25.04.2013 Views

the coking properties of coal at elevated pressures. - Argonne ...

the coking properties of coal at elevated pressures. - Argonne ...

the coking properties of coal at elevated pressures. - Argonne ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

3. Bed m<strong>at</strong>erial composition (high calcium content tends to delay, and decrease<br />

<strong>the</strong> severity <strong>of</strong> agglomer<strong>at</strong>es formed)<br />

4. Ash recycle (increased<br />

tendency)<br />

recycle <strong>of</strong> ash tends to increase agglomer<strong>at</strong>ion<br />

5. There appears to be a bed design parameter such as position <strong>of</strong> <strong>coal</strong> feed<br />

points, and distributor pl<strong>at</strong>e performance which affect bed m<strong>at</strong>erial agglomer<strong>at</strong>ion.<br />

The bed m<strong>at</strong>erial used in baseline run 2181 was 10 mesh quartz sand. Upon<br />

startup <strong>the</strong> bed was sampled every eight hours for <strong>the</strong> dur<strong>at</strong>ion <strong>of</strong> <strong>the</strong> 69 hour run.<br />

At <strong>the</strong> end <strong>of</strong> <strong>the</strong> run <strong>the</strong> system was cooled and opened to expose <strong>the</strong> inside <strong>of</strong> <strong>the</strong><br />

combustor. The bed m<strong>at</strong>erial was removed and several agglomer<strong>at</strong>es were found,<br />

which varied in size and shape, with <strong>the</strong> largest having a diameter <strong>of</strong> 6 cm. These<br />

agglomer<strong>at</strong>es were found both free flo<strong>at</strong>ing in <strong>the</strong> bed and <strong>at</strong>tached to <strong>the</strong> inside<br />

wall <strong>of</strong> <strong>the</strong> combustor.<br />

The limestone bed m<strong>at</strong>erial was tested in run 2281 using 100% ash reinjection<br />

for a dur<strong>at</strong>ion <strong>of</strong> 73 hours. The bed was sampled in <strong>the</strong> same manner as 2181.<br />

The form<strong>at</strong>ion <strong>of</strong> agglomer<strong>at</strong>es in <strong>the</strong> run was very minimal with no major agglomer-<br />

<strong>at</strong>es found. On <strong>the</strong> o<strong>the</strong>r hand, run 2481 which used a limestone bed ran for 160<br />

hours with 100% ash reinjection and had severe problems with agglomer<strong>at</strong>ion. After<br />

<strong>the</strong> run numerous agglomer<strong>at</strong>es were found loose in <strong>the</strong> bed. In addition, a large<br />

agglomer<strong>at</strong>e was found on top <strong>of</strong> <strong>the</strong> distributor pl<strong>at</strong>e <strong>at</strong> <strong>the</strong> bottom <strong>of</strong> <strong>the</strong> combus-<br />

tor. The agglomer<strong>at</strong>e had dimensions <strong>of</strong> 30.5 cm X 30.5 cm X 12.5 cm; it weighed<br />

10 kg and covered 30% <strong>of</strong> <strong>the</strong> distributor pl<strong>at</strong>e.<br />

The bed m<strong>at</strong>erial and agglomer<strong>at</strong>es were characterized by polished thin section<br />

study, polarized light microscopy, and scanning electron microscopy/microprobe<br />

(SEM) - both secondary electron (SEI) and backsc<strong>at</strong>ter electron (BEl) images were<br />

used. Bulk samples were analyzed by x-ray diffraction and x-ray fluorescence.<br />

The goals in characterizing <strong>the</strong> bed m<strong>at</strong>erial and agglomer<strong>at</strong>es are to identify <strong>the</strong><br />

stages which lead to agglomer<strong>at</strong>ion and possibly postul<strong>at</strong>e a mechanism <strong>of</strong> <strong>the</strong>ir forma-<br />

tion to <strong>the</strong>reby determine methods and procedures to control <strong>the</strong>ir growth.<br />

Quartz Bed Agglomer<strong>at</strong>es<br />

RESULTS AND DISCUSSION<br />

Agglomer<strong>at</strong>ion <strong>of</strong> quartz bed m<strong>at</strong>erial is typified by run 2181 utilizing high-Na<br />

Beulah lignite and ash injection. Samples <strong>of</strong> bed m<strong>at</strong>erial taken <strong>at</strong> various intervals<br />

during <strong>the</strong> run are illustr<strong>at</strong>ed in Figures 1 to 11 with chemical analyses d<strong>at</strong>a given<br />

in Table 3. The following four stages can be used to summarize <strong>the</strong> agglomer<strong>at</strong>ion<br />

process :<br />

Stage 1. Initial ash co<strong>at</strong>ing.<br />

Initial samples <strong>of</strong> bed m<strong>at</strong>erial have a fine co<strong>at</strong>ing, about 50 microns thick,<br />

consisting <strong>of</strong> sulf<strong>at</strong>ed aluminosilic<strong>at</strong>e particles (Figure 1). The co<strong>at</strong>ings contain some<br />

coarser ash m<strong>at</strong>erials in <strong>the</strong> outer parts and <strong>the</strong> inner parts have penetr<strong>at</strong>ed <strong>the</strong><br />

quartz grains slightly along gently curved or cusp<strong>at</strong>e embayments. The quartz<br />

grains are extensively fractured, apparently as a result <strong>of</strong> <strong>the</strong>rmal stresses.<br />

Stage 2. Thickened nodular co<strong>at</strong>ings.<br />

Longer bed usage results in <strong>the</strong> development <strong>of</strong> thicker ash co<strong>at</strong>ings about<br />

100 - 300 microns thick with nodular outer surfaces resulting from incorpor<strong>at</strong>ion <strong>of</strong><br />

larger ash particles (Figure 2). Sulf<strong>at</strong>ing, shown by lighter colored areas in <strong>the</strong><br />

SEM photographs, is common within both <strong>the</strong> finer and coarser ash particles <strong>of</strong> <strong>the</strong><br />

co<strong>at</strong>ing.<br />

176

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!