Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
PC- 119 7/CAN
S~ ~3
FIELD OF THE INVENTION
,
The present invention relates to the recovery of hydro-
carbons from tar sands.
BACKGROUND OF THE INVENTION
The increase in price and the decline in known reserves
of conventional oil have brought about an increased interest in
the recovery of synthetic crude oil from alternative sources.
One such promising source consists of tar sands, sometimes also
referred to as oil sands or bituminous ~ands, these being sand
deposits which are impregnated with bitumen. A large deposit of
tar sands in the Athabasca region of Alberta, Canada, is the only
deposit to have been exploited commercially up to now. A typical
sample of as-mined tar sands might contain, on average, 83%
quartz sand, 12% bitumen and 5% water. (Unless otherwise speci-
fied all percentages quoted herein are percentages by weight.)
The bitumen contained in the sands might have a typical assay of
83.6%C, l0.3~H, 5.5~S, 0.4~N and 0.2~0.
The only two plants which have been constructed to
exploit Athabascan tar sands commercially employ a process com-
monly referred to as the hot-water separation process. In this
process the as-mined sands are mixed with hot water (80-90C)
and wetting agents and subsequent settling and flotation sepa-
rates a sand phase from a bitumen phase. The latter is then
subjected to a coking operation wherein a Iightar hydrocarbon
fraction is distilled off at about 500C leaving a coke product
in the vessel.
An alternative scheme which has been proposed for the
treatment of tar sands i~volves a direct cokina operation
whereby sand-bitumen separation and bitumen upgrading are
achieved simultaneously. In thLs case the tar sands are heated
to a temperature`of the order of 400-650C in a vassel such as
~k
'~
... .. . .... ... . . .. . .. .
:
~25~
a rotary kiln or a fluidized bed whereby a volatile fraction
is distilled off to leave behind a mi~.ture of coke and sand.
In eithe~ of the abo~e d~scribed schemes it is highly
desirable to be able to burn coke which is produced in order
to generate at least part of the energy required by the pro-
cesses. Thus in the case of ho~ water processing coke burning
could supply energy required to heat the large amo~nts of water
used for sand-bitumen separation. In the direct coking process
it has been recommended to feed the hot sand-coke mixture into
a vessel where the coke is burned and furthermore to mix the
resulting hot sand with fresh tar sands to be fed into the
direct coking vessel, thereby utilizing the sensible hea~ in
the treated sand.
A major obstacle to the success of any commercial pro-
cess for treating tar sand is posed by the presence of a sub-
stantial amount of sulfur in the bitumen constituent of the tar
sands. The undesirable consequences that flow from ~his sulfur
presence are twofold:
i) The distillate produced in the coking operation, whether
the latter is carried out on the tar sands as a whole
or on the bitumen separated therefrom by flotation,
contains a significant amount of sulfur. Typically this
is of the order of 50 to 70~ of total amount of sulfur
present in the tar sands feed, and it has to be removed
during ~ubse~uent refining of the distillate by hydro-
desulfurization. This produces H~S which can be oxi-
dized, for example by the Clauss process, to elemental
sulfur which has to be disposed of.
ii) The remainder, which can be 20 to 50%, of the sulfur in
the feed reports in the coke produced and renders it
difficult if not impossible to burn that coke without
exceeding currently imposed limitations on sulfur diox-
ide emission.
One s~ggestion which has been made for dealing with the
problems posed by sulfur comprises resorting to a hydrodesul-
furization operation carried out prior to the coking operation.
This technique, referred to as hydrocracking, results in coke
having a lower sulfur content, but produces large amounts of
hydrogen sulfide and hence large amounts of elemental sulfur to
be disposed of. Al-together the approach is a cost~y one.
Alternatively it has been proposed to add lime, lime-
stone or dolomite to sulfur bearing coal or coke as a means of
fixing the sulfur and lowering S02 emission. This expedient does
not unfortunately provide a satisfactory solution to the problem
of using coke derived from tar sands because the resulting fix-
ation of sulfur is less than complete. Even where the lime
addition is substantially more than the stoichiometric amount
needed, the fixation is found to be typically no better than
80-90~ complete and often poorer. This is inadequate to deal
with a coke which may contain as much as half of the sulfur
present in the tar sands feed.
The inadequacy of presently known schemes for dealing
with the sulfur dioxide emission problem is evidenced by the
fact that in one of the two existing commercial operations for
treating tar sands all of the coke produced is presently being
stockpiled.
OBJECTS OF THE INVENTION
The present invention is aimed at providing a process
whereby tar sands or bitumen separated therefrom can be ~reated
to recover a distillate containing a lower amount of sulfur than
heretofore possible, while a reIatively higher proportion of the
sulfur in the feed report~ in the coke produced.
A further and at least equally important object of the
invention is to provide a process whereby coke produced from the
tar sands can be burned with ubstantially no sulfur dioxide
emission.
-3-
... ...... ....... . ... . ... ..... . . ..
BRIEF DESCRIPTION OF THE INVENTION
It has been found that the desired ob~ects can be
achieved if a sulfur fixing agent, e.g. lime, is present in the
vessel wherein coke is formed~ When the fixing agent is added
to the coking vessel it is possible to get virtually complete
fixation of sulfur in the coke, in contrast with the partial
fixation which can be achieved by introducing the fixing agent
into the coke-burning vessel as advocated in the past. The
success of the newly discovered procedure is believed to be
attributable to ~he fact that when the sulfur fixing agent is
present during coke formation an intimate mixture is obtained
of coke and fixing agent. Moreover some sulfur fixation may
occur during the coking operation so that the fixing agent may
be partially reacted (i.e. it may be in the form of a sulfide)
in the solids withdrawn from the coking vessel. Apart from
yielding a coke which can be burned without any substantial sul-
fur dioxide emission, the presence of the sulfur fixing agent
in the coking vessel has been found to lower the amount of sul-
fur which reports to the distillate recovered during the coking
operation. The result is therefore a lowering of the amount of
sulfur to be removed by the costly route of hydrodesulfuriza-
tion.
Accordingly the present invention provides a process for
recovering a hydrocarbon distillate from tar sands wherein a
feed, consisting of tar sands or raw bitumén separated from tar
sands by flotation, is introduced into a volatilization reactor
wherein the feed is subjected to a coking operation to produce
a distillate product and a coke product~ the improvement con-
sisting in the fac~ that a sulfur fixing agent, selected from
the group consisting of lime and calcined dolomite, in an amount
corresponding to between about 2 and 4% by weight of the feed
when the feed consists ~f tar sands, and between about 15 and
~ ~ _. .. .... ............... .... .
30% by weight of the feed when the feed consists of raw bitu-
menr is introduced wi~h the feed in~o the volatilization reactor,
whereby the coke product produced comprises an intimate mix-
ture of coke with the fixing agent or partially reacted fixing
agent, andis suitable for burning without any substantial
sulfur dioxide emission.
DETAILED DESCRIPTION OF THE EMBODIMENTS
The improved process of the invention can be embodied
in a scheme involving delayed coking, such as the so-called hot
water separation process. Preferably however it is embodied in
a direct coking process for treating tar sands inasmuch as this
type of process is capable of being carried out with a higher
energy recovery. The direct coking involves heating the charge
of tar ~ands and fixing agent under reducing conditions to a
temperature which should not exceed about 600C, e.g. 500-600C.
Due to the presence of the sulfur fixing agent in the volatil-
ization reactor in which coking occurs, a distillate is recov-
ered in which, typically, less than 30% of total amount of
sulfur in the tar sands feed reports. Moreover SO2 emissions
from the volatilizer are found to be negligible.
After the direct coking operation hot solids, consist-
ing of the tar sands residue and coke product having fixing
agent intimately mlxed with it~ are withdrawn from the vola-
tilizer. We prefer to reject a portion of these solids, e.g.
one quarter, and feed the remaindex to a second reactor, i.e.
a coke burner. The latter is a vessel maintained at a tempera-
ture of ~bout 800-950C, preferably about 900C, under oxidizing
conditions. Coke in the solids fed to this burner is found
to burn with substantially no SOz emission~ At the end of the
~oke-burning operation,hot sand is left in the burner. The
sensible heat of that hot sand is utilized by feeding it back
to the volatilizer to be mixed with fresh tar sands to be
treated, preferably in the ratio of 3 parts of hot sand to
1 part of fresh tar sands.
The sulfur fixing agent can be lime or calcined dolo-
mite, but limestone has been found unsuitable for this purpose.
The fixing agent is preferably in inely divided form, e.g.
particles no greater than about 150 microns in diameter. We
prefer to mix the lime or calcined dolomite with the feed at
about 60C prior to introducing the feed into the volatili~er.
It is a particular advantage of the process of the in-
vention that comparatively little of the sulfur in the tar sands
needs to be removed by hydrodesulfurization and disposed of in
elemental form. The major part of the sulfur present in the
feed reports in the discarded portion of the volatilizer solids
residue, where it is present mostly as calcium or magnesium
sulfate which is more readily disposed of than elemental sulfur.
Some illustrative examples will now be given.
EXAMPLE: 1
A lOOg sample of ~ar sands was used containing 8.1%
bitumen and having a sulfur content of 0.28~. The sample was
placed in an alumina boat and heated for one hour in nitrogen
to simulate direct coking. ~he residue from direct coking was
then heated in air for 1 1/2 hours at 900C to simulate coke
burning. Sulfur assays were carried out on the residue after
the 500C treatment and after the 900C treatment. Three
further tests were ca~ried out in an identical manner except
that lime, in amounts o~ ~, 2 and 4~, respectively, of the
weight of tar sands was mixed with the sample prior to the
coking. The re~ults obtained are shown in Table 1.
-6-
TABLE: 1
CaO Addition 50UC Resldue 900C Residue
(% of Feed Weight Sulfur Content Weight Sulfur Content
Weight) (g) % S % of S in Feed (g~ % S % of S in Feed
-
- 85.2 0.17 52 83.4 0.04 12
1 87.6 0.26 81 85.8 0.27 83
2 87.6 0.27 84 85.4 0.29 88
4 91.6 0.27 88 89.0 0.28 8~
The above results show the presence of lime at the~coke-
formation stage leads to a very significant increase in the
amount of sulfur which reports in the residue, i.e. the coke
product. In fact with this sample of tar sands even as low
as 1~ lime addition resulted in over 4/5 of the sulfur report~
ing in the coke. Moreover a comparison of the sulfur distri-
bution data before and after coke burning shows that none of
the sulfur was lost on burning of the coke in any of the tests
where lime was used. In the first, comparative, test
although the coke contained only half the sulfur in the feed,
most of that sulfur was lost upon coke burning.
EXAMPLE: 2
A second series of four tests was carried out in an
identical manner to that described in the above example,
except that the feed in this case was lOOg of tar sands having
a bitumen content of 12.8% and a sulfur content of 0.46~. The
results of this set of tests are reported in Table 2.
TAB~E: 2
CaO Addition 500C Residue _ 900C Residue
(% of Feed Weight Sulfur Content Weight Sulfur Content _
Weight) ~ % S X of S in Feed (g) % S _ of S in Feed
: -83.8 0.24 52 81.9 0.14 28
184.0 0.32 58 ~1.6 0.24 42
2~5.8 0.39 70 83.2 0.40 70
4 87.6 0.3g 77 85.0 0.44 77
In the case ~f this sample of tar sands a 1% lime addition
provided only marginal increase in the distribution of sulfur
to the coke product, and substantial though not complete fixa-
tion of that sulfur. The use of at least 2% lime addition gave
excellent ~esults in terms of both improved distribution and
complete fixation, i.e. total elimination of S02 emission.
EXAMPLE: 3
To determine the utility of limestone as a substitute
for lime in the process of the invention a set of three tests
were carried out on an identical sample of tar sands to that
used in the tests of Example 2. The results obtained, which
are shown in Table 3 belowl showed that neither the improved
sulfur distribution nor complete sulfur fixation are obtained
when limestone alone is used. Indeed the test in which 2%
limestone as well as 2% lime were added, gave similar results
to the test reported in Table 2 wherein the same lime addition
was used alone.
TABLE: 3
900C Residue
CaO Addition CaCO3 Addition Weight Sulfur Content
~% of Feed Weight) (~ of Feed Weight) (g) % S % of S in Feed
2 2 86 0.37 69
- 4 80 0.26 45
- 8 80 0.26 45 `~
It should be pointed out that the fact that as much as 45% of
the total sulfur remains in the burner residue (compared with
28~ when no fixing agent is used~ shows some benefit from
using limestone, but the benefit is clearly inferior to that
derived from using lime.
EXAMPLE: 4
A further sample of the tar sands used in Examples 2
and 3 was tested in a rotary kiln. ~he lOOg sample was mixed
-8-
~ ....... ... ~.. .... ... ... .. .... . .. ... .
, , ,. . . , ~
f~
with 2g of lime and subjected to heating for one hour at
500C under nitr~gen and the~ ~or a further hour at 900C
under air. By way of comparison the test was repeated with-
out lime addition, and the sulfur distributions determined
in both tests are shown in Table 4 below.
TABLE : 4
S Distribution (%of S~in Feed)
Product No CaO Added 2~ CaO Added
Distillate product 48 28
Coke product 52 72
Cokebur~ing residue 28 72
Coke-burning off-gases 24 N.D.*
*None detected
EXAMPLE : 5
Tests were conducted using bitumen which had already
been separated from the sands. This bitumen was obtained in-
dependently of the aforementioned tar sands and it analyzed
4.48% S. One test was conducted using CaO addition and the
other having no additions. In the test where no lime was
used the bitumen was ~acuum distilled and then coked at 600C
to give a distillate containing about 75% of the total sulfur.
The resulting coke contained 5.71% S or about 25~ of the total
sulfur in the initial bitumen. A 17g portion of this coke was
iginited as in Example 4 with the release of 2.5g SO2 whish
contains all of the coke's sulfur.
To test the utility of CaO as the sulfur-fixing
agent for raw bitumen the second test used about five times
the stoichiometric amount of lime needed to form CaSO4 with
the sulfur in the bitumen, i.e. a lime addition of about 30%
by weight of the raw bitumen. This mixture was distilled and
coked and the resulting distillate contained only about 55
of the total sul~ur. The resulting coke contained about 45
of the total sulfur and analyzed 4.58% S. This coke when
burned gave off only 1.2% of its contained sul~ur as S02.
... .... . ~ .. .. ~ . ................ ., .. ~ .. .. .... .. .
These tests demonstrate that in the case of raw
bitumen the addition of 30% CaO serves to decrease the sulfur
in the distilled oil product from about 75% to about 55~ of the
total sulfur in the bitumen and, concomitantly, decreases the
SO2 evolution during the coke-burning step from nearly 100
of the contained sulfur in the coke to almost zero.
The present invention has been described~with refer-
ences to specific embodiments thereof. It will be appreciated
that the tests described are merely illustrative. In particu-
lar while in these tests the same vessel was used for both
the coking operation and the burning operation, in commercial
practice different vessels would be used for these steps, each
of which might be operated in a batch manner or in a continuous
manner. These and other modifications which may be made to
the details of the embodiments are within the scope of the
invention which i5 defined by the appended claims.
- 10 -
.
.: . , : ~- : :
