Upgrading of middle distillate fractions of a

syncrude from Athabasca oil sands

Michael F. Wilson and Jerry F. Kriz

CANMET, Energy Research Laboratories, Department of Energy, Mines and Resources,
555 Booth Street, Ottawa, Ontario KIA OGI, Canada
(Received 29 April 7983)

Current processes for upgrading bitumen from Athabasca oil sands produce synthetic crudes which are
high in aromatics and deficient in hydrogen. As a consequence, middle distillate fractions derived from
these syncrudes produce diesel fuels of low cetane number and jet fuels which are hydrogen deficient.
Results obtained from bench-scale hydrotreating experiments indicate that quality fuels may be
produced from Athabasca syncrudes. Middle distillate fractions from this source were subjected to high-
severity hydroprocessing in a continuous-flow reactor unit using conventional hydrotreating catalysts
which were pre-sulphided by a mixture of Hz/H& Aromatic hydrogenation at high temperatures and
pressures was affected by the approach to thermodynamic equilibrium, however, at lower temperatures,
in some cases virtually 100% saturation was achieved and treated fractions were found to meet cetane
number and jet fuel smoke point requirements. Data treatment in the present study includes a model for
the hydrogenation kinetics and correlations between aromatic carbon and fuel combustion properties.

(Keywords: oil; oil sands; syncrude; middle distillate fraction)

Synthetic crude oil is presently produced in Canada from hydrotreating (which followed delayed coking in the
two commercial operations which process bitumen from commercial process) but the product still did not meet the
Athabasca oil sands in northern Alberta. In both plants Canadian cetane number specification for diesel fuel.
the synthetic crude is produced by upgrading the bitumen This study was carried out to investigate effects of
through coking processes and further refining by severe additional hydroprocessing on the product quality, parti-
hydrotreating. cularly its diesel ignition and smoke point properties. The
The hydrotreating removes sulphur, nitrogen, and also aim was to establish the extent of conversion required to
hydrogenates olefins and some aromatics. The extent of enhance the fuel combustion quality and also to correlate
aromatic saturation is a function of the process severity. cetane ratings and smoke point values for various fuels
Much of the middle distillate produced from Athabasca with their aromatic content. It was also necessary to
oil-sand syncrudes remains highly aromatic. Diesel fuels evaluate the activity of the hydrotreating catalyst since the
which are produced from highly aromatic feedstocks have literature contains little work on catalyst selection for
poor ignition qualities and jet fuels from these sources saturating aromatics in this type of feedstock. The catalyst
cause overheating in turbo engines resulting in increased activity was compared by elucidating the reaction ki-
maintenance costs. netics. Finally, the effect of the thermodynamic equilib-
Blending of synthetic with conventional crudes is the rium on the aromatic saturation process was investigated
current practice for accommodating unconventional and the optimum operating conditions were established
feedstocks in refinery operations. The amount blended is a to be used as a basis for future work.
function of the total product mix. Canadian refineries are
usually limited to a synthetic crude intake of between 20
EXPERIMENTAL
and 30% and middle-distillate fuels are usually the
limiting factor. The feedstock was a middle distillate fraction cut from a
Additional hydroprocessing of the synthetic distillates synthetic crude oil produced by delayed coking of Athab-
may be an acceptable alternative for producing quality asca bitumen. The properties of the feedstock are given in
fuels. Adding hydrogen to saturate aromatics in con- Table 1. Three different commercial catalysts were used for
ventional distillates and converting them to naphthenes hydroprocessing the feed; these were typical hydrotreat-
has been practiced by refineries for some time. It is ing catalysts, i.e., COO-MOO&-A1203, NiO-Moo,/
important to note that although saturation of aromatics y-A1,OJ and NiO-WO,/y-A1,OJ. The properties of the
may improve the hydrogen/carbon ratio of synthetic catalysts are presented in Table 2.
distillates, the disproportion of paraffins in these fuel A detailed description of the hydrotreating system has
sources is a fundamental problem. Increasing the naph- been published previously’. Experimental runs were
thene content by conversion of aromatics may only carried out using a bench-scale continuous-flow reactor
marginally improve the combustion properties of a operating in the up-flow mode. Liquid feed and hydrogen
fuel. For example, the aromatic content of the feedstock were mixed, passed through a pre-heater and then over a
used in this study was reduced significantly by severe fixed bed of catalyst in a 316 stainless steel 0.305 x 0.025 m
0016-2361/84/02019~7$3.00
@ 1984 Butterworth & Co. (Publishers) Ltd
190 FUEL, 1984, Vol 63, February
 Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz
pressure was also varied in the range 7.00-17.3 MPa.
Table 1 Properties of middledistillate feedstock
Experimental runs were performed at temperatures of
Relative density (15/l 5°C) 0.836 34@-44O”C,liquid space velocities of 0.75-2.25 h-’ and a
Carbon (wt%) 87.5 hydrogen flow rate of 530dm3 (hydrogen) drne3 (feeds-
Hydrogen (wt%) 12.5 tack) (3000 scf bbl-l). The reactor system was maintained
Sulphur (wt%) 0.06
Nitrogen (ppm) 54
at steady-state conditions for 1 h prior to and 2 h during
Average molecular weight 201 the period in which product was collected.
% Aromatic carbon by 13C n.m.r. 14.3 The hydrotreated product was analysed for aromatic
Cetane number 36 carbon content by 13C n.m.r. analysis using a Varian XL-
Distillation (D86) 200 spectrometer. Sulphur content was analysed by the
Initial b.p. “C 142
10% 181
Wickbold technique and carbon, hydrogen and nitrogen
50% 232 analysis was carried out using a Perkin Elmer 240 B
90% 283 analyser. The average molecular weight of the feedstock
Final b.p. 333 was determined with a Knauer vapour pressure osmo-
meter. The catalyst total pore volume was determined
with a Micromeritics Model 910 mercury porosimeter.
The mean reactor residence time for aromatic saturation
Tab/e 2 Catalyst properties
was calculated from a knowledge of the total voidage of
Catalyst the catalyst bed, the hydrogen flow rate and the average
molecular weight of the feed. Complete vaporization of
COO-MOO&-Al,03 Cyanamid Aero Trilobe HDS-20
feed on contacting the catalyst bed was assumed.
Containing: 10.8 wt% MO and 3.9 wt% Co Based on operating conditions determined from the
Surface area: 230 mz g-l
Pore volume: 0.43 ml g-l
work on reaction kinetics, samples of fuel were generated
Extrudates of 2.5 mm average diameter by continuous processing and tested to evaluate their
combustion quality. Cetane numbers were determined
Cyanamid Aero HDS-9A using a standard Cooperative Fuels Research test engine.
Containing: 11.7 wt% MO and 2.5 wt% Ni A standard smoke point test (one of the jet fuel quality
Surface area: 170 m2 g-l indicators) was also performed.
Pore volume: 0.39 ml g-l
Extrudates of 1.5 mm average diameter
RESULTS AND DISCUSSION
NiO-WO,ly-A1,03 Katalco Sphericat NT-550

Containing: 15.8 wt% Wand 3.9 wt% Ni

Effect of thermodynamic equilibrium on aromatic saturation
Surface area: 200 m2 g-l A high degree of aromatic saturation of a middle
Pore volume: 0.58 ml g-l distillate may be achieved using a single-stage hydrotreat-
Spheres of 0.32 mm average diameter
ing unit operating under severe process conditions. Ho-
wever, the conversion to naphthenes is a reversible
reaction and involves a chemical equilibrium which is
temperature dependent:
i.d. tubular reactor having a total capacity of 155 cm3. The
reactor furnace consisted of three separate heating zones. Aromatic + nH,#Naphthene (1)
The heat input to each zone was controlled to maintain
the catalyst bed at a constant isothermal temperature. The At high temperatures the extent of conversion may
axial temperature profile was measured by a moveable therefore be limited by the approach to thermodynamic
thermocouple located in a thermo-well mounted centrally equilibrium. Because of this limiting factor it is important
in the reactor. For each experimental run the reactor was to find the optimum operating conditions.
filled sequentially from the bottom with 42cm3 of berl Since aromatic compounds generally have higher re-
saddles, 100cm3 of catalyst extrudates or spheres and lative densities than their corresponding naphthenes it
13 cm3 of berl saddles. was expected that the relative density would decrease with
The catalyst was sulphided at 380°C and atmospheric the extent of conversion. Figure I shows a plot of the
pressure by passing a mixture of 10% H,S in hydrogen relative density of hydrotreated distillate as a function of
over the bed for 2 h. The volume of H,S passed was liquid hourly space velocity (LHSV) for each reaction
equivalent to eight times the amount of sulphur required temperature using the sulphided cobalt-molybdenum and
for the formation of sulphides. Immediately after sulphid- nickel-tungsten catalysts. The values show a steady
ing, the reactor was pressurized with hydrogen ready for decrease with reaction temperature, the effect being
the first run. The unit was run for 8 h on oil before the first greater for the cobalt-molybdenum catalyst. Thus, with
steady-state sample was taken. The catalyst loadings to increase in temperature the values do not pass through a
the reactor were 70 g for the Co-Mo/Al,O, and N&W/ minimum as would be expected from other experimental
Al203 catalysts and 81 g for the Ni-Mo/Al,O, catalyst evidence. It should be mentioned at this point, however,
respectively (these amounts were required for 100cm3 of that aromatic hydrogenation may be accompanied or
reactor volume). followed by cracking reactions such as ring opening or
For most of the experimental runs the reactor pressure chain fragmentation. Reduction of the overall mol&ular
was maintained at 17.3 MPa (2500 psig). (This condition weight would naturally be indicated by lower densities.
was chosen because the feedstock was previously hy- The cracking reactions are known to be enhanced with
drotreated in primary upgrading at 10.4 MPa (1500 psig) increasing temperatures. It is concluded therefore that the
over a sulphided Ni-Mo/Al,O, catalyst.) To investigate decrease in relative density at the higher temperatures
the effect on the thermodynamic equilibrium reaction the (40&44O”C) is caused by reactions other than aromatic

FUEL, 1964, Vol 63, February 191

Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz

0.83

0.82
t

3
z 44O'C /
6 a
0

34O’C
I I I I
0.82 360°C 360 380 400 420 4. 3
380°C Temper&we ( “C )
4OO’C
0.81 Figure 2 Effect of liquid hourly space velocity (LHSV) on
aromatics saturation over sulphided nickel-tungsten catalyst,
420°C
Pressure 17.3 MPa. A, LHSV 2.25; B, LHSV 1.75; c, LHSV 1.25;
D, LHSV 0.75
0.80 440°C

I I I I I I I
1.o
Liquid space velocrty (K’)
2.0
1°.O1
Figure 1 Plots of relative density of hydrotreated product vs
liquid space velocity for sulphided a, cobalt-molybdenum and b,
nickel-tungsten catalysts

saturation such as cracking of the distillate to lighter

products.
Figures 2 and 3 show the effect of change of liquid
hourly space velocity on the % aromatic carbon in
the hydrotreated product using the nickel-tungsten and
cobalt-molybdenum catalysts, respectively, over the com-
plete temperature range. For both catalysts all curves
show a minimum within a temperature range 38@-4OO’C.
Above this temperature the effect of the thermodynamic
equilibrium is observed as the aromatic carbon level
begins to increase again. The plots also show that, at any
operating condition, the extent of aromatic conversion
was significantly greater using the nickel-tungsten ca- 0 I I I I I
340 360 380 400 420 440
talyst. At 17.3 MPa pressure, the maximum conversion Temperature (‘C)
was achieved at 380°C and 0.75 space velocity. Under Figure 3 Effect of liquid hourly space velocity (LHSV) on
these conditions the aromatic carbon was reduced to aromatics saturation over sulphided cobalt-molybdenum catalyst,
0.41% corresponding to a 97 % conversion. pressure 17.3 MPa. A-D as for Figure 2
Although very good reproducibility for the data in
Figures 2 and 3 has been confirmed, a simple explanation
for all the patterns is not apparent. For instance, the Figure 4 shows the effect of hydrogen partial pressure
curves in the equilibrium domain (high temperature end) change on the conversion of aromatic carbon in the
should be converging together at the same aromaticity feedstock over sulphided nickel-tungsten catalyst. The
level for both catalysts, but a spread in the aromatic results cover the complete temperature range for a
carbon of l-2% is evident. The consistently higher constant liquid hourly space velocity of 1.00. It is observed
product aromaticity when using the CO-MO catalyst that over the pressure range 7.OtS17.3 MPa the degree of
(even near equilibrium) suggests some differences in aromatic carbon conversion is reduced significantly. Each
aromatic-type distributions resulting from the use of the curve demonstrates the effect of the equilibrium condition
two different catalysts. Versus saturation kinetics, the minimum being shifted to

192 FUEL, 1984, Vol 63, February

 Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz

rate expression for the reversible reaction is then given by:

16.0 -dA
1 - =k,A(P,J”-k;N (2)
dt
where: kl and k; = the forward and reverse rate constants:
and PHt= the hydrogen partial pressure.
Introducing assumption 1. equation (2) becomes:
-dA
-=krA-k;N
dt
where: /+= k, (I’“,)” = constant.
At equilibrium the forward and reverse reaction rates
are equal and therefore:

Also, by a simple mass balance:

A+N=A,+N, (5)
where: A, and N, = the equilibrium fractions of aromatic
and naphthene carbons respectively. Using equations (3),
(4) and (5) it follows that:
-dA
-= k,(l + AJN,)(A -A,)
dt
Putting k,= kf(l + AJN,) and integrating, the final ex-
pression is obtained:
ln A-4
__ = -k,t
A,-A,
where A0 = the initial aromatic concentration. The
I I I I
340 360 360 400 420 4 equilibrium aromatic fractions were estimated from the
Temperature (“C) literature data of Wilson et al.’ and Frye3.
Figure 4 Effect of hydrogen pressure on aromatics saturation Figure 5 compares pseudo-first-order-reversible rate
over sulphided nickel-tungsten catalyst, LHSV 1.00. A, 7.00 MPa; plots for the sulphided cobalt-molybdenum and nickel-
9, 10.4 MPa; C, 13.9 MPa; D, 17.3 MPa molybdenum on alumina catalysts over the temperature
range 340400°C. A good fit to the kinetic data found
gives support to the assumptions made in developing the
rate expression. Figure 5 also shows that the sulphided
the left with decreasing hydrogen pressure. It should also nickel-molybdenum catalyst has a higher activity than
be noted that at 440°C and 7.OOMPa the equilibrium cobalt-molybdenum for hydrotreating the feedstock. Fi-
percentage of aromatic carbon is 15.9 and greater than gure 6 shows a similar plot for the sulphided nickel-
that of the feedstock. It is apparent that the data revealed tungsten catalyst again indicating the superior activity of
by these plots have importance in determining the this catalyst in comparison with the molybdenum ca-
optimum operating conditions needed to produce fuels talysts. For example, at any given temperature the
which meet the required combustion standards. The reaction rates for the nickel-tungsten catalyst were found
implications of these operating conditions on fuel quality on average to be greater than those of cobalt-
are discussed later. molybdenum by a factor of two.
The pseudo-rate-constants, k,, were determined from
Kinetics of aromatic saturation the kinetic data by regression analysis and used in the
The equilibrium reaction given by equation (1) is the Arrhenius equation:
basis for elucidating the reaction kinetics. Two simplifying lnk,=lnA-E/RT (8)
assumptions are also made:
where: A = the Arrhenius pre-exponential factor; E = the
1. The forward reaction is assumed to be pseudo-first-
activation energy; and R = the gas constant.
order since the hydrogenation is carried out in large
Arrhenius plots for the three catalysts used are pre-
hydrogen excess at constant partial pressure.
sented in Figure 7. Regression analysis on the data gave
2. The reverse reaction is also assumed to be first-order in
values of activation energies and pre-exponential factors
naphthene carbon.
which are shown in Table 3. The activation energies 63 and
The aromaticity of the feedstock is defmed as the per- 60 kJ mol-’ for the sulphided cobalt-molybdenum and
centage of aromatic carbon (“/,c,) as determined by ’ 3C nickel-tungsten catalysts agree within experimental error
n.m.r. The fraction of carbon atoms which are aromatic is but the corresponding value of 83 kJ mol-’ for the nickel-
therefore given by A = %CJlOO. Similarly the fraction of molybdenum catalyst is significantly higher. All of the
naphthene carbon atoms is given by N = %C,,/lOO. The activation energies compare reasonably well with the

FUEL, 1984, Vol 63, February 193

 Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz

li 3 I I I 1 b 1 I I I I
0 2 4 6 8 0 2 4 6 8 10
Time (rnin)
Figure 5 Pseudo-first-order reversible kinetics for aromatics saturation over sulphided a, cobalt-molybdenum and b,, nickel-molybdenum
catalysts, 0, 340°C; A, 360°C; l . 38OC; 0, 400°C. Pressure 17.3 MPa

-l.,

3
I I I I
1.5 1.6
103/T (OK-‘)
3
Figure 7 Arrhenius plot of pseudo-rate constants for aromatics
saturation. A, Ni-W; l , Ni-Mo; 0, Co&lo

I I I 1 I Figure 6 Pseudo-first-order reversible kinetics for aromatics

0 2 4 6 8 10 saturation over sulphided nickel-tungsten catalyst. 0, 34VC; A,
Time (min 1 360°C; a, 380°C; 0, 4OO‘C. Pressure 17.3 MPa

‘I 94 FUEL, 1984, Vol 63, February

 Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz

depending on the nature of the catalyst used. Furimsky’

Tab/e 3 Activation energies and log preexponential factors for
indicated that, in the bulk, MoS, should be a more stable
saturation of aromatics
sulphide than WS,. This difference in stability would
Activation contradict the strong poisoning effect of H,S on the
Catalyst energy nickel-tungsten catalyst. However, it may be argued that
(sulphided) Feed (KJ mole-‘) InA
the important sulphide stabilities are those at the surface,
Co-Mo/AlzOJ this study 63 + 9 10.5 i 1 .8 i.e., at the bases or edges of crystals. Also the effect of the
Ni-Mo/AlzO, this study 83+- 3 14.7 f 0.5 promoting metals should not be discounted.
Ni-W/Al,O, this study 60 + 4 10.6 k 0.8
-
Ni-W/A1,03 Benzene 71
Effect of aromaticity on diesel and jet fuel combustion
quality
Figure 8 presents a plot of per cent aromatic carbon ( %
value of 71 kJ mol -i obtained by Voorhoeve and Stuiver4 C,) uersus cetane number for fuel products generated
for benzene hydrogenation over sulphided nickel- under various hydroprocessing conditions using the sul-
tungsten on alumina at 300-4OO”C. Figure 7 also reveals phided nickel-tungsten on alumina catalyst. In all cases
that the activity of the nickel-molybdenum catalyst the pressure was 17.3 MPa while temperature and liquid
approaches that of nickel-tungsten at the higher hourly space velocity were varied. The percentage aro-
temperatures. matic carbon and cetane number of the feedstock is also
A previous study4 of hydrogenation kinetics using indicated in the plot together with the current Canadian
nickel-tungsten catalysts demonstrated that the hydroge- diesel fuel specification (cetane number 40). The figure
nation sites can be blocked by hydrogen sulphide. In the demonstrates the severity of processing required to pro-
present study low concentrations of the gas were liberated duce good quality fuel. It is apparent from the graph that
during hydrodesulphurization. It is concluded that the the first 50 % of aromatic carbon conversion increases the
sulphur content of the feedstock has an adverse effect on cetane rating by only two points, i.e., from 36 to 38, and a
the reaction rate and it is the number of active sites on the 75 % conversion produces a fuel of ignition quality which
catalyst surface that is affected by H2S and not their is still just below the specification rating. Thus, at least a
intrinsic activity. It is widely accepted that the active sites 90% conversion is required to raise the cetane number
in these catalysts are caused by anion vacancies in the well within the specification limit. It is apparent that the
catalyst surface which expose the reactive metal centres. It remaining aromatic hydrocarbon species are very stable
is also accepted that the anion vacancies are in equilib- and resistant to saturation under these severe hydrop-
rium with the pressures of hydrogen sulphide and hy- recessing conditions. They also appear to have a signi-
drogen gas thus: ficantly more adverse effect on diesel ignition quality than
0 + H,S&S(ads.) + H2,g) (9) those which are removed under less severe conditions. It is
significant that low concentrations of these compounds
where: 0 =an anion vacancy. Increasing the hydrogen can have such a detrimental effect on fuel ignition quality.
pressure at the catalyst surface under these conditions Figure 9 presents a similar plot obtained by processing
should therefore have the effect of converting adsorbed fuel products using the sulphided cobalt-molybdenum
sulphur to H,S and creating active sites for aromatic catalyst. It is noteworthy that the correlation is not as
hydrogenation. From the experimental results presented good as in Figure 8 since the points show more scatter
above it may be concluded that, by this mechanism, which is probably caused by more extensive cracking
sulphided nickel-tungsten on alumina responds well to observed with the sulphided cobalt-molybdenum catalyst
large increases in the hydrogen partial pressure. Further over the same temperature range, as mentioned pre-
work5v6 on nickel-tungsten catalysts indicated that the viously. It may be assumed that saturation of aromatics is
active sites for aromatic hydrogenation in the sulphided the only major process occurring when using the nickel-
catalyst are W3 + ions and it was postulated that hy-
drogenation proceeds via n-bonded adsorption com-
plexes in which the hydrocarbons are bound to tungsten 16.0 1
ions in the WS2 surface.
Some discussion explaining the difference in behaviour
of the three catalysts is needed. Recently Weisser and
Jaffar’ investigated the effect of the presence of H,S on
catalyst performance for benzene hydrogenation. In the
absence of H,S a sulphided nickel-tungsten on alumina
catalyst was found to be much more active than cobalt-
molybdenum when tested for the hydrogenation of pure
benzene at 300°C and 8 MPa hydrogen partial pressure.
However, in the presence of 1~01% H,S a severe re-
tardation of the hydrogenation activity was observed. The
original high activity was restored upon removal of H$
from the feed. The impeding effect of H,S was much more
pronounced in the case of the sulphided nickel-tungsten
catalyst and the performance of the two catalysts became 01 ’ I I I I I I I I I -I
36 36 40 42 44 46
comparable in the presence of H,S. Based on the me- Cetane number
chanism of reaction (9), which involves the sulphiding of Figure 8 Correlation of product cetane number with aromatic
anion vacancies, it is apparent that these vacancies would carbon by 13C n.m.r. Reactions over sulphided nickel-tungsten
be liable to undergo sulphiding to a different extent catalyst

FUEL, 1984, Vol 63, February 195

Upgrading of middle distillate fractions of a syncrude: M. F. Wilson and J. F. Kriz

16 .O-

12
980 I
. .
z I
t\ I
L . I
0"
f,
.o -

36 38 40 42 44 4 .o -
Cetane number

Figure 9 Correlation of product cetane number with aromatic

carbon by 13C n.m.r. Reactions over sulphided cobalt-
molybdenum catalyst
o-
20 25 30
Smoke point (mm)
tungsten catalyst, the result being a smooth correlation. Figure 10 Correlation of smoke point (ASTM D1322) with
These graphs also reinforce the findings from the kinetic aromatic carbon of product by 13C n.m.r. Reactions over
work that the cobalt-molybdenum catalyst is less suitable sulphided nickel-molybdenum catalyst
for saturation of aromatics in these fuels. For this catalyst,
in spite of the reduced activity, the cetane number
specification can still be achieved. Reference to Figure 4
shows that to produce good quality diesel fuel from this adverse effect on diesel fuel ignition quality. The results
feedstock using the sulphided nickel-tungsten catalyst, a from hydroprocessing and engine testing also show that
hydrogen partial pressure of x12MPa would be fuel of acceptable quality, i.e. cetane number 45, may be
required. produced from this feedstock. Similar correlations of
A correlation of XC, versus smoke point determined on smoke point with aromatic carbon content reveal that this
fuel products from hydroprocessing at different tempera- particular jet fuel specification can easily be achieved.
tures and space velocities and 17.3 MPa using the sul-
phided nickel-molybdenum catalyst is presented in Fi- ACKNOWLEDGEMENTS
gure 10. This shows that the specification limit of 20 mm is
easily surpassed using this catalyst. The authors wish to thank Michael Stolovitsky for
The present work can be summarized as follows. carrying out hydrotreating experiments, Professor G. W.
Approximately 97% of the aromatic carbon in the Buchanan, Carleton University, Ottawa for “C n.m.r.
feedstock may be converted by severe hydroprocessing analysis, N. Kallio of the National Research Council for
over a sulphided nickel-tungsten on alumina catalyst cetane number determinations and the Alberta Research
giving a fuel product which is predominantly naphthenic. Council and Suncor Inc. for samples of synthetic crude oil.
Kinetic experiments demonstrate the superiority of this
catalyst over the corresponding molybdenum types for REFERENCES
achieving the greatest conversions. The investigation
established the optimum operating conditions for satu- Ranganathan, R., Teman, M. and Parsons, B. I. CANMET Report
76-15, January 1976
ration of aromatics given that the equilibrium reaction is Wilson, T. P., Caflisch, E. G. and Hurley, G. F. J. Phys. Chem: 1958,
an important factor in determining such conditions. 62, 1059
Cetane engine testing and correlation of ignition quality Frye, C. G. J. Chem. Eng. Data 1962,7,592
with aromatic carbon content has cast doubts on the Voorhoeve, R. J. H. and Stuiver, J. C. M. J. Cutal. 1971,23,228
Voorhoeve, R. J. H. J. Catal. 1971,23,236
efficiency of promoting fuel quality by this kind of
Voorhoeve, R. J. H. and Stuiver, J. C. M. J. Catal. 1971,23,243
hydroprocessing. It is apparent from these correlations Weisser, 0. and Jaffar, A. S. Chemicky Prumysl 1982,32/57,237
that low levels of stable aromatic species may have an Furimsky, E. Catal. Rev.-Sci. Eng. 1980, 22, 371

196 FUEL, 1984, Vol 63, February