sda lecture

PPPeeetttrrrooollleeeuuummm &&& CCCoooaaalll
IIISSSSSSNNN 111333333555---333000555555
Available online at www.vurup.sk/pc
Petroleum & Coal 48 (3), 14-19, 2006
SOLVENT DEASPHALTING OF VACUUM RESIDUE IN A
BENCH-SCALE UNIT
M. Sattarin1
, H. Modarresi2
*,H. Talachi1
, M. Teymori1
1
Petroleum Refining Division, Research Institute of Petroleum Industry,
Tehran, Iran, 2
CAPEC, Department of Chemical Engineering, Technical
University of Denmark, DK-2800 Lyngby, Denmark,*Corresponding author.
Tel.:+45-4525-2986, Fax: +45-45932906. Email: ham@kt.dtu.dk
Received October 6, 2006; accepted November 24, 2006
Abstract
The introduction of solvent deasphalting (SDA), on different refining schemes, bring to the refineries a very
promising solution for desulfurization problems and/or catalytic cracking processes in recent years. In this work,
some effective parameters of SDA process such as temperature and solvent and residues nature have been
studied on the different residues of an Iranian crude oil mixture (ºAPI 19.5) in a bench unit. All produced
deasphalted oils (DAOs) were almost free of asphaltene (<0.5 wt%) apart from the feed nature. SDA by means of
pentane as solvent reduced the heavy metal content of the feeds up to 85%, sulfur content up to 20% and CCR
(Carbon Conradson Residue) content up to 50%, where SDA with butane results in 93% reduction of heavy
metals content, 20% reduction of sulfur content and 66% reduction of CCR.
Keywords: oil upgrading; extraction; deasphalted oil; pitch.
1. Introduction
Direct hydro-desulfurization followed by hydrocracking of crude oil heavy cuts and vacuum
residues is one of the best methods of heavy residue upgrading in refining industry. But, problem
emerges when metal and asphaltene contents of residue are high. In fact, the presence of these
compounds adversely influences the activities of the hydro-desulfurization and hydrocracking
catalysts.
The solvent deasphalting process (SDA) which is based on liquid–liquid extraction by using
paraffinic solvents (C4–C7) is one of the most efficient approaches to reduce metal and asphaltene
contents of heavy oil cuts before sending them to hydro-desulfurization and hydrocracking units.
A number of deasphalting process parameters are to be considered, amongst which the DAO
process yield and the levels of demetalization and deasphalting could be noted. The important factors
influencing the mentioned parameters are solvent composition and ratio of the solvent to the feed,
temperature, pressure and the type of extractor equipment.
Increasing the solvents molecular weight increases the yield of recovered DAO by allowing more
of the heavier, more – resinous components of the feed stock to remain in the DAO. At the same time,
however, the quality of the DAO decreases because these heavier materials have higher contaminant
levels. Consequently, proper solvent selection involves balancing increased product yield and
decreased product quality. The precipitation increases substantially as the solvent/feed ratio increases
up to 10 folds. Beyond this value, precipitation increases by very small amounts
[1]
.
Contrary to the liquids solving behavior, DAO yield is reduced by rising the temperature. This is
due to the solvent density reduction with increasing the temperature. The solvent density reduction
with temperature rise, leads to the reduction of solvent tendency to dissolve heavier compounds and
improves DAO quality at the expense of reduced DAO yield. Extraction temperature must be
maintained below the critical temperature of the solvent, however, because at higher temperatures no
portion of the residue is soluble in the solvent and no separation occurs[2]
. In industrial plants,
increasing the solvent to feed ratio compensates for the DAO yield reduction with temperature rise. In
consequence, the extraction process selectivity for paraffinic oils escalates and eventually the
extracted oil will have an improved quality due to the reduction of undesirable components.

Extraction pressure which is chosen to ensure that the SDA extractors solvent–residue mixture is
maintained in a liquid state, is related to the critical pressure of the solvent used. During normal
operation, when both pressure rise results in the solvent density increase and therefore DAO yield
increases. This effect gets even more pronounced near the critical point caused by the sharp variation
of density versus pressure. Nevertheless, due to the process and mechanical restrictions, changing
the extraction operational pressure to regulate DAO yield, is not an industrial common practice [3]
.
The main aim of this study was to investigate some key factors influencing SDA process.
Therefore, the residues of a typical Iranian heavy crude oil (i.e. 450+
, 470+
, 500+
, and 520+
°C) are
deasphalted, by using of pentane and butane solvents. In addition, different operational temperatures
in SDA process (i.e. 35, 50 and 80°C) have been studied.
2. Bench Scale SDA Operation
The set of equipment employed in the bench unit (Fig. 1) comprises a jacket vessel equipped with
a mixer (as an extractor), inlet and outlet for the materials in accordance with extraction operation and
utilities such as warm and cold fluid suppliers, outlet DAO condenser and products collection facilities.
The operation is batch–wise and in each batch after being pre-diluted with the solvent, some of
the residue is introduced to a high-pressure extractor. The solvent is introduced to the extractor ten-
folds the amount of the feed. After the temperature is kept constant for a certain period of time, mixing
and precipitation operations are performed. Afterwards, DAO was withdrawn from the extractor by
means of a pipe dipped into it. After passing a condenser, DAO was collected. When DAO was
completely evacuated from the extractor, extraction residue was drawn out from the extractor bottom.
The extracted DAO phase was sent to the next step for solvent evaporation and regeneration.
As the pitch content is small in bench scale studies, solvent regeneration step was waived. But
the solvent carried along with the DAO was regenerated, primarily by being introduced to an
atmospheric evaporation vessel and then to a vacuum rotary evaporator. Eventually, the DAO and the
dried pitch were sent to the laboratory for further analysis.
Fig. 1: A schematic diagram of the bench scale SDA unit (CW: Cooling Water)
3. Results and Discussion
The results of tests, with butane and pentane solvents, are presented in Table 1 for different
vacuum residues. The effective parameters of SDA process, included in this table, are specific gravity,
asphaltene content, Conradson carbon residue (CCR) content, metals content and sulfur content, test
results. Tests with pentane solvent are carried out at 50 and 80 ºC.
To further clarification of SDA process parameters, the test results are presented in the form of
graphs. Figure (2) shows mean variation of extraction yield (DAO yield) for each solvent versus the
M. Sattarin et al./Petroleum & Coal 48(3) 14-19 (2006) 15

type of vacuum residue (SDA feed). This figure clearly shows that extraction yield increases as the
solvent molecular weight increases and, decreases as feed mean molecular weight increases. Figure
(2) shows that for vacuum residues heavier than 500+
, extraction yield does not change significantly.
The reason is that oil and light resin contents of vacuum residue are reduced when its mean molecular
weight (or boiling point) increases and after the 500+
, this amount tends to trivial. Additionally, the SDA
process temperature variation does not have a pronounced effect on extraction yield for cuts heavier
than 500+
. It should be noted that extraction yield increases directly with temperature, but here, the
results are contrary to this fact. This could be explained by referring to the concept of macro-fluid.
According to Kolmogrov’s relationship for macro-fluid, mean size of a coalescent aggregate of a same
fluid in a semi-immiscible liquid mixture can be estimated by equation (1) :
4/13
)/( εγλ = (1)
where λ is the aggregate size [m], γ is kinematic viscosity [m2
/s], and ε is mixing power per mass unit
[W/kg]. For mixture of the vacuum residue (VR) (450+
) and pentane, with a fairly high mixing power of
1 W/kg, the order of magnitude of the aggregate size (μm) can be roughly varied between 100 for low
temperature (close to ambient temperature) and 1 for temperature as high as 150 °C. Big aggregates
can hold almost large quantity of DAO, therefore, it can be considered another mechanism for solving
of VR in solvent. According to Figure (3), for low temperatures (close to ambient temperature),
solubility of DAO is controlled with size of aggregates, so it increases as temperature rises up. For
high temperatures, which are estimated higher than 120 ºC (for pentane solvent), aggregate sizes are
quite small where they cannot hold much DAO. Therefore, solubility is controlled by density variation
of solvent and it decreases as temperature rises up.
Fig. 2: DAO Yield of SDA with butane (C4) and pentane (C5) Fig. 3: Qualitative solubility of DAO in a parafinic solvent
Table 1: Some results of bench operation @ T=50 ºC for pentane solvent and T=35 ºC for butane solvent.
450+
470+
500+
DAO DAO DAO
VR C4 C5 VR C4 C5 VR C4 C5
Sp.Gr 15.56°C
Asphaltene,wt%
CCR., wt%
Sulfur , wt%
Vanadium, ppm
Nickel, ppm
1.0458
17.2
19.42
4.70
215.0
55.0
0.9912
<0.5
6.59
3.79
14.0
3.9
1.0038
<0.5
9.65
3.96
28.0
7.2
1.0525
18.3
20.5
4.79
240.0
70.0
0.9985
<0.5
6.99
3.72
16.3
4.2
1.0096
<0.5
10.07
3.92
33.0
8.8
1.0646
20.0
21.45
4.88
265.0
73.0
1.0003
<0.5
6.99
3.99
19.0
4.6
1.0007
<0.5
11.88
4.15
50.0
12.0
Figure (4) shows the heavy metal content (Vanadium +Nickel) variations of DAO in SDA process
according to the type of feed used. This figure shows, when the initial boiling point of vacuum residue
increases, the DAO metal content increases accordingly. Also, raising the process temperature results
in a better purified DAO. CCR content variations of DAO shown in figure (6) are similar to the heavy
metal variations. The DAO sulfur content )S) variation is shown in figure (7), which is almost similar to
CCR and heavy metal content graphs.
45
50
55
60
65
440 460 480 500 520
VR Initial BP ,°C
Yield,wt%
C5
C4
Bottom of
Column
Bottom of
Column
Top
Top
Temperature (°C)
Solubility

Fig. 4: Metal (V+Ni) content variation of DAO, SDA with
pentane
Fig. 5: CCR content variation of DAO, SDA with pentane
For all the tests, the DAOs produced were free of asphaltene (<0.5 wt%), regardless of the feed
type. Table (2) presents the amount of the vacuum residue impurities reduction after the SDA process
with pentane. According to this Table, extraction with pentane will reduce the heavy metal content of
vacuum residue up to 85%, its sulfur content up to 20% and its CCR content up to 50%. As expected,
because of the uniform distribution of sulfur compounds in different crude oils, no significant reduction
in DAO sulfur content comparing to the vacuum residues is detected. Schematic graph of Figure (7)
could be an understandable representation of how impurities in a vacuum residue are distributed.
Table 2: Reduction of Impurities in the SDA Process with Pentane
Parameter Impurity Reduction (%) Vacuum residue
Vanadium 87.7
Nickel 87.4
Sulfur 18.3
CCR 50.8
450+
Vanadium 86.9
Nickel 87.6
Sulfur 21.5
CCR 51.8
470+
Vanadium 85.5
Nickel 85.9
Sulfur 18.6
CCR 50.4
500+
Vanadium 85.0
Nickel 84.8
Sulfur 21.5
CCR 52.6
520+
Figures (8) and (9) depict metal and CCR content variations of the DAO product with butane
solvent. Comparing these two figures with Figures (4) and (5) reveals that the DAO produced by
extraction with butane is more purified than the DAO produced by extraction with pentane. SDA with
butane results in 93% reduction of heavy metals content and 66% reduction of CCR content of the
vacuum residue while for pentane as the solvent, these figures are 85 and 50%, respectively.
30
35
40
45
50
55
440 460 480 500 520
VR Initial BP ,°C
V+Ni,ppm
50°C
80°C
6.4
6.6
6.8
7
7.2
440 460 480 500
VRInitial BP,°C
CCR,wt%

Resins
Asphaltene
Sulfur
CCR
Metals
Impurities
Oil Asphalt
Vacuum Residue B P
Light Medium Heavy
Fig. 6: Sulfur content variation of DAO, SDA with pentane Fig. 7: Qualitative variation of impurities of vacuum residue
among its constituents
Fig. 8: Metal (Vanadium+ Nickel) content variation of DAO,
SDA with butane
Fig 9: CCR content variation of DAO, SDA with butane
4. Conclusion
In this study, SDA bench-scale tests have been carried out on vacuum residues 450
+
, 470+
, 500+
and 520+
using butane and pentane solvents. It should be noted that prior to the design of an industrial
plant, more studies have to be performed. In the other words, in order to design an industrial scale
extractor, design data from similar equipment in a pilot-plant scale are required. Having said that, the
results of the present work can be referred to as a good measure for selection the feed, solvent, and
somewhat, for the process operational conditions.
The extract and raffinate phases were fully separated in all tests, which is not common in an
industrial scale. In the other words, full separation of the two phases is hardly practicable in industrial
scale plants as it will be very costly, therefore, it can be concluded that the yields reported here are
almost the highest possible amounts.
In real industrial process conditions, part of oil and light resins enter the pitch. Nonetheless, the
result of tests carried out on back-blending of oil-resin cuts with pitch (the tests are performed in a way
close to industrial conditions), do not show any improvement in the quality of the pitch for industrial
production of commercial bitumen. Except for SDA units specifically designed to produce roofing or
paving asphalt, the asphalt product is normally considered a law - value by-product. Partial oxidation
of pitch can be produced synthesis - gas for manufacturing power, steam, methanol and many other
products, aquaform for pelletizing asphaltene from solvent deasphalters is a new technology. It is low
– cost, high capacity, continuous process for producing uniform, solid asphaltene pellets. The pellets
produced are resistant to dusting and can be easily handled, stored, and transported
[4]
. The
asphaltene pellets have a higher heating value and improved fuel properties versus petroleum coke
and thus represent improved fuel value.
16
17
18
19
20
21
22
23
24
440 460 480 500
VRInitialBP,°C
V+Ni,ppm
6.4
6.6
6.8
7
7.2
440 460 480 500
VRInitialBP,°C
CCR,wt%
3.6
3.8
4
4.2
440 460 480 500 520
VRInitialBP ,°C
S,wt%
50°C
80°C

References
[1] Gonzalez, E. B., C. L. Galeana, A. G. Villegas, J. Wu, AIChE Journal, vol. 50, No.10, Oct.
2004, p. 2552
[2] Meyers, R. A., Handbook of Petroleum Refining Processes, 10th
ed., McGraw-Hill, 2003.
[3] Ditman, J.G., Van Hook, J.P.: ACS Meeting, Atlanta, April 1981.
[4] Subramanian, M., Moretta, J., Bloom, R., Martin, M.: PTQ J., 2000, Summer, p.17

sda lecture

More Related Content

What's hot (20)

Viewers also liked (15)

Similar to sda lecture (16)

sda lecture