[0001] The present invention relates to a process for dewaxing hydrocarbons employing low-speed
scraped-surface chillers.
[0002] More specifically, this invention relates to an improved process for dewaxing hydrocarbon
oils, particularly petroleum oils, most particularly lube oils wherein said waxy oil
is introduced at a temperature above its cloud point into a first chilling zone divided
into a plurality of stages, passing said waxy oil from stage-to-stage of said chilling
zone, introducing a cold dewaxing solvent into at least a portion of said stages,
whereby a solvent-waxy oil mixture is formed, maintaining a high degree of agitation
in at least a portion of the stages containing solvent and waxy oil, thereby effecting
substantially instantaneous mixing of said solvent and said waxy oil while cooling
said solvent-waxy oil mixture, preferably at a rate of from 0.55 to 4.44°C (1 to 8
0F) per minute, as it progresses through said first chilling zone to a temperature
greater than the temperature at which the wax is separated from the oil i.e., the
separation temperature, but less than about 22.2
0C (40
0F), above said separation temperature, whereby a substantial portion of the wax is
precipitated from said waxy oil under conditions of said high degree of agitation,
and forming a solvent-oil mixture containing precipitated wax, withdrawing said mixture
containing precipitated wax from said first chilling zone, and cooling the same to
the separation temperature in a second chilling zone comprising scraped surface chillers,
thereby precipitating a further portion of said wax from said waxy oil and separating
said wax from the oil-solvent in solid-liquid separation means, the scraped surface
chillers of the second chilling zone being operated at a speed of only from 5 to 20%,
preferably 8 to 14%,of the original design operating speed. This approximately 10
fold decrease in scraper element speed results in obtaining wax filtration rates improved
on the order of 10 to 20%, preferably 15 to 20%, while heat transfer coefficients
are either uneffected or reduced by only about 15%. This loss of heat transfer efficiency
is more than compensated by the improved wax. filtration rates obtained.
[0003] It is known in the prior art to dewax hydrocarbon oilstocks by cooling an oil-solvent
solution in scraped surface heat exchangers before separating the crystallized wax
from the oil by physical means. In U.S. Patent 3,775,288 - it is taught that scraped
surface heat exchangers can be used as a secondary cooling zone for the dewaxing of
hydrocarbon oils following a primary cooling zone in which oil is cooled by contacting
said oil with a cold solvent at a plurality of points along a vertical tower while
maintaining a zone of intense agitation in at least a portion of the points of solvent
injection, such that substantially in- staneous mixing occurs at each point, i.e.,
within a second or less. This first cooling zone has become known as DILCHILL, a registered
service mark of Exxon Research and Engineering Company. In the standard DILCHILL operation,
oil is cooled by the injection of a chosen dewaxing solvent along the various stages
of the DILCHILL tower with intense agitation from the cloud point to a temperature
about 40°F above the separation temperature of the wax-in-oil typically followed by
additional cooling in scraped surface chillers to the separation temperature.
[0004] It has been discovered that the above process is improved by reducing the speed of
the rotating elements in the scraped surface chiller to a speed of from about 5 to
20
%, preferably 8-14% of the original design operation speed of the scraped surface chiller.
Operation at this reduced speed improves wax filtering rates by about 10 to 20%, preferably
15 to 20%, while not adversely affecting the heat transfer performance of the chillers.
This approximately 10 fold decrease in scraper element speed resulting in improved
wax separability (as by filtration) is accompanied by no more than a 15% loss of heat
transfer efficiency. This loss of heat transfer efficiency is more than made up by
the improved wax separation [filtration] rates obtained.
[0005] Any hydrocarbon oilstock, petroleum oilstock, distillate fraction or lube oil fraction
may be dewaxed by the process of this invention. In general, these stocks will have
a boiling range within the broad range of about 500°F to about 1300°F. The preferred
oil stocks are the lubricating oil and specialty oil fractions boiling in the range
of 550°F to 1200°F. These fractions may come from any source such as paraffinic crudes
obtained from Aramco, Kuwait, the Pan Handle, North Louisana, Western Canada, etc.
The hydrocarbon oil stock may also be obtained from any of the synthetic crude processes
now practiced or envisioned for the future such as coal liquefaction, tar sands extraction,
shale oil recovery, etc.
[0006] Any low viscosity solvent for oil may be used in the process of this invention, representative
of such solvents are the ketones having 3 to 6 carbon atoms such as acetone, methylethylketone
(MEK), and methylisobutyl-ketone (MIBK) and the low molecular weight hydrocarbons
such as ethane, propane, butane, propylene and the like, as well as the mixtures of
the foregoing ketones and mixtures of the aforesaid ketones with C
6 to C
10 aromatic compounds such as benzene and toluene. In addition, halogenated low molecular
weight hydrocarbons such as dichloromethane and dichloroethane and mixtures thereof
may be used as solvents. Specific examples of suitable solvent mixtures are methylethylketone
and methylisobutylketone; methylethylketone and toluene; acetone and toluene; acetone
and propylene; benzene and toluene; dichloromethane and dichloroethane. The preferred
solvents are the ketones. A particularly preferred solvent mixture is a mixture of
methylethylketone and methylisobutylketone or a mixture of acetone and propylene.
Another preferred solvent mixture is methylethylketone and toluene.
[0007] General operating conditions of the instant invention are recited and presented in
detail in U.S. Patents 3,775,288 and 3,773,650, both of which are incorporated by
reference. The instant application is an improvement over both of these patents by
demonstrating improved dewaxing and wax filterability by reducing the speed of the
scraper element in the scraped surface chilling units which follow the DILCHILL process
tower.
[0008] Scraped surface chillers such as those used in combination with DILCHILL towers described
in U.S. Patent 3,773,650 and 3,775,288 generally operate at rotating element speeds
from 14 to 30 revolutions per minute. This rotation is in response to the need to
remove wax from the chiller walls since build-up of wax on the cooling surfaces results
in a substantial decrease in the chilling efficiency of the units. The build-up of
wax on the chilling surfaces and internals also has the effect of effectively blocking
the flow paths of the precipitated wax/oil solvent slurry increasing the pressure
drop through the unit. It has now been surprisingly discovered that scraped surface
chillers can be run efficiently at approximately a 10 fold decrease in rotating element
scraper speed, i.e., at speeds of from 1.5 to 2.4 RPM, and that such a speed reduction
does not hamper the heat exchange ability of the chiller nor result in an unacceptable
pressure drop across the chiller. In fact, and surprisingly, it has been found that
the wax coming from such a unit wherein the rotating element is operating at the reduced
speed exhibits a surprisingly in- crease/improvement in separability (i.e., filterability)
yielding a wax cake which does not clog the filter clothes or filtering means typically
employed in a solid-liquid separating means. The frequency of wax removal from the
wall is sufficient to maintain adequate heat transfer rates, but significantly reduces
the-addition of "wall crystals" to the slurry which are responsible for reducing filtration
performance.
EXAMPLE A
[0009] The lead and lag scraped surface chillers in a typical
DILCHILL process stream had their rotating element speed reduced frp, 14 RPM to 2.3
RPM in the 6" lead chiller and from 24 RPM to 1.8 RPM in the 8" lag chiller. This
reduction in rotating element speed resulted in an increase in laboratory measured
wax filter rates by 15% on LP 150N and LP 600N oils.
[0010] A base DILCHILL process stream (sequence 1-4) and a test DILCHILL process stream
(monitored both before sequence (lA, 2A) and after sequence (3A, 4A) element speed
reduction) were compared so as to evaluate the degree of benefit obtained by the speed
reduction when handling LP 150N oil stock. Table I summarizes the results of the runs.

[0011] The outlet feed filter rate averaged 8.80 m
3/
m2 day on the base stream for the entire test period which represents 81.3% of the average
filter rate entering the base process stream. The performance of the test stream prior
to the speed reduction is similar (Sequence lA, 1B), averaging 8.97 m
3/m
2 day or 79.7% of the filter rate entering the test process stream. When the speed of
the rotating elements in the scraped surface chillers in the test process stream was
reduced (Sequence 3A, 4A), the filter rate increased 23% to 11.00 m
3/m
2 day; 89.2% of the entering filter rate. Comparing the outlet-inlet filter rate ratio
of the test process stream before (79.7%) and after (89.2%), the speed reduction reveals
a 12% increase in the throughput as the result of the slow speed scraping.
[0012] A similar test was conducted employing LP 600N as the feedstock. This test differed
from the previous one, however, in that the before and after data from the test process
evaluation stream comes from two different runs. See Table 2. In Table 2 speed reduction
is shown only for DILCHILL Tower II (Test Stream) (Sequence A3-C3). The comparison
is to a separate and distinct DILCHILL Tower I (Standard stream) (Sequence Al-Cl).
No direct comparison was run on the test stream with the scraped surface chillers
running at normal speed. At a different time, however, DILCHILL Tower II (Test Stream)
had been run with the scraped surface chillers run at normal speed (Sequence A2-D2).

[0013] Comparing the absolute outlet filter rates before (3.25 m
3/m
2 day) (Sequence
A2-
D2) and after (3.92 m
3/m
2 day) (Sequence A3-C3), the speed reduction, based on approximately equal entering
filtering rates for the different runs reveals an improvement in the test process
stream of about 21%.
EXAMPLE B
[0014] The base DILCHILL Process stream and the test process stream (before and after speed
reduction) were evaluated to determine the effect of the reduction in the speed of
the rotating elements on liquids/solids ratio for LP 150N, and for LP 600N. Based
on approximately equal liquids/solids ratios from the DILCHILL towers, it was determined
that the reduction in speed in the scraped surface chiller had little absolute effect
on the liquids/ solids ratio exiting the process streams. Tables I and II present
this data.
EXAMPLE C
[0015] The base DILCHILL process stream and the test process stream, again before and after
speed reduction, were compared and evaluated to determine the effect of the speed
reduction on heat transfer efficiency. Table III summarizes the results of the comparison
tests. For given test sequences, as when comparing the before and after values for
either LP 150N (Sequences lA, 2A and 3A, 4A) or LP 600N (Sequences Al, Bl, Cl and
A3, B3, C3) the slurry flow volumes, flow velocities and chill rates are relatively
equivalent.

[0016] The heat transfer coefficients were calculated for each sequence using average throughput
and temperature data for five 1-hour periods during each sequence and averaging the
result. For the LP 150N runs, it can be seen that very little change occurred in heat
transfer performance of the 6" scraped surface chillers after the speed reduction.
An 18% decrease in heat transfer coefficients was recorded for the 8" scraped surface
chillers after the speed reduction. For the LP 600N runs, very little difference is
seen in the heat transfer coefficients of the 6" chillers while about a 12% reduction
in heat transfer coefficient was measured for the 8"chiller. Since a majority of the
heat removal occurs in the 6" chiller, the overall debit in heat transfer is about
6 to 7% along the entire process stream.
[0017] In this patent specification, the following approximate conversions of units apply:-Temperature
in °F are converted to °C by subtracting 32 and then dividing by 1.8.
[0018] Temperature differences in
0F are converted to °C by dividing by 1.8.
[0019] Lengths in inches (in. or ") are converted to centimetres by multiplying by 2.54.
[0020] Areas in square feet (ft
2) are converted to cm
2 by multiplying by 929.0.
[0021] Amounts of heat in.British Thermal Units (BTU) are converted to calories by multiplying
by 252.0.
[0022] The feeds LP 150N and LP 600N are wax-containing hydrocarbon oils having viscosities
of 150 SUS (about 32 centi-strokes) and 600 SUS (about 130 centi-strokes) respectively.
1. A process for dewaxing a wax-containing hydrocarbon oil comprising introducing
said oil at a temperature above its cloud point into a first chilling zone divided
into a plurality of stages, passing the oil from stage-to-stage of said chilling zone,
introducing a dewaxing solvent into at least a portion of said stages, whereby a solvent-oil
mixture is formed, maintaining a high degree of agitation in at least a portion of
the stages containing solvent and oil thereby effecting substantially instantaneous
mixing of said solvent and said oil while cooling said solvent-oil mixture as it progresses
through said first chilling zone to a temperature greater than the temperature at
which the wax is separated from the oil, i.e., the wax separation temperature, but
less than about 22.220C (400F) above said separation temperature, whereby a substantial portion of the wax is
precipitated from said oil under conditions of said high agitation and forming a slurry
containing the solvent-oil mixture and precipitated wax, withdrawing said slurry containing
precipitated wax from said first chilling zone and cooling same to a wax separation
temperature in a second chilling zone comprising scraped surface chillers thereby
precipitating a further portion of said wax from said oil and separating said wax
from the solvent-oil mixture in solid-liquid separation means, the process being characterized
by comprising operating the scraped surface chillers of the second chilling zone at
a speed of from 5 to 20% of the original design operating speed.
2. The process of claim 1 wherein the hydrocarbon oil which is dewaxed is a petroleum
oil.
3. The process of claim 1 wherein the hydrocarbon oil which is dewaxed is a lube oil.
4. The process of any one of claims 1, 2 or 3 wherein the rotating element of the
scraped surface chiller rotates at a speed of from 1.4 to 2.4 RPM.
5. The process of any one of claims 1, 2, 3 or 4 wherein the dewaxing solvent is selected
from the group consisting of C3 to C6 ketones, low molecular weight hydrocarbons, mixtures of C3 to C6 ketones with aromatic compounds having from 6 to 10 carbon atoms, halogenated low
molecular weight hydrocarbons and mixtures thereof.
6. The process of claim 5 wherein the dewaxing solvent is selected from the group
consisting of C3 to C6 ketones and mixtures of C3 te C6 ketones with aromatic compounds having from 6 to 10 carbon atoms.
7. The process of claim 5 wherein the dewaxing solvent is selected from the group
consisting of methylethylketone, methylisobutylketone and mixtures thereof, methylethylketone
and toluene, acetone and toluene, and acetone and propylene.
8. The process of any one of claims 1 to 7 wherein the first chilling zone is a DILCHILL
tower wherein the oil is mixed with portions of cold de-waxing solvent in at least
some stages thereof with a high degree of agitation in at least a portion of each
stage where cold solvent is introduced.
9. The process of claim 8, wherein the wax and oil/solvent mixture are separated by
filtration.
10. The process of any one of claims 1 to 9 wherein the scraped surface chiller of
the second chilling zone is operated at a speed of from 8 to 14% of the original design
operating speed.