CA2075108C - Instrumentation for dilution of bitumen froth - Google Patents

Abstract

A refractometer is used to monitor naphtha/bitumen ratio in diluted bitumen froth containing high water and solids contents. The addition of naphtha to the froth is controlled in response to the refractometer readings.

Description

1 Field of the Invention 2 This invention relates to an improvement of the hot 3 water process for extracting bitumen from oil sand. More 4 particularly, it relates to a method for controlling light S hydrocarbon diluent addition to a bitumen froth stream, to ensure 6 that the diluent to bitumen ratio remains substantially constant 7 at a pre-determined desirable value. The method is based on use 8 of a refractometer to establish measures indicative of the 9 bitumen and diluent concentrations in a diluted froth stream and using such measurements to guide diluent addition to froth to 11 yield diluted froth having a constant diluent/bitumen ratio.

13 As stated, the invention has to do with a useful 14 technique for controlling the addition of diluent to bitumen froth. As such, it relates to only a part of an overall known 16 process referred to as the hot water process for recovering 17 bitumen from oil sand. This overall process is described in 18 detail in the patent literature. However, for purposes of 19 background it is appropriate to now give a description of the steps of the hot water process.
21 The feedstock to the hot water process is as-mined oil 22 sand.

20751~8 1 The oil sand contains relatively small amounts of 2 bitumen. Typically, it may analyze at:
3 11% bitumen 4 84% mineral solids 5% water 6 There is, however, considerable variation in oil sand 7 quality. The bitumen content may be as high as 16% or as low as 8 6%, with corresponding variation in the other components. As 9 well, there is variation in the particle size of the mineral solids. The coarser particles are silica, usually in the form 11 of quartz. They typically have a size equivalent to that of 12 grains of sand on a beach. The fine material is mostly clay 13 minerals. These clay particles are very small, for instance 1 14 or 2 micron in diameter.
As a general rule, those oil sands that most readily 16 release their bitumen in the extraction process are those rich 17 in bitumen and low in fine minerals content. But for economic 18 reasons and due to the inevitable inflexibility of mining, any 19 ores down to 6% bitumen content are processed.
The oil sand is mined using draglines and stockpiled 21 in large dumps called windrows. It is then dug from these 22 windrows and dropped on to conveyors, that transport it to the 23 extraction plant.
24 In the extraction plant, the oil sand is fed to tumblers wherein it is mixed with hot water and a small amount 26 of NaOH. Steam is added to ensure the exit temperature of the 1 product slurry is about 80C. This treatment effects a 2 prellminary dispersal of the bitumen away from the solids and 3 into the aqueous phase. Four process trains, each comprising a 4 tumbler, are used in parallel. Each tumbler produces slurry at the rate of about 1312 kg per second.
6 On emerging from the tumblers, the slurry is screened 7 to remove oversize solid rejects and diluted with additional hot 8 water. The volume of diluted slurry produced by each train is 9 about 1587 kg/s.
The screened diluted slurry is then advanced to a 11 primary separation vessel (PSV), forming part of the train. This 12 is a wide cylindrical tank having a conical bottom, said tank 13 being of large volume by comparison with the tumbler. The in-14 coming slurry thus has a long residence time in the PSV, which means, in effect, that it is held in a quiescent state. Under 16 these conditions, the different components of the slurry can 17 separate.
18 The coarse sand (density of about 2.65 kg/dm3) sinks to 19 the bottom of the PSV. It is withdrawn as a dense sand/water stream. This stream is substantially oil-free and so is 21 discarded as tailings.
22 The bitumen, meanwhile, is in the form of globules 23 suspended in the PSV contents. Bitumen density is close to that 24 of water so, left to themselves, the globules would simply hang in the suspension. However, air bubbles were entrained during 26 tumbling. The bitumen has a strong affinity for the bubbles, 1 forming itself into films around them. In this way the bitumen 2 is rendered buoyant and floats to the surface of the PSV
3 contents, where it collects as a layer of bitumen-rich froth.
4 The froth is pushed over the vessel lip as new diluted slurry continuously enters the vessel. The froth falls into a launder 6 surrounding the vessel and is collected for further treatment.
7 The froth product from the PSV is referred to as 8 primary froth. Each PSV produces at the rate of about 160 kg/s.
9 The flotation process is by no means entirely selective for bitumen. Substantial amounts of fine solids which float 11 along with the bitumen and water is also taken up in the froth 12 structure. A typical analysis of primary froth would be:
13 66.4% bitumen 14 8.9% solids 24.7% water.
16 In the PSV, some bitumen fails to float. This bitumen 17 remains suspended in the vessel contents ("middlings"), usually 18 in the form of small flecks. There is enough bitumen in the 19 middlings to warrant an add-on process to recover it.
Accordingly, a stream of middlings is continuously withdrawn and 21 fed into sub-aeration cells. In these cells, the slurry is 22 subjected to vigorous agitation and air is drawn through from 23 underneath. This treatment results in essentially all the 24 bitumen flecks being formed into a froth which is recovered. The product is referred to as secondary froth, to distinguish it from 26 the primary product of the PSV. Because of the harsh conditions 2a7sl0~

1 used in its formation, the secondary froth has a much higher 2 level of impurities. It typically analyses at:
3 23.8% bitumen 4 17.5% solids S 58.7% water.
6 The secondary froth is allowed to stand in a settler, 7 whereupon much of the loose water sinks and the cleaned product 8 is recovered. Even after settling, however, the product is still 9 considerably less pure (based on bitumen content) than the primary froth. For this reason, one seeks to operate the 11 extraction process so that as much bitumen as possible is 12 collected in the form of primary froth. Due to variations in 13 geological history throughout a lease, however, some ores yield 14 less primary froth than desired, and operators may rely on secondary extraction to collect a substantial portion of the feed 16 bitumen.
17 The froths are blended to provide a combined froth 18 product. Typically this product analyses at:
19 62.1% bitumen 9.1% solids 21 28.8% water.
22 It should be emphasized that these are typical values only. In 23 reality, the quality (bitumen content) of the combined froth 24 varies considerably depending on the nature of the oil sand from which it is derived.

-1 As the typical analysis shows, the combined froth still 2 contains considerable amounts of solids and water. The bitumen 3 is not yet suitable for advancing to downstream upgrading 4 processes. The impurities are tightly bound into the froth structure, so special processes are needed to remove them.
6 Conventionally, two stages of centrifuges have been used. More 7 particularly, the froth is first treated in a scroll centrifuge 8 to remove coarse solids, and then in a disc-type centrifuge to 9 remove fine solids and water.
In order for the froth to be cleaned in the centrifuge 11 circuit, it is necessary that the froth first be diluted with 12 light hydrocarbon diluent (normally naphtha), in which the 13 bitumen component readily dissolves.
14 Having the froth hydrocarbon in the form of a solution of bitumen in diluent serves two purposes:
16 1. The solution has a low viscosity, compared with 17 that of bitumen. Impurities can more easily 18 separate, due to the mobile nature of the 19 hydrocarbon phase; and 2. The solution has a lower density than bitumen.
21 This increases the density difference between the 22 hydrocarbon and the contaminants. When 23 centrifugal separatory forces are applied to the 24 diluted mixture, separation of the phases is facilitated.

.
1The solids and water (collectively referred to as the 2 sludge component) recovered from the centrifuge circuit are 3 rejected as waste, while the hydrocarbon (bitumen dissolved in 4 naphtha) is the product.
5Because it is a highly-refined petroleum fraction, 6 naphtha is valuable and so must be recovered for re-use. To this 7 end, the purified bitumen/naphtha solution is heated to distil 8 out the naphtha, which is then condensed by cooling. The 9 residual bitumen is in a high state of purity and ready for upgrading.
11The amount of naphtha used as froth diluent needs to 12 be kept to a minimum. There are at least four reasons for this:
131. Heat used in the distillation step is expensive;
142. The fuel for this heat comes from the hydrocarbon lSproduced in the upgrading of bitumen. In effect, 16useful product has to be burned;
173. By the use of diluent, the total volume of feed 18to the froth purification process is increased.
19But the naphtha is only a carrier. If an excess 20is used, there is less equipment capacity for the 21froth components. In consequence, extra 22centrifuges must be installed, and associated 23equipment built unnecessarily large, all of which 24leads to high capital and maintenance costs; and 20~10~
1 4. In centrifuging, there is always some loss of 2 naphtha with the process tailings. When an excess 3 of naphtha is used, such losses increase.
4 For these reasons, there is a strong inducement to avoid using an excess of naphtha. On the other hand however, 6 the naphtha/bitumen ratio (N/B ratio) must be high enough to 7 lighten the bitumen fraction so that good separation between 8 hydrocarbon and sludge can take place.
9 These competing demands require that the ratio of naphtha to bitumen be maintained within narrow limits. This, in 11 turn, requires determining the N/B ratio in actual commercial 12 circumstances in such a way that the ratio can be adjusted for 13 most effective naphtha consumption.
14 For ease of measurement, one would determine bitumen in the hydrocarbon product emerging from the purification 16 process. Unfortunately, in a large plant, there is a 17 considerable delay between the feed entering and the product 18 emerging. The centrifuge process to purify bituminous froth 19 takes at least 20 minutes. So, analysis of the product is always 20 minutes out of phase compared with the feed. Even results 21 measured instantaneously using samples of purified product would 22 be of reduced value because the composition of in-coming froth 23 often changes faster than this delay. The change in froth 24 composition arises from the constantly changing nature of the oil sand feed. Therefore purified product analyses are not desirable 26 for use as a basis for ad]usting the N/B ratio of the froth.

1 For the purpose of controlling the process, the most 2 useful results would be given by monitoring the diluted froth 3 feed. This requires a method that responds to the composition 4 of the hydrocarbon phase, but is not significantly affected by the presence of the considerable amount of contaminants.
6 Therefore, while continuous reliable monitoring of the separation 7 product composition would be advantageous, monitoring of the 8 process feed mixture composition is preferred. If the feed had 9 only low levels of contaminants, one could fall back on conventional measurements such as density. As the bitumen 11 content of a blend increases, so will the density, because the 12 density of bitumen is higher than that of naphtha.
13 Unfortunately, by the same logic, once solids and water 14 are included in the system being measured, these contaminants contribute the influence of their own densities to the apparent 16 density of the total mixture. Apparent density is the sum of 17 (fraction x density) for each component of the mixture.
18 Therefore the solids have a strong influence due to their high 19 density, and one portion of water is as influential as one portion of bitumen because both substances have a density of 1Ø
21 In practice, this means that, although density does 22 change with bitumen content, the changes have little significance 23 due to the strong influence of the contaminants relative to the 24 narrow desired range of N/B ratio. Thus, density meters have been found undesirable, even for the measurement of N/B ratio of ~075108 1 the product stream from the process. The same reasoning is true 2 if viscosity is used to monitor the diluted froth composition.
3 In the absence of a reliable N/B ratio monitor, the 4 industry has resorted to adding an excess of naphtha, to ensure that the N/B ratio is always high enough for the worst case froth 6 feed. When the naphtha is distilled out of the product for re-7 use, extra heat energy is consumed due to the excess of naphtha.
8 In the case of the present assignees' oil sands plant, it is 9 estimated that the extra cost of operating at a "worst case" N/B
ratio is at least $15 million per year and may be as high as $20 11 million at current levels of production.
12 Poor operation of the process may signal a need to 13 increase naphtha, but an excess of naphtha is unnecessary. The 14 practice has been to sample the product from the separation each lS two hours, and subject the sample to laboratory analysis for 16 determination of N/B ratio. Thus, up to four hours could pass 17 from a change in feed composition to discovery that the N/B ratio 18 is incorrect.
19 There has therefore long existed a need to identify a means that is effective for continuously and accuruately 21 indicating in timely fashion the diluent to bitumen ratio in a 22 diluted bitumen froth stream, before it undergoes purification.

2 The present invention is based on the discovery that 3 a refractometer, placed in contact with unpurified diluted 4 bitumen froth, will accurately measure the refractive index of the hydrocarbon phase, which index provides a measure indicative 6 of the diluent/bitumen ratio. Surprisingly, the solids and water 7 contents of the froth do not interfere with the measurements.
8 Having established that the refractometer will function 9 to provide an accurate indication of the diluent/bitumen ratio, the invention comprises the following combination of steps:
11 - mixing light hydrocarbon diluent with a stream of 12 bitumen froth prior to subjecting the stream to 13 purification for separation of contained solids 14 and water from the bitumen;
- applying a refractometer in contact with the 16 diluted stream to continually take measurements 17 indicative of the diluent to bitumen ratio of the 18 hydrocarbon phase of the stream; and 19 - varying the diluent addition to the stream of bitumen froth in response to the measurements to 21 maintain the ratio at a substantially constant 22 pre-determined value.

207~10~
-2 Figure 1 is a schematic showing a refractometer in 3 conjuction with a circuit for diluting bitumen froth with naphtha 4 prior to purification;
Figure 2 is a comparative plot of naphtha in product 6 samples, as determined by refractometer and by gas 7 chromotography;
8 Figure 3 is a schematic illustrating a pump loop with 9 a refractometer in the loop, for refractometer testing;
Figure 4 is a plot showing the correlation between %
11 bitumen determined by refractometer and N/B ratio determined by 12 standard lab techniques of a product stream over a one week 13 interval; and 14 Figure 5 is a plot showing % bitumen by refractometer, standard lab techniques, and the pump rate, for an in-line 16 refractometer on the process feed at the inclined plate settler.

18 Having reference to Figure 1, the invention comprises 19 providing a source 1 of bitumen froth connected by a pump 2 to a line 3 to a froth purification circuit 4. A source 5 of 21 diluent ~typically naphtha) is connected by a pump 6 and line 7 22 to the line 3, for adding diluent to the froth to form diluted 23 froth. A refractometer 8 is connected in operating relationship 24 with the line 3, downstream of the diluent addition point. The refractometer 8 should be well downstream of the mixing point to 1 ensure that the diluent and bitumen froth are well mixed prior 2 to taking a measurement. A controller 9 is connected with the 3 refractometer 8 and diluent pump 6. The controller 9 is 4 operative to control the pump 6 in response to the output of the refractometer 8, to maintain the N/B ratio substantially constant 6 at a predetermined value. Alternatively, or in addition, the 7 pump 2 for the bitumen froth source 1, may be controlled in 8 response to the refractometer output to maintain the N/B ratio 9 at the desired value.
The refractometer is a process critical angle 11 refractometer as is known in the art, preferably intrinsically 12 safe. The crystal in the refractometer should be hard enough to 13 avoid scratching by solids contained in the froth. An example 14 of a suitable hard crystal is sapphire.
At 20C, the typical refractive index ("R.I.") of 16 bitumen is 1.569, and of naphtha is 1.413. Blends of these two 17 components have an R.I. which depends on the fraction of each 18 present in the blend.
19 Readings from the refractometer may be in the form of R.I. Alternatively, the instrument may be suitably calibrated 21 to give % bitumen readings. This is achieved by setting the 22 refractometer at 0~ for the R.I of the naphtha and a full scale 23 response of 100% for the R.I. of bitumen. Thus, scale readings 24 give % bitumen in the hydrocarbon phase directly.
R.I. of the naphtha used may vary somewhat, depending 26 on the source of the naphtha. This can be dealt with by 1 adjusting the zero setting of the refractometer for the 2 particular naphtha being used. Thus, the refractometer may be 3 set to read 0% with pure naphtha. The span should then be set 4 to read 100~ with pure bitumen. Alternatively, it can set up using pure liquids of known R.I. and the readings then related 6 back to the R.I. of naphtha/bitumen blends of known composition.
7 R.I. variations due to temperature fluctuations are 8 compensated for automatically by the refractometer, as is known 9 in the art.
The refractometer is preferably mounted in such a way 11 that a portion of the sample flows against the prism face of the 12 refractometer, yet water or solids do not collect on the prism 13 or within any depression within which the prism is mounted.
14 Thus, a weir-type valve body, known for use in association with refractometers, may be used. On larger streams, some alternate 16 means of directing a flow against the prism face can be provided, 17 such as a flow deflector within a large pipe which deflects a 18 portion of the stream against the prism face.
19 Preferably the refractometer is placed in an upward flowing stream to prevent air entrapment in the sample. Where 21 this is not possible, the instrument should be mounted so that 22 the prism face is on the side of the horizontal flow line, in 23 order to ensure that water separating from the stream and flowing 24 along the bottom of the line, or gas separating and flowing along the top, are not measured by the refractometer.

207alO8 1 The invention is supported by the following examples.
2 Example I
3 This example describes laboratory runs that established 4 that R.I. readings, for known diluted mixtures of bitumen and naphtha, varied in proportion to the actual naphtha/bitumen 6 ratio.
7 More particularly, a series of 89 plant product samples 8 taken over time, which contained bitumen and naphtha with solids 9 and water removed, were used. The samples had been analyzed using standard laboratory analysis by gas chromatography (GC) to 11 establish the bitumen and naphtha contents.
12 An Anacon model 47 refractometer (Anacon Instruments, 13 Marlborough MA, U.S.A.) was used. It was first calibrated using 14 standard refractive index liquids to obtain experience with the span and zero setting required to zero the instrument for various 16 liquid R.I. and to set spans corresponding to various changes in 17 R.I. expected for the mixtures to be tested.
18 The results are illustrated in Figure 2. As shown, the 19 naphtha determinations by the refractometer closely follow those by gas chromatography. It should be noted that the sample at 21 time about 54 which does not match was later rechecked and it was 22 determined that the G.C. reading was incorrect.

23 Example II
24 This example describes runs carried out in a simple 2S test circuit to determine the effect on refractive index measured 207~108 1 by the refractometer when water or solids were added in 2 progressively increasing amounts to a naphtha/bitumen mixture.
3 A pump loop as illustrated in Figure 3 was used, 4 comprising a holding tank 10 filled with a measured amount of a bitumen-naphtha mixture 11, and agitated by a mixer 12. A pump 6 13 pumped the mixture through a conduit 14 to a refractometer 15 7 and back into the holding tank 10. A chart recorder 16 recorded 8 the output from the refractometer 15. The refractometer 15 was 9 a Liquid Solids Control Co. Model 829 refractometer (West Upton, MA, U.S.A.).
11 To calibrate the refractometer 15, its zero setting was 12 set to correspond with the measured R.I. of pure naphtha. Its 13 span was set to cover the expected range of R.I. values for the 14 bitumen/naphtha mixtures to be tested, using a linear change relationship in R.I. between pure naphtha and pure bitumen 16 measurements. Data from the initial laboratory measurements of 17 pure liquids was used to estimate the settings.
18 The refractometer 15 was mounted in the line 14 with 19 a weir-type valve body (not shown) attached to the refractometer measuring head, so that part of the liquid flow contacted the 21 prism face. This valve body was adapted to direct the sample 22 flow against the measurement portion of the prism face to ensure 23 that the instrument measured the current composition of the 24 stream passing it and further to ensure that water or solids would not collect in the depression within which the prism was 1 mounted. The weir-type valve body used was obtained from the 2 suppliers of the refractometer.
3 The circuit was operated initially with a mixture of 4 bitumen and naphtha only, as a stabllity test. A constant R.I. reading was obtained. There was only a slight drift with 6 time due to temperature drift (no temperature compensation used 7 in this example). An injection of 3% isopropyl alcohol was made 8 to ensure the refractometer was running properly.
9 There was a lack of impact in the R.I. readings from adding water to the bitumen-naphtha mixture (8% naptha). The 11 water was added in four increments up to 5.3~. Apart from an 12 initial "blip" as the slug of water came through, there was no 13 appreciable change. Similar results were obtained from the 14 addition of solids (seven increments) up to 10.6%.
Example III
16 This example describes the performance of the L.S.C.
17 Model 829 refractometer when installed on a hot water process 18 plant stream. More particularly, the refractometer was mounted 19 on a 2 inch diameter pipe forming a side stream loop in series with a B.S. and W. meter (a water content measuring instrument), 21 communicating with the product line from the final centrifuge 22 stage of purification. In this location, the refractometer was 23 tested on streams that contained only low concentrations of water 24 and solids. It was not possible at the time to locate the refractometer on the line feeding diluted froth to the centrifuge 26 circuits.

1 The refractometer was calibrated, zeroed and mounted 2 as described in Example II. Data from the refractometer was 3 continuously logged by a data logger, and the readings (as %
4 bitumen) were compared to results from laboratory grab samples (N/B ratio determined by centrifuge removal of water and solids 6 followed by density measurements of the hydrocarbons).
7 Instrument readings were taken at the nominal time the grab 8 sample was supposed to be taken, but it should be appreciated 9 that samples were in reality taken sometime within the two hours up to the nominal sample time, as convenient.
11 Representative results are shown in Figure 4. In 12 Figure 4, the left scale gives the results of the refractometer 13 readings (solid line results). The scale is given as a chart %
14 scale of 0 - 100. It should be explained that, to obtain this scale, the refractometer scale was expanded, that is the scale 16 was expanded and the zero setting adjusted upwardly in order to 17 obtain a centre scale approximately equal to the normal operating 18 point and an operating range somewhat greater than the expected 19 variations in R.I. readings over time. Thus, the chart % does not directly correspond to % bitumen from the R.I. readings. The 21 right scale gives the results of the laboratory measurements 22 (star line results) as N/B ratio by volume. Figure 4 23 illustrates the rather large uncontrolled changes in N/B ratio 24 experienced in the plant. The data further illustrate that the refractometer readings are in substantial agreement with the 26 values obtained by laboratory analyses.

`1 Example IV 2075108 2 This example describes the performance of a Liquid Solids 3 Control Model 725 refractometer on a hot water process plant 4 stream having high water and solids contents. More particularly the refractometer was mounted on the feed line to a single stage 6 inclined plate circuit operating as a scalping unit for the feed 7 to the centrifuge circuits. The refractometer was calibrated, 8 zeroed and mounted as described in Example II. A series of 9 different feed streams containing different sources of naphtha were tested.
11 The results are shown in Table I and plotted in Figure 5, 12 comparing % bitumen results from the refractometer, standard lab 13 techniques, and pump rate (calculated bitumen).
14 As shown, the % bitumen determined by the refractometer is in substantial agreement with the lab determination and expected 16 results. Thus, the refractometer worked reliably with a feed 17 bitumen stream, still unpurified and containing typical water and 18 solids contents. The refractometer was sensitive to N/B ratio 19 changes and provided measurements indicative of the ratio. In fact, in the last sample of 88-05-10 Plant 14 Naphtha, when the 21 refractometer reading was out of range from that expected, the 22 system was checked and it was determined that the pump had stuck 23 open at the wrong speed. This shows the value of continual 24 monitoring of the feed stream. The refractometers employed in the examples are based upon the measurement of the critical angle 26 therefore, they are known as critical angle refractometers.

;,,. r,.

207~108 -Date Expected Refractometer Laboratory Pump Rate calc.
/time % Bitumen % Bitumen' ~ ei tumen % Bitumen 88-03-30 NRU Naphtha 12:30 57.1 57.5 53.7 57.1 14:30 57.1 56.9 50.3 57.3 88-04-06 Plant 14 Naphtha 12:30 57.1 56.8 56.4 59.3 14:35 57.1 58.7 59.6 59.7 88-04-07 Plant 7 Naphtha 10:30 64.5 62.9 62.6 65.3 12:55 64.5 66.6 62.5 65.3 15:03 64.5 66.4 62.0 65.3 88-04-08 Plant 7 Naphtha 10:47 57.1 58.0 56.8 58.2 88-04-13 Suncor Naphtha 10:43 57.1 56.Z 58.0 54.4 12:57 57.1 56.7 58.1 57.1 14:52 57.1 58.5 58.5 57.4 88-04-14 Plant 6 Naphtha 10:42 64.5 65.5 63.9 65.8 13:00 64.5 65.6 64.0 64.4 14:27 64.5 65.1 63.7 65.2 88-04-28 Plant 14 Naphtha 10:44 64.5 61.5 61.7 62.2 12:36 64.5 62.9 62.0 62.0 L~ , I
88-05-02 Plant 14 Naphtha 10:52 57.1 57.3 59.2 60.5 88-05 10 Plant 14 Naphtha 12:40 64.5 35.2 ? 33.9*' ' Corrected for naphtha variations ' Pump control had stuck open.

Claims (4)

1. A process comprising:
mixing light hydrocarbon diluent with a stream of bitumen froth prior to subjecting the stream to purification for separation of contained solids and water from the bitumen;
substantially continuously measuring the refractive index of the resultant mixed stream containing at least four components with a critical angle refractometer to obtain measurements indicative of the diluent-to-bitumen ratio of the hydrocarbon phase of the stream; and varying the proportions of diluent and bitumen froth mixed in response to the measurements to maintain the ratio at a substantially constant predetermined value.

2. The process as set forth in claim 1, wherein the proportions of diluent and bitumen froth mixed are varied by varying the rate of diluent addition.

3. The process as set forth in claim 1, wherein the diluent is naphtha.

4. The process as set forth in claim 2, wherein the diluent is naphtha.