ACARP Project Number: C19053      Published: July 12

Note project was published in two parts-

The force that a coal exerts on the coke oven walls can have deleterious effect on coke batteries if it is excessive. The most common method of controlling oven wall pressure is blending. However predicting the expected oven wall pressure (OWP) is complex as there is no one coal property that is known to directly indicate OWP. Additionally, the oven design and charging conditions such as bulk density are important.

Previous research has demonstrated that the properties of the plastic layer influence both OWP and the structure of the final coke, thus coke strength. The plastic layer properties can be assessed by a Plastometer test that measures the thickness (Y, mm) of the plastic layer and the degree of contraction (X, mm) that occurs when a coal sample is heated unidirectionally. There are two national standards that describe this test (Russian & Chinese), in those countries the plastic properties are used in the classification of coals. This test is known as the Sapozhnikov test.

In Part 1 of this project the influence of bulk density on contraction and plastic layer thickness was studied using an automated commercial Sapozhnikov apparatus. In Part 2, a number of parameters affecting the test results of the Sapozhnikov and Chinese Plastometer test were considered and the repeatability of these tests assessed. Recommendations are made in respect of technical factors that should be considered in any future ISO standard

The Edinburgh Cohesion Tester (ECT) was used to investigate the influence of size distribution, moisture content and coal properties on bulk density. There was good correspondence between the increased bulk density caused by a consolidation load determined in the ECT and the Sapozhnikov apparatus. However, it was found difficult to predict the loosely packed bulk density of different coals based just on their size distribution and coal properties and, as the bulk density after application of a consolidation load is related to the loosely packed bulk density, that also cannot be predicted based on coal properties.

It was found that the bulk density of the packed coal has an influence on the contraction and the thickness of the plastic layer in the Sapozhnikov retort. This influence was different for each coal. There is also evidence that the bulk density influences the heat transfer from the base of the retort thus impacting on the size of the region that is undergoing plastic deformation. Again this influence may be dependent on the coal properties. As found by others the contraction as measured by the Sapozhnikov test is a good indication of the coking pressure that may be generated by a coal.  

The effect of automation was found to be minimal, with four out of the five coals tested having results within the published repeatability, assuming that the plastometer needle is aligned correctly at the beginning of the automated test.

Coal tested at 5% moisture had increased contraction compared to coal tested at air dried moisture, the magnitude of the increase in contraction varied with each coal (due to the varying effect of moistures on bulk density), plastic layer thickness was also affected by moisture but this effect was also inconsistent.  

Due to the effect that moisture and bulk density have on the results of the plastometer test it is recommended that the moisture level that the test is conducted at be specified and the packing bulk density be defined, additionally the final bulk density (after the load has been added) that the test is conducted at should be recorded.

For a coal tested at various intervals over a one month period with one subsample stored at ambient temperature and another subsample kept in cold storage there was no significant difference in the measured contraction. For samples stored for an extended period of time in cold storage there was also no significant change however a sample that had been stored at ambient temperature for an extended period showed a significant increase in contraction.

The measurement uncertainty of the automated plastometer method was investigated experimentally. It was found that for contraction the uncertainty (u) was 2.28mm and the expanded uncertainty (U) was 4.57mm. For plastic layer thickness the uncertainty (u) was 1.33mm and the extended uncertainty (U) was 2.66mm. These numbers are higher than those published in the standards, possible reasons for this are discussed.