Expanded Vermiculite as Loose-Fill, Thermal Insulation for Dwellings

Publisher’s version / Version de l'éditeur:

Technical Note (National Research Council of Canada. Division of Building Research), 1952-02-21

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

https://nrc-publications.canada.ca/eng/copyright

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la

première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à [email protected].

Questions? Contact the NRC Publications Archive team at

[email protected]. If you wish to email the authors directly, please see the first page of the publication for their contact information.

NRC Publications Archive

Archives des publications du CNRC

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

https://doi.org/10.4224/20359194

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Expanded Vermiculite as Loose-Fill, Thermal Insulation for Dwellings Hutcheon, N. B.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

NRC Publications Record / Notice d'Archives des publications de CNRC: https://nrc-publications.canada.ca/eng/view/object/?id=91255d46-174e-44ec-8d81-adfa139e8563 https://publications-cnrc.canada.ca/fra/voir/objet/?id=91255d46-174e-44ec-8d81-adfa139e8563

":/'

DIVISION OF BUILDING RESEARCH

NATIONAL ｾｅＵｅａｒｃｈ COUNCIL OF CANADA

'f

EClHI N lICAlL

NOTlE

No.

las

NOT FOR PUBLICATION

PREPARED BY • • • •

H.

CHECKED BY

FOR INTERNAL USE

APPROVED BY

f(jfl:-:

PREPARED FOR ｊｾｎａｌｙＱＮｅｴＩ .

SUBJECTt

Expande. Vermioulite

aa

Loose-Fill, Thermal

I••

ulatlon

top Dwell1g8s

A general dlscuaslon otnrtdoUllte 1s given in BUllc11ng

Iote 1\0. 1 by E. V. G1bbons, Fe};). 19S0. A furt.r a8s••s_at h.a

"been .equested aa to ,he au1tabl11t, of .,apend•• " • •leusl1'.

rop

us. 1n tbe tne.latloft

of dWl11taga

and.

otb.r • •

'.d

blll1d1rla

atpu.'Ufts.

'rb1a .NOte 1a bued 1••

6811

on

tbe . .

sult.

fit work

carzt18a out a8 tbea1..

,"jaGtaat

tlle Unlvepalty -of

aaskatoheva

in 194-8-!4-9 and not prey!••al, l'8pol'te4.

'be

JJlat.rial

. a t

00-.011.17 p"v148<t to%'

verm1

01111

t.

insulation

or

tn.'tl1l

type 18 belle..4

to

be

that

obtain.d

ｾｯ

•

•ｸｰ｡ｮ､ｾ tb.e opuabtut mineral ot tbe.l••• "hleh paaa • No.

.3

81eve cut, aN rata-ln.« Ott 1\ No.

14

ale . . (co!llROnl1 de.lgB.'.d -),

+14). In

thts

ca•• the ...

aJorlt,

or

tbf putl:C1e8

In

tbe flaa1

1nauatlon

an about

1/4

1110b

t.

41 ter. An"1 individual particle when examine" " . . .bl•• a aRl&ll toncer ina. Th any small pla ea

11lakblg up a p. .tl01. Wb10h

wen

aeparated by heatl are readily

oompl'es.ed bytbe l'lage1"8 to 1/10 or 1 s ｯｾ tbelr original

expM4ecl thiokness. Tbe 1lptnesB o£ t aterial In -Its expended

tOft! 18 ladloat1•• of 1ts value .a an insulating material, while

the

unique

81)&oe4-p1&te sta-uotUh vlth1n an, ODe particle l.a _

re8ponalble top 1ta a.tt1nl'J tor vilt... 1nllqutdtOb, 4. wUl be

dlscu••••

ｬＮＧＮｾＮ

REz

ｏＡａ､ＡｑｴｬｔＦｾＱ

Val• •

ror

tbe

ooettlcl.ent

ot conduotivity, k, 1n B.t.a.

per Ｎｱｾ ft.,

hr.,

tor. ' •.,.:rature radient

ot

lOp. per !.nch

ot

thlokn,e •• ba.... been quote. over Ii. talrl1 wide rllDge, tor tbe dr"1

. .tarlal.

TbAtGulde Qt tbe A• •loan oolety

ot

Heating and V

ntl-lattng

Engineers,

1948,

quote.

11

value or

k 01' 0.32

tor

Mat ｾｬＮｬ

hay,! a den.Ity

or

6.2

lb./ou. tt. and grad.ed. -).

+14.

A seoond

yalu 1 ｱｾｯｾＮ､ fop expanded vermioulite, no t.at details b ing

2

-'l••ta matte 1n the gu.ll1'4e. hot plate at the tJalverslt7 of Saskatchewan 1n

194€J-4<J

on material P1'ocessed 1a "West8.ra aqUa

provld.4 'he

following results.

Matell'lal

Ip,itt,l

gondl

t

lon

.l)e n81t

l

!1!

b Y .

T&

8 _ _

_....L.

1)

Insulation.

_2%

molstUN

e

lb.1

Ct'h ft.

810

.Sl

2)

If iJ%t'1 abott'

8

79

_Ｎｾｯ

)

It

Drl

7.5

80

• 6

ｾｾ

It _Dr)'

_1.1

₈₀

.46

If D17

1.2

.47

6)

ft _Dry

_7.2

_.$0

7)

PI.at.p aggregate D.r1

80

.44-8)

eonoret.

aggpegate ｄｾｹ abaut 10

71

.SS

9)

19 . Dry

9.'

80

.)8

10) rt _Dry

,.6

₈₀

.Sl

T ata 1 and 2,

.3

and

4, S

an.

6

were done in the l.a-tAoh,

24.-tnch and 7-1neh hot plates re pectlwl;y_ Thea. X'$8111t, tend

to apport the value of'

0.48

q oted

1n the

A.S.i. V.E. Chtld.. In

the abe nee of sp cute t at dat to the contrary, tbe k value

ot

fl11-t;pe expanded vermiculite s£oQId be taken

as

0.48.

On'thts

basis from

5

ｾ to ｬｏｾ g1" ate1" thickness

or

expanded 'f...ｭＱＰ｜ｴｊＮｬｾ

must be used to provide tbe s e ＱｮＮｵｬ｡ｴｩｾｶ｡ｬｵ･ as

mater!.ala

baving k values in t • r e .

32

to

0.24,

and the coat

tor

comparative pu.rpoa8 adjusted aaoor-d1ngly. Pt-ovlded. it W81"e

otherw1se ace ptable, a tud space .ompletely tille" with

.,.m1-cullte (3 '5/8 inc as thick) would have an lnaultltlng

valtle

appl'oxl..

mat.If

tn.

equIvalent

or

a 2-1neh mineral wool blank.' ha.1Dg a

k vallie o£ ｃｾＩｏＮ .

Fl111Bs and s.ttl

nt

Expanded ｙＮｲｭＱ｣ｵｬｩｴｾｊ as normal11 prooes ad

tor

t11l-t1pe t rl'ft' 1 insulation, willpour red11y to a fa1rly def1nlt

density. 'rnis char cterlst1c will frequently be

or

advantage in

cert 1n Rpplications where speci 1 conditlons 1n tilling must be

met, partio\.11arly

t os

w

re no opportunity is provided fop packing

the inSUlation aft r filling. Mater!.1 fro the same source aa

referred to abov: • in conneotion with ot plate t ta, when poured.

from the top into a simulated stud space

8

tOt. high ga densities

vaniDg .fl"Om

1.0

to

7.9

lb. per cu. ft. Subseq at settle at

oause4 b1

eoPPlnI

tbe box forming the simulat d stl d space through

1/8 !.neb on to a s()114 bloek amoun.ted to <) inob s or 11

art

1"

ZOO drops. This settlement ta le8s tban that found

tor

a

PallS-of other 1naula1ilng materials. tested

1n

the aame manner in. an

ｵｲｩｯｯｾ｡｣ｴ･､ atate asproduoed by pouring lnto plaoe, and 1$ lesa

thar" that found for Most of the other materials tested 111 a

co.paeted state. The on17 ｾｸｯ･ｰｴｬｯｮｳ _!Ie tvo insulating materials

.. .3 ..

'I• • •

tests

we:re

all 0 . . . .1.4

out

in the dry

atate.

ｗｾｮ p180.4 ill a o.nta!ner

4

inCMa 11)

41-.."e. 01 6 lnoa._ hlp

1d.tb

a scr••n 'DottOlll, and oo.,l.'.ly _tted by ."lbmeraJ.nc 1ft water, .. a.pl.of YeJltmlcu11te ahoved a settl_ment ot abc\lt Ｒｓｾ

an...

draining and subseqwn t dt-y1.ng. On the baaia or theae test. 1t

may 1 concluded th t expanded vermioullte'lnsulatltmmal M

oonaidered

ｳ｡ｴｬｳＺｦＧ｡ｾｴｯｲＷ under normal con4itlens.tl'Otft tbe

p.,l., ot

view of ••ttl ment, but that if cofta1....ahJ.. 9uU'lt1t1••

or

water

sUbsequently added. to 1 , appreciable

settle.-at .117

"all1t.

tift.tV_SQrption

aml S\lba8gl.!ent

!?nIH

Expanded ftNlo\l11tecAIUlot be .ald

t,

be hlgb11

hygro-scopl0, 1n tbe usual _aIllng ot the ,....

Sample.

talcea 11"0. .

etor• •

at

about

2S"

relative bu.ml<llt,. were I.unci

to

oontaSn 1 • •;

moisture b1 velght, 4ry bas1a. 8..,le • •ｾｯｳＮ､ tor

5

daJe to_,

at_spure ot

Ｌｾ "latlve bmd4ity

at 900F. p1ok••

u,p ...

otal

or

ｾ moisture,

dry

bald... ｃＰＱＢＢ｡ｰｯｮ､ｬｾ

mol.tun .oatent. to•.

wood

ahaylaga ••

1".

ｾ t"r. at

or...

an4 ｬ｡ｾ atter

b1pbuc4it,

tHat.at.

It 1'11&7 be concla40d ,hat the abillty or. ftmlou11t. tQ plok up

motsture

ｴｾＮ \be. ＮＮＬｯｾ

etat. 1n tbe alp

18

not

Ｘｾ｡ｴＬ

at 1•••

t u.ntl1 humS-cii'i•• approach 100". It. attialt)" to., l1qu14 ..at••

t.a.

howye•• relatlvely Yery great.

$..,1.8

in eoatat.r,s

4

behea

in d!ametfJr byo inone. high were

.et

1n 1

1/4

laoheS 01

vate.'

tor

48

hours. Other sample8

or

t 8&t11e aterlal re 8 :b jeot d

to • 11ght water spray while ttll others If re 00 let ly sub rged

in vater. The ol.tl11". g lna by t ' ven1.cullte sample. in theee

three c sea wre ｚＷｾＬ

315%

and

340%

respeotlve 17,

dP1

basia.

The

va

r r

t ntlon after dryin .tor

40

days 1n tbe same Gontainere

exposed to the laboratory ｡ｾｯ｡ｰ｢Ｘｲ･ at

25

relative humidlt7. or

le.1 vas still nearly 100% b welght, dry baala. Th1a m 7 be

oompared to two sample.

ot

shavings wbloh picked up 22O',C and ＲＴＳｾ

by

weight, dry basis.

art

r compl te submergeno and held

35%

and

48%

respectively after

40

d 11 drying time. ,

Unlik. most oft,he other materlals te.'.4. t.be

veNlou-lite pIcked up vat.el- alltlOst aa readily by wlald.ng .f.-om. '-be 1

:V4

inoh Ｎ｜ｬ｢ｭ･ｬｬＧｧｾｮＮｴＮＮ

(27.) • .s

ｴｾｯＮ the complete sub_pcenoe(ｊＴｾＩ

•

The pertor:manoe

ot

espanded ..-.-mloul1te with Hape.t

to liquid 'WAter may be ...1.\84 ､ｬｾＮ｣ＧｬＮＬ '0 the .p. . .4-plat.

etruct... of tbe

partlol...

The.e ｴＨｾｬｮＸｰ｡Ｔ ••

proVl4.

exo.l1_'

oond1 tions for capillary wio' and

top atGrage of.

l1qU4

vat•••

ｾｵＮ｢ equant drying by evaporation 18 alov,

howe_r,.

aboe ｾｨｬＮ

aotion 1. 11mlted to th thin expo••

a.

edg_

of the "a180·

ot

vatel'

bet en the individual plates c4 m

terla1.

The

Maun.,

of the

matep1al to ••ttle on wetting may ala,o be due,1n. pa!'t at 1...

t,

to the action

ot

tbe allPtaee tension to. .ea develop.db. the water

bet_en tl. platea. kn41ng to pull them together and.

tau.

to

Effect of Wat :r Conduotivity

, 'the

err

at of water on

he.,

transmlaslon thJtoupmaterlala'

in a wall 18 undoubtedly ver1 complicated _d

."0.

b. precll.'8.

aoourately on th ba 1s ｯｾ present knowledge. The probl•• _ttlti

Much 8imp} l' 1f the water did no't move in tbe . ,••tal 8de.

the intl'uence of temperature. It wou.ld .. then b. p0881\)1. to

determine ita eft'ect on conduot!v!t7 an4 to

quote

lnorea In k

"alue tor each per cent of oisture pre••nt. \4ben the tel' •••••,

howe.. r, it may do :90 by 8 pro.e.s of evaporation, ...

rutr••

an. .

as vapour, and I)oade-nsatlon. When this ooe__8. lneNa••'

'lu.'l-tie

ot

heat m y b transported ｴｾｯｵｧｬｬ tne

material..

1.eM heat. In addition t o18ture

oontents

at an7 oae ｾｯＱ｡ｴ •• not

reMain oonstant. The xact natUl"e or t1M moisture

_patio• • •

t

be k

own be:for

its :r1'.ot

on

heat

t.-.8111.alon

oan "

estimate4-.

It 1s, however, poseible to asur the ette.t OD n.at flow _d••

spec11'10 aotual cond.ttlon., without

knowln6exaetl1

What ' - ••••r

la doing. Such J..n:rormation 18 \l

t-tU,

but OD17 wbaa . .011

C... ... .

applied to ca e8 8imilar to tho for wh1ch they have been4etCtJllDllne4.

A con8ideraf

ble amount

ot

1n£ormatlon ot tbiskln4 Is 8",.11..

_1e.

If work o:t Oam.mer r (l) in ny h.a p...hapa b••n the

.oet

e.xtenaive. Ife

roWld

tbat blgroacopl0 llWla'ure,

that

1.

mot.t.ur-picked up rrO!l'l the gour at tft in the air, prodnced 11 ,t1e ertect

on h t t·ranaf'er. Additional ounts ot' vater in inorganic terial,

howey r, were found to produce fJry substantial increasea in beat

transml sion. C r's .rig .ea, wbich 1 s ys pply to all

1norganic materials, are a i'ollow-s:

_letu"

oontent

above

laborator,.

dry oondition"

%by

volume.

1

2.S

S

10

1J

20

is·

Rate of increase 01 k for

drl

_'erial

11'1 ｾ

peJl

ｾ by '9'01__ !aoMas. 1Il

mol.tuft.

These figure apply to moisture contents expre s d

as per c nt bY' volume r ther than per cent by 19h. ne

hundred per cent by vol would indio t. a complete filling

ot

the unit volume

wIth

water, 1.e.,

62.S

lb. ｰ･ｾ cu. ft. per

cent 'by volume, therofor , means .62:5 lb. v ter p r ou. :tt.,

and tor vermlcu.llte welghing about 8 lb. per eu,. ft. dry,

correspond.s to (.6Z5

+

8) 100 := ＱＮＸｾ by weight, dry ｢ｴＱＮＱｾ ..

It mal be oonoluded that hygroscopio o18ture contents

in expan4ed ve:rmloullte, up to

J%

re s ul tlng from contact with

the air alone, will baye ｬＱｴｴｾ･

efr

ct, perhaps only a smell

.. s ...

water, however. the ' 01 t e cOntent may 1.n<s as. up to ＲＰｾ or

MOl' • by _lgh • dry ba i , It eno\lgh water 1s added. Aaaumins

C erer's .figures to ｡ｰｾＱＱＮ 7Ｎ･ｾ latur ontent by ght

abov. t • laboratol"Y"dJ:Oy oondition ｯｷｲｯｳｾｯｮ､｡ to

1%

moisttnte oontent by volume, as Ｘｾ above, and w111 ".a1:11' 111.an !aeft. . . in It

of

)2%.

Simll81'ly

39%

by

velgh',

OO1l"l'8*poo.da

to ._01

",01_,

result 1J1 an iaonaae of 15% per ｾ

P1

vol.... ｏｾ a 1Sj !nOft•••

in k OTt r t at ror dry ater1 1. Wb11e 1t "oala be lUltd,.••

to

auggeat tb t C reI". flgur •

vlll

ap,l,

atrl.t1J

to • •

ｾＴ

verp'..1oul1te U8 d as.a f111

inaultS.a, It

.\let

1MaJUIU_" • •

U

."iden•• to

,he'

oont• ..,. ...

oeta!M4 that

thepre,sen..ot _to"

in ｶ･ｾＱＮｵＱＱｴＮＱｮ

quantitl••

1n exoe••

ot

the bJpo•••pl0

_tatve

oontent.

will

hay. . . . ._4 e1'reot

oa

1ta

a.netuctl

vlt,.

fbe

.faot th t this material 1 h1ghll ,.lo.klDg, 40•• Itot ､ｾ . .a411J,

and1'fes up lts tI te1'" .10wl1 by

evaporation

a.oo. .__'" \0 • •,

ｑｴ｢･ｾ

t

rial_, \Uk. tho oonsld$ra'lon

ot

po.albl. 1r.lcs:reaae

in

ｯｯｮ､ｾｯｴｬｶＱｴ

du

to moist

re an

Important

one.

aat'

".I'!!!!

'1.1,1 InIiHl1!tlJ)n. l:n Ｌｾ｟

Wal!_

Iil

hal

b••

n

$hOte 1n

a ",oent ' • •

1' (2) that UI' apace.

mwalla 4an nault

in 41atUblna_41tloatlona

to

the

u.sual

aV.1"age. unU'orm

ot)ndltloaa

or

kntporatllH en4

heat

tlovdawaed

in caloulations. In

partleular.

air space. ontbe WQ"M aide

ot

the main luulatlng ｭ｡ｾ • .t'1al we... found to pl'o4uee lJU'ke« •

•1-atlons .h'1 ｡ｴＱｾｲ｡ＮＮ

ratlre.

the wall. being muoh o.lde» at

tbe bottom than -., the top.. Alit leakage tpo1\'1

the

0014 .14. to

ｴｾ warm aide of an It1PJ'o,uJl"'ly ••ｾＮ､ blanke' in a wall vas loun4 to result in tl'laJl'ke4 inc...a . . . in beat

.ran_la.lon....

well

.8

ad.d.b1& to

tn.

te.,erature ",&ptationa p:reYlo1181yreteJl're4

to.

Tl'IUI. tilttlollitl•• with batt u4 blanket 188\11&tI0118 tdgbt be WI• •

1a 8"PPOI't

of an arsum.ont ro,

tlll1 tU1 paulation in

walle

la

pret.r.noe to batt. and blanket.. Tlwse advantage • .,..

ott..,.

howevep, oJtb1e poa:alb111tles 01' aettl.-nt and

the

dlttl,u&l'J

ot

a.hie vias tmlfol-m "enattl.. in plaeing, witb. . a t

till

1.8a-lations.'

'theM

le, 1n.

ad41tlon.the

matter

of oonde....

tloa to

･ｯｮｳｬ､ｾＺｲＮ While the Id••l 1. to

pre_nt

,be

ｦｯ｟Ｎｾｬｯｮ

or

tro.t

on the was-1ft aide or tbe outside abeathing 1%1 an

tnsaIahd

Val1,

in w1nter tbGre 18 a de:tlnlte possibility tbat a.ollllt coudou.t,loB

may

torm

in l'Uinlwalla due to wef&lttlessea 1n v.poOP

barrtol' in.tal....

lationa, abnormall7 high lnsid. aumi41tl••, and uaaaual

wall

4ealgna and oOllDlnatlone or_tertlll.

1'01'

"Mo)). the !aatall••

v8pOW? ban-:l... !U'1 be 1nadeqaaate. When th

re is an air

pao.

bet.en the oatt ｏｾ blanket 1naul t10n :and the sbea lng, the

oondensation mayt'ol'm on thes\lpfaGe

ot

the sheatbing and later

can melt Gel run to the bottom of th wall wltbGut tirst nter1ng the insulation. If' a wall beoOltJ8s wet 'by oondensation it 18

likely that .nbsequent dPy1ngout will "- tacl11tatod b1 the al.

...

(,

-Bspande4 vermioullte us.d as a

tl111Daulatlon

1ft ..

vall would, undoubtedly, absorb an w: er produced as the

rro

t

ｾＱＮｴｬ ･ｏｬＧｬＴ･ｮＸ｡ｾｬｯｮ melted. It aeO 8 fairly c rtaln a180 tbat

ｾ｢ｊ .• wa\er w?ul.d be held, alld pttev(U'lte4 fro draining to the

l'iottOlQ ·of ｴｾ wall. Sub. .

quat

-:rln,g

ou.t would. be alow bee u

or tbe valin whioh the watel" 1. helt, an' b oause of t, 1 ok

or

al. circulation ne,xt to

the

8heathlng. It haa been argued

that 1••80 damage w111 . .8U1t 11 the vater 18 held a4 kept

cUatrlbll_d \bl'ot.lgh the in.ul.t,10., to b. glvenott 810wly, t:ban

i t the vater' 18 allowed to PUJ1 dow and.

conoea.rate at

th.

bot

t_

ot

the Wall.

Wat If retaining c paclty of a te:rial must llot be

regarded aa a dlsad,v t 8e in 11 ca es. Th,ere are undoubte 11

a ｮｵｾ｢･ｾ of applioations in wh1ch t 1s oharacteristio 19bt be

usd to ad antag.

:-n

ortunat 1y the pres nt knowl dge of vatar and:

1t. aotion in 'Walls is limited d <exper.1 nt 1 proof of the

advantage nd d1 a.dvant e. of· water .to,.• • • •pae1tl b l q

in ulatlng m tar1al a:r laoklng.

The pr«t$enee of water 1n a wall has tv) main 41••dy...

tagea, inorease4 :Mat·t:Nu-urmlselOll and. a a9terloHt1Di et£eot on

tiut l1'1atarlalll and on t4ewall etruetm-e aa _ .bole. ftott1ag

produce4 01

fungi

ｌｾ､ bacter1a ｾ｡ｫ･｡ plaoemQst

rapIdly

ｵｮＴＮｾ

warm, _1st oonditions. In general ..

it

mnat be ae1tted that

1t 1$ ｨｬｧｨｬＺｴ､･ＸＱＢＮ｢ｬｾ

that

a wall be able to get rld of . .

wat.r enterug "01 ｯｯｮ､･ｮｳ｡ｾｬｯｮ in winter •• 1'apldll a8 po.alble

111 the sprlns. be.fore tbe vall 1s sUbｪ･ｯｾ･､ to the high_!'

ｴ･ｭｰ･ｾ｡ｴｕｬＢ･Ｎ of late spv1ng and

.Wlmer.

a.tereno••

_;

1. CammereI", J.S. See either of the tollovillg fOJtNSWW 1a

Englls.b of OammeHr' 8 md.n conolusion,s:

(a) Robe., G. ｍｯｬ｡ｴｬＱｾ 40ntent an4 beat...u.sulat1n,g

ｾＢｰ･ｲｴＱＮ｡

ot

oUl1<llng usat."lal$. freel.tioD

ｲｾＭＹＧ or t:n. lv.tlonal He.eareh Oounoll, Can.«a,

1949.

(0) yan Gunst, E. and D. V81l Zul1el1. fh$ latluenoe

()£ moisture COlltent an tlle tl.terlftal con4ul)t,lvlt,.

t:>r

boildlng mat.,lals. lapel" pre sentedto

Bul1titng Reaeeroh Oongr,e.sJ London, Engla.n". 19S1.

2. liandegord, G.0• and H.13. Ihlt;oheon.'l'hermal pertoJimUlce of

frame valla. Paper presented to annual ._tUg.of