Manufacture of smokeless powder:
Many of the old hands at shooting have likely read the quoted material below but it is presented here for some of the newer shooters who might not have read what constitutes smokeless powder and how smokeless powder is made:
The reference quoted below was written by Philip B. Sharpe, a noted
writer on small arms, primarily in the early 20th. Century. The author
spent a substantial amount of time at the Hercules Radford plant at
Radford, Virginia, often in areas where employees were not allowed at
certain times, to prepare Chapter II of the Supplement for shooters and
reloaders, to further their understanding of what smokeless powder consist
of and how it is made. Chapter II consists of pages 13 through 40 of the
Supplement.
The purpose of presenting the information quoted below is to state the
major operations in manufacturing smokeless powder but leaves out consider-
able detail presented by the author in Chapter II of the Supplement.
Chapter II of the Supplement, of course, does not include complete
technical detail of the physical facilities, chemical processes and
procedures required to complete the complex operations required for the
manufacture of smokeless powder.
The author stated that he studied the manufacture of smokeless powder in
the powder plants in 1944. Since that time, the process has likely changed
to some degree but it is likely that the same general process has been
updated with modern equipment, procedures and possibly propellant components
available in the latter years through 2011. HBC
Reference: Complete Guide to Handloading
by
Philip B. Sharpe
THIRD EDITION
second Revision
Supplement
FUNK & WAGNALLS COMPANY
New York
Copyright, 1937, 1941, 1949, 1952, AND 1953, BY
FUNK & WAGNALLS COMPANY
Now a subsidiary of: World Almanac Education Group, Inc
23221 Morgan Ct
Strongsville, OH 44149
216/663-8867
Quotes from: Chapter II of the Supplement, pages 13 through 40.
THE MANUFACTURE OF MODERN SMOKELESS POWDER
See the supplement of the reference book and find on page 14, following
page 463 of the INDEX of the book, the following two paragraphs which
states, as an outline, how nitrocellulose is made.
1) The inset paragraphs are quoted from the reference exactly except
that full justification is not employed below. Deviations from the
text are quoted with single marks, e.g. 'deg'. A line space in the
inset quote indicates a page change or a skip over a comment by the
author that was not required in the quote. Quotation marks within
the quoted text are as they appear in the book. Quotation marks
within introductory paragraphs are to indicate quotations from the
Supplement text:
Cotton linters are received in bales at the plant.
There they are broken and fluffed to properly dry
them and remove excess moisture. Where wood
pulp is used, it is first dried and then shredded.
From this point the manufacturing process is
identical.
The cellulose fibers are nitrated by immersion
and stirring in a mixture of nitric and sulphuric
acids, and following nitration, are freed of acids by
wringing, boiling, pulping, alkali poaching, screen-
ing, blending, and again wringing. This pure
nitrocellulose is then dehydrated, mixed with other
ingredients, dissolved in an ether-alcohol solvent,
and finally shaped into grains. In the manufacture
of double base powders, acetone is used as the basic
solvent instead of the ether-alcohol mix., and nitro-
glycerine is blended with the nitrocellulose.
On pages 14 and 15, the author describes cotton linters:
Smokeless powder manufacture begins with the
production of nitrocellulose from cellulose or vege-
table fiber. This basic ingredient was formerly
cotton linters. Just what are "linters"? In the
manufacture of cotton products, the white cotton
bolls are picked and baled for the processing mills.
Here they are fed into ginning machines which
remove the long fibers used in the manufacture
of cloth. Other machines handle the cotton seeds
from the gins and clip free the shorter fibers re-
maining. These shorter fibers are also used in
cloth, thread, felt, or batting. The clipping process
leaves very short fibers remaining, usually not more
than 1/8 inch in length. These are again clipped
free, and since they are too short to be used in
the manufacture of cotton goods, this by-product,
called linters, is disposed of to the powder and
plastics industry.
On pages 15 and 16 the following paragraph, which states the
functions of the nitric and sulphuric acids used:
The two basic acids used are sulphuric and
nitric. Nitrogen from the nitric acid combines
with the cellulose to convert it to nitrocellulose.
Since the mixture of the two acids contains a
small amount of water, and since water is re-
leased by the reaction of the nitric acid upon the
cellulose, the function of the sulphuric acid is
to combine with this water and prevent dilution of
the nitric acid during the chemical process of
nitration.
In the nitration of each pound of cellulose, one
pound of nitric acid and 1/2 pound of sulphuric
acid are consumed. Thus is produced about 1 1/2
pounds of nitrocellulose (dry weight), the exact
weight depending on the degree of nitration.
On page 17 of the supplement, it is explained that the common
term, nitrocellulose, used above is more accurately described
by the term, cellulose nitrate.
Also on page 17, two forms of nitrocellulose, S1 and S7, are
described. The paragraphs below are quoted exactly as stated
in 1) above except the subscripts for S1 & S2 in the reference
are typed below as postscripts:
No Single nitrocellulose is entirely adequate for
the manufacture of smokeless powders. Two
types are used, differing chiefly in degree of nitra-
tion. These are known as guncotton or high grade
nitrocellulose, designated by the trade as "S1" and
pyrocotton or low grade nitrocellulose, designated
as "S7". The "high" and "low" grades do not
refer to quality, but to degree of nitration. Thus
the high grade S1 is approximately 13.4% nitrated,
whereas low grade S7 is 12.6%. These two grades
are manufactured and mixed to produce a blend
having a nitrogen content of 13.15% for single
base powders and 13.45% for double base types.
It may seem that the above percentages of nitra-
tion could be secured without the preparation of
two types of nitrocellulose plus the additional work
of blending. However, a single type cannot be
used for two reasons: first, other requirements,
particularly that of solubility, makes the produc-
tion of a single grade of nitration impractical; and,
second, the blended product makes use of the com-
plementary qualities of the two grades employed.
Low grade nitrocellulose of 12.6% nitration is
99% soluble in the ether-alcohol solvent, used
later, and the high grade 13.4% is almost insoluble
(6% soluble). Therefore when the colloid known
as "smokeless powder" is formed, the soluble low
grade dissolves and acts as a vehicle carrying the
insoluble high grade in suspension. The high
grade has the greater explosive power--it produces
a large volume of gas at high pressure--and thus
supplies the high potential desired in the finished
product. Further, the high grade, being diffused
in a fibrous state throughout the soluble plastic
low grade, acts as an igniter for the low grade.
Perhaps the outstanding development in smoke-
less powder manufacture and performance has
been the evolution of the blending of these two
grades to achieve the desired results.
The process of manufacture of the two grades
of nitrocellulose is identical--the only difference
being in the nitrating time. The process of nitrat-
ing consists of stirring cellulose fibers in a pot
charged with a mixture of nitric and sulphuric
acids in proper proportions, and in agitating the
mixture at proper temperature for the required
time interval.
On page 19, following further details of nitriding wood pulp and
cotton linters, it is stated "Here the material was boiled for
50 hours to remove remaining water-soluble acids." And further in
the next paragraph "After 50 hours of this boiling, the water was
drained from the tub and replaced with fresh water. The boiling
was repeated for an additional 10 hours."
On page 19, description of the process of removing acids from the
nitrided pulp or linters is described:
From the dewatering pots the mass was pres-
sue-piped to another building where it fed into
large wooden tubs for the "poaching" process.
This additional step removed traces of acid which
might have escaped previous washes. A tub was
filled with the pulp, a mixture of soda ash and
water was added to neutralize all acidity, and the
tub contents were brought to a boil, again using
steam. The mixture was agitated by means of a
horizontal, mechanically operated propeller. After
agitation the pulp was rinsed to remove traces of
the alkali (soda ash) and piped to blenders
in the same building.
Following the description of the "poaching" process and a description
of a blending process, the following statement was made on page 20:
From the blenders the pulpy mass was hosed
into centrifugal wringers to reduce the water con-
tent to about 30%. Below this percentage,
nitrocellulose tends to catch fire spontaneously.
Further description states that the wet nitrocellulose is dehydrated
some in presses and almost simultaneously by introducing pure alcohol
into the top of the block of nitrocellulose to displace water out the
bottom. p. 21 - 22
Once the dehy presses were finished with the cake
of nitrocellulose, it became known as "powder".
After the presses, a charge of seven 441 pound blocks of nitrocellulose,
which included 91 pounds alcohol per block, is mixed with 213 pounds of
a mixture of ether and alcohol in proportions of 65 pounds ether to 35
pounds alcohol. In addition, diphenylamine, DPA, is added (0.5% to 2%)
as a stabilizer to protect against decomposition which would result from
the liberation of nitric and nitrous acids in the colloid. (see p. 23)
At this stage, dinitrotoluene, DNT was added to the mix as a deterrent
to slow the rate of burning for artillery powder. Small arms powder
is coated with DNT, later in the process.
The nitrocellulose, solvent and additives mixture was then mixed
approximately 45 minutes and then further processed in a macerator.
"This machine served only to continue the colloiding operation under
pressure. " (see p. 25)"
"From the macerator the rubbery chunks were transported to the
preliminary blocker for compression into cakes for ease in handling
in later stages of manufacture." (see p. 25)
On pages 28 and 29 it is described that two cakes of nitrocellulose,
including additives and solvent are placed in a press. The press
forces the material through two screens (a 16 mesh supporting a 40
mesh screen, p. 26) then through 49 dies to extrude 49 strands of
"green" powder with a single axial hole for small arms powder. (Some
extruded powders, e.g. those manufactured by ADI for Hodgdon, do not
have axial holes in the powder granules. HBC) The green powder strands
are collected in 49 containers to be transported to the cutting
machines.
On p. 30 it is stated, after the green powder has been cut into granule
lengths:
Green powder is now finished--but you still
can't shoot it. To transform it into a useable fin-
ished propellant, it is necessary to remove all traces
of solvent. The first step is to dry the powder, and
strange as it may seem, this drying is done with
water.
On p. 31 it is described that the majority of solvent is removed from
the green, cut powder in water sealed tanks by a flow of carbon dioxide
gas downward through 12,000 pound charges of green powder in the solvent
recovery building followed by water drying, described below. On p. 32,
coating of the powder with DNT is described:
When the powder arrived at the Water Dry
House, it still contained from 3% to 5% of the
solvent. This process was simple in that the pow-
der was aged and freed of traces of the solvent
by soaking in water for a predetermined length
of time. Open metal carts, usually of copper,
transported the powder to Water Dry, to be
emptied into wooden tanks 12 feet in diameter and
holding 50,000 to 60,000 pounds of powder, dry
weight. One of these tanks would normally handle
more than the charge capacity of an entire solvent
recovery building.
When the tank was filled with its charge of
powder, it was flooded with water heated to 65 'deg.' F.
For six days this water was circulated through the
batch by pumping it into the top and drawing it
off through the bottom. At the end of the six-
day water cure, the charge received one or two
rinses of water, and the batch was tested for free-
dom from solvent. If it passed, the water flow was
stopped and the tank permitted to drain dry. This
draining usually required from three to six hours.
The water-moist powder was then ready for the
air-dry process. Coating, a wet operation, was first
for rifle powders, so the moist powder was poured
into 50-pound canvas sacks and transferred to the
coating house.
The heavy copper coating bar-
rels took a charge of 1112 1/2 pounds of powder,
90 pounds of water, and 65 pounds of DNT. The
powder was first poured into the tumbling-barrel,
the oily brown sugar DNT was added, the water
run in, the cover sealed, and the motors started.
The powder was tumbled for one hour then washed:
The powder was gravity-
fed from a top floor through a hopper, washed
down with water, and passed through a series of
shaker screens to remove traces of dust and lumps
of individual grains which might have stuck to-
gether during the coating operation.
Following the deterrent coating, the powder was air
dried and coated with graphite, p. 24:
All small arms powders
were given a final coating of graphite, although
many cannon powders included this powdered
metal as part of the mixture. Besides its natural
lubricating qualities, graphite serves to reduce the
hazards of loading small arms ammunition. It
causes the powder grains to flow freely through
the hoppers, and since these powders, being a form of
plastic, are inclined to generate static electricity
when tumbled, the graphite serves as an electrical
conductor to dissipate this static by grounding it
through the metal hopper, the loading machine,
and thence to ground.
(On p. 24, the author referred to graphite as inert. Graphite
is not inert, it will burn. There is some free oxygen in the air
within a loaded cartridge case that likely combines with
some of the graphite. By "inert", possibly the author
meant that the graphite is not intended as a source of
propellant energy. HBC). The air dry process was described
on p. 33.:
Powder was air dried in tanks similar to those
used in solvent recovery. Here, any minor traces
of solvent remaining in the water-coated grains
was evaporated and exhausted to atmosphere.
Steam-laden air was passed through the powder
and the vapors discharged through a stack 3 feet
in diameter and 20 feet high. If the powder in
this process was of the double base type (nitro-
glycerine content) the nitroglycerine fumes were
absorbed by a tank of sodium hydroxide in the
bottom of the stack. The air within the stack was
kept preheated at all times to the high temperature
of 60 'deg.' to prevent condensation.
The air dry process took from one to five hours.
Rifle powders took more time than the larger sizes.
But at long last you had powder which could be
used. Only it needed a coating of graphite, or
"glazing," as the boys on the powder line chose
to call it.
The powder was again poured into sacks and
transported to the graphite house, a building
similar to the coating house. Here a charge of
approximately 5,000 pounds of powder was poured
into a kettle and 1 1/2 pounds of
powdered graphite sifted over the charge.
The reason for the barricaded graphite building
is that powder is dangerous before the graphite
coating is applied--static electricity could fire the
charge before the coating was uniformly applied.
Thus no employees were permitted within the
graphite house while the tumbling kettles were in
operation. But the observer was a gun nut--not
an "employee"....
After graphite coating the powder is blended. On p. 33, a
brief description of the blending process was given:
This is a rather simple operation; yet one that
is necessary, to create a final product of uniform
quality. A 5000 pound charge from the glazing
barrels was run into 100 50-pound canvas sacks.
These sacks were segregated by batches, and 100
batches were used in the preblending operation.
A sack from each batch was emptied into the
tumbling kettle. The tumbling blended these
thoroughly, whereupon they were resacked and
ready for a similar distribution in the final blend-
ing.
This was nothing but a repetition of the pre-
blending process. The 50-pound sacks were segre-
gated and a sack from each of 100 preblends was
poured into a final blending barrel. Thus, despite
the fact that a lot of IMR 4895 contained 55,000
to 60,000 pounds, the blending operation assured
little variation in the ballistic qualities of the many
lots emerging from the blending houses each day.
Beyond the steps outlined above, the powder is packaged.
On p. 36, the composition of two double base military powders are
given:
NITROGLYCERINE POWDER
manufacture
Although much of the equipment used in the
manufacture of the double base powders is like
the single base, and can be used interchangeably,
there is some special equipment, and a difference
in manufacturing procedure.
The major change began with the nitrocellulose
mix or blend. You will recall earlier in this chapter
that the low and high nitration nitrocelluloses were
blended for single base powders in proper pro-
portions to give a final blend of 13.15% nitrogen
and for double base types, 13.45%. Since cellulose,
having a nitration greater than 13.25%, is prac-
tically insoluble in the ether-alcohol solvent, but
is readily dissolved in acetone and nitroglycerine,
the later solvent was used for the double base
line. Acetone serves a dual purpose--added to the
nitroglycerine, it acts, not only as a solvent, but
also somewhat desensitizes the high explosive.
Formula for double base powders vary. Two
samples of military powders are given below to
show some of the possible variations:
MATERIALS M2 M5
Nitrocellulose 75.50% 81.50%
Nitroglycerine 20.00% 15.00%
Potassium nitrate 1.00% 1.00%
Barium nitrate 1.50% 1.50%
Diphenylamine 0.75% 0.75%
DNT 1.00% -----
Graphite 0.25% 0.25%
The familiar Hercules HiVel #2 powder is
grained somewhat differently from M5 army pow-
der, but is of the same nitroglycerine content and
approximately the same formula.
MANUFACTURING TIME, p. 39:
The time required in powder manufacture
naturally varied with the individual types. Small
arms single base powders such as IMR 4895 ran
from the raw cotton linters or wood pulp stage to
the final packing of the powder in about 16 to 17
days, the greater part of this time being used to
purify the green powder.
Summary of some ingredients of smokeless powders for small arms from
the referenced book and other sources from the internet:
Portion of
Propellant
by mass Chemical Formula
Nitrocellulose---------85%-95% C6H7(NO2)3O5 + 3H2O, approx. 2)
Nitroglycerine---------0 to about 20% C3H5N3O9 3)
Potassium sulphate-----1% K2SO4 4)
Graphite---------------0.03%-0.25% C 5)
Diphenylamine (DPA)----0.5% to 2% (C6H5)2NH 6)
Dinitrotoluene (DNT)---0.06% dry C6H3(CH3)(NO2)2 7)
2) Double base powders contain about 80% nitrocellulose.
3) A high explosive, mixed with acetone to mix with nitrocellulose
which adds a higher energy content to the nitrocellulose.
4) Potassium sulphate reduces muzzle flash slightly.
5) Graphite is an allotrope of carbon. In the text, p. 33, it is
stated that 1.5 pounds of graphite is mixed with approximately
5000 pounds powder, that is 0.03% graphite. On p. 36, the table
showing ingredients of two double base powders, M2 and M5, the
graphite content is shown as 0.25%.
6) A stabilizer to prevent the formation of acids and thus reduce
deterioration of the powder and reduce the chances of auto-
combustion.
7) DNT is a deterrent used to reduce the burning rate of nitrocellu-
lose, highly explosive and sold with a maximum moisture content of
1%. DNT solubility in water is up to 150 ppm. When mixed into
the propellant (as opposed to coating the outside surface for
small arms powder), DNT reduces the hygroscopic properties of
nitrocellulose.
On p. 32, it is described that 65 pounds DNT is mixed with
1112 1/2 pounds of powder (probably moist powder), which would
mean that 5.8% DNT was mixed with powder for coating.
On one internet source, Wikipedia, it is stated that:
"Dinitrotoluene (toxic, carcinogenic, and obsolete)"
Centralites and dibutyl phthalate are shown as deterrents.
In Sharpe's Supplement, he mentioned that the old celluloid film was nitrocellulose
but that comment could not be located again but an interesting tidbit was found on
the Internet.
A search for "Using shredded celluloid film for gunpowder" turned up the Internet
site "http://repositories.lib.utexas.edu/bitstream/handle/2152/1915/frickd15921.pdf..."
from which the following quote was copied and pasted. (A shooting friend told of African
hunters using shredded celluloid film in a similar manner as stated below and in looking
for that information, the following quote was found, copied and pasted below.):
The belligerent tribes of the Khyber Pass area of India and also those of the
Kurdish region of Iran and Iraq…used to raid the local cinemas periodically and
cart off all the movie film on hand, which they would later shred up for
gunpowder. It worked fine, and put British patrols in the tragic-comic
predicament of being decimated by an early edition of Beau Geste or The Great
Train Robbery.16
The paragraph above is quoted from Caroline Jane Frick's Doctoral Dissertation:
Restoration Nation:
Motion Picture Archives and "American" Film Heritage
Additional information about the constituents of propellants M2 and M5 plus a single base IMR propellant.
----------------------------------Information within brackets, [ ], by HBC.
From Interior Ballistics of Guns, page 310, Table 2
Edited by
Herman Krier and Martin Summerfield
Progress in Astronautics
And Aeronautics
Martin Summerfield
Series Editor-in-Chief
Volume 66
ISBN 0-915928-32-9
Copyright 1979 by
American Institute of Aeronautics and Astronautics
Constituent--------------------M2----------------- M5-----------------IMR [1)]
Nitrocellulose (NC), %-----77.45---------------81.95----------------100.00
% nitrogen in NC-----------(13.25)--------------(13.25)-------------(13.15)
Nitroglycerin, %-------------19.50---------------15.00----------------…
Barium nitrate, %------------1.40-----------------1.40-----------------…
Potassium nitrate, %--------0.75------------------0.75-----------------…
Potassium sulfate, %--------…-------------------…-------------------1.00 a
Dinitrotoluene, %----------- …------------------- …------------------- 8.00 b
Diphenylamine, %-----------…-------------------…-------------------0.70
Ethyl centralite, %-----------0.60---------------- 0.60-----------------…
Graphite, %-------------------0.30---------------- 0.30-----------------…
Ethyl Alcohol(residual), %-----2.30-----------------2.30-----------------1.50
Water (residual), %--------- 0.70-----------------0.70-----------------1.00
Isochoric flame temp.
-------------- Tv, K----------3319-[5515 F]----3245 [5382 F]----2835 [4643 F]
Unoxidized carbon, %------0--------------------0---------------------2.7
Combustibles, %------------ 47.2-----------------47.4-----------------59.2
Heat of explosion, cal/g----1080 [4521 J/g]----1047 [4383 J/g]----868 [3633 J/g]
Density, g/cm3---------------1.65-----------------1.65-----------------1.62 [density of solid NC]
a) Added. b) Glaze added.
[Isochoric: A constant volume process as in a bomb calorimeter.]
[1) It was not stated which IMR propellant, (possibly IMR 4895)]
