direct contact with the cold barrel and die of the extruder dur-

ing the extrusion period. If the ceramic is heated to a higher

temperature than the coextrusion can it will act as a source of

supply of heat to the can during the extrusion and, it is hoped,

that this will enhance extrusion. If this method produces

successful extrusions, a similar approach will be tried in the

hot extrusion of barium titanate, choosing a coextrusion metal

which has a higher extrusion temperature than aluminum. If it

becomes apparent that the cooling of the coextrusion can by the

room temperature extrusion barrel and die is greatly hampering

successful extrusion, preheating of the barrel and die will be

tried. In this case, care must be taken not to get the barrel and

die hot enough to cause them to lose their temper.

2. Herbert R. Simonds, Archie J. Weith and William Schack, "Extrusion of Plastics, Rubber and Metals," Reinhold Publish- ing Corp., New York City, 1952.

•35-

Another method which may be used if the previous methods are

not satisfactory will require a slight modification of the present

assembly. The method consist essentially of having a liner made

which can be preheated and placed Inside the extrusion barrel

just prior to the extrusion attempt. The coextrusion can will fit

inside of this liner. The liner should be made of a material

which will not flow at the extrusion temperature. It will serve

as a buffer between the cold die and the coextrusion can thus

keeping the temperature of the can from dropping so rapidly dur-

ing the time required for extrusion.

I

r

C)

-H O U

.a .«i

u t« •-< u +J > ■< O I -- C W 0) ( ^ A +J c >, i Qi o u 3 t- " - -< M <u tr u c - Q -. ,« 6) O -

«a; a» u o —i i

rt«+JoJ<nOi-,- t, (0 u O D. (0 ' 6) .J -D 6) E ^ ' u c > >>o -^ ;

« « w -o u c i

4J TJ

•a: 9 c O -H a e

0) O V

■ i i a) .

a. >, [0 O ra

i —■ < « —<

y§^

; es t i u o

S C « rt ^ -»J O o« « «a; S t- wo c -t-» U M U ft) -« -, c a t- * TJ •3 O I 3 w CO ft)

«+J to 0) C -1

ui -»J o (0 t- •« CO w -H x » s t- »; i O (fl g +J (0 > - —JO« rt 6) fti i

a. u w t. D, t- ^ g ft) cu ft) a. i O O 3 •- Q. (0 «- ; j u +J e «i

E tn 0) "> ^ ' tn ft) ft) ft) 4-> ft] ^ o) ft) +J a w » -o t- w o -o to 2

" §0

m C C

ss.

C t- G) (Ö O t-

-SKS 1 (/i <r - ^

Cij X ^ -) > < O u W I. H -4 tK • •- ^ p oo a; w IO s:

- 3 i-i ft) H ( </) L. er; Ä CC v omH ope E ^ U] i/) (JJ

<üI■-'Ct/)*>^a(l, ^ -nft-^ft)^ U+J-O

o —' -^ 0) O '^ O -H rfwuft)-^" a.e U | IM j" *J • G l-

GOJ=— tny^o-'J rtM4jroiMOC— o i_w t-oa.rf—'(u-a

u c > ^o** 2 a o ft)(ti6>-dUCSU*J

a jz v> P to « >4J-d +JQ)—'-a-Ht-

o« E4Jn]>'-'(0n) _joft> rtft)Q)rt<-<E

D, u tf) t- Q. t- 3Gft)ft)ö)0. RJ=:"

j ja-—

■H o cr o +JCO

i ö) OJ Hi««1

t-G)ft)ft)*JW ^- — <j)ft)*>D. win-t->E ■ot-tfl o-utö^Ct-ö)

•z> ** i a. c

■ « M -. -a c I C S t- C -H

, E TJ "^ *" -^ l

. u t- —■ — m ( i ft) ft) ra (0 xi f i a*J — c - | •» C ♦» Ö a • « | -< -H »H U C '

ft) <M •« 3 •

^■a M ft) o -. ft) t, c

j E C W t- « CO ) S ta £i~* Ji

J ft) T3 -■ ) U W (- f)

(ft O (0 • *o X ft) W U, S C 03

C ft) o G 1 « t. wot, * ^ Q. • C t- O : ^ ft) ft) to <-. ) « ft) t- E t-

+J I. p _ C «D ., to a) -a u -• ~ j >, 5 ft) w

U 3 3 6) CT — Oft)

) O E O <M w j a, 3 w *J S ^ C 4J M 6) •■ ,*JK)— 6)CO<ft-*

3 ft) ft) I x: < r u t~ o d) ** +> \

(0 p S c « o ^ Q. C*-> O O XI ■<

o o q t-4-> jr to ♦J * i

«P *: x u < *- 4J U] ft) - ft) P. < n . M +J o :

•o ") o ' o ft) c x " -t 5 i- w •* I

■•-> (0 T) i- •« < D. ft)

ft) E M 13 T) 1 .c o tl C C < ■»-) u o) D (0 i

•-' J

O C ft) • ft) ,

trt o) « -H i ft) -a 3 n t-

. ;■! : w

— ■ I ■ 3 ■ C « 3 C 3 - *, OD-UO -*d'-Ä ^"OcOft)'

*J .-.(u t-<M3C--t-l0—l njwt-ot-—'-* mftjm-^D.« Gx:ft)cu6)totOTJ fi -H <w-a— < t, *J -a C^+J •-. c « +J Ao a.-« o c<ftc;*Jdo «u-" ora x i

■) c 6) c ^Tc: ft) [ i y) C <u o D.^ o o

i -a ft) t- m +J w o J • C'Wuiyto'M« s-^i

-. x: o o c t- ■<" MR r .fS2l

Ij ft) • a 0) *-> o

^ O ft) x -^ -. u {-• XI ft)

n p O v •- w 6) t, w o ra ■ "a

. .-i a « to [i, 3 c « WT) Cft" OE uw'v (uot/i-iu tncnt-otuc^to .MJ3«— D-'Ct-O^t-O) -t-1

*J x:—■ ft>ft)(a*M4->(0'Oi--< (-,(-.+J(0ft)t-E t- o. ft) too 4Jt.3-Hd'i)Q)Slfl'a'0 D.« (M W ft) -a i> -H o-x; o " c c

i o ^^ u (x-a - i ö>

is.' : O C '

wnTlft) ^H(-+JU)iy—(Hi ui tJ0)I-.-t-)OJ'^0)tOtO+Jt-t- I ^«.wG tMtftS *• P "2 S ' u3pü«WI xitotnioo- ft) trt-t-otu+J+JdtO oi ^-.ftJWnJp^CWOftJOJCt- (i)xi t- a. D.-*-1 o tftxiTj E-H to-

xtr 4

fti ft) Oft) w

• sD V> u g n «) —. 3 - to « u

■ ^>— u -. eo co « a >, ft»

w rc »-H i ÜAi t

cr u c — < 3 c -. « « v. ft) O -H o --«

^ i "^ jr -^ • e *- x ^ •-' > *>

,-TI Go j:

u (0

> >, O *» 3 (0 O UJ ä (/) -i |r» o ft> -a u c e o *J

«x: in p « —

g8«ug n) ft) ■»J w x: c -< a)

CO ",

to u a: x n: vo u « t- u • -a ra c C

ft) 3 t- « d c "O IXZ UJ • ^ m -H ^ E *> M > -^ ffl n

>• -a; t- fc-l- (j w O ra S t2 0) ft) tt) o. E x: cr.

— a, ra t. 9 o c *> S « c « —

_j a; < oi^j ä-^ OP ^ — < «^ 0> cc CE t»--^

(U *J ft) f A ft) ft>4-<

»- c .c jr « o (- u z s: o-t ocr ^5 5 ■-< äi '-0 *JCO

I

r

ü] ^ to -> > O

H ^ n: • • —

l..(- u to o u S ^ -J a cn-J Ä —

«:■■ s ̂ £3 yg H 4JCO

^ A j 0)0 u

—i ED ft)

< ft» G O .C ra « +- t- «

> +J ■

S c

rJ O ft) P. t-. I | a s <

o o ? ■

g U 0) 0) ft) ft» -t-J •a t- c^ c £) >>

1 —* 3 - in -rt i- D M « W +J -^ -. nj m w D. >i ft) j *J > *-> 6 £1 x: - Irt <y o] 0) *J 3 >, O +) T) 3 t. in -H +J C « TU C — < 3 C i) o -* o —" ^ ft) ^ to a e

±i -x E o o ft»

- o G. S — ft» -a > E .c --■*-» ■ >> o *J r « o > -a o c E u -»J

o ft) c -^ •-; J= o t- ■-■ ro £3 -u CD D 3 t- ft) ^f c "a 5 +J ro > ^i cd tu

rt w o n) ^ E o t a, t- u ft) D. S x: o.

a 3 6.4*

e c « ro o t-

i

—, o ü ra .c *J — t.

« o u

: p t- c - S D — .■. i- .-.: r -O «„ « —i (n ■•ft) t. fc. 3 C j t. ^. -. cQ » i) 0) « ffl T) ß X-J — c: -♦ ■» C +1 C O ■ O

— — -* U C « l in C ft) O P- j —. -o i" O)« H -O « t- 5 C « '- »> ffl '"

^•o ra fti ft) .-, 6» U C l-

rt -H a« — 4-4-D.-X «

u u 4-> ra 6) o *-> a— Vi to o -* o >,

Q.+J (0 U

— O C — n «-> ft) o u E11-

u 3 ? » a) 6-1-

—i ft) ft) «9

10 Cl t. P i. • c *- o i

ft) ft) n) (M *J ) u E c- - 3 -H C «> « j -o o — Q-x: S ft» ft» -»J

u Q.'O M ra o «o ft» M v. c (-, ^ 5 o ^ o ^- w io 3 ft) 4^ ft) ^. l. _> : ft) M i

« —• ft) ra w <-> i <« M S — :

4» i x: « i i. O « ♦> *» c i

— Oft)' XJ "O |

o c L+J x: ««-• « C F- x: »< ^ o

«j u) f —■

. ft) *J O 3 TJ V) O «- 6) CX «o ^ t. ft) *> X M TJ t.« « a 0) e "» TJ -a « o w c c • U ft>3 « t-

T) Xi ** SO • ft'

t. >..xi ^ ; ft» _H —, o T)

TP ? C t- «

3 S C i

- »

ft) <- -I <M CD • C O < O C CD O ft) I

p ■« C O u- ■ (0 10« -O C 6» <■

* (o f- n -a «H « - ) -* ft) H« 3 C jftjuu»- — — en« 3x:o)<ywrara-o E , 4j -a c+) « c ** ) C " C -tJ c o • U H*J^) -.«-lOCft» ; ft» v -o c fti o o. ,ft) u w-i-otoO)(n

G . ~. T3 ft) u. i GECWuftJrtV. <06ira3x: — x: o

« — D. ra

— ft« X c C üit ft» E

o c t- ♦* -c ra-j ra c F- j.- x i. o

i] ft» —

a -H -. a

- ^ ra CD (- E

O G CJ) u p ' i- O >:

. C « ft» (^ to a -a u — D.X: o >-, S « « +J

u v c -a v .woraoMft»«'«

+J >> C l- (- 3 O

woe o VH to to E Q, 3 tu 4J ft»

i o « t. +J to d) --

3 <o -rj 73 -ft) S n e e i j 0»^) ra t.

— x:

I C ft) . (U ra t- ^xi •M •

I ft) —. «H o -o 1 -D 3 C U 6) i ft) O U ^ -O

p w ft) i x: io ( trt-uooj+J+JCi

'ftjro3?cft»Oft)i it,o, a-tJOtoxiTii

V~ 4r

I I