GUIDELINES FOR THE DETERMINATION OF OXIDIZING PROPERTIES
(SUPPLEMENT TO A 17 METHOD)

 

1. Applicability

Before performing test A 17 for the determination of oxidizing properties, it is important to check that this test is applicable to the substance.

Test A 17 is not applicable to liquids, explosive or highly flammable substances, it is therefore useful to perform tests A1, A 10 and A 14 before performing A 17.

Test A 17 cannot be applied to organic peroxides: these substances are classified as oxidizing on the basis of their structure.

Test A 17 need not to be carried out when examination of the structural formula establishes beyond reasonable doubt that the substance has no oxidizing properties.

Compounds which have no highly electronegative atom - oxygen, fluorine, chlorine, bromine - are not likely to possess oxidizing properties. Similarly, where these elements are present but the atoms are only bonded to carbon and/or hydrogen, then oxidizing properties are unlikely.

A substance may have oxidizing properties when:

• the electronegative atoms which are present constitute a high proportion of the molecule and are bound to elements in a high oxidation state;
• the electronegative atoms are bonded to each other or to other electronegative elements such as iodine, nitrogen, sulphur or phosphorus.

As the ability to predict the reactivity of chemicals compounds from their structure is still limited, the best approach is by analogy with existing compounds. Attached in Annex 1 is a list of oxidizing compounds and reactive groups which increase the oxidizing power. However this list is not exhaustive. If the substance meets one of the above criteria, the lack of any reactive group named in Annex 1 may not be sufficient to justify not performing the A 17 test.

For organic substances only, the oxygen balance (OB) calculation may be useful as a criteria combined with an examination of the chemical structure as a means of predicting oxidizing properties.

For an organic substance CaHbNcOdSeXf (X = halogen), of molecular weight M, the OB is calculated as follows:

OB = (d - 2a - 2e - 0,5 (b - f))	1600
	M

Although experts think that the OB calculation is a possible approach, there is currently no consensus on the criteria that should be used. For the moment, professional judgment will be required to use the OB value on a case by case basis. In any case, if theoretical considerations are used to exclude performance of the test A 17, the reasons and all relevant information should be clearly stated in the technical dossier.

2. Performance of the test A17

When the test is applicable and the theoretical approach is inconclusive, it is necessary to perform the test with cellulose according to the principles described in the A 17 method. The following quality criteria need to be observed:

1. the reference mixture which gives the maximum burning rate should contain about 50 to 70% of barium nitrate.
2. the reference mixture should have a maximum burning rate greater than 1 mm/s.
3. the use of hot wire ignition must be fully justified (the standard test requires the use of a gas flame). Therefore if a hot wire is used as ignition source, its surface temperature should be at least 1000^o C (colour indication) and sufficient area should be in contact with the test mixture (temperature and contact area should be specified in the test report) : if no propagation is observed, the point of contact of the wire with the pile should be examined to ensure that the mixture has not melted and flowed away from the ignition source.

If these criteria are not fulfilled, the test conditions should be checked (e.g. try another cellulose type, grinding and mixing method, re-dry materials) and the procedure should be repeated.

A substance is to be considered as an oxidizer when:

1. there is a vigorous reaction in the preliminary test; or
2. during the test, the fastest burning rate of the test mixtures is greater than or equal to the fastest burning rate of the reference mixture.

If the result is suspected of being a false positive (e.g. when the substance melts or decomposes at a rather low temperature or when the fastest burning rate of the substance/cellulose mixture is obtained for a proportion between 10/90 and 50/50), the test should be repeated using one, or if it is considered necessary, both of the procedures below:

• Using an inert substance, such as Kieselguhr, instead of cellulose: there is little likelihood that the test substance is an oxidizer if any of the test substance/inert material mixtures exhibit significant combustion.
• A test with cellulose in an inert atmosphere: using the mixtures which had the fastest burning rates by the standard procedure but it may be necessary to test mixtures with a range of compositions: the test substance is, in all likelihood, an oxidizer if the burning rate of a mixture with cellulose in this test is greater tan or equal to the burning rate of the fastest burning reference mixture tested in the same inert atmosphere.

The usefulness of the supplementary tests with kieselguhr and under an inert atmosphere are recognised but they have not been validated. Description of the protocols are available in Annexes II and III.

ANNEX I

Nitrates (salts or esters)	NO3- M+	02N-0-R
Nitrites (salts or esters)	NO2- M+	ON-0-R
Organic nitro compounds
Nitroalkyl	NO2-R
Nitroaryl	NO2-Ar
Fluorodinitro	F(NO2)2-C-
Metal oxides	MOn
Metal oxometallates	M+ MOn-
N-Halogen compounds	N-X
N-Haloimides	-C(O)-NX-C(O)-
Difluoroamino	- NF2
Difluoroaminopolynitroaryl	(NO2)n-Ar-NF2
Oxohalogen compounds:
Acyl hypohalites	RC(O)-OX
Hypofluorites	FO-
Bis(fluoroxy)perhaloalkanes	F3CCI(OF)2 etc
Perchlorates	CIO4-
Chlorates	CIO3-
Chlorites	CIO2-
Hypochlorites	CIO-
Perbromates	BRO4-
Bromates	BrO3-
Bromites	BrO2-
Hypobromites	BrO-
Periodates	I04-
Iodates	I03-
Difluoroperchloryl salts	F2CIO2+ Z-
Dioxygenyl polyfluorosalts	O2+ [MFn]- or 02+ [EFn]-
Interhalogen compounds:
Metal polyhalohalogenates	M+ [XX'n]-

ANNEX II

TEST PROCEDURE USING KIESELGUHR

1. PROTOCOL

The test protocol is almost exactly the same as that used in the standard A 17 test procedures.

The test substance and kieselguhr are dried to constant weight at 105^o C. If the substance decomposes at a temperature below 105^o C, it should be dried at the same temperature at which it was dried for the original A 17 test with cellulose. The dried substance is sieved (mesh size 125 µm). The fraction that is retained on the sieve is pulverized, and the sieve procedure is repeated until the complete test portion has passed through the sieve. The kieselguhr is sieved and ground in the same way as the test substance.

Nine mixtures of kieselguhr and test substance are prepared containing from 10 to 90 per cent by weight of kieselguhr in 10 per cent increments.

The standard mould, heat resistant plate and ignition system are employed, together with the normal procedures for filling the mould and forming the pile.

The test is carried out at least once with each of the nine mixtures of substance/kieselguhr. This test series suffices if any propagation rates are observed. The propagation rates observed should be compared with the propagation rates of the substance/cellulose mixtures in the original A 17 test.

If the test with the inert material is not conducted at the same time as the original A 17 test, a repeat test on the substance/cellulose mixture which gave the fastest burning rate in the original A 17 test should be conducted at the same time as the test with kieselguhr is performed; this is to confirm that the burning rates do not differ significantly from the original values. If there is significant disagreement the full A 17 procedure should be repeated.

2. INTERPRETATION OF THE RESULTS

It is confirmed that the material possesses oxidizing properties if there is no propagation of any reaction through any of the mixtures of the test substance and kieselguhr.

The material may be considered not to possess oxidizing properties and thus the results from the A 17 test may be considered a "false positive" if:

1. the mixtures of the test substance with both cellulose and kieselguhr show similar (but not necessarily identical) propagation rates; and
2. the two series of tests show a similar pattern of change in the rate of propagation as the composition of the mixtures is altered.

It is not possible to draw any conclusions as to the oxidizing properties of the test substance if a reaction does propagate on the kieselguhr but the above conditions are not met. In this case, the test with inert atmosphere should be performed.

ANNEX III

TEST IN AN INERT ATMOSPHERE

1. PROTOCOL

The tests under an inert atmosphere should be performed on the test substance/cellulose mixtures and the reference/cellulose mixtures which had the fastest burning rates in the original A 17 test.

However, as the composition of the mixtures with the fastest burning rate in the inert atmosphere may differ from that with the fastest rate in air, it is recommended that a range of sample/cellulose and reference/cellulose mixtures are tested in order to find the sample mixture and reference mixture with the fastest propagations rates.

The only differences between this procedure and the full train test are that:

1. the test is conducted in an atmosphere having less than 2% oxygen instead of 20.9%, and
2. the ignition source is a hot wire instead of a hot flame from a gas burner (as highlighted for the performance of the original test A 17, it is necessary to ensure that the mixture does not melt and flow away from the wire before ignition can occur).

In order to attain the inert atmosphere, one proceeds as follows:

• an enclosure with the same dimensions as shown in figure 1 is connected to an inert gas source (for example, nitrogen or carbon dioxide). The inert gas, which must first be dried, enters the enclosure as indicated in figure 1.
• The inert gas flow rate is set to a suitable level in order to lower the oxygen content to less than 2% within a reasonable time: for example, a nitrogen flow rate of 3 litres/minute (180 l/hr) is sufficient to reach the required oxygen level in about one hour. The oxygen content is verified in the gas exit line of the chamber by means of an oxygen meter, detector, or analyser with direct readout.

The inert gas flow rate is set to 3 litres/minute and maintained at this level throughout the experiment.

One end of the pile is ignited by the hot wire.

Measure the time of reaction over 200 mm after the reaction zone has propagated an initial distance of 30 mm. If this is not possible, measure the maximum distance covered by the reaction zone and the corresponding time. If the fumes generated prevent observation of the progress of the reaction front down the train, thermocouples may be used. In all cases, calculate the burning rate.

If a test mixture gives no reaction, at the end of the test the portion of the train originally in contact with the hot wire should be examined to ensure that contact has been maintained. Failure to ignite may arise from the sample melting and flowing away from the wire.

If problems are experienced with limited propagation of the reaction front along the reference mixture train in an atmosphere of less than 2% oxygen, then it may not be possible to assess the oxidising properties of the material under test. Under these circumstances the procedure may be repeated under an atmosphere containing approximately 10.5 ± 1% oxygen. The increased oxygen content should then be sufficient to allow propagation along a train containing an oxidising substance, but low enough to suppress any effects due to flammability of the sample. such an atmosphere may be generated by using a 1:1 mixture of air and inert gas in place of pure nitrogen or carbon dioxide.

2. INTERPRETATION OF THE RESULTS

A substance is to be considered as an oxidizer when in the test in an inert atmosphere, the fastest burning rate for a mixture with cellulose is greater than or equal to the fastest burning rate obtained with a reference mixture in the same inert atmosphere.

Figure 1:

Cover, sides and bottom of the enclosure : stainless steel
- Front : transparent plastic (polycarbonate for example)
- Pile test plate : the same as in the test air
- Hot wire : nichrome wire for example
- Removable cover