ITEM No. 17 
FILE No. XXII—17 
COPY No. 
NEW LIQUID INCENDIARY AGENT 
(C1F3), 
COMBINED INTELLIGENCE OBJECTIVES 
SUB-COMMITTEE INFORMATION ON A NEW LIQUID INCENDIARY AGENT 
ESSE 
23 April, 1945, 
Reported by 
M.F. FOOLER, 
U.S. Civilian 
CWS, Eq.ETOUSA 
JOSEPH E. SWADEL, 
Lt .Col.,Medical Corps, 
Hq. BTOUSA 
7 May, 1945. 
CIOS Black List Item - 17 
Flame and Incendiaries 
COMBINED INTELLIGENCE OBJECTIVES 
SUB-COMMITTEE 
G-2 DIVISION, SHAEF (REAR) APO 413 TABLE OF CONTENTS 
Subject Page No, 
1. Introduction... 3 
2. Properties of 3 
3. Preparation of CIF3 at Leverkusen 4 
4. Industrial Development of GIF- 4 
5. Effect of ClFj on Human Beings 5 
6. Recommendations 5 
7. Appendix I - Translation of Description of 
Fluorine Apparatus.... 7 
8. Appendix II - Translation of Operating 
Instructions for Fluorine... 8 
9. Appendix III - Translation of Operating 
Instructions for 11 
PERSONNEL OF INSPECTION TEAM 
Mr, Francis J, Curtis. CWS,Hq,ETOUSA 
Cmdr, Charles L, McCarthy USNR 
Dr. Mayor F. Fogler CWS,Eq.ETOUSA 
Lt,Col,Joseph S, Smadel MC AUS 
Lt,Co1,KamiIton Southworth.......USPHS 
Dr. Ernest H. Volwiler CWS,Hq.ETOUSA 
CONFIDENTIAL CONFIDENTIAL 
INFORMATION ON A NEW LIQUID INCENDIARY AGENT 
1. INTRODUCTION. 
a. The information contained in this report was obtained from 
Dr, L. Klebert, Director of Inorganic Department of the I.G, Farben- 
industrie plant at Leverkusen, Germany and Dr. Erich Noack, who is 
a member of Dr. Klebert’s department, during conversations on 23 
and 25 April, 1945. 
b. Prior to 1937 a Dr. Kwasnik began an intensive study of 
fluorine compounds since the I.G, plant at Leverkusen was interested 
in obtaining a good fluorinating agent for the fluorination of 
hydrocarbons. Investigations were made of halogen trifluoride 
compounds which had been described previously in the literature. 
(Ruff and Krug; (Jber ein neues ohlorfluorid - CIF3, zt. Anorg. 
Chem.1930, 190. 270; Ruff; Neues aus der Chemie des Fluors, zt.f. 
Angew. Chem. 1933, 46, 739.) Iodine trifluorid© was prepared and 
tested for the fluorination of hydrocarbons with some success. 
As a matter of scientific interest, chlorine trifluoride was pre- 
pared and found unsatisfactory for the purpose just mentioned. 
The Inxr ganic Division had found no use for GIF, and had no interest 
in it until 1937 when the German Army Ordnance inspected the 
laboratory preparing the material. The Army subsequently asked 
for samples at intervals; the first sample was sent about four 
years ago, the last in 1944. During this period the laboratory 
prepared 3 to 5 tons of GIF™. It is believed that the German Army 
intended to use GIF, in shells against aircraft and tanks. The 
unique properties of the substance would be peculiarly suited 
against such machines since organic matter would be inmiediately 
ignited and glass or plastic windows would be permanently etched 
end rendered opaque. 
2. PROPERTIES OF CIF 
a. The structure of the substance is not known. It is a 
liquid with a boiling point of 12 C, Iflhen stored in iron pressure 
cjrlinders, the material is stable and can be handled and shipped 
without difficulty. 
b. The substance combines with all organic and a number of 
inorganic compounds so vigorously that the heat generated will 
generally cause the material to burst into flame. Combustible 
organic compounds will then continue to burn in the normal manner. 
Non-combustible inorganic substances cease to flame as soon as the GIF* is dissipated. Glass, for example, would be 
badly etched and rendered opaque by contact with ClFg. 
Glass wool is said to burn with a flame in the presence of 
ClFg and the reaction with water is so vigorous as to give 
off incandescent gases. The reaction of C1B’3 with glass 
and water is illustrated as followst- 
SiOg + C1F3 SiF4 «- 02 +• Clg heat 
HgO *- C1F3 — HF + 0 + Cl2 i- heat 
3. PREPARATION OF GIF- AT LEVERKUSEN. 
a. All the necessary apparatus for manufacturing on a 
laboratory scale was contained in a room with floor dimensions 
of about 8 x 16 ft. This small plant could make 3-g pounds of 
ClFj per hour. The process of manufacturing is as follows 
b. Fluorine is produced by electrolysis in a cell- made 
of metallic magnesium. Carbon anodes are used with the 
metallic vessel serving0as the cathode. The electrolyte, which 
is KHFn) is kept at 100°C and is not dissipated since liquid 
HF is continuously fed into the cell and broken down into Hg 
and Fg, A sketch of the cell end detailed description of 
its operation have been obtained for comparison with those 
employed industrially elsewhere. 
o. The gases from the fluorine cell are passed through 
a coil insnersed in dry ice (-80°C) to remove HF, The fluorine 
gas is then mixed with chlorine gas in proper proportions to 
have an excess of fluorine at all times. A blue flame results 
at the point of mixture. The gases are then passed through a 
U shaped tube which is heated to 280°C. The C1F3 thus obtained 
is condensed in a coil immersed in a -80°C bath and colt cted 
in iron cylinders. The cylinders are vented several times to 
allow HF, Gig and Fg to escape. The liquid C1F3 is stable in 
such iron cylinders for long periods of time. 
4*. INDUSTRIAL DEVELOPMENT OF GIF,. 
a. No industrial development of ClFg was undertaken at 
Leverkusen. All information on this substance was given to the 
Laboratorium des Heereswaffen Amtea at Spandau, a suburb of 
Berlin; this is a laboratory of the Ordnance Department of the 
German Army. The Army name for the incendiary substance was 
"C-S" and the plant where it was subsequently manufactured was 
known as the "Seewerk”. 
b. The Army built a small pilot plant at Kuramersdorf 
which is an artillery proving ground about 50 kilometers south 
of Berlin, slightly west of Route 96 near a town named Speren- 
berg. This plant was of about the same size as the one at Leverkusen, and differed from it in no essential respect exoept that 
the fluoride cell was operated at a somewhat higher temperature* 
While it is not known when this plant was built, it was in 
operation in late 1943 when visited for a few hours by Dr* Erick 
N©ack and Dr, Kwasnik, both of I*G* Leverkusen* Dr* Kwasnik 
has recently moved to the I.G* plant at Bitterfeld; he is a 
specialist on fluorides and did mode of the development work on 
ClFj at Leverkusen* During its existence the pilot plant 
made several tons of ClFj. 
o* A production plant for GIF- of unknown capacity was 
built at Gottow which is part of the Kummersdorf proving ground. 
Drs. Kiebert and Noack do not know whether this plant was ever 
completed or, if completed, whether any GIF* was produced* 
They feel sure that GIF was never used by the German Army. 
V 
d. Dr, Kiebert was questioned regarding the availability 
of samples of GIF*. He said none was available and that the 
laboratory plant for GIF, at Leverkusen had not operated for 
almost a year* The plant could be reconditioned for the pro- 
duction of a few pounds of ClFg per hour within a month pro- 
vided that Dr* Kwasnik could be brought back from Bitterfeld 
to do the work* 
5* EFFECT OF GIF, ON HUMAN BEINGS* 
a* No fatal accidents occurred during the work on GIF* 
at Leverkusen. However, experience was obtained with effect 
of the material on the skin. Contact was immediately followed 
by a flash flame which did not in itself cause an appreciable 
burn. However, the resultant formation of HF at the surface 
of the skin produced a chemical burn. These were treated by 
washing with buttermilk and bandaging. 
6. REC0M&SENDATI0N5. 
On the basis of the above statements, it is recommended 
that *- 
a* Arrangements be made to produce samples of GIF5 
at the earliest possible moment to determine if properties are 
as described above. 
b. Work on this substance be undertaken by G.W.S. 
o. The CIFg plants at Kummersdorf and Gottow be considered Black List targets and be investigated at the earliest opportunity. 
M.F. FOGLER, 
U*S.Civilian 
CWS,Hq. ETOUSA* 
JOSEPH E. SMADEL-j 
Lt*Col.,Medical Corps* 
Hq.ETOUSA APPENDIX I. 
TRANSLATION OF DESCRIPTION OP FLUORINE 
APPARATUS.- 
’’’FLUORINE APPARATUS NO. 3 - for 2000 Amp. 
The cell vessel as well as the cover consists of Elektron. 
All the carbon anodes and the sheet nickel cathodes are 
connected in parallel and electrically insulated from the 
cell vessel and the cover respectively. To cool and homo- 
genize the electrolyte (especially during the introduction cf HF) 
{ rgeneration, see Appendix Il),it is pumped through a 
lateral neutral compartment* containing a compressed air 
stirrer. A corresponding neutral compartment* on the other 
end of the cell contains a cooling coil (external cooling). 
The flowing electrolyte is led back from one end of the cell 
to the other through an Elektron tube beneath the cell. 
In this tube there is also a water cooled tube (internal 
cooling). The neutral compartment with the stirrer contains 
the BF inlet tube. The neutral compartment with the cooling 
coil contains a thermometer. In order to detect stoppages 
of the gas outlet tube, which may be caused by electrolyte 
dust, a slow stream of N« is led through the electrode 
compartment so that its flow may be observed in a bubble 
bottle. The usual diaphragms are omitted; the cell compart- 
ments are separated only in the portions wet by the melt. 
The electrolyte is a mixture of lQpt, KF to 2.5 pt. EF, 
The working temperature is 80-100 C. 
Because the man who did the experimental work is not 
available at present and because there are no written records, 
the above description may contain inaccuracies or small errors 
here and there. 
* Presumably electrically neutral - i.e. containing no 
electrode". APPENDIX II. 
TRANSLATION OF OPERATING INSTRUCTIONS 
FOR FLUORINE. 
'FLUORINE PREPARATION (OPERATING INSTRUCTIONS) 
Very Confidential 
O rt 
1. -Temperature control. Keep the electrolyte at 70-75 C, 
2. Observe whether the stirrer is working satisfactorily. 
3. • Control the level of the melt at 4 cm below the cover. 
4. Put the bubble bottle (a high pressure glass vessel 
filled with K.SO.) under nitrogen pressure. Notice whether 
there is any stoppage of the pipes. 
5. Turn on the direct current machines in the order given 
by the numbers on the switches. 
6. Turn on the current gradually. If the cell operation 
is not smooth load it with a maximum of 500 amps, and hunt 
for the source of trouble (level of electrolyte too low, 
stoppage of the pipes, partially opened valves). 
7. Ydien the cell is loaded with 500 to 800 amps, start 
the internal cooling (outlet temperature 34 to 38°C), with 
higher loading using external cooling also* 
8. The cell is shut down in the reverse order: external 
cooling off, internal cooling off, current off, bubble 
bottles turned off, direct current machine shut down* 
S. The regeneration of the melt is carried out as follows: 
The HF cans are heated to 40°C thereby the vapor pressure 
of HF becomes high enough to cause liquid HF to flow into 
the melt until the HF content of the melt has reached its 
original value. This can be done nfcile the cell is in 
operation. A control analysis of the composition of the 
melt (Titration vdth N/l NaOH and phenolphthalein) is 
necessary only at intervals of 6 months. (Composition 
KF 2.4, BF 2.6). APPENDIX II. (Continued) 
10* If, for any reason, the melt freezes in the circulating 
pipe (for example when the current fails), instead of cooling 
water steam is passed through the internal cooling coil until 
the electrolyte is again liquid and is pumped through the 
stirrer. 
11, The plate built into the circulating pipe, (thermal 
expansion equalizing plate), must be continuously heated 
with gas as it will otherwise swell up with the freezing 
melt and crack. 
12, Yiihen the Works alarm sounds (high siren), the cell must 
be shut down.** * - . •. ■ • - 
. APPENDIX III. 
TRANSLATION OF OPERATING INSTRUCTIONS FOR 
ClFr 
"PREPARATION OF C.T.F.(C1F3). (DISRATING INSTRUCTIONS). 
Strictly Confidential 
The cooling coil (for removal of KF from F), the condens- 
ing coil (for C.T.F.) and the storage flasks for C.T.F. were 
cooled with dry ice. The reaction oven is heated to 280°C 
and flushed with P. As soon as fluorine can be detected in 
the cutlet tube, the flow of chlorine is started and regulated 
as followss- 
58 liters Cl per hour with 650 Amp load on F cell 
gg ii nc n n it it it it it it 
134 11 « " " " 1500 " " " " " 
It II II It II 2000 11 " ” 11 M 
The maintenance of constant temperature in the reaction 
chamber follows automatically. Yftiile the cell is in operation, 
the liquid C.T.F. must be tested every 15 minutes by looking 
through the peep-hole to see that it is colorless and flowing 
off quietly. If the flow of C.T.F. is boo slow, there is a 
deficiency of chlorine; if it is yellow instead of colorless 
there is either too much chlorine or a leak in the fluorine 
pipe. 
As soon as one storage vessel is full, the valves on 
the condensing coil (for C.T.F.) are reset so that liquid C.T.F. 
flows into the second storage vessel. 
The most important thing to do during the preparation 
of C.T.F, is to watch the apparatus to prevent stoppage of 
the pipes (formation of If a stoppage occurs, all the 
liquid C.T.F. that has collected in the cooling coil and 
pipes must be completely vaporized before repairs can be 
undertaken. If liquid C.T.F. leaks out, it is covered with 
a generous quantity of dry ice and the C.T.F. and 
allowed to vaporize off in a hood. 
As soon as a steel bottle is filled with C.T.F., it is 
closed with a degreased Hofer valve and allowed to come to room temperature* A 10 atm pressure gauge is then attached to this 
valve and a second valve added after the gauge# The excess 
pressure is then carefully removed by venting Clp^C.K.F.) 
until the pressure has dropped to 1*2 atm. The steel flasks 
are then ready for shipment# 
Leverkusen, 26 September, 1944# 
(signed) Kwesnik.11