The following is from: FBI, "Further observations on the Diphenylamine test for gunpowder residue." Law Enforcement Bulletin, 9:10-14 (1940)
In the October, 1935, issue of the FBI Law Enforcement Bulletin there appeared the results of experiments using the diphenylamine test for the purpose of detecting nitrates from the hands of persons who had preciously fired a revolver or automatic pistol and also of persons not having fired a weapon of any kind. The analysis is based upon the theory that the nitrates contained in gunpowder form oxides of nitrogen at the time of the explosion which become embedded in the skin, particularly in those areas of the hands which have been exposed to the escape of gases from the weapon at the time of the discharge. It is also believed that the nitrates unite with the salts in the perspiration to form salts which readily react with the reagents involved in the tests.
In that article the following tests were described as being illustrative of the usual results obtained with the diphenylamine reaction. Tests were conducted on both the firing and idle hands of seventeen men following a course of five shots with .38 caliber revolvers using smokeless powder. Six of these individuals, all firing with the right hand, showed negative results when the test was applied to the right and left hands. One man who fired with his left hand showed negative results on both hands. Three of the persons who fired with the right hand showed negative results on a test of the right hands, but a positive test on the left hands which were not used for firing. One man who showed a negative result on the right hand when firing with that hand, gave a doubtful result on the left hand. Two men firing with their right hands showed positive results on the right hand and negative results on the left hand. Four men showed positive results on both hands when firing with the right hand only. In addition to the above experiments there were also published in that issue the results of experiments conducted in the Bureau's Technical Laboratory directed toward the effect of tobacco residue on the hands of smokers. As a result of these tests it was found that persons smoking any considerable amount gave positive reactions to the diphenylamine test applied to casts of the fingers and palms. These results are expected, of course, in view of the presence of nitrates in the products of tobacco prepared for smoking purposes.
Recently there have been additional experiments conducted in the Technical Laboratory of the Federal Bureau of Investigation which corroborate the unreliability of the diphenylamine test as a test for gunpowder residue. These tests were conducted upon employees of the Federal Bureau of Investigation some of whom were members of the laboratory staff and others who had no connection with the laboratory.
In all of the tests in which a gun was fired, cartridges were used which contained smokeless powder. Care was exercised to prevent any contact of the idle hand with the gun in those instances where only one hand was used to fire.
The following is a description of the technique employed in making and removing the paraffin casts.
Clean paraffin, free of any nitrites or nitrates, with a low melting point, was melted and carefully poured over the hands from the wrists to the fingertips. This was continued until a considerable thickness of paraffin had been deposited over the entire surface of the hands both front and back. A two and one-half to three inch strip of white cheese-cloth was wound about the hands in contact with the warm paraffin and additional paraffin was then poured over this, the cheesecloth acting as a reenforcing material and keeping the casts intact after removal from the hands. As soon as the paraffin was set, scissors were employed to cut from the wrist to the tip of the thumb on one side of the cast and from the wrist to the tip of the little finger on the other side of the cast. The cast was then opened in halves with the tips of the remaining three fingers acting as a hinge. The opened casts were placed on a glass plate and a solution of the following was then added to the interior of the casts; diphenylamine ½ gram, sulphuric acid C.P.(free from nitrates), 100cc, and 20cc of distilled water.
The presence of nitrates is indicated by the appearance of blue colored specks or areas on the paraffin. This color reaction is a normal one for nitrates in the presence of diphenylamine and sulphuric acid. However, as will be stated later, a positive reaction obtained with the diphenylamine test cannot be interpreted as one which is conclusive for the presence of gunpowder inasmuch as any nitrate would react similarly, producing the distinct blue color reactions where specks of nitrate are present.
It was, of course, desirable to conduct parallel control tests in order to be certain that the paraffin itself did not contain nitrites or nitrates. This was done by pouring a small amount of melted paraffin from the same batch that was to be used in making the casts of one individual's hands, onto a clean glass plate and allowing it to cool. A test solution was then applied to the upper surface of the paraffin. The absence of the characteristic color changes indicated that the paraffin itself did not contain nitrites or nitrates.
There are set forth on the following page the results of experiments in tabulated form. Where a positive reaction was obtained the symbol(+) is used, and where a negative was obtained the symbol(-) is used.
RESULTS OF SHOOTING WITH BOTH HANDS Person # Right Hand Left Hand 1. (revolver used) + + 2. (revolver used) + + 3. (automatic pistol used) + + RESULTS OF SHOOTING WITH RIGHT HAND ONLY Person # Right Hand Left Hand 4. (revolver used) + + 5. (automatic pistol used) + + 6. (automatic pistol used) + + 7. (automatic pistol used) + + RESULTS OF SHOOTING WITH LEFT HAND ONLY Person # Right Hand Left Hand 8. (revolver used) + + 9. (automatic pistol used) + + RESULTS OF TESTS ON INDIVIDUALS WHO DID NOT SHOOT OR HANDLE FIREARMS OF ANY KIND Person # Right Hand Left Hand 10 + + 11 + + 12 - + 13 + + 14 + + 15 + + 16 + - 17 + + 18 + + 19 + + 20 + - 21 + + 22 + + 23 + + 24 + + 25 + + 26 + + 27 + + 28 + - 29 + +As will be seen from an analysis of the tables shown, false reactions for gunpowder may be expected as often as true actions. Strictly speaking these cannot be termed false reactions since they are undoubtedly caused by the presence of some nitrate and as stated before the diphenylamine test does not distinguish among them.
In order to more fully understand the limitations of the test, further explanation is believed desirable.
In those instances where a positive reaction has been obtained by means of the diphenylamine test upon casts of the hands of a person, the presence of these nitrates may be due to the handling of one of many different types of substances which contain nitrites or nitrates. Some of the more common compounds which contain nitrites or nitrates are set forth below. These substances will, of course, produce a positive reaction when subjected to diphenylamine and sulphuric acid as carried out in the diphenylamine test:
It is obvious, therefore, that individuals during the natural course of various occupations would be expected to have on their hands certain substances which would give the positive reaction with the diphenylamine test. For example, miners who handle explosives, farmers or home gardeners using fertilizers, employees in nitrate plants or in the handling of nitrates, employees in the fertilizer industry, persons manufacturing or handling fireworks and in some instances chemists and pharmacologists who have had occasion to handle some of these materials. Likewise persons who smoke cigarettes or other forms of tobacco which contain potassium nitrate, put there for the purpose of keeping the tobacco burning, are very apt to give positive results with this test.
In addition to the above, it is well to realize the limitations of the test which are due to the fact that in a great many instances a revolver or pistol may be fired without leaving any trace of gunpowder which might be detected by this test on the hand of the person firing the weapon. The automatic pistol, of course, is almost completely enclosed and practically all powder particles are forced away from the hand of the person firing the weapon. The revolver, on the other hand, has considerable space between the chamber and the barrel which will permit a greater quantity of gas and other products of the explosion to be freed near the hand of the person firing the gun. In spite of this possible escape of gas and powder particles, test have indicated that a person may be found to give a negative reaction to the diphenylamine test even though he has fired as many as forty rounds of ammunition. Consequently, the mere failure to find an indication of gunpowder by the diphenylamine test on the hands of a deceased person does not prove that he did not fire the weapon causing his death nor would a negative reaction of the diphenylamine test on paraffin casts made of the hands of a suspect prove conclusively that the suspect could not have fired the fatal shot.
Despite the fact that a positive reaction may have been obtained with the diphenylamine test on a paraffin cast made of an individual's hand, which reaction was due to the presence of gunpowder particles present in the skin, such a positive reaction could not conclusively determine that this individual fired the weapon in question to the exclusion of all other weapons. Even though the test were accepted as conclusive, or if substantiation were possible, such as the finding of partially burned or unburned particles of powder, it would still not be possible to show that the individual fired the weapon in question to the exclusion of all other weapons. Also there must be considered the possibility that gunpowder residue on the hand of a person might originate from the muzzle of a gun pointed at him or fired near him at close range.
In order to determine whether it is possible to differentiate various nitrate containing compounds on the basis of their reactions with the dipheylamine testing solution, the following compounds were powdered upon paraffin covered glass plates and the testing solution then allowed to come in contact with these compounds. The following is a list of the compound tried:
No differentiating features were discernible either on the basis of the shape of the specks or the blue color reaction resulting.Ammonium nitrate