Demetallation of Used Oil to Facilitate Its
                     Utilization as a Fuel
 J.  H.  Kang and A.  R.  Tarrer,  Chemical  Engineering Department,
              Auburn University, Auburn, AL   36849

Joe Kaminski, The Office of the Assistant Secretary of Defense

Jim Parrish, Defense Reutilization and Marketing Service (DRMS)

Edward R. Bates, U. S. Environmental Protection Agency

-------
                                                                                       1
                                     NOTICE
     The information in this paper has been funded in part by the United
States Environmental Protection Agency under Cooperative Agreement CR 812090
to Auburn University.  It has been subjected to the Agency's  peer and admin-
istrative review and has been approved for publication.   Mention of trade
names or commercial products does not constitute endorsement  or recommendation
for use.

-------
 ABSTRACT:  Used oil invariably contains  soluble  and insoluble metal-bearing




 compounds.   The presence  of these  compounds  complicates  the recycling of used




 oil as  a fuel,  by causing boiler maintenance problems and by making  it




 difficult to comply with  particulate emission standards  and RCRA regulations.




 Demetallating a used oil  prior to  its burning minimizes  maintenance problems




 and facilitates meeting the  regulatory requirements controlling its burning.




 Different physical  and chemical methods can be used to demetallate used oils.




 In this work, a chemical  demetallation agent was used to convert entrained




 non-filterable  metals into a form which could be effectively removed by




 filtration.   Two different types of chemical demetallation agents were used:




 metallic borohydrides (NaBH4 and KBH4)  and diammonium phosphate (DAP).   The




 activity and selectivity of these demetallation agents under different




 reaction conditions in various types and sizes  of reactors  (thereby varying




 mass  transfer rates) were  compared.  Several  types of used  oils having




 different metal  contents were examined.




      It  was observed in this study  that metallic  borohydrides were  highly




 selective with regard to the types  of metals  they removed.  The accompanying




 sludge formed also made the  separation operation  inefficient.  On the other




 hand, results obtained with  DAP were extremely favorable.  The presence of




 water during  the reaction  was found to play an important  role as a  carrier for




 the DAP, aiding  in its dispersion and reducing the importance of mass




 transfer.  This  study which also evaluated different separation techniques




 found the use of both sedimentation and filtration to be practical,  with




 filtration rates  being significantly accelerated by the addition of low



percentages of No. 2  fuel  oil.




KEYWORDS:   Waste oil, demetallation, ash content, lead content,  metallic




            borohydrides,  diammonium phosphate,  lead reduction.

-------
  Introduction




       In the United States, automotive and other oil-consuming industries




  generate about 1.1 billion gallons of used oil each year [1].  This




  voluminuous amount constitutes a serious waste oil disposal problem.  Recent




-  Environmental Protection Agency (EPA) regulations and growing public concern




  over hazardous wastes make development of a viable alternative to land-




  filling very desirable.  The objective of this work was to develop a practical




  and feasible waste oil demetallation process to remove the ash-forming




  contaminants,  especially organolead compounds, from the oil so that the




  demetallated used oil would qualify as a specification-grade fuel oil [2].




       Waste oil,  although contaminated,  has  a high energy value.   A major




  operation in reclaiming waste oil  is the removal  of water and various ash-




  forming impurities that remain dispersed in the used  oil.   The ash-forming




  impurities  consists of very  fine carbon particles,  organometallic particulates




  (e.g.,  organolead compounds),  and  inorganic materials  (e.g.,  dust).




      Reuse  of an  untreated used oil having a high concentration of lead and




  other metallic contaminants may lead to  serious air pollution and boiler




  maintenance problems.   To qualify for minimal regulatory requirements, i.e.,




  specification-grade fuel oil requirements under EPA regulations,  the lead




  content of a waste oil  to be recycled as a fuel in a commercial boiler can not




  exceed 100 ppm.   Since most of the used oil from the automotive industry




 contains several times the proposed EPA limit, it is imperative that an




 economic process be developed to reduce the  lead content of the oil to an




 acceptable level so that the reclaimed waste oil can safely be reused as a




 fuel oil without endangering public health or the  environment.




      Waste oils  were demetallated using two  types  of demetallating reagents:




 metallic borohydrides  and diammonium phosphate (DAP).   A detailed  parametric

-------
 study was undertaken to map out the process variables so as to identify the




 most efficient demetallation conditions.   An evaluation of solid-liquid




 separation techniques was also done to design an efficient solids  removal




 operation for future pilot plant studies.




 Experimental Procedure




 Materials -- All used oil was collected from the Auburn University Waste Oil




 Reprocessing Pilot Plant.  The demetallating reagents used were sodium




 borohydride, potassium borohydride,  sodium borohydride aqueous solution,  and




 diammonium phosphate (DAP).   The phase transfer catalyst used was  tri-n-butyl-




 methyl ammonium chloride.




 Equipment -- Three types of reactors were  used.   They were:  an autoclave (3785




 ml and 300 ml),  a microreactor (45  ml),  and a distillation flask (300  ml).




 The autoclave was equipped with a turbine  agitator,  a cooling coil, baffles,




 and a thermowell,  all made of 316 stainless steel.   The autoclave  was  heated




 by an electric furnace and agitation was provided by a magnedrive  system.   The




 tubing bomb microreactor (TBMR).was  constructed of 316 stainless steel tubing




 (1.9  cm O.D.  with 0.165  cm wall thickness  and a length of 20.3  cm).  Agitation




 was provided by  a shaker assembly as  shown in Figure  1.   For  a  typical run,




 the TBMR was  immersed in a preheated  fluidized sand bath.  As for  the




 distillation flask,  the  reaction mixture was  agitated using a magnetic




 stirrer, and the  reactor was heated in a constant temperature paraffin oil



bath.




Analysis -- The experimental results were verified in accordance with an EPA




approved quality assurance project plan.  The properties of the demetallated




used oil were evaluated  to assess its potential as an EPA specification-grade




fuel oil using the following test methods:   water content, ASTM D4006-81




 (Water in crude oil by distillation); ash content, a modification of ASTM

-------
  D482-80  (Ash from petroleum products);  lead content, standard atomic

  absorption (AA)  analysis.

       In  each ash analysis,  an oil specimen (about 5 g) in a preweighed

  crucible was dried first on a hot plate and then heated in a Model F-6000

  Thermolyne muffler furnace  at a rate of 373K (100°C) per hour.  Once the

  temperature  reached 873K (600°C), the crucible was allowed to stay in the

  furnace  for  an additional 12 hours.  Then, the crucible was removed and cooled

  in a desiccator.   The weight of ash remaining in the crucible was determined,

 and its percent concentration in the original oil sample was computed..

      In each lead  analysis, the ash was dissolved in an acid solution of 10 ml

 of HN03 solution (1:1) and 2 ml of concentrated HC1.   The lead content was

 determined using AA spectroscopy.   A lead working standard was prepared by

 serial dilution of a lead standard solution (1000 ± 1%  ppm)  supplied by Fisher

 Scientific.  A blind-spiked oil sample was occasionally prepared to verify the

 results.   An organic lead standard (lead cyclohexanebutyric  acid,

 [C6H11(CH2)3COO]2 Pb)  was used for this purpose.

 Results and Discussion

 Use  of Metal Borohydride As  A DemeCall ing Reagent

     The  properties of metal borohydrides, such as NaBH4, KBH4,  and SWS

 solution  (consisting of  12 wt% NaBH4, 40 wt% NaOH, 48 wt% water), are

 described in detail in technical literature published by Morton  Thiokol, Inc.,

 Ventron Division  [2,3].

    Metallic  borohydrides (e.g. sodium borohydride) are strong reducing

 agents.  The  following reaction is typical of the metal reduction that occurs

with sodium borohydride [2].

 8MX + NaBH4 + 2 H20 ----> 8M° + NaB02 + 8 HX

  where M = the metal (valence  1+)
        X = the anion

-------
    In this chemical reaction, stoichiometrically, sodium borohydride




has 8 reducing equivalents per molecule.  Based on bench-scale experimental-




results, Morton Thiokol, Inc. reported that a stabilized water solution (i.e.,




SWS solution) was an effective reagent for removing lead from used crankcase




oil [4] .  it should be noted that sodium borohydride (NaBH^.) degrades in the




presence of water or acid via hydrolysis to liberate hydrogen and a salt




(NaB02) when the pH of the solution is low.  A representative chemical




reaction is as follows:




       Water Hydrolysis:




        NaBH4 + 2H20 ---> NaB02 + 4 H2




     A parametric study was done to investigate the lead reduction efficiency




of two Morton Thiokol products: Venpure powders (NaBH4 and KBH4> and SWS




solution.  A tubing bomb microreactor (TBMR) and the reaction conditions




listed in Table 1 were used in this study.  A concentrated NaOH aqueous




solution (50 wt%) was usually added to maintain a high pH and thus a low




hydrolysis activity of the metal borohydrides.




     In a typical run, the TBMR was charged with 20 g of waste oil, the




specified amounts of 50 wt% NaOH aqueous solution and VenPure product.  The




reactor was then sealed.  Subsequently, the TBMR was attached to the vertical




shaft of the agitation assembly, as shown in Fig.  1,  and agitated at ambient




temperature for 3 minutes at 860 rpm before lowering it into the preheated




fluidized sand bath.  The heat-up time was about 60 seconds.  At the end of




the chosen reaction time, the TBMR was removed from the sand bath,  immediately




quenched in cold tap water,  and checked for any leaks.   The gases were




released slowly from the bomb under a hood and the liquid product was then




collected and vacuum filtered.   The filtrate (or product oil) was collected




for ash and lead analyses.

-------
       With  0.65 wt%  SWS  and 1.2 wt% NaOH solutions, the  lead content of  the




  Type B oil was markedly reduced from 480 to 70 ppm (Run 1) as shown in  Table




  2.  In other words, an  85.4% lead reduction was achieved.  This is well within




  the allowable maximum lead content of 100 ppm for specification grade used oil




  set by EPA.  However, a high sludge production (16 wt%) resulted from this
  run.
       Since metal borohydrides are expensive,  the charge of SWS solution




  (containing 12 wt% sodium borohydride)  was reduced to 0.43 wt% in Run 2.




  Also,  the charge of NaOH concentrated solution was reduced from 1.2  wt% to 0.8




  wt%.   The result was that the lead content of the  oil decreased by only about




  50%, to  240 ppm.   The ash content of the product oil  was 0.58  wt%.   The amount




  of sludge production was  about the same as  that from  Run 1.




      In  order  to  reduce the sludge production, in Run 3, only  0.05 wt% NaBH4




 powder was  used and no NaOH was added.  The amount of lead reduction was only




 10.4 percent; however, little  sludge was produced (2 wt%).   To enhance the




 demetallation activity, 0.8 wt% of 50 wt% NaOH was added in Run 4 and the




 other reactions conditions were kept the same.  The lead removal efficiency



 increased from 10.4% for Run 3 to 37.5% for Run 4.




      To improve the solubility of NaBH4 in the oil  phase, 0.10 wt% of phase




 transfer catalyst (tri-n-butyl-methyl ammonium chloride)  was also added to  the




 oil  (Run 5).  Comparing the results of Run  5 with those of  Run 3 (in  both runs




 no NaOH was added),  the addition of the phase  transfer catalyst did result  in




 an increase in lead removal efficiency from 10.4 to 27.1%.  With 0.8  wt% NaOH




 solution  added in  Run 6, the lead  reduction  efficiency improved slightly more




 to 31.3 wt%.  On the whole, there  seemed to be a trade-off between using




 either SWS or NaBH4 powder  as a demetalling reagent; the former performed




better in terms of lead reduction efficiency; whereas the latter produced, less

-------
  sludge.





      To determine  the reproducibility of the demetallation experiment, Run 1




 was duplicated by  Run 7, and the results were essentially the same for both




 runs, confirming the accuracy of the experiment.




      Since waste oil is derived from various sources, it is plausible that the




 lead removal efficiency using a demetalling reagent might respond differently




 to different types of waste oil.  Thus, a different oil, viz., Type A oil




 (lead content = 660 ppm, ash content - 0.59%) was used in Runs 8-12.




      In Run 8, the reaction conditions were almost the same as those in Run 1




 only a different oil was used.    The lead content of Type A oil was reduced by




 about 24% to 500 ppm;  whereas for Type B oil the lead content decreased by 85%



 to 70 ppm (Table 3).





      When the charge of SWS was  doubled (Run 9),  the lead content of  Type A




 oil was  reduced by about 58% (from 660  ppm  to 280 ppm)  as shown in Table  4.




 SWS contained 12 wt%  sodium borohydride, which  implied  that the  sodium




 borohydride  concentration in the  oil in Run 9 was 1,560  ppm.  A  sodium




 borohydride  concentration of this  level would not be  economically feasible  for



 reclaiming waste oil.




     As mentioned earlier,  the runs  with caustic  and  SWS  solution were




 characterized by high sludge production.  A gel-like product was  occasionally




 observed,  which was difficult to filter, making the determination of sludge




 content very  difficult.  The gel formation was probably due to the presence of




 sodium hydroxide which acts  to saponify fatty acids (e.g. detergents) present




 in  the used oil.  In order to inhibit the gel formation, sodium borohydride




was added  in a powder form (Runs 10 and 11)  instead of in a solution;  NaOH was




still introduced as a 50% aqueous solution to maintain the activity of sodium



borohydride.
                                                                                    , j

-------
        In Run 10, 0.15 wt% (i.e., 1500 ppm) NaBH4 w.s added, vhlch ^
   approximately the .„. amount of NaBH* added In Eun 9; the leae reduction
   efficiency was slightly lower than that of Run 9.   This was probably due to
 -  the greater solubility of NaBH* In Run 9.   In Run  11,  a ph.se transfer
 ~  catalyst  (tri-n-butyl methyl ammonium chloride) was added,  and the  lead
   content of Type A  oil was reduced  by  about 73% to  180  ppm.

       When  potassium borohydride  (KBH4) was  used in Run 12,  the  lead  reduction
  »as  59% (i.e., from 660 to 270 ppm)   ^

  above the  100 ppm allowable maximum limit.

       Examination of the «sh content of the product oil  (shown in Tables 2 »d
  4, reveals that, for both types of waste oil, the ash content of the product
  oil was not markedly decreased by metal borohydride treatment, although the
  lead content was selectively reduced in some cases.  This »ight be attributed
  to the  formation of other  fine salts which  could not be separated by regular
  filtration.  A major disadvantage of using  metal borohydrides  as a demetaUing
  reagent is  that the pH must be tept high for the borohydrides  to be stab!.,
  and this results in a very viscous  (gel) product.
       In the previous runs,  the reaction temperature used was 110°C.  Bhen the
 reaction temperature was increased fro. Il0oc Co 150oc,  keeplng ^ ^

 conditions  the same as  those used in Run 1,  the product  was a gel,  and as a
 result,  a poor lead reduction efficiency was obtained (Runs 13  and 14  Table
a cotton seed oil or a non-detergent virgin motor oil was reacted with either
SWS solution or NaBH4 powder to confirm that saponification does occur when
fatty acids are present.

     A cotton seed oil (20g)  was blended with 1.2 wt% NaOH in 50% aqueous
                                      8

-------
 solution, and then reacted with 0.65 wt% SWS in a TBMR (Run 15).  The reaction




 was conducted under reaction conditions given in Table 1 (similar to those




 used for the TBMR demetallation runs).   The reaction product was collected in




 a 2.5 cm O.D. x 15 cm length test tube.  The sludge produced was about 35




 volume % (Table 6).   This sludge was probably the salt resulting from the




 saponification of fatty acids, in the cotton seed oil, with NaOH.  The cotton




 seed oil was reacted with 0.05 wt% NaBH4 powder in the absence of sodium




 hydroxide (Run 16).  In this case no sludge was formed, confirming that the




 sludge (gel) was due to the reaction of fatty acids with NaOH.  The amount of




 NaBH4 was doubled, and still,  there was no sludge formed (Run 17).   The effect




 of NaOH on non-detergent motor oil was  observed by reacting an SAE-30 non-




 detergent motor oil with SWS (Run 18).   After the reaction,  two layers of




 liquid were formed.  The top layer was  reddish brown and the bottom layer was




 bluish green in color.   A trace amount  of suspended solids was also observed.




 In summary,  the gel product formed while treating used oil with borohydrides




 with NaOH present  was probably due to the saponification  of  fatty acids.




 Effect of Caustic  --  In an attempt to  reduce  the amount  of  sludge  production,




 two  reactions  (Runs 19 and 20,  Table 7)  were performed in which 1.5 wt% NH4OH




 was  added instead  of NaOH,  using the reaction  conditions  shown in Table 1.




 Also,  to  help stabilize  the NH4OH,  25 psig of NH3 was used (Run 19).  As




 compared  to  a 0.8 wt% NaOH addition  (Run 4), the amounts of. lead  reduction and



 sludge production were about the same.




Effect of Water — As noted earlier, the reducing ability of metallic




borohydrides can be hampered substantially by water hydrolysis.  Since the




waste oils obtained were used as-received (which had a water content of about




5 volume %) , the poor performance for those runs with metallic borohydrides,




in terms of lead reduction efficiency and ash content remaining in the oil,

-------
   may possibly be attributed to the presence of water during the reaction.   To
   investigate the effect of water on lead removal, a sufficient amount of Type A
   oil (containing 5 vol , water) was blended with 0.25 wt, and 0.50 «, NaOH
   solutions (50 «, concentrated solution)  respectively.   Water along with some
   light  ends  were then distilled off via atmospheric distillation.   The residual
   oils were termed as  Type  C and Type D,  respectively.  The properties  of  Type  c
   & D oils  are  given in Table 8.
       In Run 21.  20 g of dry Type C  oil was reacted with 0.9 wt% of SWS
   solution  in a TBMR under reaction conditions given in Table 1.  As shown in
  Table 9,  the lead content of the Type C oil was successfully reduced from 700
  to 100 PPm;  i.e. about 86% lead reduction ^ achieved   ftia ^^^^^ ^t ^

  much better derailing performance can be achieved with a dry oil (e.g., type
  C)  than with a wet oil (e.g.  Type A).
       To further minimize the  use of SWS solution,  the charge  of SWS solution
  was  reduced  to 0.4 wt% in Run  22.   As  i
  obtained.
      in Runs 23 and 24, 0.1 wt» of a phase transfer catalyst (tri-n-butyl-
 ...thyl ammonium chloride, was added to see if the demetsUing performance
 could be improved.   Compared to run 21 (without catalyst,, catalyst addition
 (e.g.,  Run 23,  did not improve the results.
      TO study the  effect of BaOH charge on lead reduction,  Type  D oil „„  us«d
 in Runs 25  and  26.  The difference between Type  C  and  Type  D oll was  the
 amount  of HaOH  in  the  oil;  the latter had  twice  the amount  of BaOH  than the
 former  (-1250 ppm, .  In comparison to Run  22, with th. same charge  of SWS
 solution, Run 26 gave a  lower lead reduction efficiency.
     On the whole.  Table 9 shows that the presence of water in'the oi! has ,„
adverse effect on demetallation when metallic borohydrides are  used.  This can
                                     10

-------
   be  avoided by distilling off the water before  the  demetallation reaction.
        It  should be noted that the results presented so  far were  obtained  for.a
   45  ml tubing  bomb.microreactor  (TBMR).  Knowing that these bench-scale
- -demetallation results were to be incorporated  in the scale-up and design of
   the Auburn University Waste  Oil Reprocessing Pilot Plant, experiments were
  extended to a  larger batch reactor, i.e., a 3785 ml autoclave, to simulate
  actual operating conditions as much as possible.
       As mentioned above, water can decrease the demetailing activity of
  metallic borohydrides.   Therefore,  the as-received Type A oil,  containing
  about 5 vol.%  of water,  was blended with  0.25 wt%  NaOH  aqueous  solution (50
  vt%) in the following runs and then distilled to a specified temperature  under
  atmospheric pressure.  The residual oil was  assumed to  be moisture-free.  The
  moisture-free  oil (or dry oil) was  then treated with metallic borohydride
  products  in the one-gallon autoclave.  The results  are  given  in Table 10  and
  discussed below:
      In Run AS1,  890 g of dry oil (lead content - 700 ppm) was reacted with
  0.9 wt% of  SWS  solution  in the 3785 ml autoclave using  the reaction conditions
  shown  in Table  11.  For Run AS1, only a 37% lead reduction efficiency was
  obtained.  The poor demetalling performance was probably due to the low
  distillation cut-off temperature (129°C),  which resulted in a certain amount
 of water being left .behind in the residual oil.   In other words,  the residual
 oil obtained in Run AS1 was not dry enough.
      Thus, in Run AS2,  the distillation cut-off  temperature was  further  raised
 to 175°C.   The  oil and SWS solution  charges  to the  autoclave were 900  g  and
 0.91  wt%,  respectively, which were about the  same as in  Run AS1.   With the
 same  reaction conditions,  the  lead content of the oil was  markedly reduced
 from  700 ppm to 25 ppm; i.e.,  a 96%  lead reduction was obtained compared to
 37% in  Run AS1.
                                      11

-------
      Although  a promising  result was obtained  in Run AS2,  the amount of SWS




 solution  (containing  12 wt% NaBH4) used was considerable,  and hence,




 uneconomical.   Thus Run AS2 was duplicated by  Run ASS except that the SWS




 solution  charge was halved.  As a result, only a 40% lead  reduction was




 obtained; this  suggested that more SWS solution was required, if a better




 demetailing performance was to be achieved.




      To compare the demetalling effectiveness  of SWS solution and NaBH4




 powder, Run AS4 was performed in which the same amount of NaBH4 was used as in




 Run AS1 (i.e. 0.10 wt% NaBH4 power) and 0.20 wt% phase transfer catalyst (tri-




 n-butyl-methyl ammonium chloride)  were reacted with 545 g of dry oil in the




 autoclave.  Coincidentally, the same level of lead reduction was obtained.




 However,  with NaBH4 powder (Run AS4)  the product oil had a lower ash content




 as compared to Run AS1 with SWS solution;  also, less sludge was  obtained in
 Run AS4.
      To study the demetalling ability of KBH4 powder,  Run AS4 was  duplicated




 (Run ASS)  except that KBH4 powder was used.   As  shown  in Table  10,  the  lead




 reduction  efficiency increased from 37%  to  51%,  but  the  ash  content of  the  oil



 was  not altered significantly.




      Since KBH4 powder is  more  expensive than NaBH4  powder,  Run AS4 was




 duplicated again in Run AS6,  in which the NaBH4  charge was increased from 0.10




 to 0.15 wt% while maintaining the rest of conditions the  same as in Run AS4  to




 determine  whether a better lead reduction efficiency could be obtained/ It




 was  found  that  the lead reduction efficiency  increased from  37  to 71%, but the




 ash  content of  the product oil  did not change  appreciably, as shown in Table




 10. Although the  amount of NaBH4  used can be further increased  to improve lead




 reduction performance,  it  is uneconomical to do so when the operating costs




 for reprocessing waste  oil  are considered.  Therefore,  another demetalling




method was explored in  an effort to seek the most economical process.




 •**""•-                             12

-------
  Use of Diammonium Phosphate As a. Demetallation Reagent




       As previously described,  the major problems  in using metallic




  borohydrides  as  deinstalling reagents were:  (1) a  soap-like sludge was usually




.  .formed after  the reaction,  which was very difficult to filter;  (2) a caustic




  oil was produced (this could result in  corrosion  problems); and (3) the




  treated oil after filtration usually had a higher ash  content than the




  original waste oil.  Because of these problems, another demetallation process,




  which had been proposed by  the  Phillips Petroleum Company as a pretreatment




  step in converting waste oil into a lubricating oil was investigated [1,5].




  In  this process, waste oil was reacted with a demetalling reagent,  diammonium




  phosphate (DAP)  (both with and without the addition of excess water).




  Different types of reactors were used in this investigation to permit dominant




  mechanisms over different operational  ranges to be identified.




       In the bench scale experiments, three  types of reactors were used:  (1) A




  tubing  bomb microreactor  (TBMR) as described previously;  (2) A 500 ml




  distillation flask equipped with a magnetic  stirrer;  and (3) A 3785 ml




  autoclave  reactor.  The TBMR was operated under closed  conditions, whereas  the




  distillation was  done under  open conditions;  i.e.  the flask was  open to the




  atmosphere  to  distill off water and trace amounts  of light  components.



  TBMR Reaction  Studies





       Effect of Reaction Time and Temperature  --  The variation in lead content




  of the DAP-treated oil with reaction time was observed  as shown in Figure 2.




 The  demetallation experiments were done at 150°C.   From an initial value of




 about 210 ppm,  the lead concentration of the product oil decreased with




 increasing reaction time to a final value of 16.5  ppm for the 90 minute run at




 an 800 cpm agitation rate.   From this figure, it can also be deduced that:




 agitation has a significant effect on the lead content of the oil.  For




 example, for a  45 minute reaction time, when the agitation rate was  decreased




                                      13

-------
. SOO to 400

 PP..  *

    .rea for r«action

               the

      To

 mmutes.   The results are shomi ln Fl
 2                               1SUre3'  W"h -»«tion temperature of
  o c,    lead „ of type B oii _ reduced ^ 4ao M ^^     ^

          rd i-> a 52% uad redu"ion-   - — - — -
            cons_tly, tt can be lnf_d from pi_ 2 and 3  that

 .     u
demetallation. as seen in Figure 4.
                        stadies  .    n    _ perf_d
          o- « - « ppn, _r „ . 5%)  ^ s m DAp
                         300 ml distillatlon

           *llowed to continue
            During the re8ctlo|
                      of, A ^^^ _i£ _ obtsiMd ^ ^

    1        ead c°Ment
                               ords> a g^
                            ash „ of the
                               to

  0- M   se           tHe                      In Run DF1
                     be
                                   by flltratlon
                              14

-------
 thus performed in which oil A alone, without adding any DAP,  was heated in a




 distillation flask for the same period of time as in Run DF1.  The result was




 that only a 3% lead reduction was obtained,  indicating that DAP plays a very




 .important role in demetallation.  In addition, without DAP, the ash content of




 the waste oil was not altered appreciably.   In an attempt to reduce the DAP




 charge,  Run DPI was duplicated by Run DF3 except that the amount of DAP used




 was cut from 8 to 2 wt%.   Again, a similar lead reduction (97%) was obtained.




 Effect of Reactor Configuration on Lead Reduction --  Table 12 shows that




 satisfactory demetallation results under mild conditions can be obtained using




 a distillation flask as a reactor.  It should be noted that the reactor was




 open to the atmosphere.  To study the effect of the reactor configuration on




 the demetallation performance,  a tubing bomb microreactor (TBMR) was used as a




 closed system reactor.   The mass transfer effect for this type of reactor was




 also studied.   A detailed discussion of the  experimental results is given



 below.




      As  shown in Table  12,  with 2 wt% DAP, 97% lead reduction could be




 obtained (Run DF3).   The  reaction time for this run was  1 hr.   To determine




 the effect of the reactor configuration on lead removal,  Run  Dl was  performed




 in which 20 g of the  same type  of oil used in Run DF3 and 2 wt% DAP  were




 charged  to a TBMR and reacted for 1 hr under  a closed system.   Only  a 70%  lead




 reduction was  obtained  in Run Dl  compared to  97%  in Run DF3.




      In  Run D2,  three steel balls  (1/4" diameter) were added  to  the TBMR in




 order to  increase the mixing effect and interfacial contact area while




maintaining  the rest of conditions the  same as in Run Dl.  The  lead reduction




efficiency  increased from 70 to 77%.  When six steel balls were added in Run




D3, the lead reduction efficiency increased further to 81%.




     To study the effect of water presence on lead reduction,  Run D3 was




duplicated by D4 in which dry oil (obtained by distilling off water from Type




.-"   _• . --                             15

-------
  A oil) was used instead of wet oil (i.e. Type A oil).   As shown in Table 13,




  the lead reduction efficiency decreased from 81 to 78%.  This difference was




  attributed to a lower interfacial contact area available for the dry oil.   It




 -should be noted that this difference was small because of the excellent




  dispersion obtained in this type of reactor;  a much larger difference was




  observed later,  using a reactor (autoclave)  that did not have as good particle



  dispersion properties.




      The inconsistent results  (Table  12  & 13)  obtained from the  two  types of




  reactors (distillation flask and TBMR) may predominantly be due  to the




  difference  in heat-up time;  i.e. ,  for a  150°C  demetallation run,  the




  distillation  flask had a much  longer heat-up time  (75 min)  than  the TBMR (1




 min).   In other words,  some  demetallation may have occurred during the heat-
 up.
      To study heat-up effects, Run D5 was performed to duplicate Run 1 (see




 Table 14). except that the TBMR was preheated in the fluidized sand bath (see




 Figure 1) so that the total heat-up time from room temperature to 150°C was




 about 75 min., (It should be noted that in a typical TBMR run, the sand bath




 alone was preheated to 150°C before the TBMR was immersed;  the heat-up time




 was then about 1 minute.)  Table 14 shows that,  with the same heat-up time as




 that for distillation (Table 12),  a satisfactory result was obtained from  Run




 D5;  i.e.  91%  lead reduction could be achieved from a TBMR run.   In comparison




 with the  lead reduction  obtained from the distillation  run,  (91% versus 97%




 lead reduction),  this  result was  acceptable  for  a  reactor with a different



 configuration and mixing pattern.




     As discussed above, with 75 minutes of preheating time and an additional




hour for reaction (i.e.  total reaction time, 2 hr), 91% lead reaction could be




obtained in Run D5.  It would be interesting to see what the lead reduction




efficiency would be with a 2 hr reaction time, 160°C reaction temperature and




                                      16

-------
 a regular TBMR 1 minute heat-up time.  Run D8 was thus performed and the




 results given in Table 14 show that about the same level of lead reduction as




 Run D5 was obtained.




      It was thus concluded that with a total reaction time of 2 hrs at 160°C,




 a desirable demetallation result could be achieved for various types of




 reactor configurations (distillation flask or tubing bomb microreactor).   This




 agreement is mostly due to the fact that thermodynamic equilibrium (=




 complete lead removable)  was achieved under these reaction conditions.




      Since the distillation flask was used for the evaporation of water and




 light components,  so that the hydroxyl phosphate droplets could shrink to




 facilitate removal of lead particulates,  an open-type reactor such as a




 distillation flask (Run DF3)  may be more  suitable for oil demetallation as




 compared to a closed reactor, such as TBMR (Run D5).




      To further evaluate  the  effectiveness of DAP demetallation with respect




 to  the configuration of the reactor used  (i.e.,  the  importance of mass




 transfer),  a certain amount of Type A oil (lead content - 660  ppm,  water




 content - 5 vol%)  and 1.5-2.0 wt%  DAP were reacted in 45  ml,  300 ml,  and  3,785




 ml  (i.e.,  1 gallon)  autoclaves,  at 160°C  respectively.  The 45 ml




 microautoclave,  as  a matter of fact,  was  the  TBMR (tubing bomb microreactor).



 The results  are  given in Table 15.




      One  should note the main difference  between a TBMR and a  regular




 autoclave  in terms of agitation and heating;  the  former was vertically




 agitated; whereas the latter  was horizontally stirred.  The mass transfer-




 rates  in the TBMR are generally better than those in the  300 ml and 3,785 ml




 autoclaves  [7,8].  Also, the  heat-up time varied a great deal; for example,the




heat-up time for a TBMR heated by a preheated fluidized sand bath was about 1




minute, whereas, for a 300 ml or a 3785 ml autoclave (heated by a electrical




furnace from ambient temperature) the heat-up time was approximately 2 hours.




                                      17

-------
    Table IS shows that regardless ., the
    lead .eduction was obtained uslng DAp as . demetalUng

    reaction condition for a reaction ti»e  of about 2  hours.   The  amount of
    solid, produced was less than 1 «,,  excluding  the  DAP added.   These
    observations^plied that DAP is nore suitable  for  use, oil destination than
    sodium borohydride.
                                 d earUer, Bater present at the early stage of .
   de..taUation reaction can faciUtate lead removal by Mklng toe surfac.
   contact area available for the DAP.

       A co^arative study „  carried out using a l-gallon autoclave    The
   effect of Water on ash and !ead re»ova!  TOS  i»vestigat.d  using the reaction
   conditions given in Table  16.  Dnlilte the earlier studies .ith a TEME  (Table-
   13). the dispersion of  the DAP vas probably much less in this reactor  (a
  stirred autoclave) .

       in Run AD!,  wlth dry oil obtained by distillin, off water fro. Type A
  oil,  and 4  «. DAP, . 74%
  the ash content of the  product oil  from this  run was  0.310  «*
      To discern the effect  of  water on  the derailing perforce , Eu» ADI
 -  duplicated  by AD2,  except  that  a wet oil  (i.e. , Type A  oil, was used   As
 - "suit, 99.7% !ead reduction was  achieved and the ash content of the product
 oil was very low (0.020 «.,.  ^ observed ^^^ ^ ^ ^^

 efficiency that resulted with the addition of water was  si.nific.ntly Urger
 than that observed b.for. with a TBMH and was  felt  to  be  due to  the difference
 in the  MSS  transfer characteristics of  th« reactors ;  na»ely.  the  desre.  to
which the DAP was dispersed.


operating  cost and sludge generation and on. part of the top of th. autoclave
»as opened to vent water and the lights to facilitate lead removal   In ,o
                                     18

-------
 doing, the lead reduction was 98.8% and ash content was 0.030 wt%.  It should




 be noted that for all the DAP runs, the sludge generation, excluding the DAP




 added, was no more than 1 wt%.  A duplicate run was done in AD4 and similar




 results were obtained as evidenced in Table 17.




      Examination of the demetallation results (Runs AD1-AD4) for Type A oil as




 shown in Table 17 reveal that: 1) much better lead and ash removal could be




 achieved when used oil contained a certain amount of water (apparently,  the




 water aids in dispersing the DAP during initial period of reaction);  2)  total




 metal contaminant content (i.e.  ash content)  was successfully reduced to an




 acceptable level (< 0.1 wt%);  3) a good quality of product oil could  be




 produced with only 1.5 wt% diammonium phosphate (DAP)  under mild reaction




 conditions;  and 4)  reproducibility of the  DAP demetallation experiment seemed



 to be acceptable.




      Other types of used oils  (Type B & E) were also  investigated using  the




 same  operating conditions  as Run ADS.   The lead contents  for Type B and  Type  E




 oils  were 480  ppm and  180  ppm, respectively and the ash contents  were  0.52 wt%



 and 0.65  wt%,  respectively.




      For  Type  B  oil, a 96%  lead  reduction was obtained in Run ADS while  a 99%




 lead  reduction was  obtained for  Type  E  oil in Run AD6.  The ash content of the




 product oil from both  the runs was very low.  Table 17 also shows that with




 1.5 wt% DAP treatment, used oil  (e.g. Type A, B, or E) seemed to be fairly



well  demetallated.




Liquid-Solid Separation Studies  --  The objective of these studies was to




evaluate  the responses in terms of separation ease to the addition of filter




aid and the use of a diluent (No. 2 fuel oil).  As has been done throughout




this work, both gravity sedimentation and filtration were studied for removing



solids.




     Two DAP demetallation runs (Runs ADS and AD9)  were performed in a 3785 ml



'..  '  .~^                              19

-------
  autoclave to study the effects of.filter aid and diluent on the settling time




  and filtration rates.   The response of ash and lead content of the DAP-treated




  oil to the settling time were obtained.   The effects of  dilution with No.  2




  fuel oil on settling time and filtration rate were also studied.   In Run ADS,




  1 wt% filter aid (diatomaceous earth)  was added to the  reactor with 700 g of




  Type A oil and 2 wt% DAP before the DAP  reaction,  whereas,  in Run AD9,  the




  same amount of the  filter aid was added  after the  DAP reaction.




       Table 18  shows  that an oil of  better quality  was obtained by adding 1 wt%




  filter  aid before the  DAP demetallation  (e.g. Run  8)  if gravity sedimentation




 was  adopted; for example,  in  Run ADS,  after  24 hour  of  settling at  60°C




 without any diluents,  the  lead content of  the product oil (i.e.,  top-layer




 oil) was greatly reduced  from 660 to 20 ppm  (i.e., 97%  lead reduction),




 whereas, in Run  9, with the addition of 1 wt% filter aid after the DAP




 reaction, the lead content of the product oil was reduced from 660 to 46 ppm




.(i.e., 93% lead reduction).  A higher ash content for the product oil was also




 obtained when the filter aid was added after the reaction (e.g., Run AD9) as




 seen in Table 18.  When filtration was  used,  both runs produced a similar oil




 quality.   In comparison with gravity sedimentation,  the  solids removal




 efficiency with filtration was higher.  However,  with gravity  sedimentation,




 the  product oil had  acceptable ash and  lead contents,  and  this  method does




have  its  advantages  in  lower operation  and maintenance costs.




      With  filtration  a  good solids-removing efficiency was obtained; however




the filtration  rates were very low for DAP-treated used  oil.  In order to




evaluate the use  of a diluent  (No. 2 fuel oil) to improve filtration rates,




another DAP  demetallation run was carried out in a 1000 ml distillation flask




under the reaction conditions shown in Table 12 (Run DF3).  After the




reaction, the reaction product including spent DAP was evenly divided into




four fractions; a specified amount of filter aid or No. 2 fuel oil or both




                                     20

-------
  were  thoroughly blended in  three  of  the  four  fractions.  Viscosity  of  each




  fraction was first measured using a  Cannon-Manning viscometer before




  filtration  (using the house vacuum).  The filtration rate for each  fraction




  was determined, and so was  the viscosity of the filtrate from each  fraction.




  The results showed that the viscosity of the product oil drastically decreased




  from 90 to 33 cP at 25°C when diluted with 20% No. 2 fuel oil by weight of




  total liquid as shown in Table 19 (No. 2 fuel oil has a viscosity of 3.5 cP at




  25°C).  The filtration rate, after dilution,  approximately doubled  (from 11.5




  to 25.0 ml/min).   With the addition of 1% by weight of the total liquid, the




  filtration rate was  further increased from 25.0 to 34.0 cm3/min; however,  the




  viscosity of the  resulting liquid increased from 32.8  to 38.4 cP due to the




  presence of extra solids (i.e.,  filter aid).



  Conclusion





      Used oil treated with NaBH4 aqueous  solution  (i.e.,  SWS  solution)  was




  characterized by high sludge production and ash content.  High sludge




  production could be circumvented by substituting SWS solution with NaBH4 or




 KBH4 powder.  However, the lead reduction efficiency decreased.  Demetallation




 performance with metal borohydride treatment could be greatly improved by




 distilling off water contained in the used oil before the reaction.




      As for DAP demetallation,  water present (about 5 wt%) during the early




 stages  of the demetallation reaction was found to be  beneficial.   This




 beneficial  effect  was  felt to be  due  to the  additional  interfacial contact




 area available for  DAP particles  to react  with  oil.   Overall,  mass transfer




 (DAP dispersion) was observed to be important; however, by extending  the




 reaction  time (. 2 hr), even  with mild reaction conditions, thermodynamic




 equilibrium  (. 100% lead removal) could be approached, thereby minimizing the



 significance of any mass transfer related parameters.




     In liquid/solid separation, a better quality of oil was obtained by




• :   --~-t    •                         21        "

-------
 filtration as compared to gravity sedimentation.   (Nevertheless, an EPA




 specification-grade oil was produced using gravity sedimentation.)   The oil




 filtration rate was doubled by dilution with 20 wt% of No.  2 fuel oil.



 Acknowledgements





      The  authors would like to thank Morton Thiokol,  Inc. for providing metal




 borohydride  chemicals.   The information in this document has  been funded in




 part  by the  United  Stated Environmental Protection Agency (EPA) under




 cooperative  agreement CR 812090 to Auburn University,  it has been subjected




 to the agency's peer and administrative review, and it has been approved for




publication.  Mention of trade names or commercial products does not




constitute endorsement or recommendation for use.
                                   22

-------
References

[1]  M.M. Johnson, "Reclaiming Used Oil by Chemical Treatment with
     Ammonium Phosphate," U.S. Patent 3,879,282, 22 April 1975.

[2]  Code of Federal Regulations, Title 40,  Part 266.40.

[3]  Process Stream Purification With Sodium Borohydride  Technical Manual and
     Users Guide, Morton Thiokol, Inc., Ventron Division.

[4]  Metal Removal & Recovery With Sodium Borohydride,  Morton Thiokol  Inc
     Ventrol Division.                                                '

[5]  Personal communication with Ventrol Division of Morton Thiokol,  Inc.

[6]  Miller, T.M., "An Investigation of the  Demetallization Chemistry
     Associated with the Re-refining of Used Motor Oil,"  M.S.  Thesis,  North
     Carolina State University,  Raleigh, N.C.  (1983).

[7]  "The Use of Disposable Catalysts in Coal Liquefaction  Activities for Oil
     Production", presented at the National  AIChE Meeting,  Houston, Texas
     March,  1981.                                                        '


[8]  Gollakota,  S.R.,  "An Investigation of Mass  Transfer  Phenomena in Coal
     Liquefaction:  Assessment of Resistances and Reactor  Types," Ph.D.
     Dissertation,  Auburn University,  Auburn,  AL (1984).
                                    23

-------
              Table I--Reaction conditions for typical  TBMR runs
Demetallation reagent
Reaction temperature,°C
Heat-up time, sec
Reaction time, hr
Agitation rate, cpm
Metallic Borohydrides
    110, 150
        60
        2.0
        860
  DAP
150-160
   60
0.5-2.0
  860
                                     24

-------
 Table 2--Demetallation results using metal borohydrides (for type B oil)
 Run No.


-50% NaOH Added to
 the Oila, wt%

 SWS,  wt%

 NaBH4 Powder,  wt%

 Phase Transfer
 Catalystb, wt%


After Reaction:

Sludge Production,
wt%
Product Oil Analysts:

Ash Content, wt%

Lead Content, ppm

Lead Reduction, %
 1.2     1.2    0.8     0     0.8    0     0.8

 0.65    0.65   0.43    0000

 000    0.05   0.05  0.05   0.05


 0      0     0      0     0    0.10.1
                            16
       16    15
0.36  0.36  0.58    -    0.77  0.40  0.53

70     69   240    430    300   350   330

85.4  85.6  50.0   10.4  37.5  27.1  31.3
aOil charge - 20g Oil Type: B (lead content
water content — 1.8 vol%).

 Tri-n-butyl-methyl ammonium chloride.
                — 480 ppm,  ash content «•> 0.52 wt%
                                     25

-------
       Table 3--Effect of oil type on demetallation
 Run No.                     i


 Oil Type                    B

 Lead Content, ppm         480

 Oil Charge, g               20

 5%  NaOH Added to
 the Oil, wt%              1.2

 SWS, wt%                 0.65


 Product Oil Analysis•

Ash Content, wt%          0.36

Lead Content,  ppm           70

Lead Reduction,  %        85.4
   8


   A

  660

   20


  1.2

0.65




 0.57

 500

24.2
                           26

-------
   Table 4--Demetallation results using metal borohydrides  (for  type A oil)
 Run No.

 50% NaOH.   Added to the
 oila,  wt%

 SWS, wt%

 KBH4 Powder,  wt%

 NaBH4  Powder, wt%

 Phase  Transfer  Catalystb,
    wt%
8 9
1.2 1.2
0.65 1.3
0 0
0 0
10
1.8
0
0
0.15
11
1.8
0
0
0.15
12
1.8
0
0.15
0
                                                            0.10
0.35
Product Oil Analysis;
Ash Content, wt%
Lead Content, ppm
Lead Reduction, %

0.62
500 280
24.2 57.6

0.56
300
54.5

0.50
180
72.7

0.68
270
59.1
aOil charge - 20 g Oil Type: B (lead content - 660 ppm, ash content - 0 S9%
water content - 5 volume %) .                                           •-=>•*,
 Tri-n-butyl-methyl ammonium chloride.
                                     27

-------
 Run No.


 Oil Type

 Reaction Temp.,  °c

 50% NaOH,  wt%

 SWS in  the
  oil,  wt%


After.
Lead Content, ppm

Lead Reduction, %
                             0.65
 Sludge  Production,  wt%     15.5
Product: Oil
                            70

                           85.4

  .  0.65




100% (Gel)




  400

 16.7

B
110
1.2
B
150
1.2
B
150
1.2
                                                                  0.65
                                                               100% (Gel)
                                   28

-------
             Table 6--Effect of caustic solution on gel formation
Run No.
Oil Type
50% NaOH in
the oil, wt%
SWS, wt%

NaBH4 Powder, wt%
15
Cotton Seed
Oila

•1.2
0.65

0
16
Cotton Seed
Oil

0


0.05
17
Cotton Seed
Oil

0

0
0.10
18
Motor

1.2

0.65
0
Oilb





After Reaction-


Sludge Production,
Vol%
                                   29

-------
                Table 7--Effect of NHjOH on sludge production
Run No.
50% NaOH, wt%           0.8
NH^OH, wt%                0
NaBH4 Powder, wt%       0.05
NH3, psig                 0

Before Reaction:
Lead Content, ppm       480
Ash Content, wt%        0.52

After Reaction:
Sludge Production, wt%   4.3

Product Oil Analysis:
Ash Content, wt%        0.77
Lead Content, ppm        300
Lead Reduction, %       37.5
 19

  0
1.5
0.05
 25
3.4
0.31
260
45.8
 20

  0
1.5
0.05
  0
3.2
0.37
400
16.7
Note:
     Oil type: B
     Oil charge: 20 g
     Reactor: 45 ml TBMR
     Agitation: 860 cpm
                                    30,

-------
                    Table 8--Properties of type oils used



Oil Type                    A        B        C         D




Water content, vol.%       5.0      1.8       0         0




Ash content, wt.%          0.59     0.52      0.75      1.02




Lead Content, ppm          660      480       700       640
 E




0.8




0.65




180
                                   31 -•

-------
   Table  9--Effect of water on demetallation with SWS
Run no.

Oila type

50% NaOH, wt%

SWS, wt%

Phase transfer
catalyst^,
wt%
9
A
1.2
1.3
21
C
-
0.90
22
C
-
0.40
23
C
-
0.90
24
C
-
0.25
25
D
-
0.35
26
D
.
0.^
                         0
Product Oil0 Analysis:

Ash content, wt%   0.62  0.53   0.68

Lead content, ppm  280   100    180

Lead reduction,%   57.6  85.7   74.3


^Oil charge - 20 g
 Tri-n-butyl-methyl ammonium chloride.
cSludge production in all the runs was high (about 15-20 wt%).
0.1


0.63

120

82.9
                                              0.1



                                              0.77

                                              300

                                              57.1
0.93

280

56.3
0.69

220

65.6

-------
 Table 10--Results from autoclave (3785 ml) runs using metallic borohydrides
 Run No.
 Atmospheric Distillation Conditions
 used for oil preparation.;
                                            AS2   AS3   AS4   ASS   AS6
                                      890   1000  1000   600   600   600
                                      0.25   0.25  0.25  0.25  0.25  0 25
                                      129    175   175   150   150   150
Type Aa oil charge,  g
50% NaOH, wt%
Distillation cut'point, °C

Reaction Mixture:

Residue13 from distillation, g
SWS, wt%
NaBlfy Powder, wt%
KBH4 Powder, wt%
Phase transfer catalyst0, wt%

Product Oild Analysis:

Ash content, wt%
Lead content, wt%
Lead reduction,  %
aWater content - 5.0%
 Lead content - 660 ppm
 Ash content - 0.59%

 Lead content — 700 ppm
 Ash'content - 0.86%

ctri-n-butyl-methyl ammonium chloride

dSludge production in all the runs was high (about 15-20 wt%).
890
0.90
0
0
0
0.74
440
37.1
900
0.91
0
0
0
0.50
25
96.4
900
0.46
0
0
0
0.51
420
40.0
545
0
0.10
0
0.20
0.57
440
37.1
545
0
0
0.10
0.20
0.56
345
50.7
545
0
0.15
0
0.20
0.53
200
71.4
                                    33

-------
Table 11--.Reaction conditions for
Demetallation Reagent
Reaction Temperature, °G
Heat-Up Time, hr
Reaction Time, hr
Agitation Rate, rpm
                     Metallic Borohydrides
                            110
                            1.5
                             2
                           2000
 DAP
 160
  2
  1
2000
                       34

-------
Table

    No.
 Type A Oil3 charge, g
 Water content in
 the oil, vol %
 Percent DAP added


 Product mib Analvseg

 Lead Content,  ppm
 Lead reduction,  %
 Ash  content, wt%
                             DF2

                              60

                               5
                               0
 640
   3
0.47
                                            18
                                            97
                                          0.06
DF3
60
5
2
DF1
60
5
8
  24
  96
0.02
aType A oil:
 Lead Content - 660 ppm
 Ash Content - 0.59 % (no filtration>>
 Water Content - 5 vol%   tiltration)

bReaction Conditions:
 Reaction temp,  °c
 Heat-up time,  min
 Reaction time, ' hr -            j_
 Reaction Pressure -.Atmospheric

                solids
                            75
                                 35

-------
    Table 13- -Effect of DAP dispersion (steel ball addition) for TBMR runs




 Run NO.                    Dl          D2          D3          D4




 No.  of Steel Balls Added   0           3           g           6a
Product Oilb Analyse-
Lead Content, ppm 200 153
Lead Reduction, % 70 77
TBMR Reaction Conditions:

127 156
81 78

Reaction Temperature     150°C




Reaction Time              1




Agitation                730 cpm




Type A oil charge         20 g




DAP added                  2 wt%






aDry Oil (lead content - 700 ppm)




bLess than 1 wt% solids (excluding DAP added) was produced in all the runs
                                   36

-------
                Table 14--Effect of heating time and reaction
                          temperature on demetallation
Run No.
TBMR Reaction Conditions15!
Reaction Temp, °C
Heat -Up Time, min
Reaction Time, hr
Agitation, cpm
Product Oilc Analyses :
Lead Content , ppm
Lead Reduction, %
Dl

150
1
1
800

200
70
D5

150
75
1
800

62
91
D8

160
1
2
800

64
90
D9a

160
1
2
800

81
88
D10

200
1
1
800

76
88
aDry Oil (lead content - 700 ppm) was used.

bOil type: A
 DAP charge: 2 wt%
 No. of steel balls used: 0

cLess than 1 wt% solids (excluding DAP added) was produced in all the runs
                                    37-

-------
 Run No.




 Autoclave Size,  ml




 Oil Charge3,  g




 DAP Added,  wt%




 Reaction Conditions:




 Reaction Temp,°C




 Heat-Up  Time,  hr




 Reaction Time, hr




 Reaction Pressure, psi




 Agitation,  rpm




 Product  Oilb Analysis:




 Lead Content, ppm




 Lead Reduction,  %
Table 15--Effect of reactor configuration on demetallation




                      D8     AD7     AD2
45
20
2
160
0
2
N/A
800
64
90.0
300
60
2
160
2
1
150
2000
21
96.8
3,785 (1 ga
554
1.5
160
2
1
atmosphere
2000
2
99.7
     used: type A




bLess than 1 wt% solids (excluding DAP added) was produced in all the runs,
                                     38

-------
Table 16--Reaction conditions for autoclave (3785 ml) run using DAP



          Temperature          - 160°C




          Heat-Up Time         - 2 hr




          Time                 - 1 hr




          Agitation            - 2000 rpm
                               39

-------
          Table 17--Demetallation with DAP using autoclave (3785 ml)
Run Noa

Oil Type
Oil Charge, g
Percent DAP Added, %
Product Ollc Analyses :
Lead Content, ppm
Lead Reduction, %
Ash Content, wt%
ADI
A*.
541
4.0

180
74.3
0.310
AD2
A
554
4.0

0
100
0.020
ADS
A
554
1.5

0
100
0.030
AD4
A
554
1.5

2
99.7
0.026
ADS
£
554
1.5

18
96.3
0.016
AD6
E
554
1.5

2
98.9
0.036
AD7
A^
560
1.5

580
17.1
0.464
aRuns ADI And AD2: The autoclave was completely sealed.




 Runs AD3-7: One port of the autoclave was intentionally opened




             to allow the water vapor to escape in an effort to




             simulate the distillation process.




bDry Oil: Lead Content - 700 ppm




cLess than 1 wt% solids (exlcuding DAP added) was produced in all runs
                                  40

-------
              Table  IS--Effect of filter aid and/or diluent on oil quality
  Run No.

 I.-Sedimentation in 60°C:
    Water bath for 24 hr
     Oil only
     Oil+1 wt% filter aid
     Oil+20 wt% No. 2 fuel oil
     Oil+1 wt% filter aid+
     20 wt% No. 2 oil

II.  Vacuum Filtration:

     Oil only
     Oil+1 wt% filter aid
     Oil+20 wt% No. 2 fuel oil
     Oil+1 wt% filter aid+20
      wt% No.  2 fuel oil
          AD 8
                         AD 9
Ash Content,
   wt%

  0.046

  0.012
Pb Content,
    ppm

     20

     10
  0.012
  0.008
      0

      0
Ash Content,
     wt%
    0.066
    0.124
    0.050
    0.066
    0.012
    0.014
    0.004
    0.002
Pb Content,
     ppm
      52
      46
      36
      32
       4
       6
       2
       4
                                       41

-------
        Table 19- -Effect of filter aid and diluent on
                  filtration rate and oil viscosity

           Viscosity at 25°C          Filtration Rate      Filtrate Viscosity
                                         (ml/min)          at 25°C (cP)
Oil Only        89.9                      11.5                 87.7

Oil +
1 wt% Filter
Aid             95.5                      12.0                 87.3

Oil +
20 wt% No. 2
Fuel Oil        32.8                      25.0                 37.3

Oil +
20 wt% No. 2
Fuel Oil +
1 wt% Filter
Oil             38.4                      34.0                 37.6
                                  42

-------
                 VARIABLE
                   SPEED
                 ELECTRIC
                   MOTOR
               FLYWHEEL
                   &
               PUSHROD

               BEARING
             TACHOMETER
               AGITATOR SHAFT
               M1CROREACTOR (TUBING BOMB)
                         FLUIDIZED SAND BATH
                       TEMPERATURE CONTROLLER
Schematic diagram of tubing bomb microreactor agitation assembly.
                 43

-------
    240
    200
 E
\
 o>
T3

O
O 8OO  CPM


A 4OO  CPM
                                                   00
                   Reaction Time, Min
         Effect of reaction time and agitation rate on lead removal
                               44

-------
                                                                                    1
 Q-
 Q.
O
O

-o
O
                                               Oil* Type B
                                               Reactor  Volume1 45 m£
                                               DAP  %  in Oil* 8%
                                               Water Added' 12%
                            30         45         60
                              Reaction Time, min.
75
       Effect of reaction time on lead removal at various temperatures.
                                      45  '

-------
o.
a.
a

-------