Page 53 - New Trends In Coal Conversion

Coal 21

by measuring the fluidity of a packed coal powder as it is heated, whereas the Audiberte
Arnu dilatometer measures the contraction and expansion of a powdered sample pressed
into a cylindrical coal “pencil”. Such properties are significant to classify coals and of
great relevance when different coals are to be blended for coke production to ensure
compatibility of the different blend components. Indeed, coal-blending strategies for
coke production are generally decided from a combination of rheologic and petrographic
parameters for individual coal samples, which are used to select coals to make up a blend
with specific coking properties.
• Sapozhnikov test. This test was developed by Sapozhnikov and Bazilevich in Russia in
1931 and adopted as a Russian standard (GOST 1186-876) to measure the coking prop-
erty of the coal. With some minor differences, China has adopted the same test (GB/T
479). The plasticity of bituminous coals is defined by the maximum plastic layer thickness
(Y index) and the contraction of the resultant coke when the coal is heated unidirection-
ally up to a final temperature of 650 C. In Chapter 9, a description of the test and its po-

tential as a guide for blending and cokemaking is given.
• G caking index test. This test was developed in China and is performed in accordance with
GB/T 5447 and ISO 15585 standards. It is somewhat similar to the Roga test developed in
Poland. It evaluates the degree of cohesion of a resulting coke in a drum tumbler when

coal is subjected to rapid heating rate up to 850 C and mixed with a fixed amount of
a specific anthracite as an inert. The test is applicable for evaluating bituminous coal
with a vitrinite random reflectance in the 0.6%e1.8% range. Values typically range
from 20 to >100, where >85 is the desired value.
2. Tests related to coking pressure control. When coals are carbonized at high temperature in
slot-type ovens, two principal plastic/fluid layers of coal are formed parallel to the oven walls,
which are linked near the bottom and the top of the charge by two secondary plastic/fluid
layers. This results in a closed envelope with an inner zone composed of layers of dry and
untransformed coal and an external zone composed of layers of semicoke and coke. As
the carbonization proceeds, the coal plastic/fluid layers form a front, which moves progres-
sively inward toward the center of the coke oven, leaving behind the coal transformed into
semicoke and coke. The meeting of the plastic/fluid layers at the oven center marks the
end of the transformation of coal to semicoke/coke. It is thought that during the process,
the volatile matter that evolves as a result of the thermal decomposition of the coal macro-
molecular structure tries to escape through the different layers formed in coke ovens. The
resistance to the passage of the volatile products generates an internal pressure against the
oven walls, which is known as coking pressure. If the generated pressure is too high, the
coking charge causes operational difficulties during coke pushing and coke oven wall dam-
age, thereby shortening the life of the coke oven. Coking coals of this type, namely dangerous
coals, are not carbonized individually but are used as components in industrial coking blends
to adjust the volatile matter and fluidity of the blend and to improve coke mechanical
strength. Low- and medium-volatile coals with similar chemical, caking, and swelling char-
acteristics can behave differently in their ability to the pressure generation. It is therefore
necessary to be able to predict and assess the danger of a coal and keep the coking pressure
below certain limits to prolong the life of the coke oven (Pajares and Díez, 2014).
Two basic methods of approach can be undertaken to determine the suitability of a
coal or coking blend: (1) indirect coking pressure measurements at laboratory scale us-
ing the Koppers and Sole heated-oven tests (SHO test) and (2) direct coking pressure
measurements by applying larger scale tests (movable-wall ovens).