Testing of industrial products - Aggregates for construction

Details of the cross-testing experiment on the Acid-soluble Sulfate test


15 laboratories were sent samples for the experiment, and 12 reported their results. It is rather surprising that as many as 3 out of 15 laboratories failed to report results: perhaps it would be advisable in future experiments in this programme to start with a larger number, say 20 laboratories.

All the laboratories were given numerical codes that were used in all the cross-testing experiments in the first year of the programme. The participants are able to identify their own results using these codes. For the purposes of this report they have also been assigned letter-codes (because single-character codes are needed for the histograms and Mandel plots).


Samples of three materials were prepared and distributed for the experiment by Partner 4. The materials were a concreting sand (Level 1), recycled concrete (Level 2) and a blast-furnace slag (Level 3). The reasons for choosing these materials were as follows. The sand was thought to have a relatively high sulfate content, high, that is, in comparison with other sands produced in the UK (but not higher than the upper limit in current specifications of 0.2% for a natural aggregate). The slag was known to contain a high sulfate content (above the upper limit in current specifications of 1% for slags). The recycled concrete had a sulfate content in between those of the other two materials.

A bulk sample of about 400kg of the concreting sand was obtained and placed in 28 ten litre plastic containers. A strip of polythene sheeting, two metres wide and about six metres long, was spread on a clean concrete floor. To mix the sand from the different containers together, sand from the first container was poured in a continuous line down the polythene sheet, sand from the next container was poured in a line parallel to the first, and the process was repeated with the remaining containers. The edges of the polythene sheet were pulled up to roll the sand into one line. The sand was then divided into 65 laboratory samples by the method of fractional shovelling described in prEN 932-2 (Proposed prEN 932-2 Tests for general properties of aggregates; Part 2 Methods for reducing laboratory samples to test portions. CEN/TC 154/TG 5 committee paper N 188. June, 1994.) using a flat-bottomed scoop made to have a capacity of about 200g of sand. Working from one end of the line of sand, 65 sampling increments were taken and each increment placed in a different polythene bag. This procedure was repeated for 30 cycles in all, so that at the end of the process each of the 65 bags contained about 6kg of sand.

A bulk sample of about 300kg of the blast-furnace slag, with a particle size range of 25/50mm, was obtained. To produce homogeneous laboratory samples from this material, it was first crushed in a jaw crusher to pass 20mm, then hand-sieved to extract the 4/14mm size fraction - the yield was about 160kg. This material was divided into 25 laboratory samples each weighing about 6.5kg by the same fractional shovelling method (and using the same scoop) as for the sand.

The participating laboratories were required to crush and grind the sand and the slag to produce sub-samples passing a 0.125mm sieve. In order that a comparison could be made of the precision of the test including this sample reduction with the precision of the analysis alone, the recycled concrete was used to prepare laboratory samples of material that passed the 0.125mm sieve. About 1kg of the recycled concrete was crushed to pass 5.00mm, then milled to pass 0.125mm. (The micro-Deval machine was used to do the milling, using a charge of five 400g steel balls and 2kg of 9.5mm diameter steel balls.) The powdered concrete was then put in a metal tray and levelled to a depth of about 20mm. Then a small spatula was used to transfer 30 increments, about 0.7g in mass, to each of 20 small plastic sample bottles, so that at the end of the process the bottles contained about 20g of material - the mass of material passing 0.125mm required by the test method. The increments were taken from positions randomly-distributed over the surface of the powder in the tray.

15 laboratory samples of each material were used for the cross-testing experiment. The remaining 50 laboratory samples of the sand have been kept for use as a reference material, or for further research on the test method if needed. The remaining 5 laboratory samples of the recycled concrete (and the unused passing 0.125mm material), and the remaining 10 laboratory samples of the slag, have been kept in case re-tests, or further tests, are required.

With the sand and the slag used for Levels 1 and 3, the participants were asked to divide the laboratory samples into two test portions, and then crush and grind the aggregate to produce a sub-sample of about 20g of passing 0.125mm material from each test portion. They then determined the acid-soluble sulfate content of two 2g specimens from each 20g sub-sample. The draft European Standard for chemical tests on aggregates (Draft prEN 1744-1 Tests for chemical properties of aggregates. Part 1 Chemical analysis. November, 1994.) contains a general requirement that two determinations should be carried out when an aggregate is tested. If it is assumed, therefore, that a "test result" is the average of the determinations obtained by analysing two 2g specimens from the same 20g sub-sample, then differences between such "test results" as obtained in the cross-testing experiment are consistent with the definitions of repeatability r1 and reproducibility R1.

With the recycled concrete, because the original samples were milled to pass 0.125mm before dividing into laboratory samples, the participants were not required to carry out the sample reduction process so it is reasonable to assume that the laboratory samples of this material were practically identical. Hence the measures of repeatability and reproducibility given by the cross-testing experiment at Level 2 are consistent with the definitions of r and R. The repeatability and reproducibility measures for Level 2 are calculated to apply to "test results" that are the average of two determinations - to be consistent with the results for Levels 1 and 3.

In practice, it is convenient to have a repeatability limit that can be compared with the difference between determinations on two specimens from the same sub-sample of passing 0.125mm material: this limit is referred to in this report as the "critical range", and values of the critical range are calculated from the data for all three levels.


When the concreting sand used for Level 1 was tested, prior to the preparation of the laboratory samples, it gave an acid-soluble sulfate content of 0.15% SO3. However, the results reported by the participants give an average of 0.06% SO3. The reason for this difference is being sought.

The method for the determination of acid-soluble sulfate does not contain any indication of how the test results should be rounded, apart from a general requirement in the document (covering all its test methods) that the mean of two determinations should be rounded to the nearest 0.01%. However, for the purpose of the cross-testing experiment, it was asked that the test results should be reported to the nearest 0.001%. This was to prevent rounding of the data affecting the assessment of the repeatability and reproducibility of the test method.

Averages and ranges

Laboratory averages are used to calculate the reproducibility of the test method, and to look for biasses that influence the results of a laboratory at more than one level.

Between-specimen, between-test-portion, and between-sub-sample ranges are used to calculate the repeatability of the test method, and to assess the repeatability of tests from individual laboratories.

The averages and ranges are shown in the histograms, and the averages are shown in the Mandel plots.

The averages and ranges are also used to test for stragglers and outliers. Where these have been found, they are indicated throughout using a single question mark (?) to indicate a straggler, and a double question mark (??) to indicate an outlier.

The Mandel plots are used to see if any laboratory suffers from biasses that affect the results at more than one level. Thus, for example, if a laboratory uses a faulty procedure, or a standardised solution that has been prepared incorrectly, then this could cause it to obtain results that are biassed in the same direction in all levels.