25.10.2012 Views

Liquefaction co-processing of coal shale oil at - Argonne National ...

Liquefaction co-processing of coal shale oil at - Argonne National ...

Liquefaction co-processing of coal shale oil at - Argonne National ...

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.

EXPERIMENTAL<br />

Wyodak <strong>co</strong>al was pretre<strong>at</strong>ed and tre<strong>at</strong>ed with steam in a semi-Row micro reactor <strong>at</strong> <strong>co</strong>ntrolled <strong>co</strong>ndi-<br />

tions to exclude oxygen. The appar<strong>at</strong>us is depicted in Figure 1. The main <strong>co</strong>mponents are a Milton Roy<br />

metering pump (29 to 290cm3/hr) which provides a <strong>co</strong>ntinuous and <strong>co</strong>nstant Row <strong>of</strong> distilled deoxy-<br />

gen<strong>at</strong>ed w<strong>at</strong>er to the re<strong>at</strong>or, a Tecam fluidized sand b<strong>at</strong>h (model SBS-4) <strong>co</strong>ntrolled with a Leeds and<br />

Northrup <strong>co</strong>ntroller (Electromax Ill), a Heli<strong>co</strong>id pressure gauge (0 to 3000 PSIG), an autoclave micro<br />

metering valve and a Picro reactor equipped with a she<strong>at</strong>hed thermo<strong>co</strong>uple (see Figure 2).<br />

The feed m<strong>at</strong>erial was an unwe<strong>at</strong>hered subbituminous Wyodak <strong>co</strong>al provided by EPRI in a w<strong>at</strong>er<br />

slurry kept in a sealed air-tight barrel (#4171). Table 1 shows the elemental analysis <strong>of</strong> samples taken<br />

from this barrel. Samples A and B were taken from the top and middle <strong>of</strong> the barrel and were suction<br />

dried for use. These were the main feed m<strong>at</strong>erial. About a 50 gm sampled <strong>of</strong> this m<strong>at</strong>erial was removed<br />

from the barrel and vacuum dried; a sample <strong>of</strong> this m<strong>at</strong>erial is repoted as sample C. A tew reactions<br />

were carried out with this m<strong>at</strong>erial.<br />

The suction dried <strong>co</strong>al was prepared in a Buchner filter, by applying suction to it tor about one hour.<br />

The top portion <strong>of</strong> the m<strong>at</strong>erial was removed from the filter and thoroughly mixed to insure a homogeneous<br />

sample for reaction. The moisture <strong>co</strong>ntent <strong>of</strong> this m<strong>at</strong>erial (z 30 wt %) was determined by weighing<br />

a sample before and after further drying in a vacuum oven <strong>at</strong> 45 O c for 6 hours. Duplic<strong>at</strong>e samples<br />

were used for moisture determin<strong>at</strong>ions.<br />

Suction dried Wyodak <strong>co</strong>al is paste-like, and is messy to bandle. A free-Rowing <strong>co</strong>al powder was<br />

obtained for reaction experiments by drying the <strong>co</strong>al in vacuum under mild he<strong>at</strong>ing. The vacuum dried<br />

<strong>co</strong>al wan prepared by suction drying in a Buchner funnel over nitrogen; the m<strong>at</strong>erial was then transferred<br />

in a jar 611ed with nitrogen, to a vacuum oven and dried for 72 hours <strong>at</strong> 45 O c. The experimental reac-<br />

tion procedure for vacuum dried <strong>co</strong>al was similar to suction dried <strong>co</strong>al except th<strong>at</strong> the separ<strong>at</strong>e moisture<br />

determin<strong>at</strong>ion <strong>of</strong> the dried <strong>co</strong>al was no longer necessary.<br />

To prepare for reaction the oven dried micro reactor (see Figure 2) was weighed and then filled with<br />

about 4 gm ot wet suction dried <strong>co</strong>al or 3 gm <strong>of</strong> vacuum dried <strong>co</strong>al and reweighed for an accur<strong>at</strong>e deter-<br />

min<strong>at</strong>ion <strong>of</strong> the quantity <strong>of</strong> m<strong>at</strong>erial charged to the reactor. The charged reactor was fitted into the<br />

appar<strong>at</strong>ns upon <strong>co</strong>nnecting tubings and fittings. The system was purged <strong>of</strong> air by Rushing with nitrogen.<br />

Enough w<strong>at</strong>er was pumped from the buret into the reactor to raise the pressure to SO pia, about 10 to<br />

nCm3 being required. The reactor was lowered into the sand b<strong>at</strong>h, the pressure adjusted to 750 psia<br />

with the aid <strong>of</strong> the micro metering valve, as w<strong>at</strong>er was pumped into the system <strong>at</strong> 0.4cm3/min. The<br />

steam gener<strong>at</strong>ed in the he<strong>at</strong>ing <strong>co</strong>il passed through the reactor, was <strong>co</strong>ndensed, and <strong>co</strong>llected in an Erlen-<br />

meyer Bask.<br />

At the termin<strong>at</strong>ion <strong>of</strong> an experiment the reactor WBB removed trom the aand b<strong>at</strong>h, and placed in a<br />

vacuum oven over night <strong>at</strong> 45 O c to remove any w<strong>at</strong>er. The reactor was then wekhed, the <strong>co</strong>ntents<br />

were removed and placed in a predried and weighed thimble. The reactor was then reweighed, the<br />

322

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

Saved successfully!

Ooh no, something went wrong!