530 I n d . E n g . C h e m . R e s .

1992, 31, 530-536

PROCESS ENGINEERING AND DESIGN

A Solvent-Resid Phase Diagram For Tracking Resid Conversion

I r w i n A. Wiehe
Corporate Research Laboratories, Exxon Research and Engineering Company, Route 22 East,
Clinton Township, Annandale, New Jersey 08801 -0998

A plot of molecular weight versus hydrogen content is capable of representing solvent-separated

pseudocomponents of petroleum resids and their thermal conversion products by unique areas that
are independent of the resid. This solvent-resid phase diagram enables one to track the path of
chemical changes that do not convert a molecule completely from one solubility class to others. In
this study petroleum resids and the thermal conversion products of the resids and of resid fractions
were separated by solubility and clay adsorption into five pseudocomponents: saturates, aromatics,
resins, asphaltenes, and coke. The combination of this solvent fractionation and the solvent-resid
phase diagram shows how the solubility classes of decreasing solubility evolve during thermal
conversion either by increasing aromaticity or by molecular weight growth of low-volatility fragments.

Since petroleum resids are complex mixtures of thou- evaporated during solvent removal and were determined
sands of compounds of low volatility, it is difficult to de- by mass loss. In the second step the resid nonvolatile
duce the chemical changes that occur during their pro- product was mixed in a flask with 15 parts toluene and let
cessing. A common approach has been to use solvents to to sit for at least 16 h at room temperature. This mixture
separate petroleum resids and their reaction products into was poured through a fine (4-5.5-pm pores) fritted-glass
pseudocomponents and then track the pathway for chem- filter. The solids on the filter were washed with at least
ical changes by the conversion of each pseudocomponent 25 parts additional solvent, and washing was continued
into others. One of the most common of such pseudo- until the solvent passed the filter without color. The
components is the asphaltenes that are soluble in aromatic toluene-insoluble solids were vacuum dried on the filter
solvents (benzene or toluene) and insoluble in paraffinic at 100 OC for at least 16 h and are called “coke”, following
solvents (n-pentane or n-heptane). However, Bunger and the convention of Speight (19701, Schucker and Keweshan
Cogswell (1981) have raised valid concerns about this ap- (19801, and Savage et al. (1985). The toluene was removed
proach because no single chemical feature distinguishes from the toluene solubles by rotary evaporation at 50 “C
species found in the asphaltene fraction from those in the followed by vacuum drying for 16 h at 50 “C. Since the
mdtenes (paraffinic solvent soluble) fraction and because unreacted resids in this study all had prior vacuum dis-
pseudocomponents can undergo chemical changes without tillation and were completely toluene soluble, the first two
being completely converted to other pseudocomponents. steps were only applied to resid thermal conversion
Here, it is shown that a molecular weight-hydrogen con- products.
tent “phase diagram” or heavy oil “map” in combination The n-heptane separation in the third step followed the
with pseudocomponent separation overcomes both of these same procedure as for the toluene separation, except the
concerns. Not only is it capable of distinguishing one n-heptane was not removed from the heptane solubles.
pseudocomponent from another, but it can be used to track The solid, heptane-insoluble product is called
chemical changes that do not move a molecule completely “asphaltenes”. In the fourth step the heptane solution is
from one solubility class to others. mixed in a flask with Attapulgus clay, 30 times the weight
of heptane solubles, for a minimum of 16 h. The Atta-
Separation Scheme pulgus clay was National Bureau of Standards qualified
In this section, the procedure is described for separating for ASTM D2007-86, a clay-gel chromatographic separa-
petroleum resids and their reaction products that was used tion test. This clay contains a controlled amount of water
in this study. Unfortunately, such separations are sensitive and meets azobenzene activity specifications of 30-35
to changes in experimental procedures (Speight et al., equiv. This assured reproducible adsorption, while only
1984). The definition of “soluble” or “insoluble” is de- with water on the clay can one completely desorb the
pendent on the amount of solvent, the pore size of the hydrocarbons off the clay. This mixture of clay, heptane,
filter, the filter material, and the time (when less than 4 and oil was filtered with a fine, fritted-glass filter. The
h) between when the solvent and resid are mixed and when clay on the filter was washed with at least 200 parts of
the mixture is filtered. However, if the procedure is kept heptane per part of solubles, and washing was continued
constant, the yield of a fraction from a given resid and its until the heptane ran through the filter without color. The
chemical and physical properties can be quite reproducible. solvent was removed from the solution so produced by
A schematic of the separation procedure is shown in rotary evaporation and by vacuum drying at 50 OC.
Figure 1. Often vacuum distillation was used to remove The fraction remaining on the clay was desorbed in the
volatile products from resid conversion products as a first fifth step by mixing with a mixture of 50% toluene and
step. When this was not done, light volatile products 50% acetone on the filter. At least 100 parts of this
0888-5885/92/2631-0530$03.00/0 0 1992 American Chemical Society
 Ind. Eng. Chem. Res., Vol. 31, No. 2, 1992 631
Table I. ExamDle Analysis of Fractions Separated from an Unreacted Resid
total %
yield/ wt% wt% H/C wt% wt% wt% ppm ppm elements/ VPOb wt% arom
fraction (wt %) C H (atomic) S 0“ N V NI wt% MW MCR C4
saturates 18 84.54 12.31 1.73 2.74 0.0 0.03 0 0 99.6 920 1.8 15
aromatics 17 81.87 10.00 1.46 5.56 0.0 0.12 0 0 97.6 613 8.7 37
resins 40 82.08 9.50 1.38 6.09 1.72 0.77 260 110 100.2 986 24.6 45
asphaltenes 25 81.93 7.94 1.15 7.50 1.55 1.15 740 230 100.2 2980 49.6 50
total 100 82.45 9.70 1.40 5.75 1.08 0.62 289 102 99.6 1040 24.2 40
starting resid 100 82.95 9.92 1.42 5.73 0.77 0.77 300 100 100.2 1183 24.0 41
Oxygen determined by neutron activation. VPO number average molecular weight measured in o-dichlorobenzene at 130 O C . Microcarbon
residue technique for measuring Conradson carbon residue (ASTM Test D4530). dAromatic carbon determined by NMR.

VOLATILES
INSOLUBLE
COKE
INSOLUBLE
ASPHALTENES
nitrogen, vanadium, and nickel and exhibits little coke-
t
f f forming tendency. The aromatics fraction (37% aromatic
carbon) is more aromatic than the saturates fraction but
RESID DISTILL-
ATION
TOLUENE
FILTRATION
HEPTANE
FILTRATION
also is relatively free of heteroatoms, other than sulfur. It
is also the lowest molecular weight fraction. The resins
fraction is sometimes called “polar aromatics” because of
AROMATICS
SOLUBLE -(TI
!F{iz$$N d
-
1 4
ATTAPULGUS
ita high oxygen and nitrogen content relative to the satu-
rates and aromatics fractions. However, the average resin
molecule has only one oxygen or one nitrogen atom out of
a molecular weight of 986. Thus, it is not as polar as many
common low molecular weight compounds that contain
oxygen and nitrogen, such as acetone or pyridine. As will
be discussed later, the present evidence indicates that
resins adsorb on clay more because they contain large-ring
t aromatics than because of oxygen and nitrogen function-
RESINS
Figure 1. Resid separation scheme. ality. Nevertheless, there is also a sharp increase in va-
nadium, nickel, aromatic carbon, and MCR in going from
mixture per part heavy oil was used to wash the clay. This aromatics to resins. On the other hand, the asphaltenes,
was followed by washing with at least the same amount as is well-known (Speight, 1980), have still higher con-
of 90% toluene and 10% methanol, which was continued centration of all the undesirable components and prop-
until the solvent mixture passed through the clay without erties: sulfur, nitrogen, vanadium, nickel, molecular
color. The solvent was removed from the resulting solution weight, MCR, and aromatic carbon. I t is no wonder that
by rotary evaporation followed by vaccum drying at 50 OC. problems in converting resids usually focus on asphaltenes.
The resulting fraction was reported as “resins”. However, since the yield of resins is much higher than the
In the sixth step, the heptane-soluble oil that was not yield of asphaltenes, as a percentage of resid the resins
adsorbed on clay was mixed with 30 parts methyl ethyl have comparable amounta of undesirable components and
ketone (MEK)and cooled to -78 “C in a dry ice-isopropyl properties as the asphaltenes.
alcohol bath. After a 4 h waiting period, this cold mixture One of the odd features of the separation of resids is that
was filtered through a fine, fritted-glass filter that was kept the fractions have nearly constant carbon content. For this
at -78 “C by adding dry ice directly on the filter. The separation all the carbon contents were within 83 f 1.5 wt
filtration was done under a flow of dry nitrogen to prevent %. The lower hydrogen content in the more aromatic
water from the air condensing and adding to the MEK fractions is compensated by the higher heteroatom content,
solution. The waxy solids remaining on the filter were keeping the carbon content nearly constant. As we shall
dissolved off with room-temperature heptane, and the see, this feature also carries over to resid thermal con-
heptane was removed by rotary evaporation and vaccum version products. Thus, the hydrogen to carbon atomic
drying at 50 “C. This fraction was called “saturates”. The ratio that tends to measure aromaticity (lower value for
solvent was removed from the solution that passed through more aromatic) is really equivalent to a constant times the
the fiiter by rotary evaporation and vaccum drying at 50 hydrogen content for these fractions.
“C, leaving the oil fraction that was called “aromatics”.
Example Fractions of Thermal Conversion
Example Unreacted Resid Fractions Products
A 1050 O F + resid was separated by the above separations The vacuum resid with properties shown in Table I was
scheme with yields and analyses of the fractions shown in reacted for 60 min at 400 “C in small, batch-tubing bomb
Table I. %ically, the mass and elemental balances are reactors under 7 MPa of nitrogen. After the gases were
good, except, because of air oxidation, the fractions total vented, the resid product was separated according to the
higher oxygen and lower hydrogen than the starting resid. separation scheme. The yields and analysis of each frac-
This does not seem to affect the coke-forming tendency tion are shown in Table 11. The gases and the fraction
as measured by Microcarbon Residue (MCR) because the that volatilized during solvent removal were not analyzed,
sum of the fractions equal the MCR of the starting resid, and the molecular weight of the coke could not be mea-
a result reported previously (Long and Speight, 1989b; sured because of insolubility in o-dichlorobenzene.
Roberts, 1989). The terms saturates and aromatics have Again the carbon content is relatively constant, with the
only relative meaning and should not be taken on an ab- carbon content of saturates actually being slightly higher
solute basis. Thus,the saturates are more saturated than than the carbon content of coke. The average molecular
the other fractions but still contain aromatic carbon and weight of each of the fractions (except coke) is decreased
sulfur. However, this fraction contains little or no oxygen, by thermal conversion. While the hydrogen contents of
532 Ind. Eng. Chem. Res., Vol. 31, No. 2, 1992

Table 11. Analvsis of Resid Reacted at 400 OC for 60 min

fraction yield/(wt %) wt % C wt % H H/C (atomic) wt % S wt%N VPO MW
gases 1.5
light volatiles 12.3
saturates 12.9 85.25 12.53 1.75 2.18 0.08 690
aromatics 25.8 83.77 10.10 1.44 5.04 0.11 470
resins 20.5 81.72 8.27 1.21 5.26 1.15 899
asphaltenes 18.3 83.16 6.34 0.91 7.23 1.50 2009
coke 8.7 82.19 5.54 0.80 7.63 1.78

Table 111. Thermal Conversion of Resid Fractions (60 min at 400 "C)
reactant product yield/(& %) wt % C wt % H H/C (atomic) wt % S wt % N VPO MW
saturates + aromatics volatiles 29.5
saturates 32.1 85.59 12.50 1.74 1.79 0.0 694
aromatics 38.4 83.95 10.59 1.50 4.17 0.0 345
resins 19.4 83.06 9.01 1.29 5.32 0.0 839
resins volatiles 10.8
saturates 5.7 84.36 12.03 1.70 2.77 0.17 670
aromatics 30.8 81.60 9.99 1.46 4.78 0.12 442
resins 30.6 82.01 8.44 1.23 4.60 2.18 804
asphaltenes 22.1 83.17 6.66 0.95 6.76 1.98 1841
asphaltenes volatiles 10.4
saturates 2.6 83.87 12.63 1.79 2.54 0.0
aromatics 14.2 81.79 10.36 1.51 4.92 0.0 422
resins 12.4 81.60 8.05 1.18 6.68 0.90 622
asphaltenes 21.0 82.00 6.18 0.90 7.96 2.18 1557
coke 39.4 82.42 5.50 0.80 7.92 1.73 7525

saturates and aromatics are not significantly changed by converted to another fraction. Again, the situation is no
thermal conversion, the hydrogen contents of resins and different if the resin or asphaltene is the reactant, a
asphaltenes are significantly decreased. This demonstrates product of another fraction, or a product of the entire resid.
one of the objections of Bunger and Cogswell (1981), that One difference is that resins formed from saturates and
asphaltenes change properties during conversion as well aromatics contain below the detectable nitrogen level.
as convert to other fractions. The conversion is more of Thus, resins need not contain nitrogen to adsorb on At-
a continuum than a discrete transformation. Meanwhile, tapulgus clay, showing that the separation is made on the
toluene-insolublecoke, not present in the resid, forms and basis of aromaticity or the size of aromatic rings.
exceeds the asphaltenes in concentrations of aromatics This is the first time that the elemental analyses and
(measured by H/C), sulfur, and nitrogen. molecular weights of five fractions of the thermal con-
A more direct way to track resid thermal conversion is version products of a resid and its fractions has been
to react the resid fractions separately and then to separate published. Much has been published on the thermal
the products according to the separation scheme, recog- conversion of asphaltenes with emphasis on the volatile
nizing that reactions and intermolecular interactions products (for example, Ritche et al. (1979) and Speight and
among the fractions may alter the chemistry when reacted Pancirov (1984)). Speight (1987) has characterized the
together in the resid. This was again performed for 60 min benzene-soluble, benzene-insoluble, and benzene-insolu-
at 400 "C in tubing bomb reactors. A summary of the blepyridine-soluble products of asphaltene thermal con-
result is shown in Table 111. In order to have sufficient version. While Speight observed a lower average molecular
quantity for reaction, the saturates and aromatics were weight (VPO measured in pyridine) for the benzene-in-
reacted together in the proportion found in the resid. soluble-pyridine-soluble product than that of the starting
In each case the thermal conversion of a fraction formed asphaltene, pyridine did not dissolve all of the benzene
the next more aromatic and higher molecular weight insolubles. In this study o-dichlorobenzene, a superior
fraction and the whole series of less aromatic and lower solvent for carbonaceous materials, was used to dissolve
molecular weight fractions. Thus, saturates plus aromatics the entire toluene-insoluble product of asphaltene ther-
formed the more aromatic and higher molecular weight molysis and show that, on average, the molecular weight
resins with a byproduct of gas and light volatiles that is higher for this product than the starting asphaltenes.
evaporate with the solvents. The resins formed more Actually, the combination of these two studies shows that
aromatic and higher molecular weight asphaltenes, with both routes to coke formation, higher molecular weight and
a byproduct of lower molecular weight and less aromatic lower molecular weight but more aromatic, happen si-
saturates and aromatics. The asphaltenes formed more multaneously.
aromatic and higher molecular weight coke with bypro- The path of saturates and aromatics to resins to as-
ducts of lower molecular weight and lower aromatic resins, phaltenes has been produced by oxidation reactions
aromatics, and saturates. Reactants of increasing aro- (Speight and Moschopedis, 1981) and postulated in a hy-
maticity and molecular weight produce an increasing yield droconversion kinetic model (Takatauka et al., 1989a).
of higher molecular weight-more aromatic byproduct at However, this is the first time that direct evidence of
the same thermal reaction conditions. thermal reactions following this path has been published,
As was seen with the whole resid, the saturates and let alone determining that each step involves molecular
aromatics fractions maintain similar aromaticity (H/C weight growth. A previous study of resin thermolysis
atomic ratio) during thermal conversion but decrease in (Pakash et al., 1980) emphasized the volatile products.
molecular weight. This is even the case when saturates Although Speight (1970) characterized the products of
and aromatics are formed from resins and asphaltenes. On thermal conversion of bitumen, asphaltenes, and maltenea,
the other hand, the resins and asphaltenes become more he carried the conversion too far (2-570 nonvolatile
aromatic and decrease in molecular weight prior to being benzene solubles) to reach conclusions as to how the in-
 Ind. Eng. Chem. Res., Vol. 31, No. 2,1992 533
termediate products evolved.
Concept of a Compositional Map
In agreement with Bunger and Cogswell (19811, this
author has found that no one property uniquely differen-
tiates one petroleum fraction from another. However,
Long (1981) has suggested that the composition of heavy
oils could be displayed in terms of a map of molecular
weight vesus "polarity". This concept has been more re-
cently extended by Long and Speight (1989a) into a map
of molecular weight versus solubility parameter. They used
desorption off Attapulgus clay with various solvents to
separate Cold Lake Crude and Arab Heavy vacuum resid i 0 130 DEGREES C

into eight fractions. Using the solubility parameter of the o 10 PO ao 40 60 eo 70 80

CONCENTRATION, DRAMS SOLUTE I KQRAMS SOLVENT
so io0

fraction and the 95 and the 5% points in the molecular

weight distribution determined by gel permeation chro- Figure 2. Asphaltene association broken by dilution is also broken
by temperature.
matography (GPC), they found that these two different
feeds gave similar molecular weight ranges for fractions
eluted by the same solvent. The molecular weight range H/C is a linear function of the fraction of aromatic carbons
increases sharply above a solubility parameter of 9 hilde- as measured by 13CNMR. He even proposes that benzene
brands ( ~ a l / c m ~ ) ' /However,
~. the relative amounts of the insolubles, asphaltenes, and n-pentane solubles each follow
fractions are different for the two feeds. a different linear function of either H/C or aromaticity
Possibly more important than the exact selection of GPC versus molecular weight as long as the fraction does not
molecular weight and solubility parameters as the inde- contain large amounts of oxygen (>12 wt %). However,
pendent variables is the concept of Long and Speight that if such a relationship exists, one could distinguish among
petroleum fractions may be distinguished with combina- the fractions with only one property, which one cannot.
tions of two properties: one that measures molecular at- Nevertheless, the use of H/C to measure aromaticity rather
traction and one that measures molecular size. The use than 13CNMR meets our need for convenience. However,
of solubility parameter as the measure of molecular at- since we have shown that the carbon content of petroleum
traction is awkward due to its indirect measurement. fractions from saturates to coke have nearly the same
While measuring molecular weight distribution is prefer- carbon content (81.6-86.5 wt %), without much loss in
red, GPC suffers from adsorption effects with petroleum accuracy, the hydrogen content can be used to measure
fractions that become more severe after partial conversion. aromaticity and, thus, molecular attraction.
Thus, to meet the objective of a convenient method to Measurement of Molecular Weight
track resid conversion, different independent variables for
a compositional map were further investigated. The measurement of the molecular weight of petroleum
asphaltenes is not an exact science (Speight et al., 1985),
Measurement of Molecular Attraction as the value depends on the technique (Dickie and Yen,
The feature that causes petroleum fractions to adsorb 1967), the solvent, and the temperature (Moschopedis et
strongly on Attapulgus clay and to be insoluble in paraf- al., 1976). The problem is that asphaltenes fall in that gray
finic liquids is their aromaticity. Since the more aromatic area between being in solution and being a colloid because
fractions have higher oxygen and nitrogen content, it is of their tendency to self-associate and to form aggregates
not easy to discount completely the influence of these that can be deteded by small-angle X-ray (Kim and Long,
heteroatoms. However, as already pointed out, resins 1979) and neutron (Overfield et al., 1989) scattering. As
formed from saturates and aromatics had no nitrogen, a result, techniques that measure molecular weight in so-
showing that it was not a necessary part of resins. In lution tend to give values that are high. On the other hand,
addition, the average resin of molecular weight 600-1000 the low volatility of asphaltenes interfere with mass
and the average asphaltene of molecular weight 1500-3000 spectrometry techniques and produce molecular weight
have one or fewer oxygen or nitrogen atoms, yet the resins measurements that tend to be low. Also, the strong
have at least 45% aromatic carbons and asphaltenes have tendency of asphaltenes to adsorb causes GPC techniques
a t least 50% aromatic carbons. Thus, while oxygen and to produce low molecular weight measurements because
nitrogen functionality may play a small role, aromaticity the higher molecular weight, more adsorbing asphaltenes
is the dominant cause of molecular attraction in petroleum are supposed to pass through the gel column first. The
fractions that cause adsorption on solids and insolubility most encouraging news is that neutron scattering (Over-
in paraffinic liquids. This conclusion contradicts the field et al., 1989) shows that asphaltenes dissociate rapidly
current view (Speight, 1980). However, if one wanted to with increasing temperature.
extend this approach to coal liquefaction products then This author has obtained the most consistent results in
a measure of molecular attraction involving oxygen func- measuring the number average molecular weight of un-
tionality would be required in addition to aromaticity and reacted and converted petroleum fractions by using vapor
molecular weight. This was the conclusion of Snape and pressure osmometry near the maximum instrument tem-
Bartle (1984, 1985) who found that three independent perature of 130 "C and using one of the best solvents,
variables, number average molecular weight, the proportion o-dichlorobenzene. An example result is shown in Figure
of internal aromatic carbon to told carbon, and the percent 2, where the apparent molecular weight of an asphaltene
of acidic OH, were needed to distinguish among benzene from a different resid than the one in Table I is plotted
insolubles, asphaltenes, and n-pentane solubles for coal- against concentration. When the measurement is done at
and petroleum-derived liquids. Thus, their "phase 70 "C, the apparent molecular weight decreases linearly
diagram" is three dimensional. with decreasing concentration. However, at 130 "C there
The simplest measure of aromaticity is the hydrogen to is little change in apparent molecular weight with con-
carbon atomic ratio (H/C). Ouchi (1985) has shown that centration, and the average molecular weight of 3400 at
534 Ind. Eng. Chem. Res., Vol. 31, No. 2, 1992
and unconverted resids. However, the resins and the as-
,o.*J phaltenea upon conversion tend to move to lower hydrogen
*oo*] Legend
content (higher aromaticity) and equal or lower molecular
weight. Since a pure compound would be a point on this
~

0
ASPHALTENES
0
z
7000-
0 RESINS diagram, the smaller the area of a pseudocomponent the

;
I
rnWCNE
-I
/\ROHATICS
0 SATURATES I
better is its approximation as a pure component. The
V COKE
1 saturates, aromatics, and resins occupy about the same
area, but the area for asphaltenes is much larger. The area
for coke is probably even larger, but one can only represent
1
coke that is soluble enough in o-dichlorobenzene at 130 "C
to measure its molecular weight.
The solvent-resid phase diagram also provides insight
on the transformation of one pseudocomponent to another.
Resins can be formed from aromatics by either molecular
o P s 7 8 ~ 1 i i ~ 1 a ~ i 6
I 4 I I I
WEIGHT PERCENT HYDROQEN weight growth or by decreasing hydrogen content by
Figure 3. Solvent-resid phase diegram. cracking off more saturated fragments or by aromatization
of naphthenoaromatics. On average, both mechanisms
130 "C agrees with the value of 3380 at 70 "C, extrapolated occur as was found with the resins formed from saturates
back to zero concentration. This is in contrast with vapor and aromatics. Likewise, asphaltenes form from resins
pressure osmometry measurements on the same asphaltene either by molecular weight growth or by decreasing hy-
in toluene at 50 "C that gives an extrapolated molecular drogen content. However, here it clearly does not occur
weight of 4900. Thus, it is concluded that either the better only by molecular weight growth, as the only asphaltenes
solvent or the higher temperature causes asphaltenes to of equal hydrogen content as resins are from uncoverted
dissociate down to a consistent molecular size. However, resids. On the other hand, coke probably does occur from
even this evidence does not ensure that the asphaltenes molecular weight growth of partially converted asphaltenes,
are completely dissociated. Asphaltenes (and coke) that as well as by decreasing hydrogen content. Of course, as
are partially converted associate even more strongly than we saw by reacting the pseudocomponents, each of the
unconverted asphaltenes. For these, both higher tem- thermal reactions produces the more desirable range of
perature and the better solvent are needed to maximize lower molecular weight-higher hydrogen content bypro-
the degree of dissociation. ducts.
Moschopedis et al. (1976) have proposed nitrobenzene
and pyridine as superior solvents for vapor pressure os- Implications and Discussion
mometry measurements on petroleum asphaltenes because The solvent-resid phase diagram points out some of the
they yield low molecular weight measurements. However, advantages and overcomes the greatest disadvantage in
observation of mixtures of these two liquids with as- using solubility and adsorption fractions for tracking resid
phaltenes between glass slides with an optical microscope chemical changes. It shows that pseudocomponents
at 600X reveals that they do not completely dissolve the characterized by solubility and adsorption can be distin-
asphaltenes. Thus, the low molecular weight measure- guished from each other by areas on a plot of molecular
ments are a result of the higher molecular weight fraction weight and hydrogen content. By being areas rather than
not being in solution. In this study, all mixtures of as- points, it also clearly shows that each pseudocomponent
phaltenes and coke with o-dichlorobenzene were checked represents a collection of a large number of different
with an optical microscope to ensure that they were in molecules and that chemical changes can occur without
solution to the limit of resolution (0.5 pm). changing one pseudocomponent into mothers. However,
even when this occurs, by measuring the hydrogen content
Solvent-Resid Phase Diagram and the molecular weight of the pseudocomponent, one can
In Figure 3, the number average molecular weight versus continue to track the path of the chemical change using
hydrogen content is plotted for fractions of eight different the solvent-resid phase diagram. This is most easily seen
resids and their thermal reaction products. Each of the for the asphaltenes in which all the asphaltenes from un-
five pseudocomponents occupy areas of the graph without converted resids lie within the dashed curve while all those
overlap with points representing other pseudocomponents. from converted resids lie outside these boundaries. This
Thus, it is concluded that the combination of number justifies the asphaltene reaction pathway presented by
average molecular weight and hydrogen content is capable Schucker and Keweshan (1980) and used by Savage et al.
of differentiating between petroleum fractions, whether (1985), in which asphaltene cores are considered to be
or not the fraction has been partially converted. For separate pseudocomponents from unconverted asphalt-
convenicence, solid curves are drawn to show that all enes. The similarity of the pseudocomponents after
species with molecular weight and hydrogen content that thermal reaction, no matter which pseudocomponent is the
lie to the right of one curve are heptane soluble and to the reactant, has often been assumed but not previously ex-
right of the other curve are toluene soluble. This, plus the perimentally observed.
unique area for each of the petroleum fractions, suggests In balance, solubility and adsorption fractionation is a
that the diagram is similar to a phase diagram but here logical experimental approach for tracking resid conversion
the different "phases" are due to their solubility behavior as long as one understands the limitations. While the
in particular solvents or due to their adsorption from so- pseudocomponents represent large distributions of mole-
lution. This is why it is called a "solvent-resid phase cules, the molecules have something in common (range of
diagram". molecular weights and hydrogen contents) and can be
The dashed curve in Figure 3 encloses all the points distinguished from each other. It one wishes to better
representing fractions from unconverted resid. The area define a pseudocomponent, he can experimentally frac-
representing convertad resids is much larger than the area tionate the pseudocomponent into two or more fractions.
occupied by unconverted resids. Actually, the aromatics However, the separation scheme presented here is already
and saturates occupy nearly the same area for converted quite labor intensive, requiring considerable motivation
 Ind. Eng. Chem. Res., Vol. 31, No. 2, 1992 535
dition of hydrogen reduces both the aromaticity by hy-
drogenation and the molecular weight growth by radical
V = VOLATILES capping. Thus, an even greater percentage of the hydro-
R w s + v S=SATURATES conversion products from each pseudocomponent are di-
rected to the right in Figure 4 toward less aromatic and

A +mAR
’ ”
R -
AR = AROMATICS
RESINS
A=ASPHALTENES
lower molecular weight products.
Acknowledgment
+

C = COKE

C *R+ AR + S + V I am grateful to Exxon Research and Engineering Co.

for permission to publish this paper and to J. L. Machusak,
W. R. Gerald, and K. Greany for careful experimental and
computer expertise.
+ *
MORE AROMATIC AND/OR LESS AROMATIC AN0
HIGHER MOLECULAR WEIGHT LOWER MOLECULAR WEIGHT Literature Cited
Figure 4. Thermal conversion reaction path. Bunger, J. W.; Cogswell, D. E. Characteristics of Tar Sand Bitumen
Asphaltenes as Studied by Conversion of Bitumen by Hydro-
pyrolysis. In Advances in Chemistry Series 295; Bunger, J. W.,
for additional steps. Another approach is to keep the Li, N. C., Eds.; American Chemical Society: Washington, DC,
number of separation steps to a minimum but model the 1981;pp 219-236.
chemical changes with a large number of molecular species, Dickie, J. P.; Yen, T. F. Macrostructures of the Asphaltic Fractions
such as the stochastic model of Savage and Klein (1989). by Various Instrumental Methods. Anal. Chem. 1967, 39,
However, one still needs to predict from the chemical 1867-1857.
structure of a molecule to which experimental pseudo- Kim, H.; Long, R. B. Characterization of Heavy Residuum by a
Small Angle X-ray Scattering Technique. Znd. Eng. Chem. Fun-
component it belongs. Savage and Klein were visionary dam. 1979,18,60-63.
enough to use criteria for their model based upon molec- Long, R. B. The Concept of Asphaltenes. In Advances in Chemistry
ular weight and H/C, albeit in detail their boundary as- Series 195; Bunger, J. W., Li, N. C., Eds; American Chemical
signments were inaccurate. For instance, they defined Society: Washington, DC, 1981;pp 17-27.
asphaltenes as having molecular weights above 300 and Long, R. B.; Speight, J. G. Studies in Petroleum Composition. De-
H/C greater than 1.0 (hydrogen content greater than 7.0 velopment of a Compositonal ”Map” for Various Feedstocks. Rev.
wt %). Figure 3 shows that such criteria would include Znst. Fr. Pet. 1989a,44,205-217.
Long, R. B.; Speight, J. G. Studies in Petroleum Composition. 2.
all resins and some saturates and aromatics while excluding Scale-up studies for Separating Heavy Feedstocks by Adsorption.
most of the asphaltenes from converted resids. Never- Znd. Eng. Chem. Res. 1989b,28, 1053-1057.
theless, now with the solvent-resid phase diagram, the Moschopedis, S. E.; Fryer, J. F.; Speight, J. G. Investigation of As-
assignments of a molecule in such a model to an experi- phaltene Molecular Weights. Fuel 1976,55,227-232.
mental pseudocomponent can be done with much greater Ouchi, K. Correlation of Aromaticity and Molecular Weight of Oil,
Asphaltene and Preasphaltene. Fuel 1985,64,426-427.
accuracy. In addition, this same diagram enables modeling Overfield,R. E.; Sheu, E. Y.;Sinha, S. L.; Liang, K. S. SANS Study
chemical changes that do not completely convert pseu- of Asphaltene Aggregation. Fuel Sci. Technol. Znt. 1989, 7,
docomponents. 611-624.
A mechanism that can rationalize much of the data Pakash, S.;Moschopedis, S. E.; Speight, J. G. Thermal Decompo-
presented in this paper is that the same aromatic moieties sition of Resins. Fuel 1980,59, 64-66.
that impart insolubility in solvents, adsorption on clay, and Ritche, G. R.; Roche, R. S.; Steedman, W. Pyrolysis of Athabasca Tar
Sands: Analysis of the Condensible Products From Asphaltene.
low volatility also impart a thermal reaction limit to the Fuel 1979,58, 523-530.
resid. As is represented in Figure 4,the saturates can Roberts, I. The Chemical Significance of Carbon Residue Data.
nearly be completely converted thermally to volatile Prepr.-Am. Chem. SOC.,Diu. Pet. Chem. 1989,34,251-254.
products. However, the resins and asphaltenes tend to Savage, P. E.; Klein, M. T. Asphaltene Reaction Pathways. V.
leave low-volatilityfragments that are more aromatic and Chemical and Mathematical Modeling. Chem. Erg. Sci. 1989,44,
lower in molecular weight after cracking off fragments that 393-404.
Savage, P. E.; Klein, M. T.; Kukes, S. G. Asphaltene Reaction
are less aromatic and lower in molecular weight. Even- Pathways. 1. Thermolysis. Ind. Eng. Chem. Process Des. Dev.
tually, these more aromatic fragments are converted to a 1985,24,1169-1174.
leas soluble pseudocomponent either by combining to form Schucker, R. C.; Keweshan, C. F. The Reactivity of Cold Lake As-
a higher molecular weight species and/or by sufficient phaltenes. Prepr. Pap.-Am. Chern. SOC.,Diu. Fuel Chem. 1980,
increase in aromaticity. On the other hand, aromatics form 25, 155-164.
higher molecular weight and more aromatic resins by direct Snape, C. E.;Bartle, K. D. Definition of Fossil Fuel Derived As-
phaltenes in Terms of Average Structural Properties. Fuel 1984,
conversion without passing through a more aromatic in- 63,883-887.
termediate. Naturally, as the aromaticity of a pseudo- Snape, C. E.;Bartle, K. D. Further Information on Defining As-
component is increased, it produces a greater fraction of phaltenes in Terms of Average Structural Properties. Fuel 1985,
more aromatic products (toward the left in Figure 4). 64,427-429.
Previously, the path to coke formation has either been Speight, J. G. Thermal Cracking of Athabasca Bitumen, Athabasca
classified completely as a polymerization process (Takat- Asphaltenes and Athabasca Deasphalted Heavy Oil. Fuel 1970,
49,134-135.
suka et al., 1989b) or completely as a molecular weight Speight, J. G. The Chemistry and Technology of Petroleum; Marcel
reduction and aromaticity increase process (Savage and Dekker, Inc.: New York, 1980;pp 192-243.
Klein, 1989). Our evidence is that it is a combination of Speight, J. G. Initial Reactions in the Coking of Residua. Prepr.-
these two paths, except that the molecular weight growth Am. Chem. SOC.,Diu. Pet. Chern. 1987,32,413-418.
is more of an oligomerization process involving the com- Speight, J. G.; Moschopedis, S. E. On the Molecular Nature of Pe-
bination of two to five molecules. troleum Asphaltenes. In Advances in Chemistry Series 195;
Bunger, J. W., Li, N. C., Eds.; American Chemical Society:
While the examples of resid conversion in this paper Washington, DC, 1981;pp 1-15.
have all been thermal conversion, the principles should also Speight, J. G.; Pancirov, R. J. Structure Types in Petroleum As-
apply to catalytic hydroconversion. The solvent-resid phaltenes as Deduced from Pyrolysis/Gas Chromatography/Mase
phase diagram is process independent. However, the ad- Spectrometry. Liq. Fuels Technol. 1984,2,287-305.
536 Ind. Eng. Chem. Res. 1992, 31, 536-543
Speight, J. G.; Long, R. B.; Trowbridge, T. D. Factors Influencing Chem. Eng. Jpn. 1989a,22, 298-303.
the Separation of Asphaltenes from Heavy Petroleum Feedstocks. Takatsuka, T.; Kajiyama, R.; Hashimoto, H. A Practical Model of
Fuel 1984,63,616-620. Thermal Cracking of Residual Oil. J. Chem. Eng. Jpn. 1989b,22,
Speight, J. G.; Wernick, D.L.; Gould, K. A.; Overfield, R. E.; Rao, 304-310.
B. M. L.; Savage, D. W. Molecular Weight and Association of
Asphaltenes: A Critical Review Rev. Znst. Fr. Pet. 1985,40,51-61. Received for review April 15,1991
Takatsuka, T.; Wada, Y.; Hirohama, S.; Fukui, Y. A Prediction Revised manuscript received September 3, 1991
Model for Dry Sludge Formation in Residue Hydroconversion. J . Accepted September 14,1991

Multivariable Controller for Distillation Columns in the Presence of

Strong Directionality and Model Errors
Alessandro Brambilla*
Dipartimento d i Ingegneria Chimica, Uniuersitci di Pisa, Via Diotisalvi, 2, 56100 Pisa, Italy

Luigi D’Elia
C O N T A S Process Control s.r.l., Via U. Visconti, 24/b, 56122 Pisa, Italy

An easy-tuning multivariable controller useful for the control of the two products of distillation
columns is presented. The controller is based on the singular value decomposition (SVD) of the
model transfer function matrix and requires only one tuning parameter, with values in the range
0-1, by which a trade-off between the nominal performance and its sensitivity to the model/real
process mismatch is accomplished. The development of this control algorithm does not require any
particular software, and the criteria for the tuning procedure are reported in the paper. The controller
can be easily implemented and on-line tuned on the modern control systems. The effectiveness of
the controller and the transparent meaning of its parameter are shown by the simulation of a liquefied
petroleum gas (LPG)splitter and a high purity distillation column.

Introduction present because of the following reasons: necessary sim-

Most of the industrial chemical processes have the plifications adopted in modeling the process; lack of
characteristic of multiinput/multioutput system (MIMO). knowledge of the process parameters; use of linear model
Among these processes, the distillation units are the ones to describe nonlinear processes; errors in assigning the
which might show to a large extent the peculiarities of the exact value of the manipulated variables to the process.
MIMO systems: the interactions between the control loops Lau et al. (1985) proposed a method to design a struc-
and the process directionality. tural compensator able to minimize control loop interac-
In this paper, after a brief recall of these characteristics tions based on the singular value decomposition. The
and their effect on the performances of a control scheme, effectiveness of such compensator in distillation has been
a multivariable controller based on the singular value verified by Bequette and Edgar (1989).
decomposition (SVD)is presented. The main advantage The relative gain array (RGA) (Bristol, 1966) is an ad-
of the controller presented rests on the possibility of its equate tool for the interaction analysis although based on
making a trade-off between performance and robustness the steady-state model of the process. Shinskey (1984) and
to model/process mismatches, by means of only one pa- Mc Avoy (1983), have used the RGA for the selection of
rameter which makes the tuning simple and transparent. the control system structure for distillation columns. The
The application of the controller to distillation columns ij element of the RGA is defined as
characterized by strong directionality in the presence of 16Yi/6mjlmK,K#j
model errors is presented. X,J = (1)
Loop Interactions. When the control structure is de- 16Yi/6mjlYK,K#i
centralized, i.e., each product composition is controlled by
the action of one manipulated variable as for single-in- where y and m are respectively the controlled and ma-
put/single-output systems (SISO),the presence of the nipulated variables. For a 2 X 2 system with the process
interactions between the two control loops may lead to a transfer function at the steady-state G(s=O) = Go, the
severe degradation of the control system performances. application of eq 1 gives
The classic approach with multivariable controllers allows
us to solve this problem at least from a theoretical point
of view (Luyben, 1970; Ray, 1981). Unfortunately, it is not 1--
g11g22
always possible to attain improvements through the ap-
plication of these multivariable techniques to real pro- where gij are the gains of the matrix Go.
cesses. In fact, conventional multivariable controllers with According to the definition, the interaction degree is
the ability to eliminate or reduce the interactions are based measured as the ratio between the process gain with all
on the process model, or more precisely on its inverse the loops open and the gain with all the other loops closed.
(“inverse-based”controllers), and their performances are Thus, the interactions are small when X i j is close to 1.
in some conditions very sensitive to the mismatch between Although the definition of eq 1is limited to the steady
the model and the real process (model errors) (Skogestad state, Mc Avoy (1983) has extended the use of the RGA
and Morari, 1988). These model errors are generally in the frequency domain.
OSSS-5SS5/92/2631-0536$03.00/0 0 1992 American Chemical Society