Embed Size (px)
Citation preview
#% ^x- rs^
EXPLOSIVE CHARACTERISTICS OF TITANIUM, ZIRCONIUM,
THORIUM, URANIUM AND THEIR HYDRIDES .
BY IRVING HARTMANN, JOHN NAGY. AND MURRAY JACOBSOM
» # « « * # * « * Report of Investlgatioas 4835
QtOS^
UNITED STATES DEPARTMENT OF THE INTERIOR Oscar L. Chapman, Secretary
BUREAU OF MINES J. J. Forbes, Director
Work on manuscript completed September 1951. The Bureau of Mines will welcome reprinting of this paper, provided the following footnote acknowledgment is made: "Reprinted from Bureau of Mines Report of Investigations 4835."
^'?3 "'Ol
December 1951
DISCLAIMER
This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.
EXPLOSIVE CHARACTERISTICS OF TITANIUM, ZIRCONIUM, THORIUN, URANIUM AND THEIR HYDRIDES
by
Irving Hartmann, L/ John Nagy,! / and Murray Jacobsonl/
COITEirS
Introduction 1 Sunmary 2 Acknowledgments , , 2 Description of samples 3 Test equipment and procedure , ...,,... 5
Inert-gas cabinet 6 Test procedure 6
Besults of experiments 10 Ignition temperature 10 Belative flammabllity. 12 Minimum energy required for ignition. 12 Minimum explosive concentration 13 Pressures and rates of pressure rise 14 Ignition by electric sparks in inert atmospheres ik
Eeduction of fire and explosion hazard l6
ILLUS'fflATIONS
Follows Fig. 1. Inert-gas cabinet 2. Laboratory furxace and spark-ignition apparatus... 3. Apparatus for determining minimtmi explosive con
centrations and minimum igniting energies of dust
k. Apparatus for determining pressures produced by dust explosions
5. Pressure-time record of zirconium, powder explosion in laboratory test bomb
1/ ^ysicistTChief, Dust Explosions and Mine Experiments Branch, Explosives and Physical Sciences Division, Bureau of Mines, Pittsbiirgh, Pa.
2/ Physicist, Head, Mine Experiments Section, Explosives and Physical Sciences Division, Bureau of Mines, Pittsburgh, Pa.
3/ Physicist, Explosives and Physical Sciences Division, Bureau of Mines, Pittsburgh, Pa.
'"^^0
Eeport of Investigations ^835 :7?3 ri02
ILLUSTRATIONS (Cont.)
Follows Fig. page
6. Spark apparatus for testing dust flammability in
7. Minimum electrical energy required to ignite dust clouds of titanium, 12
8. Maximum pressure and rates of pressure rise developed by explosions of uranitjm ll-
9. Maximum pressure and rates of pressure rise developed by explosions of uranium hydride ik
10. Maximiaa pressure and rates of presstxre rise developed by explosions of zirconium hydride.... ll-
11. Maximum presstire and rates of pressure rise developed by explosions of fine zirconium ik
12. Maximum pressure and rates of pressixre rise developed by explosions of zirconium. ik
13. Maximum pressure and rates of pressure rise developed by explosions of titanium ik
ik. Maxim\jm pressure and rates of pressure rise developed by explosions of titaniimi hydride..,,. lt-
15. Maximum pressiire and rates of pressure rise developed by explosions of thorium ,.,, 11-
16. Maximum pressure and rates of pressure rise developed by explosions of thorium hydride 11-
17. Limiting oxygen contents in air-inert gas mixtures for preventing ignition of dust clouds by elec-
'^f f\ ^
4539 - ii -
IITEOHJCTIOH
Ttie rosuJfts of the f i r s t oxperiaental study-by the Bureau of Mines on the explosive prapert ios of metal powders wore pablishou to X$k'^,kJ BXtmo t ha t time, th© oxplos ib i l i ty of nuiaorous otbor indus t r i a l dusts and powders has boon invostigated. Par t of th i s infoaiaabion has boon published, 5 / to 15 / but nos t of tho d^-ta have not boon roloasod to da t o .
fills paper suroaarizes 1310 i^guits .of oxporinonts on 22 scaaplos of t i t s n i u a , zlrooniua^' 1iioriifla.j urenlun, and hydrlaos of thcisG motals , Hie najor pa r t of tho invest igat ion was nado under a oooporativo aGS- eeuent wi"tii tiio Atoaic Biorgy. Ccxjnission*
57HaPKnn7''S^vlii57""S^pT'"'3bM ana. J&plosibi l i ty of Ifetal Powdors .• Burocu of Mines Eopt, of Investigations 5722, 1945, kk pp.
^ Hartcxann^ I rv ing , and Nagy, John, Jn&aiaaaMIlty and Sxplos ib i l l ty of Powdei-s Usod in tiio Plas t ics Industryi Buroau of Minos Eopt, of In\'©sti* gations 5751, 19hk, 58 pp.
6 / HoJtiaarm, I rv ing , and Gx'oonwcld, H, P , , BIG Btp los ib i l i ty of Motal Powder IXist Clouds: Min» and i fet , , TOI, 26, I945, pp. 551-535»
X/ Hcrixionh, I rv ing, and lagy, John,, fiffeot of Kellof Yonts on Eoduetlon of _ ' J^rossuros Dovolopod by Dust Implosions: Bux'oau of Minos Ecpt. of Inves t i gations ^ 2 4 , 1946, S2 pp.
8/Harizienn, I rving, Pressure Eoleas© for Dust .Scplos ions: Hat, l i r e Pro toot'," " Assoc. Quart. , vo l . 40, J iay 1946, pp. 47-55. £ / Hartsaaim,, Ir-ving, Bocent Bosoarch on tho l&plos ib i l i ty of Dust Dispersions:
Ind. Stag, d i aa . , vol , 40, l o , 4, April 1948, pp, T58-758. 10 / Harteiann; I rv ing , 5lie Bcploslon Hazard of total Powders, end ProK^ntive
Measures; Metals Eaiidboofc,.tooriccn Society of Motals, 1948 ed . , pp. 52"54, 1 1 / Harteann, I rving, ©cplosion and Fire Sazaxds of Cbnbustible Ifests: Ind.
Hy^cme and foxioology, Intereoionco Publishers, Ijia»f Hew York, I .X . , vol . I , chap. 15, soo, 2 , 1948, pp. 459-454.
12 / HartEiann, I rv ing, and &.gy, John, Ihe Ikp lo s ib i l i t y of Staixsh Dust: (kiesa. Ing. lews, vo l . 27, July 18, 1949, p . 2071.
1 3 / Icgy-j Jolin, Zoll inger, J . 1 . , and. Hai-tEiann, I rv ing, Pressure-Believing Capacities of Tarious 5ia,pii3ragD. Materials : Bavocs. of Minos Ifept, of lni?osttgGtions 4656, 1950, 15 PP» ' "
14/ BartE-onn, I rv ing , Bxst ©cploslons: Marks' MoOhanicol Siginoors ' Hohdboak, 5th e l . , m9^ pp. 795-800. . . .
I g / Hartaami, I rving, Cfoopor, Austin B . , and Jacobson, Murray, Booent Studies of tho Scplos ib i l i tyof" Com Starch: Aaroati of Minos Bopt, of I n v e s t i gations 4725, 1950, 9 W» ' . . .
4559 - 1 - ' -'<«
SOyifeEX
Bie investigation disclosod "ttiat dust clouds of liie powders in a i r const i tute soTero potent ia l explosion hazards, SoEie of tho sonples were pyro-phorioj taaeso were stoi'od and prepared for t e s t ing in a holiuu alaaosphore in a special cabinet .
Lajors of uranluri, uraniun hydride, and thoriun hydride iga i ted spontaneously a t room temperature within a few ainutes a f te r exposure l a air* layers of the oSier powders Igai ted i n a i r a t taaperatutea of 100^, %3 54C^ C layers of Bovoral dusts i ^ i t e d a t elovs.ted teaperatiires in, carbon dioxide and in, nitrogen*
/ ' J .^i I I^s t clDuds of ^^«fiiai, u * ^ s ^ i hydride, and ai^eoaiau i ^ i t e d a t room
tomperature ijaiaediately upon diapersion.'^ 0lou33 of the otlier po-wlers iga i ted in a i r a t ZS(P, to '^(P 0. '-Sono .da^st clouds also igaited. a t -elevated taaperatui^s in carbon dioxide, i i l l 3«st clouds could bo iga i ted i n a i r by weo^ Qleot):'lcai s p a r l s . In the presence of a h l # - ¥ p l t a g e e l ec t r i c spark, dust clmOs of / several/powders igai ted in ©sfrtea ^oXld^ , but none ignitod i n n i t ^ ^ n , a r ^ n , or.ifee±^fei., ( - _ ,
& e energy req.uired to l ^ i l t o -tttin layers df the powders i n a i r by electricalH3ondonsor d i sohar^ spares ranged Srou l ess than 1 iilorojoule to 5 n i l l i J o u l e s , and the i ga l t i ng eiKir^ for dust clouds ranged f r tp 5 to 200 n i l l i ^ou les *
_,Bio 2ow32 explosive l l t i l t s of das t clouds in a i r r e n ^ d fr<m 45 to 500 o A t t i ^ ^ ^ •'Sm liter.-'(appi-oxlnately 0,045 to 0,500 ounce per cubic foot ) , ^Scplosibn^o^ flu.st clouds in a laboratory t e s t "boub produced pressures* as h l # as 12? p0^»*p©**-Sfttsi^ inclf-Cliie abbiss-ylation p . s . i , wi l l "te usod fo r ' th i s -un l t ) i average ra tes of pressure^ r i s e up to 5,100 P | S | i t p a ^ s e c ^ a i and xmxinun ra tes to o w r 10,000 p , s , l , j e r " s ec** . * -'••*'-*. •-'
A 1iiin surface coating of o # ^ ^ - on t l t ^ f e i and 2l»©»iaa powders reduced-.the i ^ l t l o n sensitl-tttiy sonewhati but , upon i ^ i t l o n , dust cloudy of •file ©o^^ -coa t ed powders produced explosions Itiat were nearly as s t r i ng as tifose of the untreatod powders ,~
fHe study indioates th© advisab i l i ty of processing f^& honlLlng seiroral of these powders when in •&© dry s t a t e under l » 4 i m or - ^ ^ W ^2&»ti,4
ACElOWLlDOimiS
Bio cooperative agrocsiont with the Atoalo ftier-sr-Cbmalssion was arranged * by Idward J , Kehoe, chief, t i r e and Accident Branch, Heallii and Sa*3ty Wvision,
of *U3e low lorjt office of -Qi© A.E.C., and &•* Bernard lowi^, chief, feplosiires and Physical Sciences Division, Bureau of Mines, l i n e of 12ie :pjwd0rs .used in tiie study wore prepared for tiio A»1.0» by Sylvania l l e c t r l o Yrsducts Cto., Bayside, long Is land. 1 . J* Kehoo, f. L« Brcnni^n, end Paul H e r l n of the lew lork office of HM»G, and ft*. Bem£a:*d fiipeliaan of Sylvania cooperated uost helpful ly with, the Bureau of Mines throu#iout th i s phase of the work.
4559 ' - 2 -
•ix few seiaples of titoniun. and zlrconiuu and one hydride of each woro obtained t h r o u ^ tdie courtesy of Motal H^driuos, I n c . , Beverly, Mass.
Much of the labora.tpry t e s t wari: was done by Austin B. Oooper, physical science a ide , and BD^r P* ¥ i l l i a a s , laboratory mechanic, ttist Sxplosions Section, Bureau of Mines.
DISOBIPTIOH OJ" SAMPLIB
Ihe t e s t samples were numbered in the or,jer of t he i r receipt a t ti-© laboratory. The samples wi l l be d e s l ^ a t e d by these numbers in this repor t ,
tost of the powders -s^re received in 1-pound l o t s j th i s quantity i s usual ly enou#i to conduct a l l of the t e s t s . Some semplcs tha t arr ived i n the dry s t a t e were shipped in a i r , others were shipped dry under argon gas or in an evacuated container , and a few tJaae as a sludge under a l i g u i d .
» A description of the appearance, analys is , a n l pa r t i c l e - s i ze d is t r ibu t ion
of tho samples folio's^. Unless otherwise noted, the samples were tes ted in the fjinenoss in which tdioy were received.
Titanium. •* Seven t i taniwa powdeis, including two copper-coated samples, wore tes ted , Ifo. 74Q« - Eoceived in dry s t a t e ; dark-gray angular p a r t i c l e s . Speotro-
gi'aphic analysis indicated 2 to 5 pezcont meta l l ic impuri t ios. Minety-nino percent of the par t ic les wore f iner than lOO-mesh (Tyler-sieTo s c a l e ) , 68 pewsont f iner than 200-mosh, and 62 percent f iner tiian 270-mcsh. Scp los ib i l i ty t e s t s wero made on the powaor in the fineness in which i t was recoivod, and addit ional t e s t s woro mado on the fraction thK)u^ 200-mosh, Bxis i s the titanium staple reported in Eeport of Investigations 5722, For the sake of completeness, the informa,tion i s included in ttits r epor t .
Ifo,. 864. - Eecoived in dry s t a t e . AH par t ic les woare f iner tiian 200-fflosh, and 50 percent wero reported to bo f iner than 525''iDesh (45 mi<3K)ns), toalysis was given as 68.67 percent Ti^ Q-^S percent Al, 2 .4-2,8 percent S I , 0,05 percent C, 5-4 percent ?o, 1-1,25 peroontr Cu end the balance alumina (AlgOj).
Mo. 1^55. - Eecoived in moistened condition. A portion of the sample was dried a t 7 ^ 0» for 24 hours, giving off 22.7 percent mDisture, The "tests wero mado on the dried powder a f te r i t n.uk been parsed 1di3x>u^ a 200-mosh sieve to break up a^lomere.tion, Tho pa r t i c l e - s i ze d i s t r ibut ion i s given in table 1 , together with data fo r 10 other samples. Qbcmioal analysis was reported to show 9^.5 percent T i , .
Ho, 1556, - Eeoeivod i n dry s t a t e . Ihis i s a specia l ly prepared sample (supposedly from the same batch as 1555) with a th in copper coating on the surface of tiie powder p a r t i c l e s . Application of tho coating and heat treatment of the powder are in an experimental s t a ^ * Analysis Was repDrted to show 88,2 |»,rcent Ti , 7»0 percent Cu. Assuming spherical pa r t i c les of 10 microns uiamctor, a ooppor content of 2 pewont was considered t o represent a thicinoss of copper coating of 1,6 millimiorDns,
4539 - 3 - ^ / « . ^ ^ 1, ir%
Wo A l6C^# - Oiis i s an untreated ti tanium powder s imi lar to 1555* Slie saapl© was received in dry s t a t e undar a rgm gas . I t was used only for a few tests«
Bb,> l606» - Biis poi^der was taicon from tho s«mo l o t as I6O5, M t tho pa r t i c l e s contained a surface coating of oopper, (See also 1556}* I t vas reported to contain 85»6 percent Ti and 12,7 percent Oa, The sample was received dry under a r ^ n .
Mo, l648« - Eoceived i n dry s t a t e under arg>n ^ s , Bio metal was reported t» contain 0,08 percent hydro^n , 0,82 poKsent oxygon, and 0,062 pe«j^ . t nita.x>g(m. The aferago por t lo lo dlaaetor determined by the subsiev© s i ze r was giTen as 10,5 microns, 'BIB pa r t i c l§-s izo distributicm by liio photoloaetor method was reported to bo as follows: 24,0 percent , by w e i ^ t , 0-flO mioxons; 40«3 peroontj, 10-20 mi03x>nsi 35»T percent, 20-40 microns; avora-ge, 17,8 microns,
Titonimi Hydride (TiH2). - The following two samples were tes ted : Ho. 1428. - Gray powder, received in dry condition, Beportedly, over 98
percent of the par t ic les were f iner than 525-mesh, with -^.e greater portion in tli@ 10- to 25-micron range, ^le composition was reported as 95.0 percent Ti , 3,78 percent H2, 0 ,1 percent Oa, and 0,1 percent 0 .
HQ«. l64^, - Eeceived dry under argon gas • Analysis was repsr ted to show 2.83 perc®it Ha, 0.24 percent O2, O.O7I percent H2. Al thou^ ihls hydride contained only 70 percent of the theore t ica l hydro^n for TiH2^ i t was sa id to b#iave in many ways as a ful ly hydrided n a t e r i a l , Bie average pa r t i c l e dlaaeter by subsieve-sizer analysis was given as 4.9 microns. Photelometer measuraaents wore reported to give 88,1 pe remt pa r t i c l e s between 0-10 micronsj 10,0 percent, 10-20 microns J 1,9 percent , 20-40 microns; average, 5.2 microns,
airconima, - Seven zirconium powders, including one copper-coated ssaple , WOKJ t e s t e d , l b . 745, - Gray powder; received in a moistened s t a t e as a sludge. The
metal contained 2 to 3 percent metaUio impuri t ios, T5.ie par t lo les were • reported to be f iner than 525-mesh, • The t e s t data for th i s sample
msTQ reported in Hoport of invest igat ions 5722, io« 1028> - Beoeived as a sludge. Apparently f iner than 325-m@sh;
partiole-siz© dis t r ibut ion not knowi. This sample was reportedly tal:en from a wooden keg i n which zirconium i s s tored under water befo3^ i t i s used for spraying plates and otiier par ts of vacuum tubes«
io< 1 0 ^ , - Beoeived as a s l u d ^ from s^ae source as 1028. Siis powder represents the fine par t i c les tha t are car r ied away in the a i r s t r eaa of tim spraying room.
l o , 1557, - i^eoeived in dry s t a t e . Beportedly contains 98,7 percent Zr . IbTlEB^* •* Beoeived i n dry s t a t e , Biis SMiple was from same batch as
1557 5mt had a t h i n surface coating of oopper (see remarks for 1556). Beported to contain 90,5 percent Er aad 4,4 percent Ou,
Io« 1632> - Beoeived under water. Beported to contain 5,0 percent O2. The average pa r t i c l e diameter "by photelometer aeastirements was given as 3*3 microns,
Bb. 1655. - Eoceived dry under arg^n* Sie mater ia l was said to contain 0 . ^ percent Og and 0.025 percent H2« Average diameter by photelometer measuraafflits was given as 17*9 microns. This ziytjonium was prepared SccmXts hydrldg (1627J,
4559 %^liiM J u f - I j . -
ZiTOoniAME^:^im (ZrH2). " 'Eio following two samples of hydride were tes ted: Ho. 1429. - Eeceived in dry s t a t e . Over 98 percent of the powder reported
to be f iner than 525-Eiosh, with tho major portion of tho par t ic les between 10-25 microns. Bio composition was reported as 93«6 percent Zr^ 2,08 percent Hg, 0«l8 percent Oa, 0.07 poi*oont f e , 0,01 pOTtJent C, 1,0 percent Hf, 0,70 percent SiOg.
Io» 1627. - Eeceived under argon. Analysis reportedly gave 0,19 percent O2 and 0,022 percent 1I2« Average pa r t i c l e diameter by photelometer moasurcmont was 4,7 microns,
Thoriun. Ho. 1652» - terk-gray powder, rooelvod in an evacuated pyrox bulb,
Beported to contain 0«052 percent H^, 1«20 poroont O2, and 0,19 porceat l2» AvGi^go pa r t i c l e diameter by subsicvo s izor givai as 7,2 microns and "by photolomctor as 7 '4 microns.
Thorium Hydrtdo (ThHo), • ' • Ifo» 1655« - Greenish-gray powdor recoivod in an evacuated pyrox b o t t l e .
ito.alysis reported to show 0,94 pe«jont Hg, 1»07 poiCGnt Og, and 0, l6 poKsont Hg, Theoretical hydrogen poroonta^ of SMg i s 0,86, Average pa r t i c l e diamotor by subsiovo s izor was given as 3»8 micions and by ^otolcmetor moasur^aaont as 2,9 microns,
Ureniim. " ' Ho. 1625, - Silvor-gray powder, recoivod under arg^n. When liio sample
arr ived, tho upper portion of tho powdor had been considerably darkened. Kb important diffcronco was obsorvod .botwocn tho oxplos ib i l i ty of the l i g h t and tho darkened port ion. Analysis was reported to give 0,07 percent Og, and 0.008 percent Hg, Average par t io lo diameter by photolo-moter moasuraaent 10,8 micions,
Uraniwa Hydi''ido (1155). fct 1626. - Ifa.S-gray powder, received under arg^n. Ho chemical analysis
was reported. Average pa r t i c l e diamotor by subsiovo s izor was 5.6 microns and by photclomotox', 5-«3 microns.
More dotailod infomiation of the par t ic lo-s izo d i s t r ibu t ion of 11 of tiio above powdors i s given in table 1 ,
TIST sqUIPMSHT AND PIDCanJES
The oq.uipmont and t e s t prccodaro wero,for tho most p a r t , s imilar to those described in Eeport of I n v a s t i ^ t i o n s 5722 (see reference 4 / ) , Modifications and imprsvcmonts made since tiie o r ig ina l ooaastruction wi l l bo pointed ou t . Tho dry powders woro tostod in the fineness and fom. as roooived, 2b prxjpare t o s t samples from tho powders recoivod as sludge, smaH q_uantitios were washed witli alcohol, and tiio excoss l i q u i d was drained off; the i-os-ulting powdor was dried by slow heat ing. As soke of tho powders woro pyrophparf-o, groat care was taken to prevent f i res daring stprags and handling of tho dry ssmplos. They woro kept eitiior in evacuated containers or under i ne r t gas»
4559 - J K ^
TAHJK 1 , - ga r t l o l e - s i ze distr i fai t ion of powders
J lnenoss , microns
0-1 1-2 2-5 3-4 lt-5 5-6 6-8 8-10
10-12 12-14 14-16 16-20 20-50 5 0 ^ ^
Percent by woiJi tJ / Mo, 1555
27 29 22 10
5 2 2 1 1 -
Trace . --
1556 rw^ 20 15 12 11
4 5 2 -
2 *"
-„
15J1 15 24 23 15 10
4 7
t ^ * _ -1 --
- i M j 25 51 25 12
6 1 -
3
-_ ~ -
1 T. „..„.,•
. Bo'rcent by w o i ^ t S / 1625 4.0
-2.0 3.1 5.0 4 .8
14-1 16.5 15.6 13.0 9 .8 6.0 4.9 1.5
ri62^ 10,6 27.7 17.5 11.9
8.5 10.7 10,5
2.2 ,6 -----
1627 10.6 20.1 14.2 25*5 6.5 7.7 2,4 1.8 3.8 2.6 2.2 3.7 1.3
..1612 5.9
19.2 27.5 18.7 14.5
8.5 3.3
.9 1.7
--• "
~ Z »-
Ll53l^ 2.1
_
1.5 1,2 1.5 2 .8 5.0 T.O 7.6 7.8 8.5
18.5 26,8 10.0
1652 2 .8 6.5
13.9 11.6 10.9
8,3 14.0
7.4 5.6 6.2 4,2 4.5 4.1
m
1653 3.5
10.0 28,2 19.3 14.3
9 .6 8.7 4.4 2.0
^ -—
«. •«.
1/By microscopic ana lys i s . 2 / By photoloaetor anol^'sis.
- 6 -
r V -, ' *f
I n e r t Gas Cabinet
Bering the recent invest igat ion of the l a s t nine powders, a special cabinet was constructed for handling and s to r ing . Ihe cabinet consists of a main ohaater (approximately 24 x 24 x 50 inches) and an air lock or t ransfer chamber (6 x 6x8 inches) , 3ie construction i s of l /4- inch s t e e l welded a t a l l seams, J i ^ i r e 1 is a general view of the cabinet . Ihe'main chamber has a glass viewing window, 5/4-inoh thiolc and 12 inches square, in the front wall . In the back wall i s a 20-inch square vent sealed with a reinforced, p l a s t i c -impx^^aated, cloth diaphraga. She purpose of th i s vent i s to re l ieve pressure in case of an accidental explosion in the cabinet . The a i r lock i s welded to the l e f t side of the chamber; the doors of the air lock close on rubber gaskets, forming a i r t i ^ t s e a l s , I^uipment in the main chamber includes an analyt ical balance, a fluorescent l i # t , end tools and jax-s for working with tho powders. All manipulations in the cabinet are made "oj an opoi-a-tor wearing special gloves, which are sealed to portals welded to the main section*
In designing the equipment, provisions wore made for evacuating tho main chamber and air lock and f i l l i n g thorn with h i# i ly purif ied helixjm gas obtained from the U. S-, Helium Production Co., Amarillo, Tox. Gas and mass spectrograph analyses showed t h a t , i n i t i a l l y , tho system tad to bo evacuated and flushed xfitb helium, liiree tioies to a t t a in the desired inci-t atmosphere. TbD helium pressure i n ISie cabinet a t a l l times was maintainod a t between 1-1/2 and 5~l/2 ounces, per S9,uare inch @ige by means of a sens i t ive pressure switch and a solenoid valve. A moi^ury chock valvo was placed in iiie system to prevent excess prossuro.
Sallowing receipt of a sample, the or ig ina l containor was placed in the inert-gas cab ine t , and a small quanti ty of powdor was t ransferred in the iner t atffios^oro to a 2-ounco screw-top Ja r , Ihe or ig ina l container was roseoled careful ly and removea from tho cabinet for storage u n t i l i t was needed again, fhe t o s t samples wore woi#od i n tho cabinet and removed only a few minutes bofoi^ actual t e s t i n g .
Test Pxtsceanro ^ • .
Several t e s t s % ro porfotmed on tho dusts to dotermino tho ease of i g i i -t ion and the various factors tha t influence tho explosion hazard, A brief out l ine of these t e s t s follow®:
!Eao i ^ i t i o n temperatoro of tiie diist clouds was determined in a v e r t i c a l , e l e c t r i c a l l y heated cyl indrical- tube fcrnooe, the top of which i s connected to a small brass chcaabor, or dust boat , into which a weighed amount of dust can bo placed. The dust was projected by compressed a i r downward t h r o u # the fminaoo, and i ^ i i t i o n was indicated by tho appoaranco of a flame a t tho open furnace mouth a t liio bottom. The e l e c t r i c heating c i r c u i t of the furnace includes an automatic theamostatic cont ro l . Bio lowest fumaco temperature a t which i fp i t ion occurs was considered to bo the i ^ i t i o n temperatui'e of the dust cloud, 31i© fumaco i s shoiitx on th.e l o f t in figu.ro 2 .
4559 - 7 -"^^^^ # # ,
Bi© furnace was usod in a modified foaa "to determine the i g i i t i o n tomporature of tost layers* In these tos"te .12ie top was open, aad a small amount of dust was placed on an aluminum oxide disc a t tho center of tiie furnace, A slow stream of sir flowed upward t i i r o u ^ ttie famado sad aix?\ind the heated dust l ayor . fho lowest temperature a t whicJh tiio dust i g i i t e s o r glow s t r o n ^ y , o r a t which i t s tamportitaro s h o « a sudden rapid r i s e , was considered a i the igpl t ion toaporature»
Bie prooedaros for determining t^e igni t ion tomporatures of dusts in gases oliior than a i r were s imi la r , bat before the dust sample was plaood in •tti© boat or on the aluminaa oxide disk tho o i t i r e system was Cushed thoro»i#Llj with tile g^ to bo invef i t i^ tod , and the (Jtist was protected from a i r t torou^-out tho t e s t ,
Ihe TOlatiTO^,fIgggaabilii^y of tho powdor denotes ttte aaount of i n e r t . d u s t , expressed as 1iie poroonta@3 of ttio total mlxtar©, required in astetxture with tiie powder to prevent i ^ i t i o n >4ien tbo mixture i s blown t i i rou# tho c y l i n d r i ca l furnace under a standard so t of condi t ions. "Siis system of r a t i ng was o r ig ina l ly developed for cool dusts and i s s lg i i f i oan t because i t gives ^ a measure of Hie amount of rook dust required to prevent the propagation of coa l dust explosions i n mines, Ihe ine r t dost used in t i e laboratory to s t s was oalci]p,ed f i l l e r ' s e a r th . 'Bie re la t ive flccaiability of several lowders was deterciinod also in a second to s t in which a high-voltage, low-energy, continuous-induction spark was subst i tu ted for tiio heated furnaoo as an iga i t ion source, 511© pyrex spark tubo^ with a gaoscneck-shapod top can be seen on the rigjit near tho center of f ig i ro 2 ,
To measure tho m.intoua, energy reguirod for iga i t ion of dust clouds oxpori-ments were conducted wltii electr ical-condenser discharge sparks. A condenser was charged to a defini te po ten t ia l , end the formation of tho dust 'cloud in a o j l lnd r i ca l l u c i t e tube was synchronized with tho dischnr^s of tho condenser t h r o n g the primary winding of a luminous-tube step-up transfomer^ Oie eo i . f , , induced in the secondary c i r cu i t of tiie transfoimer, produced a spark betmion tiie oleotrodos in 12io dust cloud witiiin the luo i to tube. Condensers of different oapacitios were t r i e d to find tho weakest spark that would iga i t e tho sampl©, M.QXTQ 5 shovis the appar&ttos used in these t e s t s , Hie cy l indr ica l l u c i t e tube (I»D. 2,75 inches, length 12 inches) i s shomi on a stand near the center of "tiie figure', 5to the r i ^ i t of th© tube are a small ccaapressed-al^ tcmk £»d a a a ^ o t i c ' valife, t h r o u ^ which the a i r floi® to the bottom of the l uc i t e t»be and dispoises the dust upward. To the l o f t and s l i g h t l y in back of HiQ tu te i s the luminous-tube transformer. On tiie l e f t of the f i ^ r e i s a box ocaitalning the .electitmic timing and control ciBSults , A variable condenser con "be seen on top of th i s b5x.
In tbe'e:qsertoents on tho i'@iitiott of dust l a y e r s , the condenser vm d i s charged, without stop-up in iroltago, d i rec t ly t h r o u # the spark gap between a point electrocle and a gcounded metal disk on #iioh the dast was placed*
She minlmm explosive, conpefitratioa or lower explosive l imi t of tiie dust clouds was found by determiniiig tiie lowest average dust ooaoeatration in 'ttt© luc i te- tube apparatus in •^ich a dust i g i i t e d and p ropa^ ted flame throu#iout the -aloviAj _as indica.ted by the dsvelopaent of enou^i pressure ix> bi^ak tiie weak
^^^^ • 7 ' ^ - 0 1 9 ' ^ ~
• t
•p- I vo
f
£+ r O
& R- CD
fj
CO
d-
0
the the ^
e-|g
CD
d-
2. f
f-b
h
'
f-b £
3
feV
^
5"
,^
®
o
i3
K- ^
. g
.^'
. H
,
B.^
^ '"
cD
E?8
O
P^
CO
d-
H,
p.
StJ
o
P
C'i
.«
8 ^
P
ct-
d-
co
O
f.:
p;^.
G
p-g
; K
C
D
!;•'
O
tn
H
P
«<
\i <-.-
•
«±
b^
d-
K.
fi
0
o'
Si
CD
g^
.
g:>
o
s;
P
w
d-
ho
O
^2
{=r
£3
5t
3
!:i
Q
8 e
l d
-&
o
$
M
w
t!'
t;-
p
p
t^
^
Q
P ^ P-
p-
cn -
O '
d-'
f^
ir £
i-p
©•
^•
<?••'
§.
p H
-.
to
H
W C
D
H
M
,0^
d-
s; S
S
»
s:
.
Hj
m
V-
• H
,n
•
tf-'
T'-
CD J
i P^
HI
p^
s:-
CD
H-
a.
^S
- 1 •
n.
03
C33
;
8P
. •
'«
<
Mi
•^-
.-'O
-sJ
, ->
i H
-
fi
^ p
-01
(4
-
e
©
H
w
®
H-
p.
P
Cft
tK5
p' ^ "H
%
o
. H
j e O
p.
••ci
>^
p 'i
P <
P
to
CO
to
^ g t*
p
Hi
y- C3 j3 S
0 •
5+ s
: o
p-
O
p
P
"i
f
^ <=
8'
61
^
\J1
p
o
o
P
e
! p
. o
vn
P
O
H-
p'
P
CO
f.g
P:t
r&B
p-
' p.
H
- ^
'
Ii
*^
p,
C
d-
G
O
O
P
P
d-
H-
H-
P
^
H-
d-
d-
H-
O C
5
B -1
M
C
O
a
-, '•'
? "
§ ^
P
p.
PS-
H.
'-J'
O
p
. 1
M
^ p
fe'f^
'B
t?!
P-
P
'-'•
£1
"^
O
Si
g"
g o
'g- f
K
J •
• P
O
-.
H.
O
CD P
-H
O
p
.,
t-!
L-.
p
; f.
.i.
^J
. O
i p
I,;;
d-
^ t
V»
O.
cH
H
p'
s;
O
CD p
^ ta
• B
^
^ d
-'n
;-!
>^-
.^
TO
(K
! CD
2
. •-
P
P
" 0
P
H-
p
2
W
P
P
c+ P
-d
- C
"^
H
- 4
d-
C5
CD
C
CO
CO
g-
d-
d-
p.
JL-
l-j
Hj
O
O
O
C3
Hj
o p
p.
H3
>-.
P
CO
P
d
- d
-H
- P
-l-i
CD
H
P
CD
H-
S
H'.
O'
CT
H.
H-
h'
S
H
O
'^
^ d-
H
o
e
P H
,g
P
p/1
;-p.
B p
p
- CO
o
P-
c-t-
P'
|_J,
L
.^
I. J,
Q
p
P
P •=
<
^-
d
O
' 9
p
to
|3
CO s
; cb
O
p
H
K
CO
O
H.
H-
C
W
p
TO
c+
P
^ P
^
O
p.
4;i
H
-t }-
'•
KH
^ "S
B. ^
M
p.
O
st *"
F
; O
CO
o
o'
^ '^
S
o
p
^-^}
O
• P
^ p
H-
p.
(.J.
O
t-
i P
X
1
H-
^ &
•ci
H-
O
CO
g-g
S
ti'
^ (-
' C
D
P
'^^
K ^
o
o
53*
i B
| p-
^ CD
H-
P-
P
« CQ
O
&*'
!
52
03
CD
o p.-
'.
P
0 •i
d-
3Q i
^ S"
c+
H
- p
. O
CO
g^
P"
d-
CD
P
hi
g>
<i
!•'-
ri
o f 3
• o
CO
H
O
<•+
H
-t^d
o
^ 3
'^
d-
O
CD
H-
O
(-'
O
t/-J.
H-
p
M
a
V.
C
H.
. CO
p
rt
p=
!-j^
s:
c<-
<D
^5'
CD
CD
9"
5 H
- P
' c*
- H
- P
"
P
H
P
^!
^i
d-
pi
S P
E5
M
O
•r
J t-
J n
j O
e
4
O
p-
P
OJ
p
'^
- ?
ti
ri
CD
C:)
•"g
^l]
O
CO
l-J
Hj
c+
O
m
m
p
, H
-S
o
o
CO
!i
P
d-
t»
li
4
OJ
gS
?8
H-
O
P
^"
o
<W
CO
p
O
d-
1^
s:
p-
CD
H
O
b i-f
, P
' p
H-
P
S
Oi
^•'i
H
H-
O
P
H <
-«
^
P
d-
K-
P. "
* P
X
H
•E?f
e^
^8g.
p
<
<
Gi
p
O
P
(i
H-
^-^9
ti
!-
•.
P
P
Ci-
d-
'd
P-
H-
g
'8 P
-'. ^
H-
e<!
O
P
H)
c-i-
Hi
c:>
Ii
E"
C'
p O
p
. H
d
- P
g-
g-
S-•
Hi
CD
*-"-*
^'~
"f-
p-
ti
d-
-1'' 2.
ri"
'1^
p'
o
c H-
• p
. fi
] P
cK
.-'C
i-
\>-03
te
.
rs--
^i P
' g
, CO
py
o
f^
d-
P
"rs
O.
H-
ti
P
G
p.
CO
ft
H
CO
O
CD P
ti
cn
ti
C-
D
03
CD
P
•
Ui •
d-
p
H-
^g
P
p'
O
fe
^i^
!3
K-
gS-
O
d
- G
4
P
O
CO o
E '
^ b-
''' d-
_ J^
^
m
I^|B
p
O
CD C
R
4
p
H-
HI
P
-P
^f
^
'8^
^%'
'•cJ
ti
tr
3
W
O
H-
CD
CD
O
f3
P
03
O
p
- c+
P-
>'
P
p
- rt
o
d-
• b^
t-h
03
p
^;
o
H; td
S
p
p-
ti
CD ,
o
p
I-'.
CO
O
P
p
. 03
c+
d
- !--
'• p;
p
CD
0
3
l-s
rj
Ii
r+
CD
P
-i
P
p. f
cs p
tc
; P
t>
b
H
CD
p.
O
O
to
H-
a'
rt-
''^
CD b
- P
H
c^
c
Ii
^^
<
o
H
O
fi
f ro
c
p
c+
° ^
it B
01
m
o
; c
o
p
CJ
CJ
O
ti
O
O
H)
CD
P
o
o
Pi
03 h
d
O
« C
O
^•i
t;
H-
O
p.
>
O
03 .
03
p
e-
"i
H
CJ
O
O
d-
p d
r P
; CO
ti
. H
' ty
0
03
G
p
o
H-
p
CD
d- f rf-
P- s «
p
^d ^
td
O
cr*
p.
c. S
§ 8
i
'•=' g
03
O
!i
d-
O
O
^ ft>
03
O
03
03
CD
n ^'-
s
.g g
f^g
o
i-j-
P
" tc
j p
^ p
. ty
!-'•
• H
i p
c r--
t-i
03
H>
d-
Q
H-
P-
P
W
S
P'
P
l-i
CO p
. H
O
03
C
H
o
p-
p
s; •
G).
o
e
OP
O
d
- P
h-' P
i
p.
:S|
-CD
03
O
It
i O
its
H
&
P
.P
,
si
to'
03 p
P
03
P
g.
|H
^C
d
-cH
-H
- o
p
fr
b
O
d-
o
• p
. d
- „
&
HJ
P'-
tij
p p
j CD
HO
^c
fB
H.^
./f
?
t?
!•<
cf-«
<!
P
£3
O
Ki
H-
d-
. P
d
-
" p
: &
^ f
8-
^ p
03
03
H
f-"
p3 0
3
1^
'iJ
p-
d-
• d
-ii
-O
p
03 v
._>j
.<;
o
p Ii
OJ
*.
! O
H
i d
-P
o-
j <:
+
}-'
I--"
o
C-?
, H
- H
- H
- P
d
-Q
O
O
P
d
- l^
P
- rj-
-J
P
!•--'
O
o.
b
O
O
CD .
O
M
Ii
P
p
. p
p
O
.o"
d-
cr
03
p,
p.,
O
W
j O
CD
03
H
- O
B
ro
.s
: »
i P
P
S
-O
-* d
-P
O
p
0^
d-
O
, CD
03
l-
J W
r;
a
p.
, c
c5
5+
H-
I'ci
W
S
l-h
CD
S
c5
^
^ CD
CD
O
Ii
CO p
. 03
p
. p
. K
O
0^
03
'^
B
Q
^ o
td
H-
H)
CD
f3
c+ f
^ a
H
a ti
P
i to
P
p
!:
sp
oD
P3
wb
g f3
p
ti
! H
- d
- O
o
• p
p
h^
t,< •
p.
03
p
p,
c^
ro •
s;
d-
H-
p
H-
VJi
&-
3
<!
ilC!
03
P
d-
O
d-
O
«
O
d-
O
t fi
M
03
w
Pi
o
03
O
O
'-.l
, H
- H
-d
- O
P
^
. !
-••
CLi
03
p
: 03
.
(S
p^
, V
'. t
o'
O
G
P
O
03
H-
O
Ii
P3 t
-C-
P
Ii
tc!
P
«<
-d
H
5-J
03
O
H
CD <
<
O
H
CD
O
g^
g'^
^ S
^p
'^
^'^^
%n&
^-^
t
^ ^
^ ?
^-1
^
P
O
ti
d-
p
p-
0)
P
cr
e
p
O B
i
H-
g-o
^'B
'^
I'^S
;;:
p
H-
e
t-d
Ci
H.
OJ
P'
03
03
--p
. p
•
p
a
Op
03
Ht
Cii
H
!'=
<J
g
, p.
. O
g
p. o
P
P
fi
o
p
P
t5'^
.^
&8
-&
H-
G
CD
Q
0)
l-'
li.H
. °
03
CD
03
X
C
+
O'J
!
«
CD
P'
P
CO
O
H
si
03
P
H-
d-
o
S g
-g
d-
H.
a
z^%
^BS
;^
O
h::!
ti
B
P'
H
P
i5
O
03 P
to
'
^••g
v-s
P
H «
< 0
3
• ,
' 4
Id .
VI
3
d-»
" H
50 •
P
d-
-^
.c+.
CD
p
- g
p •
ti
H
Cft
CD P
i O
p,
O .
H
O
• a
ti
I-*-
d-
E
Pi-C
D
o-
g
O
ti
• o
d-
-^
Hj
-p.-
P
' o b
' ^
» g
CD
O
d
-H
©
'
- S
d-P
^ tc
f P
f P
H
©
y
O
O
W
pj
«4
p.
H-
CO
O
p
^•g-
d^
H-
ti
CB
M
p
o9
| o*
e
d-
H-
tr
B
03
•
.J. 3 -
P^
ffi
P
1^
p-s
r S
'-'•
P
H
p.
O
ja.,.
^i-^
o ,.
CD
n
<-^
o
.^.^
^ •f
O
• V
N
H
p
c+
a
d-'
«•
©
O
O
' H
P
p
m
03-i-
g v
_^
pit
i5
• 03
p
V'
.d-
>^
P
C».
p
d-
s;
o
h
p-
03 03
p
. CO
o
CD
j:
; .5
3" P
.03
3-
133
d-
H-
&
P-
P
^ !:
:;&
p
ie<
! ti
P
w
03
O
g
d-
HI
©
Q
P
M
HP
O
O
ti
P
P
H
" P
i 4
P
ta
p.
^ci
g *
H
P
O
O
T
03
K
^tJ
03
o
P
R
pi
d-
CD
O
03
' P
-
(~+i
o
.
H
w
P
d-
g^
ig
03
M
SO
P
03
CD p
.- O
K:^
*
* t;
o
p
K
P
i p
p
o
c^
P
p.
d-
O
ffl
P
.
B&
0 t^
O
d-
•<
^
-S o
P
d
-cf
P
»i
P
p..
H--
-S-'
ti
P
g
P'
g.
0^
to t
e p g, ^
, P
:-p
. g.^
5
«
i ^
.=
Si 3
-p
•
ra'
•
'-•-
' <
"•
i^£
;
-
« ^
to ,'
' 03
,p
';.
g;
P*'
tf*''"
•0
'
-!
^ -
•
p.:
-
»:
:•
'-
© :
/-:0
;;i
'•O--
-' 'fV
:\ ts
-'i
1:^
- :
cl
7 « 01
:-
=P-:
^ti^,
-^
:p>
'-
-H"'
-,p
,- -.
CD -
H
c+
P- t
i S*
P
G <:
• a
i 03
03
Xa
p
pj
P 5
Q
1.
+ t-
d
"^
. Q
y.
f"-
'^g
g'
CD CD
ti
(£3
«<
' H
-
P
P
tg-
^*-^
1
03
p
O
p"
p
Cts
g P
t-i
P <i
W
O <
CD
ti
O
i-ti ~>
+r-
s; P
d-
d- TO
,
d-
t-t
|P
CD S
"
CD 7
tS
'g
<B t
cf
ij
O
g- '
g
^ d
- P
03
iS
p
' p
. Ii
03
p
rftj
P
B
P-
P
p.
c+
?g
4
CD
M
d-
d-
o
P
iti
^ o
Hi
O B^
d-
CD
P
g&^
O
cT
P
CD
03
,
CD
H
1:3
CD P
* C
J CD
d
- 5 t-
"
8-9
p P
-.
M
d-
(,
[J-
p-
p
P
era
P'
fc
03
5
o
c*-
P
H-
ti
o
o
P
CD
03
p
tp
to
P
CD
Ii
P*
ik M
» p
!QiOse enga^d in handling the powders and in tiie actual experimental work wore proteotive ^oves j a l l operat3.ons were performed in such a manj»r as t» protect personnel a ^ i n s t dast and fwtes, .§11 i ^ i t i o n and es^losion t e s t s veve made under closed, wel l-vent i la ted hoods, and other necessary sa fe guards were tsk&a to insure sa t i s fac tory heal th conditions for -ai© exper i menters »
R:Em.!IB Of ^ f SRIMIHIS •
Jhe data obtained in the exp los ih i l i t y t e s t s are recorded below in tables and figures^ which are largely self-explanatory. Ihusual effects noted in the work end the resu l t s of special t e s t s wi l l be discussed b r i e f ly ,
I g i i t i on fggporature,
Dae iga i t ion temperature of dust clouds and of unHspersed dust layera> deteimined In a i r , carbon dioxide, and ni t rogen, i s shown, in table 2 , In t e s t s with dust clouds, the dust (usually 1 graa) was pitjjectod rapidly through the fuwiace, passing throu#i tibe hot zone in less than 1 second. In the t e s t s with undlspersed layoiB, the dust was kept a t the center of the hot furnace for several minutes (up to 15 mihutes i f no iga i t ion oocured),
Iftien saaples of iiranium, tiranium hydride, and se'voral zirconium powders were dispersed througa tiie furnace a t room tomperature, (approximately 20° o*} strong ignit ions resii l ted. In th is respect zirconium sample 1655 did not behave lil^o other untreated, zirconitmis, and the dust cloud in a i r did not i ga i t e below 550^ 0» Cfoppor-coatod zirconium 1558 bad a minimum iga i t ion temperature of 4 ^ ° 0», oomparod witii 2C^ 0. for untreated sample 1557 f rcm the swio batch. Dust clouds of eoppor-ooatcd t i tanium 155^ iga i ted a t appitsxi-matoly tho sanio temporaturo as untreated sample 1555 from tiio 66me batch, but coppor-qoatod titanium 1606 ig i i t od a t a temperature about 100© 0, h i ^ e r than untreated sgaaplo I605 from 12io Bmio batch,
33ttst c l o u ^ of four of tho nine powdex-s t es ted ih an atmosphere of carbon dioxide l i l t e d , but non© of 12ae dust clouds tes ted i n nitrogen i g i l t e d .
So iga i t io i^ resul tod witii 1 to 3 gcaas of uranium hydride exposed in layers in a i r a t jxiom taaporaturoi 1 ^am of tiie ;^wder l i l t e d a t minimum taB:^rature of l ioo 0, Vhon 5 Si:*ams of uranium hydride was placed on a watch glass a t loom tomperature, the surface darkened Imedia to ly and a few b r i ^ t spote appea334| within a few minutes a s l i ^ t pol l ing noise was hoard, a yellow flame was seen over -file siirfaoe, and the e n t i r e mass assumed a dul l - red glow; tho residue was black, Ihree ^sim of thorium hydride exposed in a i r a t xoaa tomportituro iga i tod witiiin a few minutes 1 the iga i t ion was oharacterizod by m. intense yollow ^low; liio residue was a powdery white dus t . I t i s quite possible tha t otiier metals o r hydrides ml#Lt igni te spontaneously i f large quonti t i tos of "tiie powdex^ wore exposed "to a i r . In fac t , uTGnlvm. powder did i ^ i t e when a p i lo of about 3OO grams was prepared for disposal .
4539 - 1 0 -
* I , « f :4 "^
Figure 1. - Inert-gas cabinet.
TASLE 2, " I g i i t i o n temperaturo end relativQ flgmmabilit^ of powders
Sample Ti-740
%k 1555 1556 1605 1606 1648
TiH2l428 X6k9
Zr- '745, 1028 1029 1557 1558 1632 1633
Zr l2 - l429 1627
Ih-1652 IhH2-l655 U-1625 1113-1626
I g a i t i o n tomperature of dtts
A i r ""THo
< 530 ' 470
460 j 460 ' 590 ! 330
480 • ^ 0 , 2Cg/
20 20
1 20 480 20
\ 350 350
' 450 270 260
20 20
t clouds
0D2
_ -, -
( i
1
n . i , i / 1 „
n . i . _ --
« -«
, 650 s h . i , '
" n . i .
! 740 ' n . i .
560 720
, « C ,
fe. ---"
-n . i .
~ n , i »
-""
n . i . n . i .
-n . i . n . i . n . i . n . i . n . i .
I g n i t i o n temporaturo of
A i r 460 470 480 430 460 38O 510 540 500 210 260 ^ 0 220 52a 190 500 270 340 280
20 100
20
dus t l abors
J ,00? „ ^ ^ 8 0
t
-] "* 1 i
-J 550 , 1
710 i 560
1 -
1 ~ \ "* 1 620 i 710 , , 650 ' ' 4po
340 ' 350
1 360
, °0.
Ip ._ -~ ---
760 ~
750 530
-», m
-790 n . i .
-n . i , ' 500 330 410 210
B e l a t i v e fl< poroont
\ I n furnace a t 70CP 0 .
5? 25
1 90 90
-~ 80 -90+ 90+ 90+ 90+ 90 f _
~ 1 85 1 — -
1
-t
a m i a b i l i t y . i n e r t
Spark apparatus
77 .5 37 »5 85 85
--70
90+ 90+ 90+ 90+ 90 « •» m • «. ->. -
r j i i - i . denotes no iga i t ion a t 8 5 0 ^ 0 , , tiic h i ^ o s t tomperature usod in these t e s t s ,
2/"20"in th is table donotos iga i t ion a t room temperature, approximately 200G. I t i s probable tha t in dust clouds th i s igni t ion was caused by on olootrO' s t a t i c dischargs within the cloud, a l t h o u ^ f r ic t ionol heat developed during the movement of the powder may have boon a contributing fac tor . Igni t ion of dust layors of '3M2 and IB3 a t room tempore>,turo indioatus t h e i r h i ^ l y pyropborio character .
Bie igni t ion temperature of an undlspersed layer of copper-coated zirconium (1558) was 100^ 0« h i ^ e r than for an untreated zirconium sample (1557) I'rom the same batch. In cont ras t to t h i s , both copper-coated titanium samples (1556 and I6O6) ha-d lower iga i t ion temperatures than the corresponding untreated t i tanium powders (1555 and I6O5},
At olovatod temperatures, (3ust layors of a l l samples tostod in carbon dioxido and of a l l except t w of the samples in nitrogen i ^ i t e d .
Small layers of tho nine motal powders most recently tostod wejos exposed a t 8500 C. in an ataaospbeix of 50 percent a i r and 50 pejKjont helium, by K>lume. All dusts l i l t e d within a few minutes. She t o s t was not made a|i^otiier tcmporatsires . ^ ^ . 3 ^
4539 - 11 - ^ ' \Q
BeJAtlVQ RaBimability Bie resu l t s of the r e l a t ive flaamability t e s t s conducted on laany of tiie
jowdexB are rcsoordod in tiie l a s t two columns of table 2 . Bie lowest values were obtained for dust clouds of the r e l a t i ve ly Impure titanium ^wder , 864, follo'ited by tho comparatively coarse t i tanium, &« 740, All o'ttier powders t e s t ed r e quired a mixture of a t l e a s t 70 poroont, and a t most ®-90 percent , i n e r t dust to prevent igni t ion and suppress Heaae propagation, A vtf»e of 90+ jeroent indicates t ha t BK>ro than 90 percent i ne r t dust was required for t i l ls purpose.
Minimum; S i e r ^ , JteguiTOd for I g i l t l on
Table 3 contains ttio data from the igni t ion t e s t s with condenser d i s c h a r ^ sparks. Uie energy value for dust clouds ranges from 3 to 200 miUi^oules; tiiat for dust layers ranges from less than 1 mlcrojoulo to 3 ml l l l j ou los . In t e s t s with dust clouds, considerable preliminary work was required to f ind tho optimum conditions for i ^ i t i o n , Iho minimum I g i i t i n g energy depenos, among other taotoTS, on the pa r t i c l e si2e of the dust^ tho conconti^ation of tiie dust cloud, the charac te r i s t i cs of the i g a i t i n g c i r c u i t , and the timing of the ign i t ing spark re la t ive to tho fozmation of the cloud,
lABLE 3 . - Minimum . e l p o t r i o a l i g i i t i n g energs^ ax§; lower ^oxploslye l i m i t of powdoi^
Minton. energy required for i g i i t i o n
Sjiaplo IXist cloud, l a i l l i j p u l o s l /
Bist layor , nieroJoules
[lower explosive l i n i t , , oz . .per cu, f t . £ /
K.-740-ll/' 740-2 864
1555 1536 1605 1606 1648
Til2-l428 1649
Zr-745 1028 1 0 ^ 1557 1558 1632 1653
ZrH2-1^^29 1627
Bi~l652 B1H2-I653 U-1625 Iffi:.-l626
lj"i ,iouio « i ,000'miliijoi2os^'l66^i2io'3:^oidos » O.OOO95 B . t . u . '= 0,24 0ram-calorle,
2 / 1 02., per o u . f t . i s apprxjxiriately equivalent to 1 gram per l i t e r , ^ f i-740-1 is titanium sample l o . 740 t e s t ed in tiie fineness as received;
Ti-740-8 is the throu^i 200-nosh fract ion of the above. 4 / In "SiiS t e s t less than 1 gpca of powder was used. Larger quant i t ies
i ga i t e ^ppntajc^pi^ly without a spark, as has been mentioned.
4559 m^3-m - ^ -
t
-
.000 8
200 40 60 24 -.
600 -
0,4 0.8 ! 1 !
50 1 6,4 '
240 320
64 1 4 1 6.44/ ' 1 4/ •
32ii/ '
0,070 .043 ,300 ,045 ,050 .050 .065 .045 .070 .080 .190 ,040 ,0?0 .045 ,060 .045 .045 .085 ,120 .075 .080 .060 ,060
c .
Co
a
' 1 ' ' ' ' . , ' , • 1 • , 1
I '
" ^
I 4
Figure 2. - Laboratory furnace and spark ignition apparatus.
Jlgure 7 shows the effect of dust concentration on the minimum ign i t ing energy of dust clouds of t i tanium .pawdar 1648.
Bio minimum energy value fo r samples 1555j 1536, 1557? and 1558, show. in tablo 3 , was dxjtominod teaediatoly aflxsr removal of the powders from tiie o r i g ina l containers (Ti-1555 was dried f i r s t ) . At the saiao time portions of those jwwdors wore exposed in the • laboratory c l r - fo r 6 weoM, out of d i rect s u n l i ^ t , to dotemino the offoot of oxidation or otiier surface changes. After tills ^ p o s u r o , tiie samples again were subjected to minimum-energy t e s t s , Tablo 4 show a comparison of tiie data*
TiQSdE 4 , - Effect of exposure in s i r on igni t ing energy of, powdere
Mininum energy required for j g i i t i o n , Joules
SaaplB
a i s t oloud Iks t layer
BotoTQ After
6 weeks j Bofoiro After
6 weeks
T i - 1 5 5 5 « » . . . . . . . . . 0,015 I 0.025 ; ^ 1 0 - S ^ ^ S - T 1 5 5 6 . . , . , . , . . , . t .010 ,015 '• 2x10-4 i 8x10-5
Z r - 1 5 5 7 . . . . . . . . . . . ' .005 , ,015 . 1x10-6 1x10-6 1 5 5 8 . . . . . . . . . . . I .015 ; j / , 3x10-5 r 1x10-5
XJ'MO igal t ions wereoVbainod 'at 0,025' joulo; ' exSi t ' 1 4 i H t m t " ^
Bie above values show tha t copper-coated zirconium (1558) required more energy for i ^ i t i o n than untreated ziixjonium (1557)> whether dispersed in a dust cloud or as a l aye r , Iks t layers of copper-coated t i t an luu powder (1556) also wero more d i f f i cu l t to i ga i t e than untreated ti tanium (1555)I t ^ t , when dispersed in the a i r , tho untreated powder required a s l i ^ t l y h i ^ e r energy f o r - i ^ i t i o n , possibly owing t o - i t s somewhat a^lcaaeratod character . On the other hand, a second copper-coated titanium powder (1606, table 3) vas loss sens i t ive to igni t ion by sparks in -madlspersed form, as well as in a dust cloud, than tho corresponding untreated titanium (No. I605).
Esposure of Hie four titanium powders in a i r for several weeks resu l ted _ in an increase in the energy required for iga i t ion of tlae dust clouds. Iho energy needed for igni t ion of dust layers rouainod the saao for tho untifeated titanium and zirconium powders, but the oopper-ooatod powoers wore igni ted by some-viiat weaker sparto a f te r exposure.
Minlaum Bjcplosiye Oonoontration
Bae minimum explosive concentration or lower explosive l imi t of the powdea?s in a i r is given in the l a s t colicrn of table 3 , She expertoontal values of th i s l im i t are influenced great ly by tiie fineness of the dwst, the uniformity of tlio dust cloud, the nature of the ign i t ing sourco, and other t e s t oonlit ions, Jbr th i s x^ason, data dotemined by different invost i^ to i :^ are ra re ly in complete agreoment.
4559 - 13 ^ a ^ «
Pressures and Bates of Pressure Bise
Ihe pressui^ measurements in the 1 ,23- l i t e r s t ee l t e s t bomb ( f ig , 4) were mad© a t concentrations ranging from 0.1 to 4,0 ounces per cubic foot . Bi© t e s t data for 11 samples are given in table 5, and tiie values for 9 samples are plot ted individually on fi^Ares 8 to l 6 . Comparison of the data In the l a s t four columns of table 5> which give the t e s t resu l t s by the old method of dust dispersion in the bomb with the r e su l t s of t e s t s bj the new dispersion technique, shows tha t the l a t t e r explosions consis tent ly developed higher p res sures and higher ra tes of pressure r i s e .
Ibr most of the samples p lo t ted on figures 8 to l 6 the pxBSSiires aad the ra tes developed by the explosions a t ta ined maximum values within the concentrat i o n r a n ^ invest igated; In general, there was no sharply defined optliman. l i m i t a t •jfiiioh defini te peak values wero reached. As a matter of i n t e r e s t , the stoichiometric concentrations for the oxides arc indicated near the bottom of several f igures , a l t i o u ^ i t i s not known def ini te ly which oxides are formed (when more tiian one oxide ex i s t s ) in the experimental explpsibi|S, o r whether n i t r i de s also are foanaed.
In tiie experjlments, the b l u e s t pressures were produced by explosions of titanium hydride and the n e x t - h i # e s t by zirconium hydride. She b l u e s t a v e r a ^ ra tes of pressure r i s e were produced by zirconium, zirconium hydride, titanium hydride, and uranium hydride. Scplosions of several powders developed maximum rates of press-ux^ r i s e over 10,000 p , s , i , per second,
I ^ i t i o n by Eleotrio Spax-ks in Iner t Atmosiiieres
As mentioned above, in the manufacture of h i ^ l y explosive powdars, operat ions, such as grinding, s ieving, and conveying, occasionally are performed in atmospheres having a reduced oxy@3n content . Hie data on the igni t ion temperaturss of the saaples used for tiaose t e s t s (see table 2} indicate tha t dust clouds of some of the metal powders can be i g i i t e d a t elevated temperatures in carbon dioxide and tha t layers of nearly a l l Hie powders can be iga i ted a t elevated temperatures in carbon dioxide and in n i t r o ^ n . Bierefore those gases would offer l i t t l e protection a ^ i n s t igni t ion of Hie powders by hot surfaces or hj flames. However, in many indus t r ia l processes, e l ec t r i c sparks of s t a t i c or other or ig in const i tu te the pr incipal i gn i t i ng sources,
To study the pxevention of Igni t ion by e l ec t r i c sparks, experiments were made with dust clouds of nine powders in various mixtures of a i r with carbon dioxide, ni t rogen, helium, and argon. Ihe clouds were fonaed i n tho presence of a hiJa-volta@3 e lec t r ica l - induct ion spark in Hie spark tube shown schematically in figure 6, Tho resu l t s of the experiments are p lo t ted in figure 17, As i l l u s t r a t e d , i ^ i i t i ons by e lootr ic sparks of dust clouds of the metals and of urealxm hydride could be prevented most effect ivoly in atmospheres containing mixtures of a i r and helium, Tho l imi t mixtures in these atmospheres contained more oxygon than the o the r s . Mixtures of a i r and n i t r o ^ n and a i r and argon were about equal in effect iveness, and carbon dioxide was not effeoti've in preventing ignitions«
4539 mz.^-ozf ^14 -
^' . '• • , ,. I • •
m
^ ^ ?»tS«<T!»T.'»»'C;t^'.''»--'J#w-^--'
. * • -
.-'>.. Figure 3. - Apparatus for determin ing minimum explosive concentrations and mini
energies of dust clouds. mum igniting
TABLE 5, - Maximum prqssures and ra tes of pressure r i s e developed i n eX] slosions of several powdox^
Data with new metiiod of dispersion. concentration , 0 2 , /cu, f t .
Sample < 0 ,1 i 0 ,2
1 Ti-740-l - 1
740-2 - i 864 - ; -
1555 43 56 1556 36 ^6
TIE 2-1428 30' 62 Zr-745 , - ' -
1028 281 43 1029 22 1557 ! 44 1558 1 53
42 55 49
ZTB.2--lk29 29 46
i
Ti-740-l ' -'i'i(.0-2 ' -864 - i .
1555 '1,700 2,800 1556 '1,300,2,300
TiH2-l428; 40011,850 za-745 ; -
1028 1,750 2,700 1029 , 750,2,050 1557 '2,40012,600 1558 1,400!2,350
ZrH2-l429^ 800:2,150 ) r
Ti-740-l - 1 740-2 ' - ' -864 ' r \ -
1555 4,150-7,600 1556 3^85017,600
TiH2-l428 60014,300 Er-745 ' -
1028 j4,500 1029 1,550 1557 ,5,700
-6,400 5,100 7,600
1558 '2,750 ;6,3oo ZrH2-l429 ;i,600 '3,200
0.5 1,0
Data with o ld method of dispeirsion. Oonoontration, oz , /cu .
0.1 0.2 Par t A J Moximiaa pressures , p . s . i .
-. " •
46 81 78
; 96 i
1 59 57 66 60 ^
_ »
65 86 ,80
121 ^
68 62 76 78 90
Part BJ Average ra tes •
— -
250 3,400 3,100 3,850
" 3,900 3,150 2,750 2,350 4,000
,. -
700 3,500 3,300 2,800
_ 5,100 4,300 3,050 3,050 4,000
Par t 0: Majclmum ratos
-550
over 10,000 10,000
over 10,000 -9 ,500 7,100 8,800 8,800 8,800
• .
2,300 over 10,000
10,000 9,000
-over 10,000
9,500 over 10,000 over 10,000
9,500
15 55 26 41
-_ ~ -18 44
14 20 34 16 32 -_ 19 40
of pressure
ICO 200 250 400
--_
250 1,050 100
400 1,350 150 750
_ ~ '
250 1,100
of pressure 150 400 450 850
« . «-» ^
•400,2,150 - ; 150
1,250,2,450 400,1,300
1
450,2,230
0 .5 , i
4oi 44' 33
-66 • 31 42 41 --
! 57.
f t . 1.0
43 52 45 <M«
«. 71 42 50 49
*.f
6^
rise, p.s . i . /see. 7501 750 100
-- .
2,1001 350'
1,430, 1,000!
-. - 1
2,700,
500 750 250
--
1,250 600
1,450 1,250
--
3,800
r i s e , p , s . i , / s o c . 1,100, 1,100
200
-over 5,000
1,600 3,800 jo ver 3,150 '
i
- ' over 5,000 over
1,200 1,630
500 --
4,050 2,600 5,000 4,400
-~
5,000
Ihe ineffectiveness of carton dioxide is not surprising,' Past work with other metals^ including magnesium and magnesium-aluminum alloys, and the present oxperiaents in the fumaco have shown that dust^louds of h i ^ y reactive
^539 - 1 5 - ' ^ ' ^ ^
powders can i ^ i t e and explode in car ton dioxide, Tae greater effectiveness of helium otiitently i s due io i t s h i ^ tbeimal conductivity. Helium-air mixtures are defini tely more effective in preventing explosions of these five metal powders and uranium hydride than are argan-air mixtures, Svon wilii helium*-air mixtures ^ however, the oxygen content of the atmosphere had to he 7 percent or less to prevent ign i t ions , and i t is gulte l i k e l y tha t in most applications i t might he mpre p rac t i ca l to maintain an aimosphore of pure helixam rather than attempt to prepare holium-alr mixtures,
On the other han^ carbon dioxide-air mixtures wore the most effective portection against ignit ions of dust clouds of idiorixam, zirconium, and titanium hydrides. Tests also were made in cai-hon dioxide-air mixtures with titanium hydrido l t e 8 and zirconium hydride l te9» !Ihe oxygen l i m t s wore deteaaained to he 13»0 and 8.5 peitssnt, respect ively, Holiwa-air and ni t rogen-a i r mixtures were nearly equal in effectiveness, Uiose f i n d i n g correspond closely to the experimental data for methane gas, as reported hy Ooward and Jones . ! § / Ihe greater offocti'vonoss of carhon dioxide i s prohahly due to i t s h i ^ c r specif ic heat per u n i t TOluno. HoliUEi has h i ^ thermal conductivity, hut i t s specif ic heat i s loss than half tha t of carhon dioxide. The specific heat o±" argen i s s l igh t ly higher than tha t of helium, hut i t s theaaaal conductivity is •^orj mxoh ste-ller than that of heii-um,
ESHJGIIOI OF JIBS ASD iXPLOSIOH EAZiffilB
5!he investigation has shown that tlie f i r e and explosion hazai^ds incident to the manufacture and handling of large (Quantities of tjae subject i»wdors can be considerable. !Ehc hazards are comparable t o , o r even exceed, ttxose encountered in the manufacture of f inely divided magaosium or flaked-aluminum powders. I t i s s u ^ e s t e d tha t laboratories or plants contemplating production of these powders on an Important scale should consult and follow, insofar as i s possible and per t inent , the recommendations outl ined in the Codes for the Jx^vention of JDust implosions in iiiG l&nufacture of Aluminum Bronze Powder and of Maganesi-um Powder or Dust, published by the ifationol Fire Protection itesooiation, Boston, Mass, Important roooimaendations include eliminating a l l i g i i t i o n sources near dusty processes, reducing t i e production of fine dust as much as possible , giod housekeeping to prevent dissemination of dust in the p lan t , and the use of inert-gas atmospheres where prac t icab le , Maenover possible a t l e a s t seme, i f not a l l , of the monufaoturlng operations of iiie more hazardous powders should be performed in atmospheres of helium, or argon.
M important safety meesure against s t ruc tura l damage by explosions i s the provision of pressure-re l ief vents in eguipment and other affected s tructures Bie function of those vents i s to release gases during the i n i t i a l stages of an explosion, thereby preventing the development of hx§i pressures .
Reco33Biondations on s to r ing , shipping, extinguishing f i r e s , a i d other control measures for zirconium powder were published in a Safety Digest by the National Safety Ootmcil, Ohicae^, 111, Some of tills Infoimation is also applicable to the preparation and handling of the other powders described in th i s r epor t . 16 / (foward, g.' g.,, au(L JoaQS f~'&r%lT^~Timlt&'of Ini lSimabi i i ty of Gases"and""
Tapox^; Bureau of Mines Bul l , 279, ^939, f i g . 22, p . 5^.
lj.559 ^ 2.6 -
Int.-B..o.M^es,P^.,Pa. ^OI^S-OSS
Figure H. - Apparatus for determining pressures produced by dust expl os!ons»
J ';•
1001
? 80-
m
^ 60H
m
UJ
m m £r a.
40"
20"
Max. press. = BC=90 p.s.i. Aver, rate of press. r ise=5C/AB= 1,400 p.s.i./sec. Max. rate of press, rise =EF/DE=4,000 p.s.i./sec.
/
/
7"
Sample No. 1633 Dust concentr. = 1.00 oz./cu. f t
B
0.05 0.10 TIME, SECOND
0.15
Figure 5. - Pressure-time record of zirconium powder explosion in laboratory test bcmb.
?.-?• ^ilS
Needle valve
Spherical glass bulb (500 cc. volume)
•Water manometer
Tungsten electrodes
Spark tube
Mercury manometer
Flame Gas from tank
Figure 6. - Spark apparatus for testing dust flanmability in various gases.
i /5> 1 (" t , •";
a?
0.50
3
° 40 >-
i j j
z y j
o fi . 8
z
1A .1%)
%
\
\
t
\
\
V J ^
^ ^
San nple No, 1648
1.00 2.00 3.00 4.00 DUST CONCENTRATION, OZ. PER CU. FT.
5.00
Figure 7. - Ninfmum electrical energy required to ignite dust clouds of titanium.
100
80
. 60 03
m m S 40 Q_
20
Maximum pressure
1.00 2.00 3.00 DUST CONCENTRATION, OZ. PER CU. FT.
10,000
8,000 £
o y j m y j
6,000
4,000
- c 2,000
4.00
m o: UJ a. m ™j
y j m cc Ly a:
i y j IT a. u. O y j
Figure 8. - Maximum pressure and rates of pressure rise developed by explosions of uranium.
120
100
m 80 sc m a. oi
i ^ y j DC
40
20
1
/
/
I /
/
A Y
r
/
/
/ /
)
^
v^ 1
v ^ ^
A%/r 'f\¥«;
Maximum ra 1 - » • **
Ml
^
rage ri
c
iximurr
31'1'A
lie
[6
press jj"™""™"™"
' —
B « • * • *
jre
" )C
Sample No. 1626
12,000
1.00 2.00 3.00 DUST CONCENTRATION, OZ. PER CU. FT.
10,000 ^ y j m DC y j a.
»
8,000 -c CO QC y j a. m
6,000 ^ . o: y j cr
4,000 a.
O y j
2,000
4.00
Figure 9. - Maximum pressure and rates of pressure rise developed by explosions of uranium hydride.
u/ -^^6
120
100
Qi y j a. £Q - J l y
CO CO
yj
a.
80
60
40
20
t
\
I
/
/
r 1,,,' 1
/
/ •
/
H }
/
i \
>
/ / k
/ r
A\
^
^ ^
^aximi
/erage
Maximum pressure ^
jm rati
rate
\ «
)
\ ^
fi ^ ^ tefc.
\ > ^
^
Sample No. 1627 1 ZL, 1
12,000
10,000 o CO cc y j p-
8,000 5
6,000
4,000
CO cc y j a. m
ir y j
y j cc
2,000
1.00 2.00 3,00 4.00 DUST CONCENTRATION, OZ. PER CU. FT.
Figure 10. - Maximum pressure and rates of pressure rise developed t^ explosions of zirconium hydride.
100 10,000
CO
cr UJ
a. m
m IT
0 1.00 2.00 3.00 4.00 DUST CONCENTRATION, OZ, PER CU. FT.
Figure I ! . - teximum pressure and rates of pressure rise developed by explosions of f ine zirconium.
^?P-iJ ' '"'^a
100 10,000
c CO QC UJ
o, m y j
CO CO y j
cc
1.00 2.00 3.00 4.00 DUST CONCENTRATION, OZ. PER CU. FT.
Figure 12. - Iteixlmum pressure and rates of pressure rise developed by exploslois of zirconium.
' - a ' ^
100
. 80
CO
y j O-
60
cc 3 m m y j
^ 4 0
20
Maximum pressure
O O eaO
Average rate
Sample No. 1648
10,000
o UJ CO
8,000 £
<y m
6,000 2 m Ly
( CO
cc '4,000 ^
3
y j cc a.
2,000 O
1.00 2.00 3.00 DUST CONCENTRATION, OZ. PER CU. FT.
4.00
Figure !3. - teximum pressure and rates of pressure rise developed by explosions of titanium.
u>
%
14,000
12,000
10,000
8,000
6.000
o UJ m cc UJ
a.
2
CO
tr. UJ
m
y j
I y j
a. 4,000 O
UJ
2,000
1.00 2.00 3.00 DUST CONCENTRATION, OZ. PER CU. FT.
Figure m. - Maximum pressure and rates of pressure rise developed |y explosions of titanium hydride. O ^ ^ , ^
10,000
8,000
o y j CO cc UJ a.
CO
6,000 ^
4.000
2,000
a. m UJ
m (T UJ DC
I UJ IT a. y . O y j £C
1.00 2.00 3.00 4.00 DUST CONCENTRATION, OZ. PER CU. FT.
Figure 15. - Maximum pressure and rates of pressure rise developed by explosions of thorium.
,j ;?P
/
A ^
I 1 I
1
1
f
V /
/ ^
/
/Ma>
^
^ " ^
« « « * - * ^ '
^
0 __
imum
Avs «„««—^
pressu
srage r ^ h
Maximum ra
re
ate
^
^
te
B S K a m ^ ^ ^ = m , » ^ ^
Sample No. 1653
12,000
1.00 2.00 3.00 DUST CONCENTRATION, OZ. PER CU. FT.
T 10,000 o C/3
UJ
a. 8,000 - :
6,000
4,000
m
Q_
m
Ly m Q: y j
m m y j
a. O Ly
2,000
4.00
Figure 16. - Maximum pressure and rates of pressure rise developed by explosions of thorium hydride.
%
(1625)
Th , (1652)
Zr ^ (1632)
Zr , (1633)
Ti (1,648)1
UHg
(1626)
ThHg (1653)
ZrH2 (1627)
TiH 2
)
k
a
N ^
^
\
A V s
/
k ^
Argon #
N
M
^ ^ L
. - ^ yr
f ^
\
1 . f
v \
y
^
\ '
•leliur
V \\
s
^
9
\
n
iitrog
\
- C
^ ^
\
5n-
arbon
N ^
f-d
" * *» ,
k \
dioxi de
>
K i»-
(1649) 0 4 6 8 10 OXYGEN IN MIXTURE, PERCENT
12 14
Figure 17. - Limiting oxygen contents In a i r - iner t gas mixture for preventing ignition of dust clouds by electr ic sparks.
i/,>. ^•;-
' • ^
