UNCLASSIFIED

AD NUMBER

CLASSIFICATION CHANGESTO:FROM:

LIMITATION CHANGESTO:

FROM:

AUTHORITY

THIS PAGE IS UNCLASSIFIED

ADB803014

UNCLASSIFIED

CONFIDENTIAL

Approved for public release; distribution isunlimited.

Distribution authorized to U.S. Gov't. agenciesand their contractors;Administrative/Operational Use; 16 SEP 1943.Other requests shall be referred to Office ofScientific Research and Development,Washington, DC 20301.

OSRD list 31 dtd 5-9 Aug 1946; OTS index dtdJun 1947

Reproduced by

AIR DOCUMENTS DIVISION

l:il

i i,!'-. J=i:i'*■ ■ i!-u",i■ •' ■"■

.Iilliil'l! hllb; m .. ii'.'M i

!!i;i!i

"Hill

liii!! tiiüiiiiiiiiiilii .;;;■!■':.. ii! i' '

Pli.1'"'

■£■*■

'•■■&*• "V-I

HEADQUARTERS AIR MATERIEL COMMAND

WRIGHT HELD. DAYTON, OHO

'* V

■ »■•::• •*•'■

'■f:

.. ' 'if

.M *§l

^.GOVERNMENT IS ABSOLVED

FROM ANY LITIGATION WHICH MAY

ENSUE FROM THE CONTRACTORS IN -

FRINGING ON THE FOREIGN PATENT

RIGHTS WHICH MAY BE INVOLVED.

/—[

I

nnNRDFWTfÄL

■>

Division 8

NATIONAL DEFKSE RESEARCH COMMITTEE

of the

OFFICE OF SCIENTIFIC RESEARCH ADD DEVELOPMENT

•

: ■ i

I

1

053D Ko. 1804.

Cony Mo. 42

:

nONFJDFNT i*i

DIVISION 6

NATIONAL DEFENSE RESEARCH COMMITTEE

OF THE

OFFICE OF SCIENTIFIC Rl&EARCH AND DEVELOPMENT

Progress Report on "The Preparation of Pentaerythritol"

to "

September 1, 2.945 ,.'.,■■•;'■ by

Robert C. Elderfield

Report OSRD No. l8o4

■. Copy No. /J4

WWDENTIAL

0atw September 16, 1043

Copies No. l Copy No. 29 .

Copy No. 30 -

Copy No. 31 .

Copies No. 32

34 - 3" 3

NDRC

?f: U 39

Copy No, Copy No. Copy Ho. Copy Bo. Copy No. Copy No.

Copy No. 4o - copy No. 41 . Copy No. Kz . Copy No. £.3 _ Copy No. 24 - Copy No. 45 _ Copy No. 46 _ Copy No. 47 .

through 28 - Dr. Irvin Stewart Office of the Chief of Ordnance, Attnt Liaison Officer

Office of the Chief of Ordnance, Attni Dr. L. R. Littleton

Commanding Officer, Plcatlnny Arsenal, Attnt Technical Croup

and 33 - Chief of the Bureau of Ordnance, Attnt Ammunition Research Section Division 8 Files Dr. G. B. KistiaJcowsky Dr. Ralph Connor Dr. H. C. Elderfield Dr. Russell McGill

Office of the Chief of Ordnance, Attnt 'Mr* J. B. Washnock

Copy No Copy No 49-

Copy No. Copy No. Copy No. Copy No. Copy No. Copy No. Copy No. Copy No. Copy No.

50 51 52

Total Number of Copies - 58

11

CONFIDENTMff?

. ,,. Aft; ■■'& ~l

>

CONFIDENTIAL OSHD No. lSOij.

DIVItlOW 8

NATIONAL DEFENSE ifflBnffflH CO.IIMITTEE

OP THE

OFFICE OF SCIENTIFIC RESEARCH AND DEVELOPMELJT

Section 8.2

Progress Report on "The Preparation of Pentaerythritol'1

Service Directive OD-58

.

••

I • ■, * « ■. ■

. ■

1

■

Endorsement (1) Dr. Ralph Connor, Chief Section 8.2 to Dr. 0. B. Kistiakowsky, Chief Division 8. Forwarding re- port and noting:

"In connection with the work of Division 8 on OD-58, it seemed probable; that the by-products formed with pentaerythritol might be important in determining the thermal stability of PETN and Pentolite. The initial research on this work, there- fore, dealt with methods of preparation of pentaerythritol and methods for its - analysis.' This report describes these experiments, which were not, in themselves, significant but which contributed to the further study of this subject which has been reported in OSRD 1J4.I4. and will be further described in subsequent reports."

(2) from Dr. G. B. Kistiakowsky, Chief Division 8 to Dr. Irvin Stewart, Executive Secretary of the National Defense Research Committee. Forwarding report and concurring in the above endorsement.

This is a final report under Contract 8-19!;, OEHsr 221; with Columbia University.

,

CONFIDENTIAL 1

. 4*

0$® «Jfl^

' TAELü OF OOETENW

.

■

; • ! , v < •

•

•

Introduction »Pago 1,

Dlacuaaion.. • •• «.Pago 2«

Exporlmonta on tho Prop-rr-tlon of Pontaorythritol... «Pago 8«

A 1 V 9

i •>

r

l-V-^'-'l

<.'t ' " ' I

co^° IEK^4-

mmopgcTiON

It has been noticed on several oeoaslons by those con-

cerned that the quality, in particular the stability, of

PSTN .and Pentolite is apparently adversely affected by the

presence of impurities in the pentaerythritol used for the

preparation of PSTN* Information at hand concerning the

nature of such impurities, and tho specific effect of

Individual impurities is scanty and conflicting* The present

report describes work directed toward the production of good

quality pontaorythritol by variation in tho experimental

conditions and reactions used»

In available studies on the preparation of pontaory-

thritol (British Report A.C. 117; Org»/Ex. 27; NDRC No. B-88; ■ ■ ■

British Report, R.D. Woolwich D. 906; NDRC No. B-171; NDRO

Serial No. 86), tho yiold of product has boon arrivod at

by estimation of tho amount prosont in solution at tho

ond of tho roaction and, with but a fow oxcoptions, tho

quality of tho product with rospoct to difficultly removablo

by-products has had no attention» In tho prosont study

tho product has boon lsolatod in all oasos and its purity

dotermlnod by a variety of methods*

NnL.'tWW

-4^ff

I V •

•\

;,

, t I

8.

DiacusaioM

Analytical Method»

A number of analytical methods have been used in the

past Tor estimating the purity of pentaerythritol. The

method finally adopted as standard in this work la the so-called

bensylidene method described In detail In British Report A.C.

117; Org./Ex. 27* In order to secure satisfactory reproducible

results slight modifications havo bean necessary. As finally

worked put, the procedure is as follows:

A sample of between 0.4800 and 0*5200 g. of pentaerythritol

(Note I) in a 125 ml. Erlenmeyer flask is dissolved in

5 ml« of water by warming, and the solution is cooled

to room temperature* To this solution is pipetted

10 ml* of a solution made up by dissolving 20 ml* of

freshly distilled bonsaldehydo in 100 ml* of 87J6"

alcohol (9*2 ml* of water addod to 100 ml* 95jf alcohol) '■

and 20 ml* of hydrochloric acid* sp* gr* 1*16 (22*5 ml*

of wator addod to 100 ml* of hydrochloric acid, sp* gr*

1*19* This reagent was made up fresh every sooond day*

Within a fow minutes, the benzylidono derivative of

pentaerythritol separates. The tightly stoppered flask

is allowed to stand for four hours (Note 2) at room

temperature with occasional shaking* To the crystal mush

is then added 30 ml, of potroloum othor (Skolly Solve B>

.&'-&'-»

■\

nnd the ndxturo is shaken very vigorously by hand»

Tho tomparaturo of tho mixture is thon brought to

250+1° (Koto 3) and tho procipltato of bonzylidono

dorivativo la filtorod by suction on a tared frlttod

glass Ooooh oruoiblo and waohod vdth 150 ml» of wator.

Aftor drying for 1 hr. at 95° tho procipltato is woighod«

From tho wolght of tho bonzylidono compound tho weight

of pontaorythritol la calculated with tho aid ot tho

graph shown in Figuro 1, (Hoto 4) and tho poroont of

pontaorythritol in tho sample is oalculatod. Tho

accuracy Is +0.4 porcont«

Mo to 1» It has boon found that if tho samplo taken bo outaldo

this rango, variable results aro obtained«

Noto 2, Four hours is tho minimum timo roquirod to give

satisfactory results« Hoto 3« It is important that tho procipltato bo filtorod

at this tonporaturo. Deviations thorofrom givo variable

rosuits. Chilling tho crystal muah to 0° in loo as has

boon rooommondod by Wolfrom ha» boon found to lead to

variablo rosults, oausod apparontly by varying timo of

filtration from dotornination to dotormination* In oasos of

slow filtration, tho ohillod solution warms up. This loads

to low rosults by virtue of inoroasod solubility of tho

prooipitato. Since an arbitrary onpirioal oorrootion for

!•' Igtrfl—!*

4.

•>

solubility ia applied, It Is therefore Important to filter

at the sane temperature consistently.

Note 4« This ourve differs somewhat from a similar one given

In British Report A.C. 117; Org./Ex. 27, and has been obtained

by repeated analyses of mixtures of known composition.

The above procoduro has boon used in somo 200 determina-

tions, all dono in duplicate. Tho results oro within the

limit given above«

I

Prior to the adoption of tho benzylidono method an

attempt was made to apply a conductorst/c method baaed on tho

report of Boosokon (Roc. Trav. Chim. 30, 400 (1911); Bor»

46. 2619 (1913) ) that dipontaorythrityl othor doos not

inoroaso tho oonduotlvity of a solution of boric aoid whoroas

pontaorythritol doos to a throo or four fold oxtont. In

sovoral oxporimonts, no significant difforonoo in oonduotlvity

of borio aoid eould bo notod on addition of oithor of tho

abovo substanoos.

Wo havo also usod tho tost nitration assay method and tho

spent aoid stability mothod suppliod by tho DuPont Company

in a fow oaaos. In this mothod tho yiold and molting point

of PETN is dotormlnod aftor a standard nitration and

purification procoduro. Tho spont add is hoatod at 60° C.

and tho time to tho appoaranoo of gas bubblos on tho walls

~ 'V&l $"":'• — \t" ■

■>

':''■'.

■v

of tho tu'oo is noted» Tho proooduro la at follows i

Ton grams of flnoly powdorod pontaorythritol is

addod to 46*8 g, of 95% whlto nitric acid at such a

rato that tho tomporituro doos not oxoood 32° C,

with good stirring« Extornr.l cooling Is nooossary«

This takos about 10 min. Tho mixturo Is thon stlrrod

for an additional 10 min., coolod to 80°, filtorod

on a sintorod glass funnol and suekod as dry as

possible* Tho spont add is troatod as

doscribod bolow. Tho PETN is washed on tho funnel

with two portions of wator, onoo with 6% sodium

oarbonato solution and finally with wator« Tho orudo

PFTN is dissolvod in hot aootono (about 100 ml«) and

0*5 g« of anhydrous sodium oarbonato is addod« While

still hot« 3 volumos of wator aro sprayod in to the

well stlrrod aootono solution« Aftor cooling to

80°, tho PETN is collected and dried in a desiccator.

Tho yield and molting point aro dotorminod«

Tho volume of spont acid is dotorminod and 6 ml»

aro withdrawn, placod In a tost tube and hoatod in

a boaleor of wator at 60°+l°. Tho tin» in minutes

to tho appoaranco of gas bubblos on tho walls of tho

tubo is noted«

Tho remainder of tho spont aoid la drowned in

4 volumos of loo« Aftor standing, tho somowhat gummy

'*•. A ■*»*#

» A

U- "">

6.

. i

preoipltate 1« filtered on a tared fritted glase

Ooooh crucible, dried In a deslcoator and weighed.

The comparative results of analyses of a number

of samples of pontaorythrltol by tho three methods

are given In Tablo 1, In which tho samplos Include

synthotlo mixturos, a commercial product suppllod by

DuPont, and sovoral laboratory batches which are

Identified In Tablo II. It appoars that some incon-

sistencies botwoon tho raothods oxist, from itoich wo

foel that tho benaylidono method gives the only

aoourato indication of tho percentage of pontaorythrltol

In a givon sample. On tho othor hand, tho spont acid

stability may bo dooidodly misloadng (oomparo Run 19a

with PE-dlPB containing 6% of di-PE). Likewise, tho

residue obtained on drowning tho spont add appoars to

bo variablo and is possibly lnfluonood by tho nature

of tho impuritios proaont.

^■■■pJ'fV-

J V • a v • • n

7.

>

.

f

I

■ •

. i

■.

Pontaory- M.P. thritol of• Usod P.E.

Synthotio 252-4° mixturo

contg; 5% dl-PE.

Ditto ' 231-40 contg«

di-PE

Run 20 246-520

Run 19a 258-61°

Nlacot • 255-8° Chora. Co. Lot No. 300-570 Nitratod at Ropauno to PETN Lot No. 891.

Run 14 256-8°

Run 21 236-48°

LFH No. 2 256-8°

Tablo I.

Yield M.P 4 Spont Spont Bed.' of of Acid Acid Anal.

rocry- PETN Stabil- Rosiduo stal ity in (# of PETN rain. PE (in usod) points 1

227 139.5- 141. 40 25,0 4.6

201 139-141.5°

211 139.8-141.2°

228 140.3-141.3°

222 140.8-1420

11«5 10«2

• • 7.5 5.6

1 I

11.0 1.0

7,0 4.4

229 140.2-1410

201 140-141°

219 ,139.5-1400

88. 8#

97.25g

95,1%

3.0 2.4 92. 3#

15.5 7.4 87. 3#

2.9 95#

r ■\

8.

EXP3RIMEHT3 OH THB PREPARATION OP PBNTAERTTHRITOL

.«..

■

A largo numbor of oxporinonts havo boon porformod In ordor

to ascortain tho offoot of tho following varlabloo on tho quality

of tho PB produced: Ordor of addition of roactants, i«o.f

aldohydoa to lino or lino to aldohydos; uso of calcium oxide

vs. calcium hydroxide; uso of acotaldohydo as such or as aldol

or orotonic aldohydo; uso of fornaldohydo as formalin or

parofornaldohydo; influonco of tho so-callod "sooond reaction"

and tho tino olapsod aftor tho onsot of tho "sooond roaction"

boforo neutralization of tho roaction nixturo*

In all easos tho method of working up tho roaction mixturo

and of isolating tho product was tho sane: Sulfurlc acid

oquivalont to 96 porcont of tho calciun prosont was addod to

tho woll-stirrod roaction nixturo* Aftor filtration and thorough

washing of tho procipitatod calciun snlts, tho filtrato was

ooncontratod undor roduood prossuro until oxcossivo bunping .

causod by separation of PE provontod furthor concentration,

Aftor rofrigoration, tho PE was flltorod off and tho filtrato

furthor oonoentratod until tho final voluno was about 4 porcont

of tho original* A sooond crop of PE was ronovod and combined

with tho first* Tho orudo PE was rocrystallisod onoo fron tho

nininun amount of boiling wator with uso of docolorizing carbon*

Yiold, analysos and othor data woro takon on this onoo roorystalllz@4

material oxoopt as notod*

mh—

•\

'■«.• ■-■..

In all runs the proportions of reaotants ware kept

eomtant and in the ratio or 5.5 mola of formaldehyde, 1 • • • mol of aoetaldehyde and 0*6 mols of lime in 1.5 L. of water«

In oases where formalin was used the formaldehyde oontent

was determined; when paraformaldohyde was used, it was assumed

to be equivalent to 100 percent formaldehyde; aldol and

orotonic aldehyde were assumed to bo equivalent to 100

percent aoetaldehyde. The runs were of threo sizes, those

through No. 17 except No. 14 boing done in 3200 oo. of water, a ■

from 18 to 85 and No« 14 in 10 L. of water« In order to get

a line on a still larger sized run, No. 86 was done in 58 L,

of nator.

The results of a numbor of typical runs are shown in

Table II. F refers to poraformaiaohydo, P to formalin«

Ao to aoetaldehyde, Al to aldol and Cr to orotonio aldehyde«

Comments

1. Order of addition of roaotantsi It appears from Runs

80 and 81 that addition of tho mixed aldehydes to the limo

water rosults in a somewhat highor yiold of PH. but that tho

quality of tho product la dofinitoly lower«

8. Calcium oxldo vs. calcium hydroxide: In all oasos

finely powdered limo was usod, so that it presumably hydrated

substantially instantaneously. No study of the uso of lump

limo was made«

yi. «i*--,**»

¥W

•>

.V

t," t - ' I

10.

5. Acot-.aldohydo vs. aldnl vs. erotonlo aldehyde» All throo

give zooc. qaillty r£ although tho yiold obtalnod wiMi c-otonio

aldfifyte is nerkodJ.y loss than with acotaldohyäo or sldol»

For oonvonionco la handlir.3, nldol Is much to to preformed,

4^ Porafennaldehydc cr forrallr: Without further work

this point cannot to an&v;orod doflnitoiy, although tontatively

It can IXJ said bhat To./ dlfforoncos arico from bho uso of ono

or tho other with rospoct to qxallty. Tho yield Is battor

with paraformaldohydo, Howovor, tho uso of paraformaldohydo doos prosont ono

striking advantago from a practical point of vlow. Apparontly

tho dopolymorlzatlon of paraformaldohydo Is acoomponiod by

oonsldorablo absorption of ho at, which roduooa materially

if not oomplotoly tho rofrigoration roqulrod to control tho

reaction when formalin is usod. This is brought out by tho

dotails of Runs 22 and 23, In both oaaos tho roaction was

eorriod out in a 22-U Pyrox precipitating Jar, tho mixturo

being woll stirred mechanically and a ooil of copper tubing

was usod for tomporaturo control when nooossary. 3.0 L. of

wator, 288 g. of aldol, 1080 g. of paraformaldohydo (or an

oquivalont amount of formalin) and 246 g. of calcium oxido

wore usod. Tho limo was addod to tho mixod aldohydos gradually

as lndioatod»

!F¥

3.1»

Tino In tiiin« Tonpo Run 22

Tornp«. Run 23

0 27° 250 addition of 11,no sfcartod

2 23r5 24.5

4 23 27

6 24 30

8 27 33

12 35 42

14 43 —

16 46 -I [ Uno all in) 48

30 48.5 50.5

40 48.3 51 -(lino all in)

45 59

50 53 - 2nd r*o action 62 - ovor nddod HgS 04

Snd ro act ion ovor addod H„S0„

2 4

It is obvious that a riuch lowor tomporri-fcuro provailod In

Run 22. VVhilo no cooling was noooaaary in oithor run. on ex.

larger scalo a run sicilcr to Run 23 would probably roqairo

cooling. In Run 26. givon in dotail bolow, lnoating was nocossary

in ordor to raise tho tonporatviro sufficiently for tho second

roaotion to start«

Run No-. 26

Since this was tho largest run attonptod and can bo oxpoctod

to givo somo idoa or what can bo oxpoctod on a still largor scale,

f A i J W

■

■.

I «I » » ■

! ■

. ■

■

12.

tho rocord ia givon Jr. dotail.

Tho vossol wa3 a 25 gal. crook oquippod with good moohanical

agitation and a coppor coil for cooling or hoating. ->

Matorialot Si'. L, wator ""~~~~ 53CiC jjMe parafor-ialdohydo

15C0 gai aWol Jtochnlcol) ■ ' 1879 gn. oalciuri oxido, powdorod C.P.

Tho lino was addod gradually to tho 3lurry of aldohydos»

Tino in Hin.

0

5

10

20

30

40

50

60

70

75

90

110

115

117

120

To:.ip.

22°

20

«1

23

23

26

31

36

40

41

49

50

53

55

55.5

Ronorlu

Start adding lino.

All paraforn had di3solvod loaving a oloar solution.

lino all in.

Stoan put through coil.

Stoan turnod off.

Sooond roaotion 3tartod.

Stoppod by adding HgSO

• .'.

18.

"1

Tho flltrato from tho caloiun salts was workod up as

doscribod abovo, oxoopi; that bocause o.f tho mount of natorial

it was noccssary to connen'jrico in batoncs.» Tbo jriold of not

rocrystalllzod natortil .TH 2850 3. or fc-i to:-contr This was

roerystallizod fron 8 L. cf wator ;loic'i.tvg 1523 3. of PS which

noltod at 258-9° and c>nf.yjl 96, S J.. rO:ic r:oshor liquors woro

ooncontratod and gavo 824 ö-. noro PD •v'v.ch noltod at 248-54°

and as3ayod 91#. & third -;rop of r.ia\orinl -.loiting at 228°

and woighing 450 g. wa3 obtained on j/ur-rho.« concentration of

tho raothor liquors«

5. Tho "socond roaction": It appears tc to q-uito

dofinitoly oatablishod that unloss tho 30cond reaction ttJcos placo,

tho quality of tho PE suffors vory groatly and tho yiold docroason»

Whilo tho chonistry involvod in tiio sooond roaction i3 obscuro,

tho fact ronains that in no caso whoro tho socond roaction was

not allowod to tako placo, was a satisfactory product obtainod.

It is roconnondod that, if opportunity prosonts itsolf, a dotailod

study of this roaction bo undortakon.

Tho tino olapsod aftor tho onsot of tho socend roaction

until working up appoars to bo relativo?y -jiiinpor-car.t» Tho

roaction is charactorizod by a suddon riso in tonporaturo followod

shortly by tho dovolopnont of a yollow color in tho roaction nixtuii

It nakos littlo diff oronco whothor tho roaction is worked up a

fow minutos lator or as nuch as two hours later»

c « a<H j O 4»

L3 •H C „ ■p« H

dT?o ON ra □ *■»*■» O EH O-P

to

a TH

CM fc

•d

J

J*

CVJ

c a a £ T* "I" O IO lO

I i o m IO (O-

•• ■

IO N (0 tO i-H rH (0 8 IO

B c c;

j I Ulill B ID U) IO tt OJ ö w

.1 «o

■

I '•

; .

;'••;

■•■ .

• ■

!<§ 5 ! &

o o_

£ sss I I 8 § I ö 115* I *S 8 aas B EH *«

N

CO

pq

■ i

IO

4. ! S '

■ i tO «0

\A •«. •«. ** >*. ^A 'A t*

i i i I - •■

CM O I o> CO I

o u N oo to o> c- o to

o> c- co t- en to to to 8 cococQooooocncn

co co a? •P CM

o o. «0 « ^

2S S 8 $ 8 S 8 ^1 >J U) IO io «o

„a GINN oioioioioioi oi

a «o co

c3 (0

<< IO 01 01

I H ** I «H C- I S3 01 I i M, «. **. t*. X*. ^*

O» H lO IO ID IO

■5Ä IO

E 8 CO

to

01 01 01 o w S "3" *"

^

to co e- 10 «J IO

01 01

"G" 'S"

«0

LO c»

\Ä XA t«. \Ä Vi. *» CM to

O O O P4 + + PH . 00_._i

XMrfftHH 111! o Ü

O O H <j < rj u + + Ö o

0101 Ol CM P* tM &< H-i

OO OOOOOtlll

PH Ä o o o o o o o o «*

r r Pn (x, (i< H* +>

* r f

%

o Q o Q o + d d d d o o &, o Ü O O W d

io to co o> n n 4 io <o co o> o> o H a » s o "; ! H H H H H 3 H H 01 01 01 CM 01 01 U »3

- . -

■\

It

I « f

15,

Wotoa to Tablo II

if

. i t

■' r * i V.

1. Crudo product "was sliny and oould not bo roadtly roorystdllzod« Yiold waa dotonnlnod in solution,

2. Hydrochloric acid usod to noutraliza instoad of sulfurio«

3. Twico usual eoneontration of all ro act ants.

4. Matorial fron Run 19 roorystallizod a sooond tino»

?f?4$

>

I I -. I Iff '

.

■ • ir.r-'

- *' -. §

A *

ALrn;' 2992:~.-- . 11

... ~. ~---=tr--ce: :~:omr: L

TITLE: Progress Report on "The Preparation of Pen!Aerythritol" to September I, 1943 (NQJt.JS,-.1

1-""=••""· •""•"'"''(:,:~r.:,J•~-~ AUTHOR(S): Elderfield, Robert C. • ORIGINATING AGENCY: Columbia University, New York, N. Y. PUBLISHED BY: Office of Scientific Research and Development, NDRC, Dlv B PUIIUSHINO AGENCY NO,

OSRD-1B04 COUNnY

u.s. I I PAOU IIUUS~~~HS 16 ~<raohs

ABSTRACT:

The quality;'and in particular the stability, of PETN and Pentolite Is adversely affected by the presence of Impurities in the pentaerythritol used for the preparation of PETN. A description Is given of the work directed toward the production of good quality pentaerythritol by variation In the experimental conditions and reactions used. Experiments on the yield of pentaerythritol have shown that, the addition of the mixed aldehydes to the lime water results in a somewhat higher yield of PE, but that the quality of the product is lower. Aldol is preferred to acetalde­hyde and crotonic aldehyde for ease of handling, and the yield obtained with acetaldehyde and aldol Is greater than that obtained with crotonic aldehyde. The yield was found to be greater with paraformaldehyde than formalin although there were few differences with respect to quality.

DISTRIBUTION: Copies of this report obtainable from Air Documents Division; Attn: MCIDXD

ClNcO~m"~menL~J -~-@J- I SUBJECT HEADINGS: Explosives- Production (34511); 6 = D 0 O o 11 Explos}ies - D~velop'fent (34504. 7) ·~ ,.,... j'

0 0 }1---{"C ~f-1 . I~ f( 11~~11~11~~11~111 ~~~~~ IIIII ~~1~111~11 ~~~,,~~~ ==DEX W•lghi·P·~:~::. ~~~.~·"· ~~ -J