US3542666A - Adjustment of ph in the filtration of tar sand solvent-water systems - Google Patents

Description

Nov. 24, 1970 w. c. SIMPSON 3,542,666 ADJUSTMENT 0F u IN THE FILTRATION 6 OF TAR SAND SOLVENT-WATER SYSTEMS Filed March 20, 1968 INVENTORI WARREN C. SIMPSON ms ATTORNEY United States Patent 3,542,666 ADJUSTMENT OF pH IN THE FILTRATION 0F TAR SAND SOLVENT-WATER SYSTEMS Warren C. Simpson, Berkeley, Calif., assignor to Shell Oil Company, New York, N.Y., a corporation of Delaware Filed Mar. 20, 1968, Ser. No. 714,692 Int. Cl. Clflg 1/00 U.S. Cl. 20811 7 Claims ABSTRACT OF THE DISCLOSURE In the solvent extraction of tar from tar :sands by slurrying the tar sands in a hydrocarbon solvent and a minor amount of water and then filtering the fat solvent from the tar-depleted sand, the filtration rate of fat solvent through the tar sand bed is greatly accelerated by adjusting the pH of the aqueous phase in contact with the tar sands by the addition of a base.

This invention relates to a method of removing tar from tar sands. More particularly this invention relates to a method of removing tar from tar sands by solvent extraction with subsequent solvent recovery.

Large deposits of bituminous containing sand are found in various locations. Some are relatively soft and free flowing while others are very hard or rocklike. The tar content of these sands may vary over a wide range and presents an attractive source of supply of crude petroleum. For example, one of the largest deposits of tar sands thus far discovered lies in the Athabasca district of Alberta, Canada, and extends for many thousands of square miles.

Various methods have been proposed in the past for the recovery of tar from these tar sands but none of the methods proposed has been entirely successful as far as the economy of the operation is concerned. Tar sands suffer the disadvantage of requiring additional processing steps over conventional forms of oil recovery. It is, therefore, essential that for any process to prove commercially feasible it must be competitive in cost with other petroleum sources. It has been proposed to recover the tar content from tar sands by mixing the sands with water and separating the sand from the mixture formed. This process suffers the disadvantage of forming emulsions of water and oil which have been very difficult to break, thereby resulting in considerable loss of product. Other methods utilize solvent extraction techniques but have been slow to filter, presumably due to an excess of fine particles in the sand. Still other methods have required the use of a multisolvent system whereby the tar sand is subjected to a series of solvents before it is finally recovered.

More efficient methods proposed recently in Ser. No. 537,902, filed Mar. 28, 1966 and in Ser. No. 537,903, filed Mar. 28, 1966, comprise extracting the tar from the sand by filtering the extraction solvent through a bed of tar sand thereby removing the tar from the sand and recovering the solvent from the tar depleted sand by means of steam stripping. The product recovered from the tar sands by these methods is dependent upon the extraction solvent utilized. In general, hydrocarbon solvents con taining little or no aromatics will leave all or part of the asphaltene content on the sand while solvents containing a higher degree of aromaticity will extract all of the tar from the sand. Hydrocarbon solvents having from to 9 carbon atoms which have an aromatic content of from O to 100% by weight may be utilized.

One of the disadvantages of the above-mentioned process is that not all tar sands filter uniformly. While the reasons for this are not exactly clear it is felt that the tar content of the sand and the amount of the fines or small particles of sand greatly influence the rate at which solvent will pass through the sand.

Tar sands having relatively high tar content filter more slowly than do those of lower tar content due to the viscosity of the fat solvent. One method of solving this viscosity problem might be to use more solvent which would not be as economical. Another method would be to completely dissolve the tar off the sand prior to the filtration step. This latter step would require some sort of premixing operation such as slurrying. Even if this operation were successful it would not eliminate the problem of fines or other small particles which tend to plug the filter bed and impede filtration.

The problem of slurrying and fines in the filter bed are interrelated. It has been postulated that the fines and other small particles contained in tar sands are contained in water envelopes that surround individual grains of sand.

These water envelopes are in turn surrounded by the tar or bitumen which is the desired product. In theory, if the tar sand is allowed to dry out or if the water envelope containing the fines is broken, these small particles are released and tend to plug the movement of liquid through the sand bed in any subsequent filtration operation.

Another theory is that, although the fines are not encapsulated, a small amount of water is necessary to bind the fines particles to the larger sand grains. Since sand particles are by nature more hydrophilic than oleophilic, water tends to wet the sand particles and act as a cement to bind them together. When the sand dries out the fines are no longer attracted to the larger sand particles but tend to migrate through the sand bed.

Slurrying of the tar sands has not been considered to be practical due to the breaking of the water envelopes surrounding the grains of sand followed by the release of fines and other small particles into the slurry. Upon subsequent filtration of the slurry the bed would plug. Leary et al., US. Pat. 3,117,922 propose to solve this problem by slurrying the tar sand under conditions sufiiciently gentle that the water envelopes containing the fines are not broken but which are suflicient to dissolve the tar or bitumen from around the water envelopes. This process required a two-solvent system comprising a heavy solvent and a light solvent.

While the process of Leary et al. may prove practical for freshly mined tar sand in which all of the fines are encapsulated in a water envelope or attached to larger sand particles, it would not be applicable to a process in which the water had previously been dried out. This drying out may occur in a number of ways. For example, during mining, transportation, and crushing operations prior to the slurring step, the tar sands tend to lose certain amounts of originally contained water. Moreover, this process requires a multisolvent system.

In Ser. No. 566,039, filed July 18, 1966 it was stated that the tar could be effectively and efficiently removed from the sands containing the same by a one-solvent process which comprised slurrying the tar sands in a solvent in the presence of an added amount of water followed by filtration of the slurry and recovery of the solvent from the sand bed.

Presumably, the addition of water to the slurry causes the fines and other small particles to either become encapsulated in water or be bound to larger sand particles so as to prevent their disadvantageous migration and collection within the filter bed and to agglomerate them into masses that behave as larger particles.

This method allows tar sands which heretofore were thought to be impossible to process to be slurried and filtered at wholly acceptable rates.

All of the prior art processes offer some advantages in processing tar sand by solvent extraction. However, regardless of the solvent used or the amount of water added to the sand in a slurry step some tar sands have proven to be diflicult, if not impossible, to process. Presumably, this is due to the presence of fines in the filter bed which have not been encapsulated in water or bound to other sand particles by the addition of water.

It has now been found that even the most difficult tar sand can be effectively processed by a single-solvent extraction process by slurrying the tar sand in a solvent containing an added amount of water wherein the added amount of water in contact with the tar sand is maintained above a specified pH.

Most freshly mined tar sands have a pH of 5 and above and if processed immediately upon mining may be filtered at an acceptable rate without pH adjustment; however, upon standing exposed to air, tar sands tend to oxidize and become more acidic in nature. Also some freshly mined tar sands possess a low pH, i.e., below 5, and are diflicult to filter.

It is not known exactly why tar sands having a low pH are difficult to filter but it has been demonstrated that when suflicient caustic has been added to the aqueous phase in contact with the tar sand to raise the pH to 5 or above the filtration rates of solvent through the sand bed increases rapidly. The role of the water and base is still being studied and under a microscope a filterable bed appears to consist of coarse particles surrounded by a layer of water which either encapsulates fines or causes them to be bound to each other or to larger sand particles. In addition the bed has a porous structure. Unfilterable beds appear structureless with spaces between par ticles filled with large globules of water which do not wet the sand particles. While not being limited to any one theory of operation it is postulated that the role of pH adjustment improves the Wetting of the sand thereby allowing the fines to become encapsulated in Water or bound together.

It has also been demonstrated that the point at which a sand bed gives the most rapid filtration rates is the point at which the net surface charge on the sand particles is lowest. This so called zero point of charge is different for different types of sand. For example, the minimum for feldspars is at a pH of about 5-5.5 and the minimum for alumina is at a pH of about 7-7.5. It is, therefore, thought that by adjusting the pH of the aqueous phase in contact with the sand to give the lowest net surface charge on the sand, the sand particles, and in particular the fines, will either become encapsulated in water or be bound together via a water bridge.

Any suitable basic material may be used to raise the pH of the aqueous phase in contact with the tar sand. The oxides, hydroxides and carbonates of alkali and alkaline earth metals are most suitable. Because of availability, caustics such as sodium and calcium hydroxide are preferred. The amount of base added will depend upon the amount of water in the aqueous phase and the pH of the tar sand. All that is required is that the pH be raised to a level at least above a pH of 5. There may be optimum pH levels above 5 at which the sands will filter at an even more accelerated rate. This would necessarily have to be determined by experimenting with the particular tar sand being processed and is within the skill of one having ordinary skill in the art having possession of the knowledge contained herein.

The attached'drawing shows a schematic flow diagram of one method of operating the process.

The amount of water to be added to the slurry may vary once a critical level has been established. Presumably this level is established at a point at which all or substantially all of the fines contained within the tar sand bed become encapsulated in water or are bound together and may vary depending upon the amount of water already in the sand and upon the amount of fines contained in the sand. In general amounts of added water between 1 and 7% by weight, basis tar sand, are sufficient although more may be added if desired.

The extraction solvent consists of a volatile hydrocarbon fraction containing from about 5-9 carbon atoms per molecule and may be aromatic or non-aromatic depending upon the amounts of tar to be dissolved. Tarrich solvent obtained from filtering the slurry may also be used in place of or in addition to the C C hydrocarbon as extraction solvent. In general, non-aromatic solvents such as heptane will not dissolve the asphaltene content of the tar while aromatic solvents such as benzene and toluene completely dissolve the tar from the sands. It may be preferable to utilize a solvent such as heptane or other petroleum solvent fortified to a desired aromatic content by recycled tar-rich solvent, benzene or toluene. The solvent preferably contains an aromatic content of from 10 to 100% and more preferably between 15 and It is also desirable that the extraction solvent be made up of both fresh solvent and recycled tarrich extract from the filtrate. The amount of solvent added to the tar sand to form a slurry is generally proportional to the amount of tar on the sands. Solvent to original tar ratios between 2:1 and 6:1 by volume may be used. The solvent, base, Water, and tar sand may be mixed together in any manner desired to form the slurry.

It may be desirable to add only part of the total solvent to form the slurry and to add the remaining solvent to the filter bed once the slurry has been distributed on the filter. In this manner the latter solvent addition will serve to displace the tar-rich extract from the bed and to wash out and dissolve additional amounts of tar on the sand not extracted in the slurry.

The invention may best be described by reference to the drawing which shows a schematic flow diagram of a method of operation.

The tar sand is first stripped of overburden and mined by appropriate means and brought to an extraction plant for removal of the Oil and bituminous materials. This sand generally contains about 5 to 25% by weight of tar and 1 to 13% by weight water. The mined tar sand is then fed into appropriate machinery wherein it is crushed, broken or ground into proper size for solvent extraction.

The crushed tar sand is passed via line 1 into a slurry vessel 2 wherein it is contacted in any desired manner 'with solvent entering vessel 2 through line 3. The amount of solvent used will depend upon the tar content of the sand as earlier mentioned. An appropriate amount of water, generally between 1 and 3% by weight basis tar sand, enters vessel 2 through line 4. Sufficient base is added through line 5 to raise the pH of the aqueous phase to at least 5. The tar sand, solvent, base and water are mixed in 'vessel 2 in any appropriate manner such as by propeller stirrers, paddles on fins, to form a slurry.

During the slurrying operation the tar is substantially all dissolved from the sand and the fine particles are agglomerated in the pH adjusted water. It is this part of the process that greatly enhances the filtration rate of the slurry once it has been deposited on a filter. Certain patterns of behavior have been noted during this step of the operation. For example, slurrying poorly filtering sands without the addition of water produces a thin, smooth flowing slurry. This is consistent with observations in slurry flow where addition of the fine particles makes a slurry more free flowing and smooth. Apparently, the fine particles act as a lubricant and give the slurry a thinner consistency than a slurry made up of larger particles. When filtered, this free flowing slurry deposits a layer of claylike material at the top of the filter bed. Local pressure gradient measurements indicate that, in a majority of cases, the bulk of the pressure drop occurs in the upper layer of the bed; however, in some cases plugging at the bottom of the bed has occurred first. Agitation of the bed surface produces immediate improvement in filtration rate which ceases when agitation is stopped. This clay-like layer is absent at high filtration rates.

In contrast, those sands which slurry with difficulty, producing thick, grainy suspensions which ooze rather than flow smoothly have high filtration rates. However, air drying of such a sand prior to slurrying produces a thinner slurry which filters poorly as described below.

The addition of water to these smooth flowing thinner slurries produces a dramatic thickening of the slurry provided the aqueous phase has a pH above 5 and eliminates the clay-like layer atop the settled filtration bed and significantly improves filtration rate.

The slurry formed in vessel 2 is then passed via line 6 to a filter 7. This filter may be a continuous belt filter, moving pan or rotary pan filter and the like or a series of such filters. The slurry is uniformly distributed on the filter thereby forming a bed of a predetermined thickness, e.g., between 3 to 12 inches. A pressure drop of from about 1 to p.s.i. is maintained across the filter bed by applying a vacuum or other appropriate means at the bottom of the filter bed or by means of applied pressure above the bed.

The extract phase, containing the solvent and dissolved tar, is forced out of the bottom of the bed through line 8. If desired, additional solvent may be fed onto the filter bed 7 by means of line 9. The rate of filtration of the solvent through the bed of sand is a function of pressure drop across the bed, fat solvent, viscosity and particle size of the sand in the bed. Without the adjustment of the pH of this water added to the slurry to aggolomerate the fines and other small particles the filtration rates tend to be much slower than for some tar sands. However, the addition of water to the slurry and adjustment of pH produces the above defined thick, grainy slurry which has a relatively high filtration rate. Generally, filtration rates between 2 g.p.m./ sq. ft. and 10 g.p.m./sq. ft. are to be expected. The extract phase entering line 8 from filter 7 is passed into a solvent recovery zone 12 wherein the solvent is separated from the tar and the tar is passed through line 13 for further processing and subsequent refining. If desired a portion of the fat solvent may be recycled through line 10 back to line 3 to be used in the place of or in addition to the hydrocarbon solvent. In this manner a more efiicient use is made of the solvent. The recovered solvent is passed via line 14 into solvent storage tank 16 for reuse. The solvent-rich, tar-depleted sand remaining on filter 6 is passed through line 18 to solvent recovery zone 19. This zone preferably consists of the same type of filter apparatus used in filter 7 and can, in a batch operation, be the same apparatus. In solvent recovery zone 19 the solvent may be separated from the sand in any manner desired. Preferably it is stripped oif by means of steam entering zone 19 through line 20. The steam is pushed or pulled through the sand bed by means of pressure drop across the bed, thereby stripping off the solvent. The solvent and condensed steam leave the solvent recovery zone through line 21. The moist sand from zone 19 is pumped or conveyed through line 22 to a tailings area or otherwise disposed of. The recovered solvent from line 21' is fed into solvent storage tank 16 where it joins recovered solvent from solvent-tar separation zone 12. Any water contained in the solvent phase separates and is withdrawn via line 24. The recovered solvent in tank 16 may then be recycled through line 25 to join fresh solvent in line 9 for further slurrying and tar extraction.

The following examples are indicative of the way in which the adjustment of pH of the water in a tar sand extraction solvent slurry greatly enhances the filtration rate of the solvent and water through a filter bed of tar sand formed by depositing the slurry on a filter.

EXAMPLE I A freshly mined tar sand containing about 7.5% by weight tar with a pH of 8 was slurryed in toluene solvent at 46 C. in the presence of a minor amount of water.

The slurry was placed on a filter and the filtration rate of the fat solvent was measured to be 5.0 g.p.m./ft.

EXAMPLE II The sample of the freshly mined tar sand of Example I was allowed to remain exposed to the atmosphere for a period of two months at which time the pH of the material had dropped to 4.6. In a slurrying operation as carried out in Example I, the filtration rate was only 0.3 g.p.m./ft. However, when the pH of the slurry was raised to 7.7 by the addition of sodium hydroxide the filtration rate again increased to 5.0 g.p.m./ft.

The above results show that if the pH of a mined tar sand drops below an acceptable level in that the filtration rate of solvent through the sand is unsatisafctory, the filtration rate can be restored to its original value by neutralization with a base.

EXAMPLE III A tar sand slurried with a hydrocarbon solvent, having an aromatic content of about 75% by weight, and water had a creamy smooth consistency and possessed a pH of about 3.0. This slurry filtered at the rate of only about 0.5 g.p.m./ft. Upon adding suificient calcium hydroxide to raise the pH to above 5.0, the creamy consistency of the slurry changed to a granular sludge caused by the agglomeration of fines which then filtered at the rate of about 4.5 g.p.m./ft. This shows that a condition which agglomerates the fines also opens up the porosity of the filter sand bed.

EXAMPLE IV In order to better determine the role base addition to a slurry of tar sands plays in increasing the filtration rate of the solvent through the tar sand, the following runs were made.

Sample A was a tar sand which would filter adequately without the addition of caustic wherein Sample B was a tar sand from both Samples A and B, which had been extracted with toluene in a soxhlet extractor, to determine sand wettability. Both sands from A and B behaved similarly, having filter rates .of 0.7-1 g.p.m./ft. without water addition. Upon the addition of a minor amount of water the filtration rate increased to 1-2 g.p.m./ft. although the sand from Sample A appeared to wet somewhat better than the sand from Sample B. With the addition of both water and caustic the pH of the slurry was 67 and the filtration rate increased to 23 g.p.m./ ft. At the higher pH the water wet the sands from both samples very effectively.

In contrast the unextracted tar sands of Samples A and B filtered at less than 0.01 g.p.m./ft. without water or caustic addition. With water addition the slurry of Sample A filtered at about 4 g.p.m./ft. while Sample B showed practically no improvement. Sample B had a pH of about 3.0. Sufiicient sodium hydroxide was added to the slurry of Sample B which then filtered at the rate of 8 g.p.m./ft. This data shows that the increase in pH which improves the wettability of the sand particles greatly accelerates the filtration rate of solvent through the sand.

EXAMPLE V To further demonstrate the eflFectiveness of the present invention the following runs were carried out in duplicate on tar sands having a low pH. One run was made without pH adjustment and the other was made with the addition of sodium hydroxide.

The runs were made by slurrying the tar sand in a partially fat toluene solvent in the presence of a minor amount of water and then charging the slurry onto a filter so that there was about 50 lb. of sand per square foot of filter area, giving a sand bed depth of 7 inches. A constant pressure drop was maintained across the sand bed and the time required for breakthrough was measured as was also the filtration rate at breakthrough. The results are summarized in the following table.

The results demonstrate both the increase in filtration rate as a result of pH adjustment and also the reduced time it takes to draw the tar-rich solvent through the sand bed.

tar sand is selected from the group consisting of sodium hydroxide and calcium hydroxide.

4. The process according to claim 3 wherein the base is sodium hydroxide.

Filtration rate at break- Time to breakthrough, Percent through, g.p.rn./ft. N 0. minutes Tar Tar in sand 1 lb. NaOH 1 lb. NaOH on free ex- No per 1,000 lbs. No per 1,000 lbs sand Water tract phase NaOI-I tar sand NaOll Run 26104 511157020011 mb 20 0 0 0 0 0 0 0 0 0 0 L2 0 0 0 S334475508196499928 om 34433333323239.22232 2 3 29.980545240000055? 2 0 0 L0 Lfim5 7 6 2 3 w6 &0 5 3:m .1 11111 1 6. The process of claim 1 wherein the aromatic content of the solvent is from 10 to 100%.

7. The process of claim 6 wherein about 1 pound of drocarbon solvent is between land 7% by weight on the 35 sodium hydroxide is added per 1000 pounds of tar sand. basis of the tar sand and the ratio of solvent to original tar is between 2:1 and 6:1 by volume, and (b) depositeferences C ted ing the slurry on a filter zone to form a filter bed and UNITED STATES PATENTS 9/1931 Langford et al. 208-11 8/1966 Simpson 208-11 XR 8/1969 Benson 20811 FOREIGN PATENTS 639,050 3/1962 Canada.

5. The process according to claim 4 wherein about 1 1. In a process for the recovery of tar from tar sands pound of sodium hydroxide is added per 1000 pounds of which comprises the steps of (a) combining the tar sand I Only 0.5 lbs. NaOH per 1,000 lbs. of tar sand was added.

I claim as my invention:

tar sand.

2. The process according to claim 1 wherein the base is selected from the group consisting of alkali and alkaline earth metal oxides, hydroxide and carbonates. PATRICK GARVIN Pnmary Exammer 3. The process according to claim 2 wherein the base P. E. KONOPKA, Assistant Examiner with a hydrocarbon solvent containing added water to form a slurry of tar-rich solvent and tar-depleted sand, wherein the amount of added water contained in the hywithdrawing therefrom the tar-rich solvent as filtrate, the improvement which comprises increasing the filtration 0 rate by adjusting the pH of the aqueous portion of the slurry by the addition of sufiicient base to raise said pH to at least 5.

Images