EPA/600/R-92/000
                                             March 1993
 METHODS FOR DEMETALJLAT1ON OF WASTE OIL
                     by

      Dr. A.R. Tarrer and Dr. N.P. Dhuldhoya
Auburn University Chemical Engineering Department
                230 Ross Hall
       Auburn University, Alabama 36849
        Contract No. EPA-CR814635-01
               Project Officer

              Edward R. Bates
        Alternative Technologies Division
     Risk Reduction Engineering Laboratory
      U.S. Environmental Protection Agency
            Cincinnati, Ohio 45268
  This study was conducted in cooperation with
       The U.S. Department of Defense.
 RISK REDUCTION ENGINEERING LABORATORY
  OFFICE OF RESEARCH AND DEVELOPMENT
 U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO  45268

-------

-------
                                             EPA/600/R-92/000
                                             March 1993
 METHODS FOR DEMETALLATION OF WASTE OIL
                    by

      Dr. A.R. Tarrer and Dr. N.P. Dhuldhoya
Auburn University Chemical Engineering Department
                230 Ross Hall
       Auburn University, Alabama 36849
        Contract No. EPA-CR814635-01
               Project Officer

               Edward R. Bates
        Alternative Technologies Division
      Risk Reduction Engineering Laboratory
      U.S. Environmental Protection Agency
            Cincinnati, Ohio 45268
  This study was conducted in cooperation with
        The U.S. Department of Defense.
 RISK REDUCTION ENGINEERING LABORATORY
   OFFICE OF RESEARCH AND DEVELOPMENT
  U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO 45268

-------

-------
                                      DISCLAIMER
       The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency under contract number EPA-CR814635-01 to Auburn University. It has
been subjected to the Agency's peer and administrative review, and it has been approved for publication
as an EPA document. Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.

-------

-------
                                         FOREWORD
       A variety of research efforts have clearly established that used oil is a valuable,  renewable
resource, and that proper implementation and management of viable technology can ensure its effective
reutilization.

       Over the past few years, researchers with the Chemical Engineering Department  of Auburn
University have been working to develop economically and environmentally sound technology for
reprocessing used oil.  The result of these  efforts is a systematic demetallation procedure, whereby
soluble metals are converted to insolubles and removed with other impurities. The final product is a high-
grade, EPA-specification used fuel oil which can be used for heating or as a reusable base stock for
lubrication and hydraulic fluids.

       The product meets EPA air quality regulations and can be burned in most industrial boilers
without the requirement of a special bum permit. Also, an EPA tracking system is not required for the oil
beyond the demetallation procedure.  In addition, liability risks are lowered for users, since the product
qualifies as an ASTM No. 5 grade fuel oil with an ash content of less than .1 weight percent; thus,  no
mixing of the product with other fuel oils is needed to meet paniculate emissions requirements.

       For several years, the Chemical Engineering Department and its primary research unit, the Auburn
Waste Oil Reprocessing Laboratory, have interacted with a number of industries and military installations
to develop and  provide useful reprocessing technology.  This report offers  an account of recent
developments in the Department's ongoing goal of developing the finest state-of-the-art oil reutilization
technology.
                                           A.R. Tarrer
                                           Director, Auburn Waste Oil Reprocessing Laboratory
                                           Chemical Engineering Department
                                           Auburn University
                                              HI

-------

-------
                                          ABSTRACT
       Under a contract from the Department of Defense (DoD), work was initiated by the Chemical
Engineering Department of Auburn University to develop technology for converting waste oils into EPA
specification grade fuel oil. DoD is particularly concerned with the development and utilization of this
technology because it is one of the largest generator of waste oils in the U.S.

       A process objective was to dehydrate the waste oil, reduce its total ash content (< 0.1 wt%), and
remove the impurities (such as sludge,  carbon particles, acids, etc.) present in used oil.  Particular
emphasis was placed on demetallization and neutralization of waste oils by a chemical treatment. It was
found from bench-scale studies that diammonium phosphate and triethanol amine were very effective in
deashing, as well as  in precipitating sludge and impurities from used oil.  An extensive study was
performed to determine the effect of various parameters on the ash and lead reduction of  waste oil.
Based on the bench-scale studies, the pilot plant was developed at the Auburn Waste Oil Reprocessing
Laboratory (AWORL) to transferor) waste oil into specification grade fuel  oil.  The pilot plant has the
capacity  to treat 5 gallons/minute of used oil.  So far, a total of 7,000 gallons of waste oil  has been
reprocessed. The ash and lead content of the processed oil was reduced to less than 0.1 wt% and 100
ppm, respectively.  This process was successful in converting waste oil into specification grade fuel oil.
                                              iv

-------

-------
                                   TABLE OF CONTENTS
DISCLAIMER	    »
FOREWORD	   '»
ABSTRACT	    *
TABLE OF CONTENTS	v
USTOFTABLES.	   viii
LIST OF FIGURES	  Xl

    I.  INTRODUCTION.	1
   II.  LITERATURE REVIEW	4

          Recovery Processes	4
            Acid-Clay Process	5
            Solvent Extraction-Distillation Process	5
            Chemical Deashing-Hydrotreatment Process.	9
          Classification of Used O9	  13
            Resource Conservation and Recovery Act	13
            Final Rule, Burning of Used Oih	16
            Proposed Rule, Recycled Used Oil Standards	 16
            Identification and Listing of Used Oil, Proposed Rule	16
            Waste Oil Categorization	17
            Standards Applicable to Generators, Processors, Marketers, Burners	17
            Recent Developments in Classification of Used Oil	19

   III.  CHARACTERISTICS OF USED OIL	  24

          Origin of Base Stocks	  25
            Lubricant Additives (27, 28, 29)	  28
            Antiwear and Extreme Pressure Agents	28
            Corrosion and Rust Inhibitors	,	28
            Detergent and Dispersants.	28
            Antioxidants	29
            Metal Deactivators and Rust Inhibitors	29
            Industrial Lubricants. .  .	29
            Mechanism of Oil Oxidation	31
            Metals in Used Oil	31
            Naturally Occurring Metals	32
            Metals Incorporated  in Additives	32

-------

-------
                         TABLE OF CONTENTS (CONTINUED)
         Metals Introduced During Use.	32
         Chemical and Physical Nature of Metals in Oil	32

IV.  EXPERIMENTAL PROCEDURES	35

       Tubing Bomb Micro-Reactor (TBMR)	35
       Distillation Flask Reactor	38
       Autoclave	38
       Three-Neck Flask.	38
       Materials	44
       Analysis	44

 V.  USE OF METALLIC BOROHYDRIDE AS A DEMETALUZ1NG REAGENT.	45

       Results and Discussion.	 46
         Effect of Reaction-Temperature	 48
         Investigation  of Gel Formation	  52
         Effect of Caustics	  52
         Effect of Water	52
         Autoclave Study	56
       Conclusions	58

VI.  USE OF DIAMMONIUM PHOSPHATE AS A DEMETALUZATION REAGENT	59

       Mechanism of Demetallization Reaction.	, 59
         Experimental Procedure	60
         Solids Removed by Sedimentation	60
       Results and Discussion.	60
         Effect of Reaction Time	60
         Effect of Reaction Temperature	62
         Effect of Ash  Content	  62
       Solids Removed by Filtration	  69
         Effect of Diammonium Phosphate on Ash  and Lead Reduction	  69
         Effect of Water on Ash and Lead Reduction	  69
         Effect of Temperature and Residence Time on Lead Reduction	73
       Mechanism for  Lead and Metals Removal. . .  .	73
       Limitations	73
       Conclusions	78
                                        vi

-------

-------
                           TABLE OF CONTENTS (CONTINUED)
  VII.   USE OF TRIETHANOLAMINE AS A FLOCCULATING REAGENT	80

            Effect of Triethanol Amine on Ash and Lead Reduction	81
            Effect of Temperature and Time on Ash and Lead Reduction	84
            Effect of Water on the Demetallization of Oil	84
          Mechanism of Metals Removal	      89
            Solid-Liquid Separation	    89
            Filtration	89
            Centrifugation	91
            Gravity Sedimentation	92
          Conclusions	   92

 VIII.   PILOTPLANT	94

          Demetallizatkxi	   94
            Lamella	   94
            Sludge Treatment	98
          Results.	98
          Conclusions	101

   IX.   BURN TEST.	  102
   X.   BIBLIOGRAPHY.	104

APPENDIX

       A.  QUALITY ASSURANCE ACTIVITIES	108
       B.  QUALITY ASSURANCE/QUALITY CONTROL PLAN - RECLAIMED
          OIL TRIAL BURN	117

          Results and Discussion	117
            Comparison of Measured Emission-Reclaimed Fuel Oil and No. 5 Fuel Oil.  . .  118
            Reclaimed Fuel Oil	118
            Commercial No. 5 Fuel Oil	118
            Sampling Procedures	118
            Quality Control	  123
            Quality Assurance Objectives	125
            Sampling Procedures	125

                                          vii

-------

-------
UST OF TABLES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.

Characteristics of Hazardous Wastes . 	 	
Contaminants in Extract 	 	 	
Specification Used Oil 	
Industrial Oil Sales for 1975 	
Elemental Analysis of Metals in Used Motor Oil 	
TBMR Reaction Condition for Venpure Product Treatment 	
Effect of Metal Borohydride on Lead Reduction for Type B Oil 	
Effect of Oil Type on Lead Reduction Efficiency 	
Effect of Metal Borohydride on Lead Reduction for type A Oil 	
Effect of Reaction Temperature on Sludge Formation 	
Influence of Caustics on Formation of Gelatinous Mass 	
Effect of NaOH on Sludge Production and Lead Reduction 	
Influence of Water on Lead Removal 	
Results of Autoclave Runs 	
Effect of Reaction Time on Lead Removal 	 	
Effect of Reaction Temperature on Lead Reduction 	 	
Effect of Reaction Temperature on Lead Reduction (1 Hour) 	
Effect of Ash Content on Lead Reduction 	
Effect of Filtration on Ash and Lead Reduction 	
vui
	 14
	 15
	 17
	 30
	 34
	 46
	 47
...... 49
	 50
	 51
	 53
	 54
	 55
	 57
	 63
	 65
	 65
...... 67
	 67


-------

-------
                                LIST OF TABLES (CONTINUED)
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
A-1.

A-2.
A-3.
A-4.
Effect of DAP on Lead Reduction	  70
Effect of Water on Ash and Lead Reduction	   72
DAP Demetallization Study Using a One-Gallon Autoclave	77
Effect of Diluent and Filter Aid on Filtration Rate	79
Effect of TEA on Ash and Lead Reduction	82
Additive Metals Content of TEA-Treated Oil	  82
Effect of DAP on Ash and Lead Reduction (0.5 wt. %)	83
Effect of DAP on Ash and Lead Reduction (1.0 wt. %)	83
Effect of Reaction Time on Ash and Lead Reduction	85
Effect of Reaction Temperature on Ash and Lead Reduction	85
Effect of Water on Oil Recovery	86
Effect of Different Modes of Separation on Ash Content	  91
Metal Contents of Used Oil Processed by Chemical Treatment	93
Metals Content of Oil Processed at Pilot Plant	  100
Volatility Test of Organometallic Lead Compound in Muffle Furnace at 600°C
for 12 Hours	
112
Accuracy of Lead Analysis Procedure	   114
Determination of Ash and Lead Contents of Virgin Motor Oil	   115
Precision Test for Lead and Ash Analysis	   116

-------

-------
                             LIST OF TABLES (CONTINUED)
B-1.    Comparison of Measured Emission Rates
B-2.    Reclaimed Fuel OH Emission Data....
B-3.    Commercial No. 5 Fuel Oil Emission Data.
B-4.    Summary of Duplicate Analyses	
B-5.    Summary of Spiked Sample Results . . .
119
119
121
126
126

-------

-------
                                       LIST OF RGURES

 1.     Waste Oil Generation and Disposal	2
 2.     Acid-Clay Process	6
 3.     KTl Vacuum Distillation and Hydrofinishing Process  .	7
 4.     Recyclon Process	8
 5.     Demetallization Section of the Phillips Process	10
 6.     Hydrotreating Section of the Phillips Process	11
 7.     Waste Oil Categorization	18
 8.     Compliance with Generator Regulations	20
 9.     Compliance with Processor Regulations	21
10.     Compliance with Marketer Regulations	22
11.     Compliance with Burner Regulations	   23
12.     Production of Paraffinic Base Oils	26
13.     Production of Naphthenic Base Oils	       27
14.     Drawing of Tubing Bomb Micro-Reactor	36
15.    Agitation Assembly for TBMR	37
16.    Diagram of Filtration Apparatus	39
17.    Drawing of Distillation Unit	40
18.    Cross-Sectional View of the Autoclave	   41
19.    Schematic Diagram for the Autoclave Reaction System	42
                                             xi

-------

-------
                               LIST OF FIGURES (CONTINUED)

20.    Three-Neck Flask Reactor	43
21.    Effect of Reaction Time on Lead Reduction		61
22.    Effect of Reaction Temperature on Lead Reduction	64
23.    Effect of Reaction Temperature on Lead Reduction (1 Hour)	  66
24.    Effect of Ash Content on Lead Reduction.	68
25.    Effect of Reaction Temperature on Lead Reduction	   74
26.    Effect of Residence Time on Lead Reduction	   75
27.    Proposed Mechanisms for Additive Metal and Lead Removal	76
28.    Effect of Settling Temperature on Oil Recovery	87
29.    Effect of Settling Time on Oil Recovery	   88
30.    Proposed Mechanisms for Metal Removal	90
31.    Flow Sheet for Waste Oil Recovery Pilot Plant	  95
32.    Flow Sheet for Demetallization Section	96
33.    Lamella Clarifier	97
34.    Solid Sludge and Wastewater Pretreatment System	99
35.    Project Organization Chart.	103
A-1.    Lead Analysis Procedure	109
A-2.    Ash Analysis Procedure	110
A-3.    AA Calibration Procedure	111
                                            xii

-------

-------
                              LIST OF FIGURES (CONTINUED)





B-1.    Location of Sampling Forts	119



B-2.    Location of Traverse Points	119
                                           xui

-------

-------
                                          SECTION I
                                       INTRODUCTION
       In the United States, automotive and other industrial sources generate about 1.2 billion gallons
of used oil each year. These figures, although less than 1% of the nation's petroleum consumption, are
substantial and create a significant problem in the area of waste oil recovery both for conservation of
resources and for the protection of the environment.

       The major sources of waste oil recycling are automotive and industrial lubricating oil, which
account for 46% of waste oil generation as shown in Figure 1.  Only a small amount of used oil is
rerefined for its original use, while the rest is reprocessed  as a fuel.  The rerefining industry has
experienced a recent decline for both technical and economic reasons. For instance, lubricating oils have
become more complex with the blending of 15 to 20 chemical additives, and rerefined lube oil markets
have suffered somewhat from the controversy over whether the quality meets current-day specifications.
Also, extended drain periods have concentrated additives and impurities in waste oil, making it more
difficult to rerefine; collection costs have increased; and disposal of the sludge by-product has become
more difficult and expensive.  Further uncertainty regarding classification of used oil as hazardous or
nonhazardous wastes has placed oil recyclers, transporters, service stations and other companies that
handled used oil into a dilemma.

       Waste oils are in some respects superior for lube manufacture compared with crude oil. They are
generally rich in lubricating oil (65-80%),  compared  to 10-17% from  most crude oils used  in lube
manufacture.1-2  Also, less than 30% of waste oil is rejected as residue, compared with 45-65% rejected
in regular lube manufacture. Waste oil, although contaminated, has a high energy value, between 15,000
to 20,000 BTU/lb., and burning it  as a fuel is a major outlet for used oil. Burning offers an efficient and
inexpensive alternative to dumping, land spreading, road oiling or incineration. Unfortunately, used oil
contains high concentrations of metallic contaminants, and combustion of the oil without removing these
contaminants can cause adverse  environmental effects. From the public health point of view, lead is the
most signfficant contaminant. If all waste oil was disposed by burning, the EPA estimates that this would
account for as much as 5% or even 10% of all atmospheric lead.  However, amounts have substantially
decreased in recent years due to  the elimination of leaded gasoline. Due to the toxicity of heavy metals,
the use of recycled oil as a fuel is subject to certain regulations, unless the fuel meets a specification for
toxic contaminants and the flash  point is exempt from  regulation.

       One major obstacle to the reuse  of used oil in many applications is the presence of various
ash-forming impurities that remain dispersed in the oil due to the very effective dispersant characteristics
of the additives in a modern-day lubricant system. Materials contained in a typical sample of used oil that
are considered to contribute to the ash content of the oil include sub-micron size carbon particles, metal

-------
or
LLJ
LLJ
LL1


CO
                                    CO
                                    O
                                    a.
                                    CO

                                    Q


                                    O
                                    UJ

                                    CO

                                    I:
 ca
 CO
 o
 Q.
 to

b
                                               O
                                               0)


                                               
-------
particles, inorganic materials such as atmospheric dust, and lead and other metal compounds originating
from fuel combustion.3 In addition to lead, appreciable amounts of zinc, barium, calcium, magnesium,
phosphorous and iron are generally present in used motor oil.  Examination with an optical microscope
reveals that the particle size of the patticulates ranges  between 0.1  and 1 microns with no sign of
agglomeration.4

       A number of rerefining processes have been used in the past.  The major processes used for
waste oil recycling are: (1) acid-clay treatment and (2) solvent extraction-distillation.

       The acid-clay treatment produces a rather large amount of acidic sludge,  which could pose a
larger pollution problem than that caused by used oil. A relatively low process yield (50-60%) and almost
intractable  residue problems have resulted in a firm recommendation against its use by the rerefining
industry. Solvent extraction requires a large ratio of solvent to oil, and metal contaminants still remain
present after treatment due to the  high dispersant characteristics of additive compounds.  Solvent
extraction and distillation is mainly used for rerefining due to the high cost of the treatment.

       In principle, many methods could be employed for removing the contaminants, but economic
considerations frequently dictate using chemical treatment as a viable alternative in producing fuel oils.
Although demulsifiers, flocculating aids, contaminant oxidizer and conditioning reagents (caustics) are
available commercially, chemical treatment of waste oils has not yet been demonstrated on a large scale.
However, low capital and operating costs, high product yields, potentially wide application, low energy
treatments and especially low residue production make chemical treatment an attractive choice for waste
oil processing facilities involved in fuel oil production.

       After reviewing the literature on existing processes, it seems that chemical treatment is a viable
and more suitable technology for converting waste oils into EPA specification grade fuel oil (see Chapter
2).   Under a  contract from the Department of Defense (DoD), work was initiated by  the Chemical
Engineering Department at Auburn University to develop related technology for fuel oil production.  DoD
is particularly concerned with the development and utilization of this technology because it is one of the
largest generators of waste oils (and may be the largest)  in the United States.

       A process  objective was to dehydrate the feed waste oil,  reduce its total ash content (0.1 wt%),
and remove impurities (such as sludge,  carbon particles, acids,  etc.) present in-used oil, so it can be
burned as a fuel without harming the environment.  Moreover, through further upgrading techniques, the
treated oil can be reprocessed to  produce a high-quality  lubricating oil.  The project was to be focused
on simple,  high product yielding, low capital intensive, and efficient processing methods.

       To  achieve these goals, the project was divided into three major tasks:

       Taskl: Demetallization and removal of impurities from waste oil by chemical treatment.

       Task 2: Design and development of a pilot scale plant based on the bench-scale results obtained
from Task 1.

-------
                                          SECTION II
                                     LITERATURE REVIEW
        As previously mentioned, the waste oil rerefining industry has suffered a sharp decline in activity
 over the past few years. There are presently fewer than 10 rerefining firms and about 250 fuel processors
 in the U.S.  Many of the fuel processors once made lubricants from waste oil, but have since ceased
 because it costs more to produce lubricants than to simply clean the used oil to use as industrial fuel.
 Among the major problems faced by the industry are:

        (1) rising costs in collection and purchase of used oil;

        (2) the growing complexity of additives blended with lube oils, thus making removal  more
 expensive;

        (3) existing rerefining technology resulting  in hazardous wastes, the disposal of which  is an
 expensive undertaking as well as a threat to the environment; and

        (4) uncertainty regarding classification of used oil.

        The purpose of this chapter is to review the  existing recovery processes and discuss the merits
 and limitation of commercial processes. The second part of this chapter deals with classification of used
 oil, which is a determining factor in the future growth of the waste oil industry.
RECOVERY PROCESSES
       There are a large number of physical and chemical processes which have been developed for
reclamation of industrial and automotive lubricants.  Some of these processes are directed toward
upgrading the oil slightly for use in the same machines, while other processes are very complicated and
produce a whole range of petroleum products.

       The processing of crankcase oils from diesel and gasoline engines has developed over the years
from simple settling, filtration and dehydration processes into complicated petrochemical-type processes
producing a multiple product line.  Because of the complexity of crankcase oils, their renovation and
separation generally is more complicated. There are many proprietary processes available for waste oil

-------
recycling, but the major commercial processes generally fall into three categories: (1) acid-clay treatment,
(2) solvent extraction-distillation, and (3) chemical deashing hydrotreatment.

Acid-Clay Treatment

       The acid-clay process dominated the rerefining industry during the 1960s.  A flow sheet for the
traditional acid-clay process is shown in Figure 2.
        In this procedure, waste oil is dehydrated and stripped of light fractions prior to acid treatment.
Dry oil is reacted with 4-6 percent of 93% sulfuric acid in the reactor at 100°F. Although fresh acid is
generally used, several rerefiners use spent alkylation acid from petroleum refineries. The acid sludge.
containing oil contaminants and ash, separates from the oil and is drawn off from the reactor bottom.

        The acid-treated dehydrated oil is then transferred to the steam-stripping/clay-treating operations,
in which the temperature raised from 550-660°F. The stripping operation removes the remaining light fuel
fractions and odorous compounds which may be present.

        The clay, often a 50% mixture of activated clay and diatomaceous earth, is mixed into the oil,
where it removes  color bodies  and  colloidal  carbon by adsorption.   The clay dose  should be
approximately 0.4 Ib/gallon of oil. The hot oil containing the clay is filtered through a plate and frame filter
press. The filter cake, a mixture of clay, impurities, and oil, is then discarded by landfill.

        Acid-clay treatment is not a mild treatment, as evidenced by product yields of only 45 to 60% (6),,
It also generates a very large amount of acid sludge, which can be difficult to dispose. Odors can be a
problem with acid-clay rerefining. However, metals were effectively removed by the treatment. Acid-clay
treatment has become obsolete in the rerefining industry due to the low yield and disposal problems of
by-products acidic sludge.

Solvent Biraction-Distillation

        Distillation as a tool for rerefining used crankcase oil is not an innovative approach. In  Europe,
Kinetics Technology International (KTI) has developed the process based on distillation and hydrofinishihg.

        In the KTI process, distillation removes water and gasoline.  Vacuum distillation produces an
overhead in the lube oil range and  a  heavy residue containing metals, polymerization products and
asphaltenes. The process diagram is shown in Figure 3.5  KTI uses the refinishing process developed
by Gulf Science and Technology Company to improve the color and odor of the overheads. The lube oil
is mixed with a hydrogen-rich gas, heated and passed through a reactor holding a fixed catalyst bed.
To obtain  a product with the right specifications, the treated  oil is either stripped with steam or
fractionated into different lube cuts and then dried in  a vacuum column.

       Another commercial  distillation process, the Recyclon process, uses sodium to remove oil
contaminants from waste oils.  The process flow diagram of the Recyclon process  is shown in Figure 4."
Waste oils are first distilled to remove water and light ends. Sodium particles from 5 to 15 micrometers

-------
Waste oil
Lube oil^HM
•«••
Oewatering
i ,
Acid
treatment
I
Clay
treatment
1
Fractionation
*i^—-t Acid tar
Spent clay
Oil
•••^ Filtering •nMann^ residue
Figure 2. Acid-Clay Process.

-------
    I
    i    i    r    i    i
uunioo BuiXjp pue 6utddiJjs
    I    l'   I-    I '  i
                                                       CO
                                                       (0
                                                       8
                                                       2
                                                      a.
                                                       en
                                                      I
                                                      T3
                                                       CO
                                                       g
                                                      I
                                                      1
                                                      b

                                                       I
                                                       u
                                                      CO
                                                      £
                                                      2J
                                                      u.

-------
                                              CO
                                              CO
                                              Q>
                                              u
                                              p
                                              o

                                              I
                                              CC
                                              I
                                              o>
8

-------
in size are then dispersed into the dry oil in amounts less than 1 wt.% of oil.  Typical reactions which
occur in the oil are: polymerization of unsaturated olefins, conversion of halogen compounds to sodium
salts, and conversion  of  mercaptans into materials that remain in the residue.   Next,  low-boiling
compounds are flashed in thin-film evaporators at a pressure lower than 1 millibar. The Recyclon process
yields over 70% rerefined lubricating oil, while the rest is fuel.

       Although distillation of used oil appeared more effective than any other treatment in reducing
contaminants, the  oil in the distillation  bottoms forms a gelatinous mass with considerable cracking.
Without pretreatment, distillation is slow, and excessive distillate must be left to keep the residue fluid.
This produces undesirable hydrocarbon compositional changes selectively removing di- and pplyaromatics
as well as polar materials that are required for good lubricity and oxidation stability.7

       Coking and fouling problems encountered with direct distillation of used lubricating oils indicate
the need for a pretreatment step in rerefining processes which employ distillation. A number of solvents
have been used to pretreat waste oil before the distillation; examples of solvents used in rerefining
processes include hydrocarbon as propane8; alcohol as  butanol,9 isopropanol,10 and tetrahydrofurfuryl
alcohol11; and ketone as methylethyl ketone12.  A systematic, detailed study of pretreatment by solvent
precipitation was reported by the Bartlesville Energy Technology Center in Oklahoma.13-14'15'16 Out of 64
tested solvents, the  optimum system was determined to be a mixture of 2-propanol, 1-butanol, and
methylethyl  ketone in a ratio of 1:2:1 by volume.  The pretreatment step removes coking and fouling
compounds; the resulting sludge is separated from the oil/solvent mixture in enclosed vertical solid-bowl
centrifuges.  The solvent  pretreatment,  performed in conjunction with dehydration, distillation, and a
finishing process, produces a high-quality lube oil.  It was reported that more than 99.5% of the solvent
can be recovered from the process.

       Among the disadvantages of the BERC process are: (1) the pretreatment requirement is energy
consuming, and (2)  the lack of definitive full-scale demonstration plant data, which could be made
available to  potential licensees of the process.

Chemical De-Ashing-Hydrotreatment Process

       The Phillips Rerefined Oil Process (trademarked PROP) is a recent, innovative development.17'181S
This process uses a new method of waste oil demetallization in which oil is reacted with a slurry of
diammonium phosphate.  Since the reaction conditions are base in pH, metal reaction vessels may be
used, and the metals in the oil react to form insoluble metal phosphates rather than the acidic sludges
that form in some other rerefining processes.

       Another innovation offered by the Phillips process is the low volume of hazardous by-products
produced.  By-products of the process include kerosene, spent clay and charcoal, a filter cake, spent
caustic, spent catalyst and off gases. The metals contained in the filter cake, clay, and charcoal after use
are bound in an inert a form.  PROP also offers up to an 80% recovery, as usable rerefined oil, of the oil
actually present in the waste oil feedstock.

       The PROP plant consists of two sections: (i) the demetallization section and (ii) the hydrotreating
section.  Figures 5 and 6 show a schematic of the PROP plant. The demetallization section of the PROP

-------
                                             0)
                                             (0
                                             0)
                                             o
                                             o
                                             CO

                                             Q.
                                            Q.


                                             CD
                                            g

                                            "ts
                                            0)
                                            CO
                                            (0
                                            N




                                            1
                                            0)
                                            Q

                                            id


                                            I
                                            o>
10

-------
 tO

 CO
 CO

 Q.
 
-------
plant consists of three stirred tank reactors in a series and a filtration step.  Each reactor has a slightly
different function and operates at a different temperature. Reactor No. 1 operates at 210°F; the waste oil
feedstock and the diammonium phosphate are mixed and the reaction begins here. In reactor No. 2,
which operates at 225°F, demetallization continues while the higher temperature distills most of the water
and some of the lighter hydrocarbons found  in the oil.  The third reactor operates at 310°F; at this
elevated temperature, ammonium salts of organic acids formed in the diammonium reaction begin to
break down into ammonium and organic acids. Also, the remaining water and light ends are removed
in this reactor. After the oil passes through the third unit, it is filtered through diatomaceous earth.  The
filtration step removes the metal phosphates  formed in the reaction as well as any remaining large
particulates.

       The demetallized oil is then percolated  through a series of two clay beds and one charcoal bed.
The remaining ammonium salts of sulfuric and phosphoric acids formed in the demetallization sttep are
broken down in the beds. Ashless detergents (containing no metals) are also broken down in the  clay
and charcoal beds. Tetra-ethyllead, present in the oil due to contamination with gasoline, is broken down
and removed by the beds.  This not only serves as a final step in the demetallization process but  also
protects the hydrotreating catalyst from being poisoned by heavy metal contaminants.

       Hydrotreatment of the demetallized oil is the next step. The hydrotreatment occurs at 670°F and
615 psig pressure.  Reaction of the oil with hydrogen removes the remaining (nonmetal) portion of the
oil additives.  Compounds containing Cl, S, O, and N are converted to analogous hydrocarbons and  HCI,
H2S, H2O, and NH3, respectively. A series of strippers and scrubbers remove these  product gases from
the hydrogen stream, which is then recycled.

       After hydrotreatment and removal of unwanted reaction products, the oil is blended with fresh
additives to create a 30-weight motor oil. This oil meets or exceeds all applicable API and SAE standards.

       Among the major disadvantages of the PROP process:

       (1) filtration is very slow and cumbersome to operate;                         ;
       (2) the process is energy intensive; and
       (3) large amounts of solid waste are generated.
steps:
       In summary, current commercial waste oil reclaiming processes utilize many of the following basic
       (1)  Removal of water and solid particles by settling.
       (2)  Solvent pretreatment to remove gums, grease, sludge, asphaltenes, etc.
       (3)  Alkaline treatment to neutralize acid.
       (4)  Clay contacting to bleach the oil and adsorb impurities.
       (5)  Stripping to drive off moisture and volatile oils.
       (6)  Filtering to remove clay and other solids.
       (7)  Hydrotreatment to improve the oxidation stability of lubricating oil.
       (8)  Blending to specification.
                                              12

-------
        In view of today's environmental concerns, it appears that an economically viable rerefining
process is needed which is able to:

        (1) Eliminate or minimize potential by-product pollutants (spent acid, spent caustic, spent clay,
           SO2, etc.).
        (2) Produce acceptable products.
        (3) Remain flexible enough to adapt to changing additives and/or environmental considerations.
CLASSIFICATION OF USED OIL
       The legislative curb on industrial practices for disposal of hazardous waste has become more
stringent in recent years.  To understand the existing regulatory compliances, one must comprehend a
host of related statutes and environmental policies and programs of the U.S. Congress, as well as state
and local governments.  A basic understanding of major federal statutes and various state and local
regulations is regarded as essential in today's litigious society. Notwithstanding the need for a knowledge
of various regulations, the most heated topic of environmental policies and programs centers around
hazardous materials.

       Congress delivered a strong environmental message in 1976 when  it passed the Resource
Conservation and Recovery Act (RCRA). The mandates formulated under RCRA affect all manufacturing
and business operations that dispose of waste materials on  land.  Before discussing the regulatory
aspects of used oil management, a brief overview of the RCRA is warranted.

Resource Conservation and Recovery Act

       Since its enactment in 1976, RCRA has been amended several times. Subtitle C of this act, which
deals with hazardous waste, remains the centerpiece of RCRA.  EPA regulations define a solid waste as
hazardous waste if it exhibits any of the characteristics listed in Table 1.20 Under EPA regulations,  certain
types of solid wastes (for example, household waste) are not to be considered hazardous wastes.

       Besides the four characteristics of hazardous waste given in Table 1, the EPA has also established
three hazardous waste lists: hazardous wastes from nonspecific sources (such as spent nonhalogenated
solvents), hazardous wastes from specific sources (for example, bottom sediment sludge  from  the
treatment of wastewater from wood preserving), and discarded commercial chemical products  and all
off-specification species, containers, and spill residues.

       The EPA has provided certain regulatory exemptions based on very specific criteria. For example,
hazardous waste generated in a product or raw  material storage tank or transport  vehicle or in
manufacturing processes,  as well as samples for monitoring and  testing purposes, are exempt from
regulations.  Used oil being recycled also would not be listed as a hazardous waste under RCRA.
                                             13

-------
                     TABLE 1. CHARACTERISTICS OF HAZARDOUS WASTES

                                          Ignitability
Any liquid with a flash point of less than 140°F.
Any nonliquid capable of causing fire through friction, absorption of moisture, or spontaneous chemical
changes that, when ignited, bums so vigorously and persistently that it creates a hazard.
Any ignftabte compressed gas or oxidizer (see DoT transportation regulation 49 CFR 173.30 or 173.151).
                                          Corrosivity
Any aqueous liquid with a pH less than or equal to 2.0 or greater than or equal to 12.5.
Any liquid that corrodes steel at a rate > 1/4 in./yr.
                                                                            \
                                          Reactivity
Any waste that is normally unstable and readily undergoes violent changes with detonations.
Any waste that reacts violently with water.
Any waste that forms potentially explosive mixtures with water.
Any waste that, when mixed with water, generates toxic gases, vapors, or fumes in a quantity sufficient
to present a danger to human health or the environment.
Any waste capable of detonating or undergoing an explosive reaction if it is subjected to a strong initiating
sources or heated under confinement.
Any material readily capable of detonation or explosive decomposition or reaction at standard temjjerature
and pressure.
                                           Toxicity
       Any waste material whose extract contains any of the contaminants at concentrations equal to or
greater than the values given in Table 2. (The extract consists of the liquid component of a solid waste
and deionized water at pH 5.0 that has been continuously brought into contact with the solid phase of
the waste for a 24-hour period.)
                                              14

-------
       TABLE Z CONTAMINANTS IN EXTRACT
Contaminant
Max. Concentration
(mg/L)
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
Endrin
Lindan
Methoxychlor
       4.0
     100.0
       1.0
       5.0
       5.0
       0.2
       1.0
       5.0
       0.02
       0.4
      10.0
Toxaphene (C10H10CI8, Technical      0.5
Chlorinated Camphene, 67-69% Chlorine)
2,4-D (2,4-Dichlorophenoxyacetic
Acid)                             10.0
Silvex (2,4-5-Trichlorophenoxypropionic
Acid)                              1.0
                        15

-------
       The act specifies that a hazardous waste management program is to be based on the "cradle to
grave" concept, so that all hazardous wastes can be traced and fully accounted for.  All treatment,
storage, and disposal (TSD) facilities are required to satisfy certain minimum recordkeeping requirements.
In addition to using a manifest for transporting hazardous waste, the facilities must keep records of vital
information about the waste, such as the results of waste analysis,  trial tests and inspections,  a
description and the quantity of each hazardous waste received, and methods and dates of treatment,
storage, and disposal.

       Under RCRA, the states are authorized to develop and carry out their own hazardous waste
programs in lieu of the EPA-administered federal program.  Several states have already done so.
                            •
Final Rule, Burning of Used OQ

       The EPA published its final rule related to a hazardous waste management system (burning of
waste fuel'and used oil in boilers and industrial furnaces)  in the Federal Register on November 29,1985.
These regulations were described in 40 CFR, Parts 261,264, 265,266, and 271. On January 11,1985,
the  EPA  proposed  under subtitle  C of the  RCRA to begin regulation of hazardous  waste and
off-specification used oil burned for energy recovery in boilers and industrial furnaces.  The proposal
included administration controls for marketers and burriers of hazardous waste and used oil fuels. The
final rule prohibits the marketing and burning in nonindustrial boilers of both hazardous waste fuel and
used oil that does not meet specification levels for certain hazardous contaminants and flash point. It also
provides administrative controls to record marketing and burning activities; these include notifying the EPA
of waste-as-fuel activities and using a manifest, or an  invoice system, for shipments and recordkeeping.
Also, hazardous waste fuels are subject to storage requirements.

Proposed Rule, Recycled Used OH Standards

       This proposal meets the RCRA requirement for the EPA to provide standards for used oil that is
recycled.  The standards are proposed for generators and transporters of recycled oil and owners and
operators of used oil  recycling facilities. This proposal includes tracking requirements when used oil is
shipped off-site for recycling. Recycled oil used as fuel would be subject to certain regulations, except
that fuel meeting a specification for toxic contaminants and flash point would be exempt from regulation.
Uses of recycled oil that constitute disposal would be prohibited completely. The proposal defined used
fuel specifications (applying only to used  oil that is not mixed with hazardous waste), which  are listed in
Tables.

Identification and Listing of Used OD, Proposed  Rule

       In this regulation, the EPA announced its determination that used oil being recycled should not
be listed as a hazardous waste under the Resource Conservation and Recovery Act (RCRA). The EPA
intends, however, to issue recycled oil management standards and is conducting studies to determine
whether standards are appropriate under Section 3014 of RCRA. The EPA is also conducting certain
studies to determine  whether used oil being disposed (i.e., not being  recycled) should be listed as a
RCRA hazardous waste, or whether it should be regulated instead under different statutes.
                                              16

-------
                             TABLE 3. SPECIFICATION USED OIL
                     Constitute/Property
Allowable Level
                     Arsenic

                     Cadmium

                     Chromium

                     Lead

                     Flash Point
5 ppm maximum

2 ppm maximum

10 ppm maximum

100 ppm maximum

100°F minimum
                     Total Halogens
4000 ppm maximum*
                     * Used oil containing more than 1000 ppm total halogens must
                     be shown not to have been mixed with hazardous waste.
                     This is referred to as the 'rebuttable presumption.'
WASTE OIL CATEGORIZATION
       Figure 7 offers a categorization of waste oils.  Waste oil containing more than 1,000 ppm of
halogens should not be mixed with hazardous waste, otherwise used oil is classified as a hazardous
waste.  If the energy content of hazardous waste is more than 5,000 BTU/lb., then the waste can be
burned as a hazardous waste fuel. Waste oils meeting rnetals and halogens specifications are classified
as specification  used oil fuel.

Standards Applicable to Generators, Processors, Marketers, Burners

       A detailed listing of the requirements for these various waste oil handlers is given in 40 CFR.  Part
262 deals with generators, while parts 264 and 265 deal with processors, marketers, and burners (i.e. TSD
                                             17

-------
         Waste Oil
         Sample
        >1000ppm
        Halogens
No
               Yes
        1) Inorganic
         (e.g. Nad)
2) Not Mixed * with Hazardous
          Waste
  3) Mixed * with Hazardous
          Waste
    Are Metals
   and Halogen
specifications met?
                                            I
Yes
                           No
                                      Non-Specification
                                           Fuel
         1or2
                                            i
                           OR
                     Blend to meet
                 metals specifications
                                                 Specification
                                                 Used Oil Fuel
     Hazardous Waste
           I
          Does
       The Waste
   OH In Its Generated
  Form Have A Sufficient
     BTU Content To
     Justify Its Use
       As A Fuel?
       >5000BPJ/lb
      Hazardous Waste
           Fuel
             :5000BTU/!b
     Disposed As A
    Hazardous Waste
                   " Ifiwd mean* thw* hM ton post-gcroretlon adulteration
                   el th* WMt* straw with • hazardous waste.
             Figure 7. Waste Oil Categorization Decision Tree.
                                     18

-------
facilities).   Figures  8, 9, 10,  and 11 schematically  depict the various steps to compliance with the
regulations for the different waste oil handlers in a systematic form.

Recent Development in Classification of Used Oil

       The possibility of hazardous waste classification of oil has loomed larger ever since a recent U.S.
Court of Appeals decision. The District of Columbia court overruled a 1986 EPA determination not to
classify used oil as a hazardous waste because that stigma would hurt recycling efforts.  The court
ordered the EPA to decide whether recycled oils meet technical criteria for listing as hazardous wastes
and then to treat them accordingly. The agency is now studying how to list oils on the basis of technical
requirements.

       As a result of a lawsuit, Repi Ike Sketton and 44 other U.S. House members introduced legislation
that would direct EPA to set standards for waste oil recycling and require the agency to not list used oil
as a hazardous waste.21  Federal agencies would be directed to give preference to  lubricants and fuels
made from waste oil. The second used-oil bill before Congress is the Consumer Product Recovery Act,
also known as the Torres bill.  The bill sets management standards for the handling of waste oil by
transporters and  recyclers.  It also seeks to encourage reuse of waste oil through recycling credits.
Producers of oil products would be required to ensure that a percentage of their products are recycled.
Oil  producers could meet that standard by rerefining oil into lubricants or purchasing  credits from fuel
processors or rerefiners.

       Affected by any regulation would be more than 1.2 billion gallons of waste oils generated annually.
A concern expressed by many is that hazardous waste classification will discourage service stations and
other garages from accepting oil from do-it-yourself (DIY) operators, considered by oil change firms to be
a major source of waste crankcase oil in the U.S.  The result would be improper disposal  of more oil,
rather than less.  If oil is listed as a hazardous waste, liability insurance would undoubtedly increase, and
service station operators would be discouraged from accepting oil from the general public.22 The National
Oil  Recyclers  Association (NORA) opposes a hazardous waste  classification.   Its  position is that
categorizing oil as a hazardous waste will hurt the waste oil business and cause many customers to avoid
utilizing it.

       However, members of the Hazardous Waste Treatment Consul and the Association of Petroleum
Rerefiners believe that waste oil should be regulated as a hazardous waste to ensure its proper treatment.
They feel that a regulated system discourages illegal mixing or environmentally unsound practices.

       With the recycled oil industry divided over the merits of classifying used oil as hazardous, politics
may be the deciding factor. If the Skelton bill were passed, handling standards would be raised, but the
critical label of "hazardous* would not be applied. If the bill were not passed, the EPA would then judge
used oil to be hazardous.23
                                              19

-------
                                       GENERATOR'
            Quality
             known
                           Determine
                            quality
                                           Quality
                                           unknown
                                                                           Assume
                                                                         Hazardous
                                                                           Waste
     Used OiI  Meets Specs
      for Used Oil Fuel
                       Used Oil Meets
                         Definition
                      of Oil-Spec Fuel
                                        Waste Oil is
                                       Hazardous Waste
   No Storage Requirements
    1
 Generator
   Burns
Generator/
 Burner
  Must
 Assure
 Quality
 Control
   No Storage Requirements
Transfer to
  Burner
                                            JL
                                                    Must Be Stored as a
                                                      Hazardous Waste
 Transfer to
 Other Than
  Burner
Generator ,
or Burner •
  Must
Notify of
Spec Fuel
Activity
and Assure
 Quality
 Control
Transfer  to
  Burner
,  Consult
Regulations
    On
 Intended
   Usage
  Generator
    Burns
Generator
  Must
Comply as
 Marketer
 Generator
   Must
Comply With
Appropriate
  Burner
Requirements
                                          Must obtain EPA I.D.
                                          numbers unless generates
                                          less than 100 kilograms
                                          per month.
 Transfer to
 Other Than
   Burner
 Must Comply
    With
 40 CFR 262
 Hazardous
   Waste
Requi rements;
 Including
Notification
 Manifest
 Transport
  Record-
  keeping
               Figure  8.  Compliance With Generator Regulations.
                                           20

-------
                      Receives Analyzed Oil From Marketer
                       PROCESSOR
\
|
f
Specification
Used Oil *


I
Off-Specification
Used-Oil
lOr
Blend To Used
Oil Specification



>
^
f
Hazardous Waste
i
f

Processing of Treatment
Activities/Operations
No Processing
 Regulations
                      Notify EPA
            Provide Certification of
             EPA ID To Marketers
        For
   Off-Specification
      Used Oil
                                                   Receive EPA ID
 Keep Records of
Invoice and Manifest
                  No Storage
                 Requirements
                                                                 For    '
                                                          Hazardous Waste
                       Receive
                   Treatment Permit
              Must notify if first
              to claim that the oil
              meets specifications.
                      Provide Certification Of
                       Permit & EPA ID To
                      Generator & Marketer
   Facility Must Comply
 With Storage Regulations
 in 40 CFR 264,265 & 270
RCRA Storage
Facility Permit
Requirements
                                                      Facility Must Comply
                                                  With Treatment Requirements
                                                    In 4O CFR 264, 265 & 270
             Figure 9. Compliance With Processor Regulations.

                                    21

-------
                                   Receives Waste Oil
                                    From Generator
                                     MARKETER
                                                                                      II
                                                     Hurt Know Quality of Waste Oil       |
       [
                                                         Hazardous Haste Fuel
         J
1
f


4r
1

May Blend 1
to Heat Specs |
4
Distributes
Used 01 1 Fuel
1

Must Keep Records
to Document That
Fuel Heeta Specs
i
r
No Transportation
Requirements
i

No Transaction
Requirements



i
r
Must Certify
to Burner of
Notification
Waste as Fuel
Activities





I
Must Have
EPA Storage Permit
i
Must Be Stored as
Hazardous Waste
i
Hay Not Blend For
Distribution as
Spec Fuel Blend;
Dust Be Distributed «o
Certified Industrial User
i


Distributes
Hazardous Waste Fuel
or Off-Spec
1
IHust Have Certification
of EPA ID From Burner
Off-Spec
Fuel Oil $
1

Marketer Must
Certify to Buyer
That Notification
Has Been Hade

Hazardous
'Waste fuc
Must Manifest
If Hazardous
Waste Fuel
                                               Must Invoice
                                               Manifest if
                                              Off-Spec Fuel
                          Must Check
                           Burner's
                        Storage Permit
Figure 10. Compliance With
             Marketer Regulations.
No Transportation
   Requirements
                             I
Must Be Licensed
Hazardous Waste
  Transporter
                                         22
               Must Keep
               Records of
              Transactions

-------
                                    Receives Analyzed
                                    0
-------
                                         SECTION III
                               CHARACTERISTICS OF USED OIL
       The characterization of lubricating oils plays an important role in establishing the quality,
consistency and additive response of a given oil.  Lubricating oil fractions contain the largest and most
complex hydrocarbon molecules found in petroleum crudes.  These hydrocarbons may contain from 20
to 70 or more carbon atoms per molecule, with corresponding molecular weights of 250 up to 1,000 or
more.  An idea of their complex structure can be gained from the fact that literally millions of different
molecular arrangements of carbon and hydrogen atoms are possible in hydrogen molecules of this size.

       Many investigators have studied the composition of crudes used as feedstock for lubricating oils,
and Georgi has articulately theorized on the general types  of hydrocarbon. They are constituted as
follows:

       (1)  Straight Chain Paraffins - These are saturated compounds and are essentially high mefting
point paraffin waxes which must be removed from paraffinic lubricating oil fractions by dewaxing.

       (2)  Branched Chain Paraffins - There are two types:

               (a) Long parent chains with short side chains - These are probably similar to straight
chain paraffins and have high melting or pour points.

               (b)  Parent chains with long side chains • These possess high  viscosity indices,  low
volatility, optimum stability and adequately low pour points. These are desired constituents in lubricating
oil.

       (3)  Naphthenes - There are also two types of these:

               (a)  Naphthenes having only a few rings per molecule and an abundance of paraffin
chains of considerable length; these have properties similar to 2(b) above and are desirable fractions of
lubricating oils.

               (b) Naphthenes having a larger number of rings per molecule and/or paraffin side chains
of shorter length; these have lower viscosity indices. The desirability of this fraction decreases as the
number of carbon atoms in the ring form increases in relation to those in attached paraffin chains.

       (4)   Aromatics - The unsaturated  ring structure of the alkyl-aromaties .tends to be readily
susceptible to oxidation. Aromatics also tend to have low viscosity indices.  Asphalts and resins are
                                              24

-------
highly condensed aromatics containing numerous fused rings.  Aromatics are undesirable in lubricating
oils, and refining methods are usually directed toward their removal.

        (5) Mixed Aromatic-Naphthene-Paraffin:

               (a) Types having only a few aromatic rings combined with a number of naphthene rings
and paraffin chains; these are similar, in their general properties, to the straight naphthenes as per 3(a)
and (b) above.

               (b) Types having more than 2 or 3 aromatic rings and correspondingly fewer naphthenes
rings and  paraffin chains. These are similar to straight aromatics as per (4) above.24

        In addition  to  the major constituents discussed above,  petroleum oils may contain small
proportions of olefins as well as nitrogen, oxygen and sulfur compounds.


ORIGIN OF BASE STOCKS
        Natural lubricating oils are separated from crude oils to form a group of lubricating oil base stocks.
By blending different additives  with these base stocks, finished lubricating oils are formed.  The
preparation of these base stocks from crude oil is a complex and expensive procedure, and also depends
on the nature of the crude oil.

        The first step in the production of lube oil base stocks is the vacuum distillation of crude oil. The
heavy oil resulting from vacuum distillation bottoms can be treated to remove asphaltic materials by two
major commercial processes, propane deasphalting and Duo Sol.25-26 Most modem plants use the former,
Base stocks which have been propane deasphalted are commonly referred to as bright stocks. Figure
12 shows the different processing steps for lube oil base stock formation for paraffinic base oils.  Both
solvent extraction and dewaxing are necessary when processing paraffinic oil only. The primary purpose
of these steps is to improve the viscosity-temperature characteristics, as well as the oxidative and thermal
stability of the base stocks. A finishing step is frequently required to remove trace contaminants and to
improve color and the oxidative and thermal stability of base stocks. Most refineries use hydrotreatmerrt
as a finishing step on a continuous basis.

        Naphthenic base oils are derived from special naphthenic crudes and require less processing than
paraffinic  oils. Figure 13 shows a typical process scheme used to produce naphthenic base oils. The
principal advantages of naphthenic base oils are their low production costs and their naturally occurring
low pour points, while their disadvantages include poor oxidative stability and rapid increases in viscosities
at low temperature.

        Naphthenic oils usually can be identified by their  low API gravity and their low viscosity index.
Naphthenic base oil can be used to make motor oil, but the majority is used in formulating rubber process
oils, refrigeration oils, shock absorber fluids, etc. The production of paraffinic stocks is expected to
increase, while the production of naphthenic oils is declining. By understanding what oils use naphthenic
and paraffinic base oil, a rerefiner can control his feedstock to ensure a high quality finished product.

                                              25

-------
Atmospheric Distillation
           Gas
                . ___^. Gasoline
                . — —   »  Solvents
                                            LUBE OIL BASE STOCK FORMATION
                                             (for paraffinic base oils)

Atmospheri
Still

Crude on



c
*






^


..
•••
Ml •
— — ^- Naphtha





. — —.-. — ^ Kerosene
._____*. Gas Oil












_ fl Va
n / LUI
i
u ;z:

Propane


Residiun

I
Deasphaltin
tower



j

-*
9
1.





r-
i
1
-11


i
cuum Still
be Distillates
~"1§T
""I}

deasphalting



_-d
stripper

Bright stock
Oil

i —
|> Propane f
j| Stripper
~V s

4
Asphalt
Hydrogen finishing


i



I
1
1




f
1
i
L
4
	 j
r
urfural |
tract
ripper
1
^
*-





A



V
1
J

Extract


Furfural


I
•-*
Oountercurrent
extractor
Raffinate
stripper
Solvent
dewaxing
j Solvent
| Mixer
1



1
»f



1
1
1
i
\
\
	 J A
1 Hydrogen
Hydrogen 1 "\
>*x
1
i


Reactor
4— ••

eparator
i
i
j


1
jg
^


y
• mm



.
1
1
1
\
m
V
|



f\
i tniij-eripi
(Solvent stripper 121
"H Filters
i
( -__/^\< 	 	 	 ,_J
t—


L[3*~- — •*
Comoressor
1
i


        4
Finished lube oil base stock
 to blending
                                               Wax
                                           Solvent stripper
         Figure 12. Production of Paraffinic Base Oils.
                               26

-------
O
 O  » 9
 C  *> C •-«
•H  C O -H
iH  o> a O
 §58.
3S23
iHi
n
o
s
CO
                                                                   m
                                                                   _o
                                                                   Q.
                                                                   (0
                                                                   O
                                                                   |5
                                                                   T>

                                                                   £
                                                                   CO

                                                                   I
                                                                   O)
                                                                   iZ
                           j
                          27

-------
Lubricant Additives (27,28,29)

       Additives are used for three main reasons: (1) to protect the surface being lubricated; (2) to
Improve the performance of the lubricant; and (3) to preserve the lubricant.

       An understanding of additives will help the rerefiner to improve processing and make better
reformulated lubricants. The characteristics and purpose of different additives used in automotive engine
lubricants are described below.

Antiwear and Extreme Pressure Agents

       Purpose: To reduce friction and prevent wear and metal-to- metal contact.

       Action: A chemical reaction takes place with the additive on the metal surface, which forms a film
and separates the two metal surfaces.

       Typical Composition: Zinc dithiophosphates, phosphates, and organic sulfur.

       Zinc dithiophosphates (ZDP) additives provide both antiwear and antirust protection and are used
in a wide  range of products, including engine oils, hydraulic oils, gear lubricants and many others.  ZDP
are intended to form films of zinc sutfide on moving parts, and so it is reasonable to expect that zinc
sulfide will occur in the oil.  It also provides special problems for the rerefiner. When heated above their
decomposition point (approximately 300°F), toxic sulfur compounds are emitted.

Corrosion and Rust Inhibitors

        Purpose: To prevent corrosion of metal parts.

        Action: Chemical adsorption of polar functional groups.  Acids are also neutralized.

        Typical Composition: Zinc dithiophosphates, basic sulfonates, metal phenolates, fatty acids and
amines.

Detergent and Dispersants

        Purpose: To disperse insoluble contaminants and control deposits at high temperatures.

        Action: Sludge and varnish precursors are neutralized and kept in suspension.

        Typical composition: Barium, calcium  and magnesium phenolates, organic phosphates, and
sulfates are used as detergents, while polymeric alkylthiophosphates are  considered dispersants.

        The detergent/dispersant  package and the  dirt, water, wear metals and lead salts that are
suspended in waste oil present the greatest problem for rerefiners.
                                               28

-------
Antioxidants

        Purpose: To inhibit oxidation of oil and additives.

        Action: Free radical termination and peroxide decomposition.

        Typical Composition: Zinc dithiophosphates, aromatic amines,
hindered and sulfurized phenols.

Metal Deactivators and Rust Inhibitors

        Purpose: To reduce oxidation of lubricants and rust and corrosion of metal parts.

        Action: Deactivate the catalytic effect of such metals as copper, lead and iron by forming a film
on the particle or surface.

        Typical Composition: Weak organic acids, organic complexes of amines, sulfides and phosphites.

        The additives for the viscosity index improver and pour point depressants do not contain metals
and should not create any problems for the rerefiners, as they can be left in the rerefined oil. The most
commonly used antifoam additives by far are the silicon compounds. These silicon polymers are very
stable and, even if they are not removed from the oil, they improve the antifoam properties of a rerefined
oil.

Industrial Lubricants

        Industrial lubricants include a wide range of special products for specific purposes.  Common
uses include: turbine oils, compressor oils, industrial gear lubes, transformer oils, and metalworking fluids.
Table 4 shows the breakdown of industrial oil sales for 1975. Many of the additives discussed previously
are used in industrial oils along with a special purpose additive, or in lesser amounts to formulate a
special  lubricant.

        The oils  used most frequently in industry  are hydraulic oils, followed by metal working  oils,
quenching oils, gear oils and greases, and such  specialty oils as compressor, air conditioner,  and
transformer oils.

        Most of the hydraulic oil used is the petroleum-based type, although synthetic nonflammable ester,
water-glycol, and water-in-oil emulsion products are used in special operations. Metalworking fluids range
from low viscosity  oils to high viscosity oils with fatty oils, sulfur, chlorine,  and phosphorus additives,
among others. Emulsifiable petroleum oils, often with extreme pressure and antiweld additives, are widely
used.

        Used industrial oils also often contain lower flash point solvents such as kerosene, Stoddard
solvents (naptha) and chlorinated solvents such as trichloroethylene. Industrial oils seem to present fewer
problems in fuel use than used automotive oils. For example, they have a lower ash content.  However,
                                              29

-------
        TABLE 4. INDUSTRIAL OIL SALES FOR 1975
Hydraulic oils
Metalworking oils
Railroad engine oils
Aviation and other oils
Gas engine oils
Process oils
Electrical oils
Refrigeration oils
(Million Gallons)
      314
      145
       58
      145
       58
      340
       62
       11
Total
     1133
                            30

-------
there are other constituents which may also be undesirable from an operational standpoint, such as sulfur,
chlorine, and fatty oils.

Mechanism of Oil Oxidation

        All hydrocarbon  oils will react with  oxygen  upon  exposure to  air at sufficiently elevated
temperatures for sufficiently long periods of time. The internal combustion engine is an ideal oxidizing
machine  since the motor oil is violently agitated in the presence of air, frequently at quite elevated
temperatures, and for lengthy periods of time between the customary crankcase drain intervals. Over the
range of temperatures developed in engines, the rate of oil oxidation approximately doubles for each 20°F
rise of temperature.   In addition, metals act as powerful oxidation  catalysts or accelerators, with iron,
copper and lead being particularly active. The rate of motor oil oxidation may be increased accordingly,
as much as 100-fold at any given temperature, due to exposure to engine metal surfaces, metal particles
resulting from normal engine wear, and  contamination with  combustion chamber blow-by solids and
airborne dust.

        Although all lubricating oil hydrocarbons are susceptible to oxidation under adverse conditions,
little is  known about the oxidation products. It is probable that the first oxidation products formed are
organic peroxides, which act as catalysts for further oxidation of lubricating oil components.  Peroxides
are also vigorously corrosive to engine bearings.30'31'32

        Paraffinic hydrocarbons apparently combine with oxygen at the carbon atoms near the end of the
chains, and the mechanism of the oxidation has been suggested as:

        (1)  paraffinic hydrocarbons—> primary alcohols—> aldehydes—> acids.        ^
        (2)  paraffinic hydrocarbons—> secondary alcohols—> ketones—> ketonic acids. •

        Thus, oxidation of paraffinic hydrocarbons tends to yield primarily acids, corrosive products, and
secondary viscous compounds of complex composition. The acids formed during the  service life of
lubricating oil oxidize the lead, copper and cadmium bearings. These metal oxides are soluble in the
acids formed by oxidation of lubricating oil.

        Naphthenic hydrocarbons oxidize and yield oxy-products in a manner similar to paraffins, and
oxidation end products are largely of the oil-soluble type.  The aromatic hydrocarbon constituents  of
lubricating oils tend to be the most readily oxidized, possibly because of the sensitivity of  the hydrocarbon
atoms  attached to carbon atoms in side chains adjacent to the aromatic nuclei. The end products are
very complex condensation and polymerization products and tend to be insoluble  in the oil; these
products constitute the sludges, resins and varnishes.  In particular, detergent/ dispersant additives are
 used to disperse oxidative products and other contaminants, and to keep them in suspension.  These
 impurities clog the filter media and slow down the filtration rate considerably if filtration is a part of the
waste oil recycling process.

 Metals in Used Oil

        Metals in used oil generally are categorized according to their sources: (1) naturally occurring
 metals; (2) metals incorporated in additives; and (3) metals introduced during use.

                                               31

-------
 Naturally Occurring Metals

        Crude petroleum contains metals in low concentrations, principally zinc, iron, nickel, and
 vanadium.35 Since these metals can interfere with catalytic refining processes, they are removed during
 the refining operations.  Consequently, these metals are present at negligible levels in lubricating oil base
 stock.                                                                            •

 Metals Incorporated in Additives

        The purposes and actions of different additives in lubricating oil have been discussed previously
 in this chapter.  Metals such as K, Na, Ca, Ba, Mg, Zn, P, and Si are due to additive compounds.

        The additives are surface active compounds and are, in their most common form, constituted of
 a hydrocarbon portion and a polar or ionic portion.  The hydrocarbon portion, which may be linear or
 branched, interacts very weakly with the water molecules.  However, the polar or ionic portion of the
 molecule, usually termed the head-group, interacts strongly with the water via dipole-dipole or ion-dipole
 interactions and is solvated. Consequently, the head-group (metals) is said to be hydrophilic.

 Metals Introduced During  Use

        Metals are introduced in the oils due to wear and tear, and due to corrosion of machine parts
 during the service life of lubricating oils. As previously discussed, oxidation products of lubricating oils
 are highly corrosive, and these oxidation products oxidize the lead, copper and aluminum bearings when
 they come in contact with lubricating oil.  As a result of oxidation these metal oxides are soluble in the
 acids formed. Both metallic iron and iron oxides are formed as wear debris from ferrous metals, so it is
 likely that at least some of the iron in waste oil is present as:

                                        Fe2O3 and Fe3O4

 Chemical and Physical Nature of Metals in Oil

        Information about the nature of metal-containing species in waste oil is inferential, primitive, and
fragmentary. Knowledge of the chemical and physical forms in which metals occur in oil is important in
at least two ways.  First, the cost and effectiveness of recycling processes in removing metals from used
oils depends to some extent on the specific form of the metals in the oil.  Second, the precise form of
metals in  used oil will be significant from the viewpoint of impact on the environment.

       Because very little is known about the  chemical nature of metal-containing species in used oil,
it is necessary to draw inferences from the fragmentary data that do exist.  Data presented by Belton R.
Williams36 from his rerefining operations have proven useful. In these operations, water is separated from
oil prior to filtration and distillation. This aqueous phase contains large amounts of Na, Zn, Ba, Ca, Fe,
P, Mg, B, and Pb.  These data suggest that the metals mentioned occur to a significant extent  in
water-soluble forms.
                                              32

-------
      '  In another study dealing with the toxicity of waste oil to marine life, the oil was extracted with water
 (9:1 ratio for 18 hours). Large amounts of lead and zinc were found in the water phase. Filtration studies
 indicated that more than 90% of the lead, cadmium, and zinc in the water passed through a filter with a
 pore size of 0.05 micrometers or less in diameter.37

        Vacuum distillation of dehydrated crankcase  drainings results in a product with a low metal
 content. The data indicate that some lead and phosphorus is present as volatile species and that other
 metals occur in forms with little or no volatility. The lead in the distillate could either be entrained or
 present as tetra-ethyl lead. The latter possibility appears unlikely, since tetraethyl lead is decomposed
 by heating with aqueous acids, organic acids, and phenols.38

        The use of physical techniques such as settling, centrifugation and ultrafiltration to reduce metal
 levels have been studied, and results indicate that significant quantities of metals, specifically about 50%
 of the lead, occur in particulates.39'40

        Fractionation and infrared analysis of an SAE 30 additive-containing lubricating oil indicated that,
 after use in an automobile over 3500 Km, calcium dialkylnapthalene sulfonates were converted in part to
 inorganic calcium compounds and organic products. The filtration and centrifugation studies indicate thai:
 many metals, notably lead, zinc, calcium, barium, magnesium, and iron are present in used oil mainly as;
 small-sized particles. Another obvious form  in which metals may occur is as original  additives.  Zinc
 dialkylthiophosphates are  intended to form films of zinc sulfide on moving parts, so  it is reasonable to
 expect that zinc sulfide will occur in the oil.  Alkali earth metal sulfonates may hydrolyze slowly to produce
 hydroxides of calcium, magnesium, and barium.41

       Table 5 shows the metal concentrations in used motor oil. This motor oil was collected from the
 service stations in the vicinity of Auburn, Alabama  The most interesting result was the lead content of
 used motor oil; lead content of used motor oil is declining in general, and was found to be between 100
to 200 ppm, compared with more than 1,000 ppm in the early part of the 1980s.

       In conclusion, information on the chemical nature of metallic species  in used oil is very scant. A
 definite need exists for research to obtain more succinct characterization of the physical and chemical
forms in which metals occur in used oil.
                                              33

-------
TABLE 5. ELEMENTAL ANALYSIS OF METALS IN USED MOTOR OIL
Metal
(ppm)
K
Ca
Mg
Ba
Zn
P
Mn
Cu
Concentration
120
1570
900
150
1800
1375
9
350
Metal
(ppm)
Fe
Pb
Al
Co
Mo
Cr
Si
B
Concentration
225
150
25
0.5
11
3
35
5
                         34

-------
                                        SECTION IV
       Waste oil, although contaminated, has a high energy value, and burning it as a fuel is a major
outlet for used oil. Unfortunately, used oil contains high concentrations of metallic contaminants, carbon
particles, and other oxidation products. Burning the oil without removing the contaminants can cause
adverse environmental effects.

       The  major objective of the  bench-scale  studies was to develop a process  to convert
off-specification used oil to specification grade fuel oil;  hence, most experimental work was directed
toward reducing the ash and lead content of the used oil.  The experimental work was divided into two
parts: (1)  demetallization of  waste oil  by chemical  reagents (metal borohydrides and diammonium
phosphate),  and (2)  removal  of  oxidation products,  coking and fouling percussors,  and  acids
neutralization by alkonal amines.

       A detailed parametric study for demetallization reaction was done in a tubing bomb micro-reactor
and a distillation flask. Experiments were also extended to a larger batch reactor, i.e. a 3785 ml autoclave,
to simulate actual operating conditions as much as possible, while most of the experiments in the second
part of the project were carried out in a 1000 ml three-neck flask. The experimental setups are discussed
in detail below.


TUBING BOMB MICRO-REACTOR (TBMR)


       It is constructed from 316 stainless steel tubing, with 3/4-inch (1.905 cm) O.D., 0.065 inch (0.165
cm) wall thickness and a length of 9 inches (22.86 cm), as shown in Figure 14.  One end of the reactor
was sealed with a Swagelok cap. Gas  (if used) was introduced through a connection unit consisting of
a Nupro fine metering valve, 1/4 inch (0.635 cm) O.D. 316 stainless steel tubing having a length of 2.75
inches (6.985 cm), and Swagelok reducing  unions.

       In a typical run, the TBMR  was charged with 20 gms of waste oil and a specified amount of
reagents.  The reactor was then sealed and checked for leakage.  After charging, the TBMR was attached
to the vertical shaft on the agitation assembly and agitated at ambient temperature, for 3 minutes at 860
rpm. The TBMR was immersed in the preheated fluidized sand bath (Techne Inc., SBL-20)  as shown in
Figure 15. A Techne TC4D temperature controller was used to maintain the temperature to within 2°C.
The heating time was about 1 minute.  At the end of the chosen reaction time, the TBMR was removed
from the sand bath, immediately quenched  in tap water, and checked for any leaks. The liquid product
                                             35

-------
                           3/4" 0.0. SS Tube,
                           8" Long
                         Swage!ok Cap (3/4")
Figure 14. Drawing of Tubing Bomb Microreactor.
                     36

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

             TACHOMETER
            AGITATOR SHAFT
            MICROREACTOR (TUBING BOMB)
                        FLUIO1ZED SAND BATH
                    TEMPERATURE CONTROLLER
                        _ X
Figure 15. Agitation Assembly for TBMR.
                 37

-------
was then collected and filtered via a vacuum filtration unit as shown in Figure 16. The filtrate (or so-called
product oil) was collected for ash and lead analysis.


DISTILLATION FLASK REACTOR


       A 300 ml distillation flask was connected with a condenser via a distilling head.  The entire
assembly is shown in Figure 17. In a typical run, the flask was charged with 60 gms of waste oil and
demetallizing reagents.  The oil was agitated and heated by a magnetic stirrer and a heating mantle,
respectively.  Heating was controlled by a power transformer.  Water and light ends evaporated during
the reaction were condensed and collected in a flask. After the reaction, the oil as filtered immediately
by a vacuum filtration system (Figure 16).  The filtrate was collected for lead and ash analysis.


AUTOCLAVE


       The reactor used was a 1-gallon autoclave built by Autoclave Engineers Inc. A cross-sectional
view of tiro reactor and a schematic diagram of the reaction system are provided in Figures 18 and 19,
respectively.  The autoclave is equipped with a turbine agitator, a cooling coil, and a thermowell made
of 316 stainless steel. It also features ports for a gas outlet and inlet, a pressure gauge tap, and a rupture
disc safety device. All feed and exhaust lines are 1/4-inch O.D. 316 stainless steel tubing. The autoclave
was heated by an Autoclave Engineers 3.5 KW electric furnace, and reaction temperature was controlled
by a proportional-integral controller (model 523C Barber Coleman).  The controller has a dual input from
the specimen thermocouple in the reactor and the furnace.  Both thermocouples were type-k (i.e.
chromel-alumel). The reaction temperature was recorded on a Fisher Recordall series 5000 and displayed
simultaneously on an Omega model 199 digital thermometer. Agitation was provided by a magnedrive
stirrer with a 3/4 HP Impak V.S. drive system from the Reliance Electric Company.
THREE-NECK FLASK
       Reactions were carried out in a 1,000 ml three-neck flask as shown in Figure 20. In a typical run,
the reactor was charged with 400 gm of waste oil and desired amounts of chemicals.  The reaction
mixture was heated and agitated by  a heating mantle and a motor-driven agitator.   Reaction was
maintained at the desired temperature by a power transformer.  After the reaction, oil was decanted into
a 500 ml graduated cylinder and allowed to settle in a constant temperature oven. The clarified oil above
the sludge was analyzed for ash and lead content.
                                              38

-------
            -BUCHNER FUNNEL
                                        'VACUUM CONTROL
                                        VALVE
                                 DEWAR FLASK
-VACUUM FILTRATION
 FLASK
        Figure 16. Diagram of Filtration Apparatus.
                          39

-------
                                                   "C

                                                   .i
                                                   i
                                                   1
                                                   b
                                                   D)
                                                   Q

                                                   h-'



                                                   §
                                                   O>
                                                   LL.
UJ
2
O
2
cc
UJ
UJ
       cr
       UJ
          40

-------
        TO PRESSURE £
        GAUGE      •r
TO HYDROGEN
  CYLINDER
                                                                H
VENTING
VALVE
                                                             COOLING COIL
                                                             THERMOWELL
                                                             STIRRER
               Figure 18. Cross-Sectional View of Autoclave.
                                    41

-------
lal

!
Sj lit
si
K *•
o- a£
                  z u
                              i O
42

-------
         Motor
                                   fyr«x Thr««-N«ck Flask
Thermometer
                 Heating Hand*
               Figure 20. Three-Neck Flask Reactor.
                                 43

-------
MATERIALS
       Used oil was collected from the Auburn Waste Oil Reprocessing Laboratory and from various
service stations in the Auburn area
ANALYSIS
       Experimental results were verified in accordance with an EPA approved quality assurance project
plan.  The properties  of the demetallized used oil were evaluated to assess its potential as an EPA
specification-grade fuel oil by 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 by standard atomic absorption analysis, and other metals content by (CAP (Inductively Coupled
Argon Plasma) spectroscopy.
                                             44

-------
                                         SECTION V
               USE OF METALLIC BOROHYDRIDE AS A DEMETALUZING REAGENT


       Process stream purification (PSP) with sodium foorohydride (SBH) is extensively used within the
chemical processing industry to improve product quality and to lower operating costs (42).  SBH can
effectively control color, odor and product stability problems, most recently in natural fats and oils. There
are three major impurities which can cause quality problems in organic chemicals: carbonyls, peroxides
and metal compounds.  Although these impurities or contaminants are normally present only at the
parts-per-million level, they are costly and difficult to remove by traditional purification methods such as
distillation, hydrogenation, absorption or oxidation. Since these compounds are present in used oils, SBH
may offer the rerefiners an effective, low-cost, highly efficient alternative.  It is a versatile reducing agent
and ligand for inorganic reactions. Reduction of toxic or valuable heavy metals in process waste streams
is an important industrial application for sodium borohydride; the substance can reduce metal compounds
to a lower valence state or to the metal's elemental form.

       Metallic borohydrides (e.g. sodium borohydride) are strong  reducing agents.   The following
reaction is typical of the metal reduction  that occurs with sodium borohydride (43).
8MX
NaBH4 + 2H2O = 8M° + NaBO2
Where M = Metal (valence 1+)
          X - Anion
                                                               8HX
       In this chemical reaction, stoichiometrically, SBH has eight reducing equivalents per molecule,
Based on some bench-scale experimental results, Morton Thiokol Inc. reported that stabilized water
solution (an aqueous solution of 12 wt.% SBH and 40 wt.% sodium hydroxide) was an effective agent for
removing lead from used crankcase oil. It should be noted that SBH degrades at low pH in the presence
of water or acid, via hydrolysis, to form salt and to liberate hydrogen.  The  representative chemical
reactions are
       Water hydrolysis:
NaBH4 + 2H2O = NaBO2
                                                         4H
                                       Acid hydrolysis:
                                 NaBH4 + 4H+ = NaBO., + 4H2

       Hydrogen provides a source of hydride ion H:" for NaBH4. This hydride ion acts as a nucleophile,
so most of the reduction reactions occur by nucleophilic addition reaction mechanism.  NaBH4 reacts
                                             45

-------
when it comes in contact with a proton source (such as water or an acid); hydride ion unites with a proton
to form hydrogen gas
H+ +
                                              H2 (gas)
       A parametric study was done to investigate the lead reduction efficiency of two Morton Thiokoi
products: Venpure powders (NaBH4 and KBHJ and stabilized water solution (SWS). The study was done
in a tubing bomb microreactor (TBMR) at the reaction conditions listed in Table 6. 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 gms of waste oil,  the specified amounts of 50
wt.% NaOH aqueous solution, and Venpure product.   At the end  of the  chosen reaction time, liquid
product was collected and vacuum filtered.  The filtrate (product oil) was collected for ash and lead
analysis.
RESULTS AND DISCUSSION
       The lead content of the type B oil was markedly reduced from 480 to 70 pprn (Run 1), as shown
in Table 7, with 0.65 wt.% SWS and 1.2 wt.% NaOH solutions.  This is well within the allowable maximum
lead content of 100 ppm for specification grade used oil. However, high sludge production (16 wt.%)
resulted from this run.

       To evaluate the effect of SWS and NaOH solution on the demetallization of waste oil, the charge
of SWS and NaOH was reduced (2/3 of Run 1) to 0.43 and 0.8 wt.% in Run 2. The result was that the
lead content of the oil decreased by only about 50%, to 240 ppm. Ash content of the product oil was 0.58
wt.%, while sludge production was about the same as that for Run 1.

       It was thought that the large amount of sludge production was due to sapon'rfication of fatty acids
or due to a  polymerization reaction promoted by NaOH in the oil.  In order to reduce  the sludge
production, only 0.05 wt.% NaBH4 powder was  used in Run 3. The amount of lead reduction was only
10.4%; however, less sludge was produced (2 wt.%) compared to previous runs. It can be  concluded
         TABLE 6. TBMR REACTION CONDITION FOR VENPURE PRODUCT TREATMENT
                            Reaction Volume             45 mi
                            Reaction Temperature        110°C
                            Reaction Time               2 hrs.
                            Agitation Rate               860 cpm
                                            46

-------
TABLE 7. EFFECT OF METAL BOROHYDR1DE ON LEAD REDUCTION
                       FOR TYPE BOIL
Run No.
Oil charge,
50% NaOH

gms
added
1
20

2
20

3
20

4
20

5
20

6
20

  to the oil, wt.%       1.2    0.8    0

  SWS solution, wt.%   0.65   0.43   0

  NaBH4 powder, wt.% 0      0

  Phase Transfer
  catalyst, wt.%       0      0


  After reaction:

  Sludge Production,
                    0.8    0      0.8

                    0      0      0

             0.05   0.05   0.05   0.05
                           0.1
                           0.1
  wt.%
16
15
  Product Oil Analysis:

  Ash content, Wt.%    0.36   0.58   -      0.77   0.40   0.53

  Lead content, PPM   70    240   430    300    400    289

  Lead reduction, %    85.4   50    10.4   37.5   16.7   39.8
  As received lead content of type B oil = 480 PPM and
  ash content = 0.52%
                              47

-------
from the results that lead reduction depends on NaOH rather than NaBH4. Since most of the lead is
present in a paniculate form, it may be removed due to an entrainment of these particulates in the sludge.
It seems that lead removal from the oil is a physical  phenomenon rather than due to chemical (reaction
with NaBH4.

        In order to enhance the demetallization activity, 0.8 wt.% NaOH (50 wt.% solution) was added in
Run 4, while other reaction conditions were kept the same as in Run 3. The lead removal efficiency did
increase from 10.4% for Run 3 to 37.5% for Run 4.  Hence, addition of NaOH did reduce the lead content
of used oil.

        To  improve the  solubility of NaBH4  in oil  phase, 0.10 wt.%  of  phase transfer catalyst
(tri-n-butyl-methyl ammonium chloride) was added to the oil (Run 5).  Comparing the results with those
of Run 3, the addition of the phase transfer catalyst did result in an increase in lead removal efficiency
from 10.4 to 16.7%, while addition of 0.8 wt.% NaOH in Run 6 increased the lead reduction efficiency to
39.8%.  To  determine the reproducibility of the demetallization experiments, some of the runs were
duplicated, and the results obtained were essentially the same.

        Waste oil is usually collected from various sources, and it is possible that lead removal efficiency
using a demetallizing reagent might respond differently to various types of waste oil.  Thus, a different oil
(Type A) was used for the demetallization studies. The lead content of Type A was higher than for Type
B oil, so the amount of metal borohydride and NaOH required for demetallizing Type A should be more
than required for Type B.

        In Run 8, reaction conditions similar to Run  1 were used to study the effectiveness of an equal
amount of caustic and SWS solution in demetallizing the two types of oil.  Table 8 shows that with the
same charge of caustic and borohydride, the lead content of Type B oil was reduced by 85%, whereas
only 24% reduction was achieved for Type A.

        Because of the 24% lead reduction obtained in Run 8, the charge of SWS solution was doubled
in Run 9 to observe its effectiveness while the reaction conditions of Run 8 were maintained (Table 9).
The lead content of type A oil was reduced by only 58%.  However, the high concentration  of sodium
borohydride reduces the economic viability of using SWS solution in demetallizing waste oils.

        In Run 10, the charge of NaBH4 and NaOH was the same as in Run 9, but they were added
separately. The lead reduction efficiency was more or less the same for both runs, which shows that lead
reduction is more dependent on NaOH than on NaBH4. The addition of phase transfer catalyst increased
the lead reduction efficiency, but the lead content of the product oil was higher than lOO.ppm (Run 11).
When potassium borohydride (KBHJ was used in Run 12, the lead content was 270 ppm; hence, the lead
content of Type A oil was not successfully reduced  to an acceptable limit.

Effect of Reaction Temperature

        In all the previous runs, the reaction temperature was 110°C. When the reaction temperature was
increased to 150°C, the entire  product oil became  a gelatinous mass (Table 10).  Hence, increase in
reaction temperature had an adverse effect on the recovery of oil and demetallization reactions.
                                             48

-------
TABLE 8. EFFECT OF OIL TYPE ON LEAD REDUCTION EFFICIENCY
Run
No.
1
8
Type of
• Oil
B
A
Lead content
PPM
70
500
%Ash
content
0.36
0.55
%lead
reduction
85.4 N
24.2
Oil charge = 20 gms
50% NaOH charge = 1.2 wt.% of the oil
SWS solution charge = 0.65 wt.% of the oil
Lead content of the type A oil = 660 ppm
Lead content of the type B oil = 480 ppm
                            49

-------
TABLE 9. EFFECT OF METAL BOROHYDRIDE ON LEAD REDUCTION
         FOR TYPE A OIL
  Run No.            8      9
  Oil charge          20     20
  50% NaOH, wt.%    1.2     1.2
  SWS solution, wt.%  0.65    1.3
  KBH4 powder, wt.%  0      0
  NaBH4 powder, wt.% 0      0
  Phase transfer
  catalyst, wt.%       0      0
10     11      12
20     20     20
1.8    1.8     1.8
000
0      0      0.15
0.15   0.15    0

0      0.10    0.35
  Product oil analysis:
  Ash content, wt.%   -     0.62   0.56   0.50   0.68
  Lead content, PPM  500   280   300    180   270
  Lead reduction, %   24.2   57.6   54.5   72.7   59.1

  Lead content of the type A oil = 660 ppm and ash content = 0.59%
                             50

-------
TABLE 10. EFFECT OF REACTION TEMPERATURE ON SLUDGE FORMATION
Run No.
Oil Type
Reaction Temp., °C
NaOH wt.%
SWS wt.%
1
B
110
1.2
0.65
13
B
150
1.2
0.65
14
B
150
1.2
0.65
     After Reaction:
     Sludge Production, wt.%    15.5
                         (gelatinous mass)
     Product Oil Analysis:
     Lead Content, ppm        70
     Lead Reduction, %        85.4
100%
100%
                                51

-------
 Investigation of Gel Formation

        Several experiments were conducted in which a cottonseed oil and a non-detergent virgin motor
 oil was reacted with either SWS solution or NaBH4 powder to confirm that saponrfication does occur when
 fatty acids are present.                                                          !

        A cottonseed oil was blended with 1.2 wt.% NaOH (50% solution) and then reacted with 0.65 wt.%
 SWS in a TBMR (Run 15).  The reaction was conducted under the same reaction conditions as given in
 Table 6. The oil was collected in a test tube after the reaction to measure a volume of the sludge. The
 sludge produced was about 35 volume % (Table 11), and was probably due to the salt resulting from the
 saponrfication of fatty acids present in the cottonseed oil. The cottonseed oil was reacted with 0.05 wt.%
 NaBH4 powder in the absence of NaOH (Run 16). In this case, no sludge was formed, confirming that
 the sludge was due to the reaction of fatty acids  with NaOH. The amount of NaBH4 was doubled;
 however, there still was no sludge formed (Run 17).  The effect of NaOH on nondetergent motor oil was
 observed by reacting a SAE-30 nondetergent motor  oil with SWS (Run 18). After the reaction, two layers
 of liquid were formed. The top layer was reddish brown, while the bottom layer was bluish green in color.
 A trace amount of suspended solids was also observed.

        In summary, the gelatinous mass formed while treating used oil with SWS was probably due to
 the sapontfication of fatty acids or the polymerization reaction promoted by caustics at higher temperature.

 Effect of Caustics

        It was thought that sludge production was due to the reaction products of NaOH and used oil.
 In order to investigate the effect of NaOH on sludge production and lead reduction, several reactions were
 performed in which different amounts of NaOH were reacted with the oil without SWS or Venpure
 products. The reaction conditions were the same as those listed in Table 6.

        Results from Table 12 clearly show that sludge generation depends on the amount of NaOH
 reacted with the oil.  A higher amount of NaOH produced a large volume of sludge. One noteworthy
 aspect of these results was that the lead content of oil depends on the amount of sludge generation.
 Lead reduction is increased when a large amount of sludge is produced due to the entrapment of lead
 particulates in the reaction  products.

 Effect of Water

       As discussed earlier, the  reducing ability of metallic borohydrides can be hampered substantially
 by hydrolysis in the presence of water.  Since waste oils normally  contain 1-10% of water, the poor
 performance of metallic borohydrides in terms of ash and lead reduction can be  due to hydrolysis of
 metallic borohydrides.  To investigate the effect of water on demetallization of waste oil, a sufficient
amount of Type A oil (containing 5 vol.% water) was blended with 0.25 wt.% NaOH (50 wt.% solution).
Water, along with light ends, were then distilled by atmospheric distillation.             ;

       In Run 22, dry oil was reacted with 0.9 wt.% of SWS solution in a TBMR under reaction conditions
given  in Table 6. As shown in Table 13, the lead content of the oil was successfully reduced from 700
                                             52

-------
TABLE 11. INFLUENCE OF CAUSTICS ON FORMULATION OF GELATINOUS MASS
Run No.
Oil Type
NaOH, wt.%
SWS, wt.%
NaBH4
After Reaction:
15
Cottonseed (CS)
1.2
0.65
0

16
CS
0
0
0.05

17
CS
0
0
0.10

18
Motor Oil*
1.2
0.65
0

     Sludge Production,
     Vol.%
35
                                   trace"
      •GRC, G-100 nondetergent motor oil, SAE-30, made from Guriey
      Refining Company, Memphis, Tenn. 38101.

      Two phase were formed with the top layer in reddish brown and
      bottom layer in dark bluish green with trace amount of suspended
      particles.
                                  53

-------
TABLE 12. EFFECT OF NaOh ON SLUDGE PRODUCTION AND LEAD REDUCTION
Run No.
Oil Type
Oil Charge, gm
NaOH, wt.%
19
B
20
1.75
20
B
20
1.25
21
B
20
0.75
Before Reaction:


Lead Content, ppm  480

Ash Content, wt.%   0.52


After Reaction:
Sludge
Production, wt.%    15
Product Oil Analysis:


Ash Content, wt.%   0.5

Lead Content, ppm  80
480

0.52
12
0.55
480

0.52
0.59

330
                                 54

-------
      TABLE 13. INFLUENCE OF WATER ON LEAD REMOVAL
Run No.
Oil Type
Oil charge
50% NaOH, wt.%
SWS solution, wt.%
Phase transfer
catalyst, wt.%
9
A
20
1.2
1.3
0
22
C
20
-
0.9
0
23
C
20
-
0.4
0
24
C
20
-
0.9
0.10
25
C
20
~
0.25
0.10
Product Oil" Analysis:
  Ash content, wt.%  0.62   0.53   0.68   0.63   0.77



  Lead content, PPM  280    100   180   120    300



  Lead reduction, %  57.6   85.7   74.3   82.9   57.1










"Sludge production in all the runs was high (about 15-20 wt.%).
                             55

-------
to 100 ppm; i.e. about 86% lead reduction was obtained. This suggests that a much better demetallizing
performance can be obtained with a dry oil than with a wet oil (Type A).

       To further minimize the use of SWS solution, the charge of SWS solution was reduced to 0.4 wt.%
in Run 23. As a result, about 74% lead reduction was obtained. In Runs 24 and 25, 0.1 wt.% of a phase
transfer catalyst (tri-n-butyl-methyl ammonium chloride) was added to see if the demetallizing performance
could be improved. Compared to Run 22 (without catalyst), the addition of a catalyst did not improve the
results.

       On the whole, the presence of water in the oil has an adverse effect on the demetallization of used
oils. This can be avoided by distilling the water before the reaction.                  :

Autoclave Study

       It should be noted that the results presented so far were obtained in a 45 ml tubing bomb
microreactor (TBMR).  Knowing that these bench-scale demetallization results were to be incorporated
in the scale-up  and design of  the Auburn Waste Oil Reprocessing Laboratory, it was clear that
experiments would have to be extended to a larger batch reactor, i.e. a 3785 ml autoclave, to simulate
the operating conditions as much as possible.                                     j

       As mentioned above, water can decrease the  demetallization activity of metallic borohydrides.
Therefore, Type A oil (containing 5 vol.% water) was blended with 0.25 wt.% NaOH (50 wt.% solution) and
then distilled to a specified temperature under atmospheric pressure. The residual oil, assumed to be
moisture free, was treated with metallic borohydrides in the 1-gallon autoclave. The results are given in
Table 14 and are discussed below.

       In Run AS1, 890 gms. of  dry oil was reacted with 0.9 wt.% of SWS. For this run, only 37% lead
reduction was  obtained, compared to  96% for Run AS2.  The poor demetallizing performance was
probably due to the low distillation cutoff temperature  (129°C) for Run AS1, which resulted in a certain
amount of water being left behind in the residual oil.

       Thus, in Run AS2, the distillation cutoff temperature was further raised to 175°C. The oil and SWS
charges to the autoclave were about the same as in Run AS1.  Under the same reaction conditions, the
lead content of the oil was markedly reduced from 700 ppm to 25 ppm.  Although a promising result was
obtained in Run AS2, the amount of SWS used was considerable and uneconomical. The Run AS2 was
duplicated by Run ASS, except that the SWS charge was halved. As a result, only a 40% lead reduction
was obtained; this suggested that more SWS was required, if a better demetallizing performance was to
be achieved.
                                                                              i
       To compare the demetallizing effectiveness of  SWS solution and NaBH4 powder,; Run AS4 was
performed with the same amount of NaBH4 used in Run AS1. Coincidentally, the same level  of lead
reduction was obtained. However, with NaBH4 powder (Run AS4), the product oil had a lower ash content
and a low sludge production compared with the Run AS1.
                                             56

-------
           TABLE 14. RESULTS OF AUTOCLAVE RUNS
Run No.
AS1    AS2    AS3    AS4   ASS   AS6
Atmospheric Distillation
conditions for Type A Oil
Oil charge, gms

NaOH wt.%
(50% solution)  1.2

Distillation
cut point,°C
890    1000   1000   600    600   600
0.8
Reaction Mixture:

Residue" from
distillation, gms       890

SWS, wt.%           0.9

NaBH4 powder, wt.%   0

KBH4 powder, wt.%    0

Phase Transfer
catalyst, wt.%         0
0.8    0      0.8
129    175    175    175    150    150
900
0.91
0
0
900
0.46
0
0
545
0
0.1
0
545
0
0
0.1
545
0
0.15
0
       0
       0.2    0.2     0.2
Product Oilb Analysis:

Ash content, wt.%     0.74   0.50    0.51   0.57   0.56   0.53

Lead content, PPM    440    25     420   440   345   200

Lead reduction, %     37.1    96.5    40.0   37.1   50.7   71.4
aLead content = 700 ppm, Ash content = 0.59%
"Sludge production in all the runs was high (about 15-20 wt.%)
                              57

-------
       To study the demetallizing ability of KBH4 powder, Run ASS was made with KBH4 instead of
NaBH4.  As shown in Table 14, the lead reduction efficiency increased from 37% to 51%, butt the ash
content of the oil was not changed significantly. Since KBH4 is more expensive than NaBH4, the final Run
(AS6) was made with higher amount of NaBH4 (0.15 wt.%). It was found that lead reduction efficiency
increased from 37% to 71 %, but the ash content of the product oil did not change appreciably. 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 demetallizing method was explored in an effort to seek the most economical process.


CONCLUSIONS
                                                                 I

       (1)  Examination of the ash content of product oil reveals that ash content was not reduced by
metal borohydride treatment.

       (2)  Lead content was selectively reduced in some cases.  It seems that lead reduction depends
on NaOH rather than NaBH4.

       (3) A major disadvantage of metal borohydride as a demetallizing reagent is the high pH required
to prevent hydrolysis.                                                          •

       (4)  A large amount of sludge is generated due to the reaction of NaOH with fatty acids and
oxidation products.

       (5)  Reaction is difficult to control  because NaOH promotes polymerization reaction at  high
temperature.                                                                  \
                                            58

-------
                                        SECTION VI
             USE OF D1AMMONIUM PHOSPHATE AS A DEMETALUZAT1ON REAGENT
       Many researchers have proposed a system by which used hydrocarbon lubricating oil is contacted
with an aqueous solution containing one or more anions, such as phosphate, sulfate, chromate, etc.,
which form water-insoluble salts with metals to be removed from  used oil.44-45 In these processes, the
cation  of the aqueous solution is exchanged with the metals of the metals- containing components of
used hydrocarbon lubricating oils. In order to avoid the entry of another metal into the treated oil, which
would increase its ash content, the cation of the aqueous solution is preferably ashless, and as such can
be  hydrogen, ammonium,  hydrazine, hydroxylamine, etc. The quantity and the concentration of the
aqueous solution is of little importance as long as anions are still present in the aqueous phase after
treatment.

       The anions in the aqueous solution form salts with the metals in the oil, the equilibrium of this
reaction being dependent on the solubility of the metal salts in the aqueous solution.45 These metal salts
can be separated from the lubricating oil by distilling water from it during the reaction and then filtering
the oil  after treatment.

       Once the initial assessment of the different salts that can be used for the demetallization of used
oil is made, it was decided that ammonium salts, preferably diammonium phosphate, would be a better
reagent for removing the metals from waste lubricating oils.
MECHANISM OF DEMETALLIZATION REACTION
       Lead and other metal contaminants are present in waste oil in organometallic form. The DAP
            react with organometallic compounds to form insoluble and easily separable products. A
reaction of this type has been discussed by Miller46:

                         (RS03)aCab(OH)c(C03)d + excess (NHJ2HPO4 =
                           Cas(PO^3(OH) + NH4RSO3 + CO2 + H2O

       Based on the above reaction, the calcium overbased sulfonate detergent reacts with DAP to form
a metallic hydroxyl apatite (Cas(PO4)3OH), which is insoluble in aqueous and oil phase.  The  reaction
products remain at the interface, so the removal of water during the reaction facilitates the separation of
metal phosphates from the oil. This reaction mechanism is general for Ca, Ba, Mg, and Zn additive
                                             59

-------
compounds. In general, group II metal compounds are highly reactive with DAP, and the reaction goes
to completion even at low temperature.

Experimental Procedure

       A detailed parametric study was undertaken to map out the process variables so as to identify
the most efficient demetallization conditions for diammonium phosphate. Different ammonium salts were
used to evaluate the effectiveness of the process in terms of ash and lead reduction.  An evaluation of
solid-liquid separation techniques was also conducted to design an efficient solids removal operation for
future pilot plant studies.  Different types of reactors  were used in this investigation so that dominant
mechanisms over different operational ranges could be identified.

        In the  bench-scale experiments, three types  of reactors were used: (1)  a tubing bomb
microreactor (TBMR); (2) a 500 ml distillation flask; and (3) a 3,785 ml autoclave reactor. The TBMR was
operated under closed conditions, whereas distillation was carried out under open conditions; i.e. the flask
was open to the atmosphere for distilling water and trace amounts of light components. The quality of
product oil, in  terms  of ash  and lead  content, depends on the solid-liquid  separation techniques.
Therefore, bench-scale studies were divided into two parts: (0 solids separated by sedimentation and (ii)
solids removed by filtration.                                                      .

Solids Removed by Sedimentation                                                ;

        In a typical  experiment, TBMR was charged with 20 gms. of oil and 1.6 gms. of diammonium
phosphate (30 wt.% aqueous solution). An excess of DAP was used, so the reaction between metals and
DAP was not limited by anions (i.e.  phosphates). At the end of the reaction time, TBMR was quenched
in water and reaction  products were collected in a graduated glass tube. The oil-water emulsion was
allowed to settle for a day at 60°C to facilitate the separation  of reaction products and water from the oil.
The clarified oil was analyzed for ash and lead content.
RESULTS AND DISCUSSION
        Demetallization studies were conducted by varying parameters such as reaction time, temperature,
and agitation rate for different types of oils. A discussion of the results follows.

Effect of Reaction Time

        Several experiments were carried out to observe the effect of reaction time on lead reduction at
various reaction temperatures. The variation in lead content of the DAP treated oil (150°C) was observed
as shown in Figure 21.   From an initial value of about 210 ppm, the lead content of the product oil
decreased with increasing reaction time to a final value of 16.5 ppm.  From this figure, it can 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 from 800 to 400 cpm, the lead content of the treated oil was
                                              60

-------
250
                     ooooo 800  cpm
                     AAAAA 400  cpm
                        I—'—'—'—'—I  'r
              25        50       75
               Reaction Time (min)
100
   Figure 21. Effect of Reaction Time on Lead Reduction.
          (DAP: 8 wt.%; reaction temp.: 150°C)
                    61

-------
 greater.  An increase in agitation is thought to increase the interfacial contact area for the reaction
 between metals and aqueous reactant.

       To investigate the effect of reaction time on other types of oil, several experiments were performed
 where Type B oil and DAP were reacted at temperature of 200°C and 250°C. Table 15 shows that the
 lead content of oil decreases with an increase in reaction time.  Consequently, it can be inferred that
 greater reaction time favors the lead reduction from used oil.                         •

 Effect of Reaction Temperature

       Demetallization experiments have been carried out at different reaction temperatures to determine
 the effect on lead content of  used  oil.  As seen from  Figure 22 (Table 16), an increase in reaction
 temperature from 150°C to 400°C greatly increased the lead reduction from 63% to 96%. The lead content
 of Type B oil is also reduced significantly with an increase in the reaction temperature (Figure 23, Table
 17).  The  overall lead content of used oil depends on the reaction temperature and is reduced
 substantially at high temperature.

 Effect of Ash Content
                                                                                I
       Although lead reduction is strongly dependent on reaction temperature and time, other factors
 such as the ash content of used oil (i.e. total metal contaminants) may also affect lead reduction.

       In order to determine the effect of ash content on lead removal, experiments have been carried
 out with  different oil  samples,  each  having a different ash content.  The ash content of each oil was
 determined in accordance with a modified ASTM procedure  (see Appendix B).  As shown in Table 18
 (Figure 24), lead content is reduced  with an increase in reaction time for oil sample 1.  However, for oil
 Sample 2,  an increase in reaction time from 45 to 90  minutes showed no significant effect on lead
 removal.

       It may be noted that although the difference in lead content of both samples is only 40 ppm, the
 total amount of metal competing for DAP is not necessarily the same, as evidenced by their different ash
 contents.

       The major problem with TBMR is immediate vaporization of water during the reaction, which allows
 very little time for the reaction between metals and anions (i.e. phosphates) at the interface of the oil and
 aqueous phase. In spite of this drawback, TBMR studies were useful in providing vital information about
the demetallization reactions. To prevent the immediate vaporization of water, another type of reactor,
 i.e. a distillation flask, was used to study the demetallization mechanism. In this reactor, waste oil and
 aqueous solution of DAP were mixed and heated from a room temperature to a specified temperature.
The heat-up time was about 70 minutes, and the reaction was allowed to continue for an additional time.
 During the reaction, water and trace  amounts of light components were distilled off.

       Two experiments were initially performed in which waste oil was reacted with aqueous solution
of DAP at 150°C.  After the reaction, one batch of oil was allowed to settle at 70°C for a day while the
                                             62

-------
TABLE 15. EFFECT OF REACTION TIME ON LEAD REMOVAL
Reaction Time
Min,
15
30
60
15
30
60
Lead Content
PPM
Reaction Terno. 250°C
190
66
31
Reaction Temo. 200°C
300
263
217
%Lead
Reduction
60.4
86.2
93.5
37.5
45.2
54.8
   DAP: 8 wt.%, Lead content: 480 ppm
                       63

-------
  100

   95-E

   90^

   85-E

o  80-E
o
   65^

   60

   55
      100   150   200   250   300   350  400  450
                Reaction Temperature (C)
  Figure 22. Effect of Reaction Temperature on Lead Reduction.1
          (Time: 45 min.; 8 wt.%; Lead Content: 210 ppm)  |
                        64

-------
TABLE 16. EFFECT OF REACTION TEMPERATURE ON LEAD REDUCTION
        Reaction Temp.
            °C
Lead content     % Lead
  PPM         Reduction
150
200
300
400
77
46
15
7.5
63.3
78
92.8
96.4
        Time: 45 minutes; DAP: 8 wt.%
        Lead content of used oil was 210 ppm.
 TABLE 17. EFFECT OF REACTION TEMPERATURE ON LEAD REDUCTION
Reaction Temp.
°C
150
200
250
Lead content
PPM
325
235
30
% Lead
Reduction
32
51
93.7
        Time: 1 hr.; DAP: 8 wt.%
        Ash and Lead content of used oil was 0.52 wt.% and 480 ppm.
                             65

-------
   100
    90-
    80-
 e  70 H
 O
    60-
•o  50-
€0


J  40^1
    30-1



    20
125
                                              i  i i
             i  I i  i i  i I i  i i  i I i  i i  i I  i i r^

             150     175    200     225    250    275

                Reaction Temperature  (C)
  Figure 23. Effect of Reaction Temperature on Lead Reduction.

          (Time: 1 hr.; 8 wt.%; Lead Content: 480 ppm)
                        66

-------
   TABLE 18. EFFECT OF ASH CONTENT ON LEAD REDUCTION
Reaction Time
Min.
0
X
45
90
Lead
Oil sample 1
210

70
16.5
content, PPM
Oil sample 2
251

100
95
Temperature: 150°C, DAP: 8 wt.%
Ash content (wt.%): Oil sample 1 = 0.4755
                 Oil sample 2 = 0.8325
 TABLE 19. EFFECT OF FILTRATION ON ASH AND LEAD REDUCTION


Mode of      Ash Content   Lead Content  Lead Reduction
Separation       wt.%          ppm         %
Filtration        0.02

Sedimentation   0.4
24

350
96

47
Temperature: 150°C, Time: 1 hr., DAP: 8 wt.%
Ash and Lead content of used oil was 0.6 wt.%, 660 ppm.
                             67

-------
   300
  250
s
OH
  200 -
fl  150 -J

o
o



S  100-
    50
     0
       0
1  i

20
     40      60      80

Reaction Time (min)
100
     Figure 24. Effect of Ash Content on Lead Reduction.

              (Temperature: 150°C; DAP: 8 wt.%)
                        68

-------
other batch was immediately filtered. As shown in Table 19, the ash and lead content of the settled oil
was much higher than the filtered oil. The high-quality product oil is produced by filtration.  This may be
due to the high viscosity of used oil, which prevents the separation of fine particulates by sedimentation.
Hence, filtration  performs better than sedimentation  in the separation of reaction products (metal
phosphates) from the oil.
SOLIDS REMOVED BY FILTRATION
       The Environmental Protection Agency (EPA) has set the allowable maximum lead content in fuel
oil to be less than 100 ppm.  The purpose of the demetallization study is to find the optimum operating
conditions for lead removal which will later be used to design a demetallization unit for the Auburn
University Waste Oil Reprocessing Lab (AWORL). Therefore, several parameters were studied to maximize
the lead reduction and,  at the same time, to find the optimum reaction conditions, which  will be
economical on a larger scale.

       It is believed that more than 50% of lead is present in a paniculate form, so some part of the lead
can be removed just by filtering the used oil. To determine the possibility of lead reduction by filtration,
several batches of used oil were preheated (150°C) prior to filtration. After filtration the oil was analyzed
and  it was found that the ash and lead content of the filtered oil was not significantly changed (less than
5%)  in any of the batches. This may be so because the lead particulates present in waste oil are very
fine and pass out with the filtrate.37

       In all the previous experiments, an excess of DAP (8 wt.%) was used to demetallize the waste oils.
The  problems with using large amounts of DAP are: (0 generation of solid waste, which results in a
disposal problem; (ii) handling  of solids in the plants; (iii) clogging of the filter media; and (iv) the high
cost of the reagent.

Effect of Diammonium Phosphate on Ash and Lead Reduction

       Several experiments were performed in which waste oil was treated with different amounts of DAP.
Table 20 shows that, although a higher amount of DAP favors the demetallization reaction, high ash and
lead reduction is still obtained with a lower concentration of DAP (2 wt.%). Since the goal of this project
is to reduce the ash and lead content of used oil to below 0.1 wt.% and 100 ppm respectively, it seems
that  there is no need to produce very high quality oil, which would otherwise increase the cost of the
treatment.  It should be  noted that  the amount of DAP required for the demetallization of used oils
depends on the initial ash content of the oils.  It was found that 2 wt.% DAP was sufficient to reduce the
ash content to below 0.1  wt.% for used oil, with an initial ash content between 0.5 to 1.0 wt.%.

Effect of Water on Ash and Lead Reduction

       The reaction between DAP and metals occurs at the interface of the oil and aqueous phase. The
reaction products remain at the interfacial region unless water is removed during the reaction. It suggests
that water removal during the reaction favors the agglomeration of metallic hydroxyl apatites, which would
                                              69

-------
 TABLE 20. EFFECT OF DAP ON LEAD REDUCTION
% of DAP
w.r.t oil content
8
5
2
•\
0.5
0
%ash
ppm
0.02
0.03
0.08
0.25
0.40
0.58
Lead
Content
24
24
60
230
462
650
%lead
reduction
96
96
91
65
30
<2
Reaction Temperature: 150°C; Residence Time: 1 hr.
Ash and Lead content of used oil was 0.6 wt.%, 660 ppm.
                      70

-------
later facilitate the solid-liquid separation. Hence, when reaction temperature was less than 100°C (Table
21), the ash and lead content of filtered oil was not reduced significantly compared with the oil treated
at a higher temperature (more than 100°C). This may be due to the presence of water, which prevents
the agglomeration of metallic hydroxyl apatites.

       It was very clear that water is required initially for the reaction to occur between metals and anions
(phosphates) at the interfacial region of the oil and aqueous phase. However, the removal of water during
the latter part of the reaction facilitates the agglomeration of metallic hydroxyl apatites, which can be
separated by filtering the oil.

       The additives in the oil are group  II metal compounds. These compounds are more ionic and
reactive compared with lead compounds. Essentially the reaction between additive metals and DAP goes
to completion and is least dependent on the temperature and time.

       Lead is present in a different form in the used  oils. If lead is due to the contamination of blow-by
products from a combustion engine,  then  it may be present in a  halides form.   Organic bromine
compounds are added to gasoline to scavenge lead  as  lead bromide, and it has been suggested that
some lead dibromide may be present  in used oil.  However, X-ray emission spectroscopic analysis of
several samples of crankcase oil disclosed a lead/bromine ratio of 0.62-0.68:1.47  If all  the lead were
present as a dibromide, a 2:1 ratio would obtain. However, chlorides or mixed salts may be present. It
has been also asserted that significant quantities of metals, and particularly about 50% of the lead, occur
in particulates form.

       Lead may also be present in an oxide form due to oxidation of lead-copper bearings by organic
acids and peroxides during the service life of lubricating oils. These lead compounds (halides, oxides)
are very stable, and it is unlikely that they  react with DAP; however, they strongly interact with water via
dipole-dipole or ion-dipole interactions.

       At a workshop on recycled oil,  Belton R. Williams presented data from his rerefining operations
which are useful in partially characterizing some of the metallic  species.36  Water (which arises as a
product of fuel combustion, condensation, and seepage into storage tanks)  is separated from the oil prior
to filtration  and  distillation.  This  aqueous  phase contains a large amount  of  sodium, zinc, iron,
phosphorus, magnesium, boron,  and lead. These data  suggest that the metals mentioned occur to a
significant extent in water-soluble forms.

       Lead compounds present in the oil are too stable to react with DAP; hence, their removal is not
dependent on the chemical reaction. But due to ionic attraction or ion-dipole attraction, lead compounds
are attracted toward the aqueous phase.   So, when water is removed  during the  reaction,  lead
compounds are entrained in the agglomeration of metal hydroxyl phosphates and are removed along with
other reaction products by filtration. Since lead is not directly removed by chemical reaction with DAP,
its removal is not independent of the temperature or the residence time.
                                              71

-------
  TABLE 21. EFFECT OF WATER ON ASH AND LEAD REDUCTION
Temp.
°C

70
80
90
120
Ash
Content
wt.%
0.4
0.42,
0.36
0.15
Ash
Reduction
%ppm
33
30
40
75
Lead
Content

607
600
550
150
Lead
Reduction

8
9
17
77




i
Temperature: 150°C; Time: 1 hr.; DAP: 2 wt.%
Ash and Lead content of used oil was 0.6 wt.%, 660 ppm.
                            72

-------
Effect of Temperature and Residence Tune on Lead Reduction

       Figure 25 suggests that lead reduction is a strong function of temperature. Only.9% of lead
reduction was obtained at  80°C compared with 97% at 200°C.  Even a moderate increase in reaction
temperature (120°C) reduced the lead content very significantly. This may be due to the removal of water,
which facilitates the agglomeration of metal hydroxyl phosphates and lead compounds.

       As seen in Figure 26, increasing the residence time has a substantial effect on the removal of lead
from used oil.  When residence time was 60 minutes, more than 90% of lead reduction was obtained,
compared with 60% without any residence time at the final temperature. Hence, high reaction temperature
and time favor lead removal from the used oil.

       A DAP demetallization study was done in an autoclave to verify the results obtained so far. Table
22 shows that the final ash and lead  content of the product oil agrees with the previous results.
MECHANISM FOR LEAD AND METALS REMOVAL
       The additive metal compounds are present in waste oil in organometallic form. The additives are
surface active agents and, in the most common form are constituted of a hydrocarbon portion and a polar
or ionic portion.  The hydrocarbon portion, which can be linear or branched, interacts only very weakly
with the water molecules in an aqueous environment.  However, the polar or ionic portion of the molecule,
usually termed the head group, i.e. metals, interacts strongly with the water by dipole-dipole or ion-dipole
interactions, and is solvated.

       The organometallic additives in waste oil react with DAP in the interfacial region (Figure 27) to form
metallic hydroxyl apatite.  The additives are groups I and II metal compounds and generally are highly
reactive and more ionic in nature. These metals react with DAP, and the reaction goes to completion even
at low temperature.

       Lead is present in a form that is physically different from additive compounds in the used oil. The
lead compounds present in used oil are too stable to react with  DAP, but are attracted toward the
aqueous phase due to ionic-ionic or ionic-dipole attraction. The ionic attraction is enhanced when water
is removed during the reaction.  The  lead particles are entrained with the agglomerates of metallic
hydroxyl  phosphates and are separated by filtration.
UMfTATlONS
       The oil obtained by treating waste oil with DAP has a very low ash and lead content. However,
the major problem associated with this process is the separation of solids (reaction products) from the
oil. With filtration, good solids-removing efficiency is obtained compared with sedimentation; however,
filtration is an extremely slow and cumbersome method to operate. In general, used motor oil is highly
                                             73

-------
   100
a
O
    50-
13

05
    25-
     0
                     O
          i r
       75
100    125    150     175    200

 Reaction  Temperature (°C)
  Figure 25. Effect of Reaction Temperature on Lead Reduction.

          (DAP: 2 wt.%; Residence Time: 1 hr.; L.C.: 660 ppm)
                         74

-------
100
 50
     0
20         40         60
 Residence Time  (min)
  Figure 26. Effect of Residence Time on Lead Reduction.
          (DAP: 2 wt.%; Reaction Temp.: 150°C)
                       75

-------
Q
£s

&
SS
) =J
o

y



o
5n
o£
s§
83
OH-


                       to

                      ;'s
                      •5
                       0)
                      T3
                      0)
                      (0
                      O
                      0
   76

-------
TABLE 22. DAP DEMETALLATION STUDY USING A ONE-GALLON AUTOCLAVE
Run
no.
A1*
A2
A3
A4
A5
A6
.%0f
DAP
4.0
4.0
1.5
1.5
1.5
1.5
Lead content
ppm
380
0.0
0.0
20
2.0
2.0
Ash content
wt.%
0.51
0.02
0.016
0.016
0.04
0.06
    Reaction Temperature: 160°C; Residence Time: 1 hr.
    Ash and lead content of used oil was 0.6 wt.%, 660 ppm.

    Note:

    'Dry oil: Lead content = 700 ppm
    Runs A1 and A5: The autoclave was completely sealed.
    Runs A2 to A4 and A6: One port of the autoclave was
                  intentionally opened to allow
                  the water vapor to escape in an
                  effort to simulate the distillation
                  process.
                                  77

-------
viscous and contains insoluble suspended sludge, resinous compounds, asphaltenes, carbon particles,
varnishes, and other impurities. These compounds, along with reaction products, blind the filter media
during the filtration cycles.  Rerefiners,  which employ filtration as a part of their process, have also
encountered similar problems in the past.
                                                                                 I
        The filtration rate can be improved either by increasing the temperature of the oil or by adding
diluent to the oil. In order to evaluate the use of diluent (No. 2 fuel oil) to improve filtration rates, another
DAP demetallization experiment was carried out in a 1000 ml distillation flask. After the reaction, treated
oil was divided into four fractions; a specified amount of filter aid or No. 2 fuel oil or both were thoroughly
blended in three of the four fractions. Viscosity of each fraction was measured using a Cannon-Manning
viscometer. As seen from Table 23, the filtration rate approximately doubled after dilution. The rate was
further increased with the addition of filter aid; however, filter aid increases solid waste generation. The
filtration rate, like the lead removal, is directly related to the processing temperature.  The filtration rate
increases uniformly with temperature and increases greatly when the oil is held at the final temperature.
In a process which includes a filter, the filtration rate is likely to be the limiting factor.

        In summary, problems encountered with filtration are: (0 poor filtration rate; (iQ frequent changing
of filter media (due to blinding); (Hi) high temperature requirement; (iv) addition of the filter aid increases
solid waste generation; and (v) cumbersome operation.


CONCLUSIONS


        (1) The organometallic additives in waste oil react with DAP in the interfacial region to form metal
phosphates.
                                                                                 i
        (2)  Lead is present in a different form compared with additive metals and is not directly removed
by chemical reaction with DAP.                                                    !

        (3)  Lead reduction is dependent on the reaction temperature and residence time, and is greater
at higher temperature and time.

        (4)  Water is required in the early stages of the demetallization reaction, while in the latter stages,
removal of water facilitates the agglomeration of rnetal hydroxyl phosphates and lead compounds.

        (5)  Pressure can be avoided in the reactor by venting out the water vapor and light components
dun'ng the reaction.

        (6)  Filtration produces high-quality reclaimed oil, in terms of ash and lead content, compared with
sedimentation.

        (7)  Filtration is  cumbersome to operate and increases the  cost of the  process due to high
temperature requirements.                                                        '

        (8)  More than 90% of lead  is  removed by treating waste oil with DAP at 150°C for an hour.

                                               78

-------
TABLE 23. EFFECT OF DILUENT AND FILTER AID ON FILTRATION RATE
«
Oil
Oil + 1 wt.%
Filter Aid
Oil + 20 wt.%
No. 2 Fuel Oil
Oil + 20 wt.%
No. 2 Fuel Oil +
1 wt.% Filter
Aid
Viscosity
(cp)
89.9
95.5
32.8
38.4
Filtration Rate
(ml/min)
11.5
12.0
25.0
34.0
Filtrate
Vis. (cp)
87.7
87.3
37.3
37.6
 Viscosity was measured at 25°C.
                             79

-------
                                          SECTION VII
                   USE OF TR1ETHANOL AMINE AS A FLOCCULATING REAGENT
        The quality of DAP treated oil depends on the solid-liquid separation. Filtration produces high
 quality oil in terms of ash and lead content. Although DAP reduces ash and lead content signii-ficantly,
 it does not remove the oxidation products present in used lubricating oils.

        Oxidation products in used oils include peroxide, carbonyl compounds, organic acids, etc. These
 compounds are highly corrosive and difficult to remove by traditional purification methods such as
 distillation, hydrogenation, or absorption. These oxidation products can be removed by treating ithe used
 oil with a strong reducing agent. A better approach would be to use the reagent, which is similar in
 nature to an antioxidation additive in motor oils.  Typically, zinc dithiophosphate is used as an antiwear
 and antioxidant additive; however, recently amines and phenolic corn-pounds have been widely used as
 antioxidants.                                                                    '

        The outstanding chemical properties of amines are due to the unshared pair of electrons on the
 nitrogen atoms. This makes amines both  bases and nucleophiles.48 The ability of amines, especially
 alkanolamines (e.g. triethanol amine), to neutralize organic acids is beneficial not only  in removing the
 oxidation products but also in neutralizing sulfonated compounds  (dispersant additives). The resulting
 oil is free of organic acids, peroxides, insoluble sludge, varnishes,  and other impurities.

        Alkanolamines are bifunctional molecules, having both amino and alcohol functional groups. As
 a result, they undergo a wide variety of useful reactions common  to amines and alcohols.  The amino
 group may be either primary, secondary or tertiary,  and may display typical  reactivity of a primary,
 secondary or tertiary amine.  Since an alkanolamine combines both the amine and alcohol groups in one
 molecule, it is capable of many reactions which an amine or an alcohol is not. In general, the hydroxyl
 group is regarded as able to reduce vapor pressure and increase water solubility, while the amino group
 provides the alkalinity in the  water solution required to neutralize acids.48

        In the past, amine compounds were tried as a flocculating agent to precipitate the sludge and
 impurities from used oil.50-51'62 From an initial experimental work, we found that triethanol amine (TEA) was
 a better flocculating agent in precipitating the impurities from used oil. While other alkanolamines such
 as monoethanol amine, diethanol amine, isopropanoi amine could  be used, the separation of impurities
were not as good as with TEA.

       A detailed parametric study was undertaken to map out the process variables in  order to identify
the most efficient reaction conditions. Experiments were performed to investigate the effect of different
 parameters on demetallization of used oil.
                                              80

-------
Effect of Triethanol Amine on Ash and Lead Reduction

       Used motor oil was reacted with TEA and allowed to settle for a day.  As shown in Table 24, the
ash content of clarified oil was not reduced significantly compared to the lead content. The analysis of
oil samples for different metal contents presented in Table 25 shows that the concentration of additive
metals more or less remained the same.

       It was thought that additives in the motor oil were so finely dispersed that they do not precipitate
with  the sEudge and impurities, while particulates lead were entrained in the reaction  products and
removed along with sludge.

       In recent years, lead contaminants in used oil have been due to corrosion of lead-copper bearings
rather than blow-by products of combustion (i.e. not due to tetraethyl lead).  The acids formed during the
service life of lubricating oils oxidize the metal bearings and other machine parts. These metal oxides are
soluble in the acids; hence, neutralizing the acids facilitates the separation of metal oxides from the used
oil.

       Since additive metals still remained present in the TEA treated oil, it was thought that the addition
of DAP would facilitate their removal along with the neutralization products.  Table 26 shows the ash and
lead reduction with respect to DAP. Significant points emerging from the data of Table 26 are:

        (0 without DAP, only a 24% ash  reduction is obtained, because additive metals like Ca, Ba, Mg,
and  Zn still remain dispersed in the oil.  The low ash reduction is due to the precipitation of paniculate
metals and metal oxides with sludge; and

       (ii) ash reduction seems to increase with an increasing amount of  DAP, up to 1 wt%. A further
increase in the amount of DAP  has very little effect on ash and lead reduction. This may be due to the
limitations of solid-liquid separation achieved by sedimentation.

       In all these experiments, a different concentration of DAP solution (20 wt%  aqueous solution) was
mixed with TEA (0.5 wt%) prior to the reaction. The addition of TEA separately also has the same effect.

       The amount of TEA required to flocculate the impurities and heavy metals (lead) depends on the
used oil. The highly oxidized oil (used motor oil) requires a higher amount of TEA to neutralize the acids.
Table 27 shows the effect of TEA on ash and lead reduction, while keeping the concentration of DAP
constant.

       Without TEA, the separation of reaction products (metal phosphates) and impurities was so poor
that  hardly any oil was recovered.  Normally, 0.5 wt% TEA is sufficient enough to flocculate the impurities
from used motor oil.  The higher amount of TEA favors the ash and lead  reduction but  decreases the
recovery of oil.
                                              81

-------
   TABLE 24. EFFECT OF TEA ON ASH AND LEAD REDUCTION
TEA
wt%
0.2
0.5
1.0
1.5
2.0
Ash Content Reduction
wt% %
Before After
0.88
0.86
0.88
0.90
0.89
0.80
0.65
0.67
0.66
0.67
9
24
24
27
25
Lead
Before
130
135
136
135
134
Content
ppm
After
114
55
45
45
43
Reduction
%
12 ,
59
67
67
i
68 ;
Reaction Temp: 80°C; Reaction Time: 1 hr.
Settling Temp: 75°C; Settling Time: 24 hrs.
   TABLE 25. ADDmVE METALS CONTENT OF TEA TREATED OIL
Metals
K
Ca
Mg
Ba
Zn
P
Used oil
ppm
120
1630
950
175
1900
1395
Treated Oil
ppm
108
1525
825
147
1734
1251
       Reaction Temp: 80°C; Reaction Time: 24 hrs.
       Settling Temp: 75°C; Settling Time: 24 hrs.
                            82

-------
   TABLE 26. EFFECT OF DAP ON ASH AND LEAD REDUCTION
DAP   Ash Content   Reduction
wt%     wt%           %
     Before  After
                                Lead Content
                                   ppm
                               Before    After
Reduction
0.0
0.5
1.0
1.5
2.0
0.86
0.89
0.89
0.80
0.82
0.65
0.47
0.11
0.10
0.11
24
47
88
88
87
114
91
109
89
69
50
25
6
4
2
56
7
94
96
97
TEA: 0.5 wt%, Reaction Temp: 80°C, Reaction Time: 1 Hr.
Settling Temp: 75°C, Settling Time: 24 Hrs.


    TABLE 27. EFFECT OF DAP ON ASH AND LEAD REDUCTION

TEA   Ash Content   Red.  Lead Content  Red.   Recovery
wt%     wt%        %       ppm        %
     Before   After        Before   After
0.0
0.5
1.0
1.5
2.0
0.87
0.89
0.88
0.88
0.89
—
0.11
0.10
0.07
0.07
—
88
89
92
92
112
110
118
122
123
-
7
4
2
2
-
94
97
98
98
-
89
85
70
70
DAP: 1.0 wt%; Reaction Temp: 80°C; Reaction Time: 60 min.
Settling Temp: 75°C; Settling Time: 24 hrs.
- The run without TEA resulted in poor separation of .impurities.
                             83

-------
Effect of Temperature and Tune on Ash and Lead Reduction

       The additives are group I  and II metal compounds and are highly reactive and more ionic in
nature.  The reaction between DAP and additive metals goes to completion and is least: dependent on
time and temperature. The same is true for the reaction between TEA and acids. The reaction between
oxidation products and TEA occurs even at low temperature within a short period of time. As seen from
the Table 28, a reaction time of 15 minutes is sufficient to reduce the ash and lead content to 0.12 wt%
and 10 ppm, respectively.  A further increase in reaction time does not improve the ash and lead
reduction.                                                                       i

       As discussed in the previous chapter, the lead content of DAP-treated oil depends on reaction
temperature and residence time, but with the addition of TEA, high lead reduction was obtained at low
temperature due to the following reasons:  (0 neutralization of acids by TEA causes the precipitation of
lead compounds, and these neutralization  reactions occurs at low temperature; and (ii) TEA flocculates
the particulates lead along with reaction products and impurities.  Table 29 shows the effect of the
reaction temperature on ash and lead reduction.  The ash and lead content is reduced between 85 to
95% at 80°C, while there is not  much difference in ash and lead content  with further increase  in
temperature.
.                                                                               I
       Although ash and lead reduction is  least dependent on reaction temperature and time, a recovery
of oil is greatly  influenced  by settling temperature and  time.  The viscosity of oil depends on the
temperature and is lower  at higher temperatures to facilitate the separation of sludge from oil.  Figures
28 and 29 show the effect of settling temperature and time on oil recovery. The oil recovery is biased on
the volume % of recovered oil with an ash  and lead content below 0.15 wt% and 10 ppm. respectively.
It can be seen from these figures that 90% of the oil is recovered at 80°C (settling time 24 hours;), which
Is very high considering the fact that used motor oil contains roughly 5 to 8% of sludge and additives.

Effect of Water on the Demetaffization of OH

       Trie presence of water is absolutely necessary for the reaction to occur between phosphate ions
and metals.

       It should be noted that a large quantity of water increases the load on the separation unit and on
the wastewater treatment  plant. To evaluate the effect of water on the demetallization reaction, various
concentrations of aqueous solution were reacted with waste oil. In all these experiments, the amount of
DAP and TEA was kept constant.  As seen from Table 30, too much dilute  solution reduces the  oil
recovery. This may have been caused by the poor separation of the reaction products, or it may be that
a large quantity of water  inhibits the flocculating power of TEA.  It was found that a concentration of
aqueous solution between 20% to 30% essentially provided the same  level of oil recovery. A further
increase in the concentration of DAP solution was not possible due to the recrystallization of DAP by the
addition of TEA.
                                             84

-------
    TABLE 28. EFFECT OF REACTION TIME ON ASH AND LEAD REDUCTION

            Reac.   Ash Content   Red.   Lead Content   Red.
            Time      wt%         %      ppm         %
            (min.)  Before   After        Before    After
7
15
30
45
60
0.86
0.85
0.89
0.85
0.89
0.45
0.12
0.11
0.10
0.10
48
86
88
88
89
100
105
109
104
108
45
10
8
6
5
55
90
93
94
95
            DAP: 1 wt%, TEA: 0.5 wt%; Reaction Temperature: 80°C
            Settling Temperature: 75°C; Settling Time: 24 Hrs.
TABLE 29. EFFECT OF REACTION TEMPERATURE ON ASH AND LEAD REDUCTION

            Temp.    Ash Content   Red.   Lead Content      Red.
                         wt%       %        ppm           %
                    Before   After        Before   After
50
80
100
120
0.85
0.89
0.89
0.89
0.45
0.11
0.13
0.11
47
88
85
87
75
109
108
114
30
6
8
6
60
94
93
95
            DAP: 1 wt%, TEA: 0.5 wt%; Reaction Time: 1 hr.
            Settling Temperature: 75°C; Settling Time: 24 hrs.
                                  85

-------
        TABLE 30. EFFECT OF WATER ON OIL RECOVERY
Aqueous   Ash Content      Red.   Lead Content  Red. Recovery
Solution        wt%          %       ppm       %    (vol%)
 (wt%)    Before      After         Before  After
10
15
20
25
30
0.87
0.85
0.89
0.85
0.89
0.10
0.10
0.11
0.09
0.11
89
88
88
89
88
102
104
107
103
108
9
10
9
7
6
91
90
92
93
94
71
77
89
89
88
DAP: 1 wt.%; TEA: 0.5 wt.%; Reaction Temperature: 80°C
Settling Temperature: 75°C; Settling Time: 24 Hrs.

Note:

The amount of DAP is same for all the run but concentration of
aqueous solution is different.
                              86

-------
   100 q




    90^




    80^




^70^


 U     -

 o  60-
 o     I

03 .    -

-  50-
•1-4
O     I


    40^




    30-
    20
       30         46        62        78        94

               Settling  Temperature (°C)
Figure 28. Effect of Settling Temperature on Oil Recovery.

          (DAP: 1 wt.%; TEA: 0.5 wt.%; Settling Time: 24 hrs.)
                         87

-------
      100
       90-
       80-
   CD
   >
   o
   o
   K   70-

   «—H
   • p—I
   O




       60-
      50
0
                 I
                5
10    15    20    25    30   35

 Settling  Time (hr)
Figure 29. Effect of Settling Time on Oil Recovery.             i
         (DAP: 1 wt.%; TEA: 0.5 wt.%; Settling Temperature: 80°C)
                           88

-------
MECHANISM OF METALS REMOVAL
       The organometallic additives in waste oil react with DAP in the interfacial region, as shown in
Figure 30. The hydrophilic metallic parts are oriented toward the aqueous phase, while the organic chain
is oriented toward the oil phase.  The anion (PO^  in the aqueous solution forms salts with the metals
in the oil.

       Triethanoi amine (CeH15NOa) reacts with organic acids to form neutralization products. TEA has
three hydroxyl groups and one amino group.  In general, the hydroxyl group is viewed as serving to
reduce the  vapor pressure and increase the water solubility, while the amino group provides the
necessary alkalinity in water solution to cause the neutralization of acids. A typical reaction between
tertiary amines and carboxylic acids is as follows53:

                               R3N + R1COOH	> RaNH+OOCR1

       Based on the above reaction, tertiary amines react with carboxylic acids to form salts, which are
highly complex and insoluble in  the oil and aqueous phase. TEA also reacts with a wide variety of
peroxides and forms insoluble products. Due to the neutralization of organic acids by TEA, the heavy
metal oxides (e.g. lead oxide) which are soluble in these acids also precipitate with the agglomeration of
metal hydroxyl phosphates and neutralization products from the oil. Lead particulates are entrained in
the agglomeration of reaction products and are removed along with the sludge. The flocculating power
of TEA facilitates the separation of impurities from the used oil.

Solid-Liquid Separation

       In precipitating the impurities and reaction products from the oil, three separation techniques,
including filtration, centrifugation,  and constant temperature settling, were studied. A detailed discussion
related to these separation techniques follows.

Filtration

       Solid-liquid separation is much better by filtration,  compared with any separation techniques.
Filtration is very cumbersome to operate,  needs frequent  supervision, and  increases the cost of the
process due to high temperature requirement.

       It appears that a combination of separation techniques may be better for overcoming the problem
associated with filtration.  Sedimentation, which precedes filtration, removes the bulk of impurities from
used oil, which in turn improves the filtration rate and is less troublesome from an operational point of
view. The ash content of clarified  oil (0.1-0.12 wt%) was further reduced (0.05-0.08 wt%) by filtration.
Hence, the process scheme for the pilot plant should include sedimentation before any other separation
techniques.
                                              89

-------
UJ
O
              CO

                                                                 1
                                                                 (0
                                                                 0>
                                                               !  8
                                                                 a
                                                                 o
                                                                 8

                                                                 £

                                                                 O)
    CO
 5 a:
                       90

-------
Centrifugation

       It increases the mass forces on particles and thus extends sedimentation to finer particles that
are normally stable in the gravity field. In sedimenting centrifuge, the centrifugal acceleration is much
larger than the acceleration due to gravity.  Centrifugal force overcomes the brownian diffusion forces,
which in gravity sedimentation hinder or prevent the settling of very fine particles. As the separation
efficiency is mainly affected by the behavior of the smallest particles in the system, it is common to
assume that centrifugation might work better, in terms of separating the fine particles, in comparison with
gravity sedimentation. However, in terms of ash and lead reduction, both  separation techniques i.e.,
centrifugation and sedimentation, work equally well (Table 31).

       In order to verify the results, several samples of waste oils were sent (after the treatment) to a
commercial centrifuge manufacturer for their evaluation. The oil was cerrtrifuged under the force of 8,000
G (acceleration was 8,000 times more than  gravity).  The  ash and  lead  content of the oil after
centrifugation were in the same range as that obtained through gravity sedimentation. This indicates that
flocculation increases the particle sizes, which facilitate the separation of impurities from oil.   Hence,
gravity sedimentation takes longer than centrifugal sedimentation, but the overall separation of the solids
in terms of ash and lead reduction is the same.
         TABLE 31. EFFECT OF DIFFERENT MODES OF SEPARATION ON ASH CONTENT
                     Samples   Ash Content   Red.
                                  wt%         %
                               Before   After
                        Lead Content     Red.
                             ppm           %
                        Before  After
                       1

                       2

                       3
0.89    0.11     88

0.89    0.12     87

0.89    0.10     89
104     9

105    10

104     9
91

90

91
                     DAP: 1 wt.%; TEA: 0.5 wt.%; Temp: 80°C; Time: 30 Min.

                            Note:

                     Sample 1 was allowed to settle at 75°C for a day.
                     Sample 2 was centrifuged in the lab.
                     Sample 3 was sent to the Sanbom Inc. for their evaluation.
                                             91

-------
       The major advantages of centrifugation over gravity sedimentation are high throughput capacity
and smaller size of the equipment. However, centrifuges are high maintenance items and can handle only
a certain amount of solids. Since waste oils contain high concentrations of solids, commercial centrifuge
systems include a combination of a decanter (for the separation of coarse particles) and a high-speed
centrifuge (for the separation of fine particles). This increases the cost of the process significantly; hence,
centrifuge was not considered for the separation of impurities from used oil.           j

Gravity Sedimentation

       The results obtained so far suggest that gravity sedimentation performs just as well in separating
solids (reaction products) from waste oil.  The clarified oil has a low metals content, as can be seen in
Table 32.  In addition to additive metals, lead was also significantly reduced by gravity separation.

       The efficiency of gravity sedimentation depends on temperature, time, settling area, and particles
size. Settling area and time are interrelated, because a large settling area requires less time for particles
to settle, and vice versa  Temperature is an important factor, because the viscosity of liquid is reduced
at higher temperature, which reduces the viscous force.  Particle size, especially for fine particles, affects
the effect of gravity separation. But in the case of TEA-treated oil, particles sizes are much larger due to
flocculation, and hence a nearly complete separation of solids is achieved by gravity sedimentation.


CONCLUSIONS


        (1) The organometallic additives in waste oil react with DAP in the interfacial region and
form metal phosphates.

        (2) Triethanol amine is very effective  in neutralizing acids and flocculating impurities from used
oil.                                                                             :

        (3) Lead is present either in particulates or in oxide form, and its removal is least dependent on
temperature and time.

        (4) Lead particulates are entrained with the sludge, and are  removed along with  the reaction
products, while neutralization of acids cause the precipitation of lead oxide from used oil.

        (5) More than 85% of the metals are  removed at 80° under atmospheric pressure.

        (6) Reaction conditions of this process are very mild as compared to other waste oil recovery
processes.

        (7) Solid-liquid  separation by gravity sedimentation is  comparable  with other separation
techniques, and produces specification grade fuel oil.

        (8) This method of treatment can be used as a pretreatment for the  Distillatiori-Hydrogenation
process.                                                                        ,

                                               92

-------
  TABLE 32. METAL CONTENTS OF USED OIL PROCESSED BY
           CHEMICAL TREATMENT
     Metal  Used Oil          Processed Oil        Sludge
     Cone,  (ppm)    Sedimentation     Filtered
K
Ca
Mg
Ba
Zn
P
Pb
Cu
Fe
Mn
Mo
Al
Cr
B
Si
Total
119
1570
905
150
1805
1367
150
338
225
9
11
24
5
5
34
6710
9
157
11
2
92
240
<2
268
9
<1
<1
20
<1
<2
5
815
9
120
7
2
44
180
<2
236
5
<1
<1
18
<1
<2
3
625
231
12297
7187
1160
13500
26623
1264
400
1106
73
52
120
30
33
57
64143
DAP: 1 wt.%; TEA: 0.5 wt.%; Temp: 80°C; Time: 30 min.
Settling Temp: 75°C; Settling Time: 24 hrs.
                          93

-------
                                         SECTION VIII
                                         PILOT PLANT
       Based on the bench-scale studies, the Chemical Engineering Department of Auburn University
has developed a pilot plant to reprocess waste oils into specification grade fuel oil. The pilot plant (Figure
31) has the capacity to treat 5 gallons/min. of used oil. The pilot plant studies wil! be very useful in
determining the feasibility of the  process.  It will also  identify engineering problems so they can be
corrected before any future upgrading of the facility takes place. The plant is divided into three sections,
namely demetallization, separation and sludge treatment.


DEMETALUZATION


       Waste oil is preheated to 85°C and is reacted with DAP and TEA in the clemetaliation unit (Figure
32). The demetallization unit is equipped with a dual turbine type agitator with 45° pitch blades. The
additive metals in used oil react with DAP to form metal hydroxyl phosphates, while TEA neutralizes the
acids and flocculates the impurities from used oil. The reaction is carried out for 30 minutes. After the
reaction, demetallized oil is transferred to the agglomeration unit.  Agitation in the demetallization  unit is
very vigorous so the reaction is not mass transfer limited, while agitation is very slow in the agglomeration
unit to facilitate the  agglomeration of metal hydroxyl phosphates, neutralization products and  heavy
metals. Both tanks are well  insulated to maintain the temperature during the reaction.  Demetallized oil
passes through a lamella clarifier  to separate solids and impurities.
                                                                                 i
Lamella

       The unit is shown in Figure 33, and is a development of the Axel Johnson Institute for Industrial
Research in Sweden.54  It consists of a series of inclined plates in close proximity to one another so that
the effective area becomes the horizontal projected area of each plate. The incoming slurry is introduced
either directly into the feed box or into a flash mix and flocculation tank. The slurry  is introduced to the
lamella plates through a bottomless rectangular feed curtain, from which it flows to the plates, as shown
in Figure 33, exiting at the top of the tank through flow distribution orifices.  The solids settle against the
inclined surfaces of the  plates and slide downwards into the sludge hopper, where additional densifying
is achieved by the action of a low-amplitude vibrator located inside the hopper.       !

       The  unit at  the AWORL  does not have  a flocculation tank or vibrator, and  slurry is directly
introduced into the feed box. The lamella clarifier offers an advantage in that it increases the settling area
and decreases the settling distance for the solids. The distance  is reduced to the separation between
                                               94

-------
                                                                          (0

                                                                         CL
                                                                         a:
                                                                          ^

                                                                          §
                                                                          u

                                                                         !
                                                                         6
                                                                          
-------
                                            o
                                           '•G
                                            0)
                                           CO

                                            o
                                           1
                                            
-------
       Flow distribution orifices


Overflow box
      Overflow
       Lamella
       plates
Discharge flumes

       Feed box
                                                      Rocculation tank
                                                                  Flash mix
                                                                  ank
                                                             Thickener-
                                                             scraper
                                                             drive
                                                             Picket-fence
                                                             thickener
                                                             Sludge
                                                             scrapers
                                                        Underflow
                                                        sludge
                      Figure 33. Lamella Clarifier.
                                    97

-------
the two plates. This unit is finding wider usage beyond its original coal and metallurgical applications.
They are now used extensively in the treatment of gas scrubber effluents, mill scale solutions and fly ash
suspensions. This unit has been in use for the past several years at AWORL, mainly for removing water
and suspended solids from used oil.


       The demetallized oil (a mixture of oil, water, reaction products, etc.) is passed through a lamella
clarifier to separate the impurities. Another advantage of the lamella clarifier is that settled solids are
washed  in  water before they are removed from the   bottom.  The clarified oil from  the top  is
steam-stripped or flash-vaporized to remove light hydrocarbons, chlorinated solvents and water from the
oil. Depending on the ash and lead content of oil, it is either filtered or bypassed and stored in a product
tank. This oil can be used as a specification grade fuel oil or as a feedstock to the refinery.

Sludge Treatment

       The bottoms (solids, water, and oil) from the lamella clarifier is transferred to the skim tanks, where
loose oil from the top is sent back to the feed tank for reprocessing (Figure 34).  Water from the skim
tanks is treated in a waste water treatment plant, while oily solids from the bottom are washed with light
distillate or Stoddard solvents. The oil-solvent mixture is either centrifuged or filtered to remove inorganic
solids and neutralization products. The solids are dried and stored in the aggregate bin. Research is
presently being conducted on how to use these solids as an asphalt aggregate.
RESULTS
       At this point, 7,000 gallons of waste oil had been reprocessed at the pilot plant. The used oil was
a mixture of different kinds of lubricating oils from Fort Rucker, Alabama. Since this project was funded
by DoD, which is very interested in developing technology to convert used oil generated at their facilities
into specification grade fuel oil, the oil selected for the pilot plant studies was from Fort Rucker. this oil
was similar to the kinds typically generated at many DoD facilities.                     \

       Before the pilot plant evaluation, the oil was treated in a laboratory to determine this ash and lead
reduction achievable by this treatment. The predictive demetallization test provides  a quantitative
measurement of ash and lead reduction possible by the above process. Waste oil from Fort Rucker was
treated with a solution of DAP (1 wt.%) and TEA (0.5 wt.%) at 80°C for an hour. The oil was allowed to
settle overnight at 75°C.  The ash content of the clarified oil was reduced to 0.1 wt.% from 0.48 wt.%.

       Table 33 shows the metals content of the oil processed at the pilot plant. As seen from this table,
the ash and lead content of the processed oil is reduced to less than 0.1 wt.% and 100 ppm respectively,
while  pentane insolubles are negligible due to the flocculation of impurities by TEA.

       The problem encountered during the pilot plant studies was in removing the chlorinated solvents
from the used oil.  Waste oils from DoD invariably  contain small amounts of solvents due to improper
mixing. The objective of this project is to remove the chlorinated solvents from used oil and to reduce
                                              98

-------
           tv
t
u.
                       g

                     «8
  Ol
  c:
                /
                          J,
                        ib
                  Ul
                                                      CO



                                                      
                                                      •
                                                o

                                               I
                                               i

                                               ?
                                               (0

                                               (D
                                               O>
                                               T3

                                               CO

                                               jg

                                               "o
                                               CO


                                               CO


                                               §
                                               0)
                          99

-------
TABLE 33. METALS COMTEhfT OF OIL PROCESSED AT PILOT PLANT
                         Used Oil      Processed Oil
Ash Content, wt.%
Pentane Insolubles, %
Metal Content (ppm)
K
Ca
Mg
Zn
Ba
P
Cu
Fe
Mn
Mo
Al
Co
B
Si
Pb
As
Cd
Cr
0.48
0.201

19
500
210
375
251
700
40
74
4
3
35
<1
2
5
140
<1
2
3
0.092
0.018

5
76
10
10
10
150
32
18
<1
<1
5
<1
<1
1
41
<1
<1
<1
     Total
23603
360
                           100

-------
ash and lead content. The project was very successful in ash and lead reduction, and in converting waste
oil to specification grade fuel oil.

       To determine the effectiveness of different methods in removing the solvents, a small batch of
used oil (700 gallons) was mixed with a known amount of 1,1,1-trichloro^thane and tetrachloroethylene
in a tank equipped with an agitator and a recirculation pump. The mixture (oil and solvents) at 275 F and
150 psi was flash-vaporized to the atmospheric pressure. The overheads were condensed and analyzed
for chlorinated solvents. Since this used oil contained a large amount of light distillates (roughly 30 wt. A
with a boiling point of less than 250°F), this temperature and pressure were not found sufficient to
vaporize the chlorinated solvents along with the light components.  Mainly 1,1,1-trichloroethane (boiling
ooint  165°R  was distilled off with light  components, while tetrachloroethylene (boiling point  250 F)
remained in the oil.  Hence, care should be taken in designing the capacity of a flash vaporization  unit.


CONCLUSIONS


        (1) The pilot plant studies demonstrated that this process is feasible and can convert waste oil
to specification grade fuel oil.

        (2) This process can also be used as a pretreatment for making lubricants out of used  motor oils.

        (3) This treatment has many applications in the field of oil purification and sludge treatment.

        (4) This process was chosen by the Department of Health Services in California as an innovative
 process for waste oil reprocessing.
                                               101

-------
                                          SECTION DC
                                          BURN TEST


        It is well established that oil reclaimed as a result of the pilot plant studies can be analyzed in
 detail to ascertain whether it meets the 'EPA specification for a used oil fuel* and whether it satisfies the
 ASTM specification for a No. 4 or No. 5 fuel oil.  However, this  analysis will not provide ariy insight as to
 the fate of particulates generated by the burning of reclaimed oil.  ASTM specification for a No. 4 and No.
 5 fuel oil allow an ash content of 0.1 wt.%; a goal of the developed process is to reduce the ash content
 of used oil to below 0.1 wt.%.

        The particulates formed by burning used oil are derived from two sources: very fine (between 0.1
 to 1 microns) insoluble residual particulates remaining in the reclaimed oil, and soluble metals bearing
 compounds (additives  such as detergent/soaps, rust inhibitors, etc.).   In general, the size of the
 particulates gene-rated by the combustion of waste oils are assumed to be so fine that they are emitted
 completely from the stacks. This is believed particularly true for lead-bearing particulates $5). The overall
 objective of the bum test is to establish how paniculate emissions of a reclaimed waste oil compare with
 those of a No. 5 fuel oil, each having nearly the same ash content (<0.1 wt.%).

        Paniculate emissions are very critical in high density population areas. In such areas, regulatory
 agencies usually prohibit or severely limit the burning of No. 6 fuel  oil (e.g. this is true in Birmingham,
 Alabama).  The  main performance criteria of a bum test are the gaseous and particulates products  of
 combustion (degree of pollution) and the condition of the boiler (amount of residue) after firing. The bum
 test will also establish the equivalency of  treated oil with ASTM grade fuel  oils, which will help  the DoD
 obtain a future permit for air emission discharges.                                   !

       The boiler trial bum is limited to particulate matter and lead because the DoD Is mainly interested
 in particulate and lead emissions from the stack; whereas moisture,  CO, CO2, and O2 are measured to
 determine the efficiency of the combustion  process (i.e. burner in the boiler).   The emission  of other
 parameters such as HCI and SO2 can be found by analyzing oil for the respective parameters (S and Cl
 content of the oil) prior to burning.

       Roy F. Weston, Inc. (WESTON) was retained by the Chemical Engineering Department of Auburn
 University to conduct stack emission testing during the trial bum. Two series of emission tests will be
 conducted. The first series was conducted with the boiler burning No. 5 fuel oil, while the second series
will be conducted on the reclaimed oil. Approximately 1,000 to 2,000 gallons  of each type oil will be
 burned during the tests. The boiler at Fort Rucker, Alabama, is a typical industrial boiler found  at many
DoD facilities; hence, it was selected for burning the reclaimed oil.  Figure 35 illustrates the proposed
organization for the project.                                                        j
                                             102

-------
                    AU PROJECT MANAGER
                     Hay T*rr«r,  Ph.D.
                 1
                  AO PROJECT COORDINATOR
                     Himi«b Dhuldhoy*
WESTON TECHNICAL DIRECTOR
  Bruc« Ferguson, Ph.D.
J
                 J
WBSTOM PROJECT MANAGER
    Jo* Ov«n, P.E.

                                  NBSTON TEST TEAM LEADER

                                      WBSTOM TEST TEAM

            Figure 35. Project Organization Chart.
                             103

-------
                                         SECTION X
                                       BIBLIOGRAPHY
Berry, Reginald (1979, April 23).  Re-refining Waste Oil, Chemical Engineering.

Brinkman, D. W., Cotton, F. O., & Whisman, M. L (1978,  December).  Solvent Treatment of Used
       Lubricating Oil to Remove Cokino and Fouling Precursors (Report No. BETC/RI-78/20).

Brinkman et al. (1980, January).  Pilot-Scale Used Oil Re-refining Using a Solvent Treatment/Distillation
       Process (Report No. BETC/ RI-79/14/XPS).

Brinkman, D. W. (1980, June). Engineering Design of a Solvent Treatment/ Distillation Used Lubricating
       Oil Re-Refinery (Report No. DOE/BC/10008-9/XPS).

Brownawell, D. W., & Renard, R. H. (1972, December 1). Refining of Used Lubricating Oils. U. 8. Patent
       No. 3,639,229.

Burwell, A. W. (1938).  Industrial Oxidation of Petroleum Hydrocarbons.  In A. E. Dunstan (Ed.), Jhe
       Science of Petroleum. London: Oxford University Press.

Canada, J. R. (1983, July). North Carolina Used Motor Oil Re-Refinino Program: Economics and Logistics:
       and Operating Results. Topical Report 3 (Report No. DE84001448/XPS).

Coyle, T. D., & Siedle, T. R. (1979, September).  Metals in Oil: Occurrence and Significance for Reuse of
       Spent Automotive Lubricating Oil (Report No. NBS SP 556).

Cutler, E. T. (1976, May).  Conserve Lube Oil: Re-refine. Hydrocarbon Processing, p. 86.
                                                                               i
Defives, D. et al. (1974). Regeneration Des Huiles Usages Par Ultarfiltration (Proc., First World Congress
       on Ultrafiltration, Paris) (p. B15).  New York: John Wiley and Sons.

Degering, F. (Ed.) (1945). Organic Nitrogen Compounds. Ypsilanti, Ml: University Lithoprinters.

Denison, G. H. (1944, May).  Oxidation of Lubricating  Oils, Industrial Engineering Chemistry. 36, No. 5,
       p. 477.                                                                  ;

Denison, G. H., & Condit, P. C. (1945, November). Oxidation of Lubricating Oils, Industrial Engineering
       Chemistry. 37, No. 11, p. 1102.
                                             104

-------
Duo-Sol Solvent Extraction Process (1947).  Process Handbook.  Max B. Miller & Co.

Ferrell, J. K. etal. (1984, June). North Carolina Used Motor Oil Re-Refining Program: Plant Operation and
      ' Quality Control Tests. Topical Report 4 (Report No. DE84014456/XPS).

Fiedler, K., et al. (1971, March 18). German Patent No. 2,030,609.

Fletcher, L C. et al. (1978, November 7). Distillation and Solvent Extraction Process for Rerefinina Used
       Lubricating Oil. U. S. Patent No. 4,360,420.

Fung et al. (1978, November 7).  Process for the Reclamation of Waste Hydrocarbon Oils. U. S. Patent
       No. 4,124,492.

Georgi, C. W. (1951). Motor Oils and Engine Lubrication. New York: Reinhold Publishing Corporation.

Gilson, J. R., & Massicotte, C.  R.,  (1967, February 21).  U. S. Patent No. 3,305,478.

Graves, E. F., U. S. Patent No. 2,568,583.

Griffith, W. C. (1984, February). North Carolina Used Motor Oil Re-Refining Program. Final Report (Report
       No. DE84009104/XPS).

Hedtke, F. (1974, April).  The Effect of Waste Oil on Freshwater Aquatic Life (U.S. EPA Report No.
       PB-257-693), p. 50.

Henley, Moke (1989, April 10).  Oil Recycling Bill Introduced. WasteTech News, p. 1.

Henley, Mike (1989, April 24).  Viability of Oil Recycling Seen Hinging on Label, WasteTech News, p. 11.

Henley, Mike (1989, August 14).  Used Oil Handlers Await Congress Decision, WasteTech News, p. 1.

Hess, L Y. (1979). Reprocessing and Disposal of Waste Petroleum Oils. Park Ridge, N. J.: Noyes Data
       Corporation.

Johnson, C. B. (1983, October 25).  Process for Removing Contaminants from Waste Lubricating Oil by
       Chemical Treatment. U. S. Patent No. 4,411,774.

Johnson, M. M.  (assigned to Phillips  Petroleum Company) (1975, April 22).  Reclaiming Used oil by
       Chemical Treatment with Ammonium Phosphate. U.S. Patent No. 3,879,282.

Johnson, M. M. (1976, January 6). Reclaiming Used Motor Oil. U. S. Patent No. 3,930,988.

Johnson et al. (1981, January 27). De-Ashing Lubricating Oils. U. S. Patent No. 4,247,389.
                                             105

-------
Jordan, T. B., & McDonald, J. W. (1973, October 2). Method of Reducing the Lead Content of a Used
       Hydrocarbon Lubricating Oil bv Adding Methvlethvl Ketone to Separate the Resulting Mixture into
       a Coagulated Insoluble Phase. U. S. Patent No. 3,763,036.

Kohl, A., & Riesenfeld, F. (1985). Gas Purification.  Houston: Gulf Publishing Company, i

Larsen, R. G. et al. (1942, February).  Oxidation Characteristics of  Pure  Hydrocarbons, Industrial
       Engineering Chemistry. 34, No. 2, p. 183.

Ueberman, M. (1974, February). Combustion and Heat Recovery of Air Force Waste Petroleum Oils and
       Lubricants (Report No. .AD 774 563).

Majuumdar, S. B. et al. (1990, May).  Regulatory  Requirements and Hazardous  Materials, Chemical
       Engineering Progress, p. 17.

Mattox, W. J. (1978, August 8). U.S. Patent No. 4,105,538.

Milde, R. L, & Beatly, H. A. (1959). Chemical Reaction of Tetraethyl Lead. In Metal-Organic Compounds
       (p. 3060. Washington: American Chemical .Society.)

Miller, T. M. (1983).  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.

Process Stream  Purification with Sodium Borohydride.   Technical Manual and Users Guide. Morton
       Thiokol, Inc., Ventron Division.                                             '

Propane Deasphalting Process (1947). Process Handbook.  M. W.  Kellogg Co.

Re-refining Lures Oil Firms (1975, March 31). Chemical  Engineering, p. 64.

Salusinszky, A. L (1981, February 10).  Treating Used  Hydrocarbon Lubricating Oils. U.S. Patent No.
       4,250,021.

Schilling, G. J., & Gordon, S. B. (1977).  Fuel and Lubricant Additives - II, Lubrication. 63,  No. 2.

Sedell, G. W. (1988).  Evaluation  of the B.E.S.T. Solvent  Extraction  Sludge Treatment Technology
       Twenty-Four Hour Test (Report No. EPA/ 600/2-88/ 051).

Smallheer, C. V.  (1976, December).  Chemistry of Lube Additives, World Petroleum, pp. 45-55.

Smallheer, C. V., & Hadley, H. A. (1967).  Lubricant Additives.  Lezius-Hiles Co.

Smith,  I. C.  et al. (1975).  The Role of Trace  Metals in Petroleum.  Ann Arbor: Ann Arbor Science
Publishers.
                                             106

-------
Sodium Borohydride. Technical Manual and Users Guide. Morton Thiokol, Inc., Ventron Division.

Stossel, E. (1945, July 21). Oxidation of Paraffins, Oil and Gas Journal. 44, p. 44.

Svarovsky, L (1981). Solid-Liquid Separation, Butterworths Monographs in Chemistry and Chemical
       Engineering (p.  141). London: Butterworths.

The  Alkanolamines  Handbook.   Midland,  Ml: The  Dow Chemical Company,  Chemicals & Metals
       Department.

Weinstein, N. J. (1980, September).  Feasibility Study for the Retrofitting of Used Oil Re-Refineries to the
       BETC Solvent Treatment/Distillation Process (Report No. DOE/BC/10044-8).

Williams, B. R. (1977, August). Automotive Crankcase Drainings Used for Fuel (Report No. NBS SP 488).

Yen, T. F. (Ed.), & Smith, I. C. et al. (1975). The Role of Trace Metals in Petroleum. Ann Arbor: Ann Arbor
       Science Publishers.
                                            107

-------
                                        APPENDIX A
                              QUALITY ASSURANCE ACTIVITIES
       The experimental results of oil demetallation as presented earlier were verified in accordance with
the Quality Assurance Project Plan (QAPP). The procedures and results are described in detail later in
this appendix.

       As mentioned earlier, the oil demetallation performance was evaluated in terms of ash and lead
contents remaining in the treated oil. Thus, the reliability of ashing and lead-determining procedures as
well as the AA calibration procedure was very important.

       To determine the lead content of the oil sample, about 5g of oil sample was ashed first in a muffle
furnace at 600°C for 12 hours. The ash (i.e., total ash-forming contaminants) content was determined as
described in Figure A-1. Lead was assumed to remain in the ash.  The ash was dissolved in an acid
solution and then diluted to an appropriate concentration range for AA analysis as described in Figure
A-2. The AA calibration procedure used is described in Figure A-3.                  ,

       Since the oil sample was ashed in a  muffle furnace at 600°C for 12 hours, it is questionable
whether some of the lead is lost due to evaporation while ashing.  To verify this,  a known amount of
organometallic lead compound (lead cyclohexanebutyric acid, ICeH^ChUJaCOO];, pb) was weighed in a
crucible and ashed as described above. The lead content in lead cyclohexanebutyric acio" is 37.8%. This
compound was obtained from Kodak Laboratory and Specialty Chemicals.  As shown in Table A-1 , after
ashing, the remaining amount of ash amounted to 38.97% of the original compound charged to the
crucible.  This percentage is close to the  lead content of the original organometallic lead compound.
Following the ashing, the ash was dissolved in an aqueous nitric acid solution as described in tine QAPP
and the lead content of the solution was determined using atomic absorption spectroscopy.  The AA
analysis showed that the lead content in the aqueous solution was 176 ppm which is Very close to its
estimated lead content (176.5 ppm) based  on the known compound  composition and  amount of
compound ashed (Table A-1).  This observation suggests that the lead in tie  organometallic lead
compound is not volatilized during ashing.

       To determine the accuracy of lead analysis, a blind spiked oil sample was prepared. To prepare
the blind spiked oil sample, a known amount  of organolead standard, [CeH^CHygCOOk pb, was first
dissolved in an appropriate amount of xylenes (heated  if necessary)  and then blended with a specified
amount of virgin SAE 10-30W motor oil.  Regular procedures for ash and lead analyses as described in
Figures A-1, A-2, and A-3 were followed. The results are given in Table A-2.
                                             108

-------
FIGURE A-1. LEAD ANALYSIS PROCEDURE (MODIFIED ASTM 2788-72).
             After completing ash analysis
     Add 10 ml (50 vol% HNO,)  + 2 ml (con.  HC1
     to each crucible.  Heat the crucible on at
     hot plate gently for about 10 min.	
      Transfer the hot acid solution to a 100
      volumetric flask.  Dilute with distilled
      water to 100 ml and shake vigorously.
                      AA Analysis
     AA Spectrophotometer:
     Lead hollow cathode lamp current:
     Wavelength:
     Spectral Band  Pass:
     Optimum working  range(ppm):
     Working conditions  (fixed)
          Fuel:
          Oxidant  (support):
          Flame stoichiometry:
VARIAN AA-475
5 MA
217.0 NM
1 NM
5-20

Acetylene
Air
Oxidizing
                           109

-------
 FIGURE A-2. ASH ANALYSIS PROCEDURE (MODIFIED ASTM D482-81).
                   Sampling as per
                    ASTM D270-1981
Heat: three crucibles at 750°C for
15 minutes.   Cool  to room tempera-
ture in  a  desiccator and weigh  to
nearest  0.1 mg.  (W0)
                               Shake the oil sample vigor
                               ously before transferring
                               5  grams  to each crucible
                               and weigh again.  (W,) ;
Heat the crucible/sample  on  a hot
plate cautiously  in a hood until a
carboneous in a hood until a car-
bonaceous residue is obtained.
                              Heat the  residue in the
                              muffle  furnace  at 600*0
                              until all the carbona-
                              ceous residue has been
                              reduced to ash  (about 12
                              hours).
Cool the crucible/ash to room temperature  in the  dessicator
and weigh to nearest 0.1 mg. (W2)

          Ash content, wt.% =    fW=-W0l   *   100
                                 (WrW0)
                           110

-------
          FIGURE A-3. AA CALIBRATION PROCEDURE.
Prepare three lead Standards (5,
10, and 15 ppm) by diluting a
Fisher inorganic Lead standard
(1000 + 1% ppm)
Set absorbance zero
Aspirate the standards:
5, 10, and 15 ppm in
the absorbance mode and
note the readings.
                              Set-up and turn on AA.
                              Align the instrument.
                              Optimize flame conditions
                              using 5 ppm standards.
                           Construct a calibration curve
                           by plotting absorbation  vs.
                           concentration (ppm).
                           Aspirate the unknown samples
                           for lead analysis and note
                           the readings. (The absorbance
                           should be bracketed by those
                           of standards or dilution is
                           required).
   Obtained the concentrations of unknown samples
     form calibration curve.  (C ppm)
     Pb content, ppm =   C  fppm) * 100 ml * D.F.
                            Wt. of oil sample

     D.F = Dilution Factor
                           111

-------
TABLE A-1  VOLATILITY TEST OF ORGANOMETALUC LEAD COMPOUND* IN
             MUFFLE FURNACE AT 600PC FOR 12 HOURS.
    Before Ashing:
          Crucible weight
          Crucible/lead compound weight
          Lead compound weight
          Lead weight
16.4060 g
16.4527 g
 0.0467 g
 0.01765 g
    After Ashing:
          Crucible/ash weight
          Ash weight
          Ash content
16.4242 g
 0.0182 g
 38.97%
    Atomic absorption analysis:

          Lead content

          Theoretical lead content
          calculation
176 ppm in aqueous nitric
 acid solution
176.5 ppm in aqueous nitric by
 acid solution
    * Lead cyclohexanebutyric acid
     Structural formula: [C8H11(CH2)aCOO]2 pb

     Molecular weight: 545.68
     Lead content: 37.8%
                                      112

-------
       To determine the repeatability of ash and lead analyses, the blind spiked oil sample was poured
into four crucibles; each crucible contained about 5g of spiked oil.  As shown in Table A-2, the ash
contents of the four oil samples were 0.568, 0.585, 0.582, and 0.594 wt.%, respectively. The resultant
average measured ash content was 0.582 ± 0.011 (or 0.582 ± 0.019%) weight percent, which was
acceptable.  The lead contents of the four oil samples were 230.7, 243.6, 231.9, and 242.1  ppm,
respectively. The resultant average measured lead content was 231.1  ± 6.7 (or 231.7 ± 2.9%) ppm,
which was also acceptable.

       Knowing the amount of organolead, xylenes, and motor oil used and the lead content in the
organolead standard (37.8 wt.%), the lead content of the spiked oil could be calculated, assuming that
no lead was present in the virgin motor oil.  To confirm this assumption, the ash and lead contents of the
virgin motor oil were measured. As shown in Table A-3, no lead was found in the motor oil. The true lead
concentration in the spiked oil was 242.7 ppm as shown in Table A-2 as compared to the average
measured value of 237.1 ± 6.7 ppm. The percent accuracy of lead determination was about 2.3%, which
was acceptable.

       To determine the precision of ash and lead analyses, a demetallated oil, treated with a metallic
borohydride solution (i.e., SWS solution), was  used in which eight aliquats (5g each) of this oil were
analyzed for ash and  lead contents. The ash and lead analysis results are given in Table A-4.  The
average measured ash content for this oil was 0.548 ± 0.009 (or 0.548 ± 1.64%) wt.% and the average
measured lead content was 74.99 ±  0.90 (or 74.99 ± 1.20%) ppm. Both of them are acceptable.

       In summary, the Quality Assurance Project Plan has been executed properly, and the results have
been found to be satisfactory.
                                             113

-------
     TABLE A-2.  ACCURACY OF LEAD ANALYSIS PROCEDURE,

Crucible No.         1            2            3           4       I    Average
Before Ashing:
Crucible wt, g 44.8785
Crucible/oil
sample* wt., g 49.8643
Oil sample wt., g 4.9858
After Ashing:
Crucible/
ash wt., g 44.9068
Ash wt., g 0.0283
Ash content, % 0.568

44.0060 42.2435 42.4106
49.0146 47.2462 47.4090
5.0086 5.0027 4.9984 4.9989±0.0097
;
44.0353 42.2726 42.4403
0.0293 0.0291 0.0297 0.0291 ±0.0006
0.585 0.582 0.594 0.582±0.011
AA Analysis:
 F3b content, ppm    230.7
243.6
                                           231.9
                          242.1   237.1 ±6.7
"lead-containing oil sample was prepared as follows:

Lead cyclohexanebutyric acid  : 0.0321 g
(Lead content: 37.8%)
Xylenes                    : 4.0056 g
SAE 30 hd motor oil          : 45.9802 g
Lead content (by calculation)   : 242.7 ppm

Percent error: (242.7 - 237.1) x (100%) = 2.31%
                242.7
                                   114

-------
TABLE A-3. DETERMINATION OF ASH AND LEAD CONTENTS OF
            VIRGIN MOTOR OIL*
          Before Ashing:
                 Crucible weight
                 Crucible/oil weight
                 Oil weight
  17.4812 g
- 22.4771 g
=  4.9959 g
           After Ashing:
                 Crucible/ash weight
                 Ash weight
                 Ash content
= 17.5123 g
=  0.0311 g
=  0.62%
           Atomic Absorption Analysis:

                 Lead content
=  0 ppm (nd)
                 * Motor oil used: Quaker State SAE-30HD
                   Refined from Pennsylvania Grade Crude Oil
                                115

-------
TABLE A-4. PRECISION TEST FOR LEAD AND ASH ANALYSIS.
  Sample No.
Ash Content
   (wt.%)
Lead Content
    (ppm)
      1
      2
      3
      4
      5
      6
      7
      8
   0.551
   0.541
   0.555
   0.549
   0.540
   0.543
   0.541
   0.567
   75.26
   75.04
   75.26
   75.10
   74.89
   73.84
   73.84
   76.68
    Average
   0.548±0.009
  74.99±0.90
                             116

-------
                                       APPENDIX B
                       QUALITY ASSURANCE/QUALITY CONTROL PLAN
                                RECLAIMED OIL TRIAL BURN
       Roy F. Weston, Inc. (WESTON) was retained by the Chemical Engineering Department of Auburn
University to conduct emission testing on an oil-fired package boiler at Fort Rucker Army Base at Fort
Rucker, Alabama

       Two series of tests were conducted for paniculate, lead, and carbon monoxide emissions in the
stack gas. The first test series was conducted during the burning of reclaimed oil from an experimental
oil reclamation process being developed by Auburn University.  The second test series was conducted
during the burning of commercially available No. 5 fuel oil.  The purpose of the tests was to compare the
emission rates from the two fuel oils.

       The emission testing was performed on July 25-26,1990, by a WESTON test team comprised of
Mr. Greg Sims, Mr. Arnold Seidl, Mr. William Kelly, and Mr. Jeff Hollingsworth.  Mr. Joe Oven was the
WESTON Technical Director and QA Officer.

       Auburn University personnel collected boiler operational data and samples of each type of oil
during testing.  Oil analyses were conducted by Auburn University.

       Dr. Ray Tarrer was the Auburn University Project Manager.  Mr. Joe Hayes of Auburn University
and Mr. Ron Leatherwood of Fort Rucker were the WESTON technical contacts at the plant.

       The following section of this report details the emission test results. The section afterward outlines
the sampling and analytical procedures used to conduct the testing.
RESULTS AND DISCUSSION
       This section presents the results of the emission testing on the reclaimed fuel oil and the No. 5
fuel oil.  Because all runs were being conducted on the same boiler, the runs were sequentially
numbered from Run 1 through Run 7.

       Runs  1, 2, 3, and 7 were conducted on the reclaimed fuel oil while runs 4, 5, and 6 were
conducted on No. 5 fuel oil.
                                           117

-------
Comparison of Measured Emission-Reclaimed Fuel OH and No. 5 Fuel Oil

       Table B-1 presents a summary of results of the paniculate, lead, and carbon monoxide (CO)
emission testing on the package boiler at Fort Rucker, Alabama, performed on July 25-26,1990. Results
of the individual test series on each type oil are discussed in the following two sections.

Reclaimed Fuel Oi

       Table B-2 summarizes the results of the paniculate lead and carbon monoxide emission testing
performed while the boiler was burning reclaimed fuel oil on July 25-26,1990.  A total of four runs (Runs
1,2,3, and 7) were conducted on the reclaimed fuel oil. After completion of Runs 1,2, and 3, the client
informed WESTON that the boiler was set at an incorrect air/fuel ratio during Run 1 and that a fourth run
would be necessary to obtain three runs under optimum boiler conditions.  It was mutually agreed upon
to conduct the fourth run (Run 7) following the tests on the No. 5 fuel oil. Run 1 was voided. The mean
paniculate emission rate for Runs 2, 3, and 7 was 1.20 1b/hr and 0.061  1b/MMBTU.  The mean lead
emission rate was 0.041b/hr and 0.0021b/MMBTU. The mean carbon monoxide emission rate was 108
ppm and 1.92 1b/hr.

Commerciai No. 5 Fuel Ofl

       Table B-3 summarizes the results of the paniculate lead and carbon monoxide emission testing
performed while the boiler was burning No. 5 fuel oil on July 26,1990. The mean paniculate emission
rate was 2.06 1b/hr and 0.135 1b/MMBTU.  The mean lead  emission rate was 0,05 1b/hr and
0.0031 b/MMBTU. The mean carbon monoxide emission rate was 160 ppm and 2.76 1b/hr.

Sampling Procedures

       Effective sampling procedures were used in conduction of the tests  with the exception of the
following deviations:                                                           •

       « The QA/QC plan indicated that each test run would be two hours in duration, to make sure that
enough paniculate would be collected on the EPA Method 5 sampling train filters. However, due to the
heavy filter loading and high vacuum in the sampling train after the first run, it was decided to reduce the
test run time. The test run sampling durations were 64 minutes (4 minutes per point) or 80 minutes (5
minutes per point) for various runs.

       • After each test run, the EPA Method 5 sampling nozzle, probe, and filter holder were rinsed
with 0.1 NHNO3 following the acetone rinse. This was done to make sure all traces of lead were collected.

       « Dr. Len Nelms replaced Dr. Bruce Ferguson as Technical Director and QA Officer in the project.

       • Carbon monoxide sampling was done by collecting integrated samples in bags, foltowed  by
analysis in WESTON's laboratory. The sampling program described in the QA/QC plan anticipated on-line
CO analysis.  However, due to the calibration sensitivity of the CO analyzer and the open roof location,
it was decided not to use the CO analyzer for testing in the field.
                                            118

-------
                TABLE B-1. COMPARISON OF MEASURED EMISSION RATES.
       Paniculate
       Lead
       Carbon Monoxide
Reclaimed Fuel Oil
MEAN TEST VALUE

1b/Hr 1b/MMBTU   ppm

1.20     0.061      —
0.04     0.002      -
1.92     —        108
                                                          No. 5 Fuel Oil
                                                          MEAN TEST VALUE

                                                          1b/Hr  1b/MMBTU ppm
      2.06    0.135
      0.05    0.003
      2.76    —
           160
                    TABLE B-2. RECLAIMED FUEL OIL EMISSION DATA.
Date

Time Begun

Time Ended
                          RUN 2
             RUNS
RUN?
MEAN
07/25/90      07/25/90     07/26/90

1928         2055        1547

2038         2203        1707
Stack Gas

 Temperature, °F
 Velocity, ft/sec
 Moisture, %
 Oxygen, %
 Carbon Dioxide,
335
17.9
13.4
4.5
12.5
348
16.2
13.6
4.5
12.5
343
16.5
12.9
3.2
13.0
342
16.9
13.3
4.1
12.7
                                        119

-------
               TABLE B-2.  RECLAIMED FUEL OIL EMISSION DATA (CONTINUED).
Volumetric Flow Rate

 At Stack Conditions
       tfVmin
 At Standard Conditions*
       tfVmin
                           RUN 2
RUN 3
RUN?
                                                                   MEAN
7,580
4,370
6,890
3,900
7,000
4,030
7,160
4,100
Paniculate

  Isokinetio Sampling
       Rate, %             93
  Concentration at
  Standard conditions*,
       gr/ft3                0.035
  Emission Rate, 1b/hr       1.29
  Emission Rate,
       1b/MMBTU           0.066
96
0.033
1.10

0.056
97
0.035
1.19

0.061
95
0.034
1.20

0.061
Lead
Concentration at Standard
conditions* , gr/ft3
Emission Rate, 1b/hr
Emission Rate,
Ib/MMBTU
Carbon Monoxide"
Concentration, ppm
Emission Rate, 1b/hr
0.001
0.04

0.002

86
1.64
0.001
0.04

0.002

72
1.22
0.001
0.04

0.002

165
2.90
0.001
0.04

0.002

108
1.92
                            •68°F and 29.92 inches of mercury.
                            "Corrected for CO2 in gas stream.
                                           120

-------
                  TABLE B-3. COMMERCIAL NO. 5 FUEL OIL EMISSION DATA.
Date
Time Began
Time Ended
Stack Gas
Temperature,°F
Velocity, ft/sec
Moisture, %
Oxygen, %
Carbon Dioxide, %
Volumetric Flow rate
At Stack Conditions
te/min
At Standard Conditions"
tfVmin
RUN 4
07/26/90
1005
1128

354
16.8
13.7
2.0
14.2

7,140
4,020
RUNS
07/26/90
1155
1317

344
16.0
13.2
2.0
14.2

6,780
3,890
RUN 6
07/26/90
1350
1510

363
16.9
13.7
3.2
13.2

7,180
4,000
MEAN
—
—
—

354
16.6
13.5
2.4
13.9

7,030
3,970
Paniculate

 Isokinetic Sampling
       Rate, %              98
 Concentration at Standard
    Conditions', gr/ft3        0.061
 Emission Rate, 1b/hr        2.09
 Emission Rate,
       1D/MMBTU            0.137
96

0.064
2.14

0^140
97

0.057
1.96

0.128
97

0.061
2.06

0.135
                                           121

-------
           TABLE B-3. COMMERCIAL NO. 5 FUEL OIL EMISSION DATA (CONTINUED).
                           RUN 4
             RUNS
             RUNG
             MEAN
Lead

 Concentration at Standard
       Conditions', gr/ft3     0.002
 Emission Rate, 1b/hr        0.05
 Emission Rate,
       Ib/MMBTU           0.003
             0.002
             0.06

             0.004
             0.001
             0.05

             0.003
             0.002
             0.05

             0.003
Carbon Monoxide"

 Concentration, ppm
 Emission Rate, 1b/hr
177
3.10
213
3.61
 91
1.59
160
2.76
                           *68°F and 29.92 inches of mercury.
                           "Corrected for CO2 in gas stream.
                                          122

-------
Quality Control

       Throughout the entire project, a high level of quality control was maintained to ensure the
accuracy of the data.  The test personnel are experienced  in the use  of the instrumentation, the
procedures involved, and the quality control requirements. The following paragraphs briefly summarize
the quality control associated with the project:

       General. All data were recorded at the time of collection on preprinted data sheets. All samples
were prepared for shipment, and chain-of-custody was maintained from the sampling technician to the
analyst. Calculations were performed (where possible) with pre-programmed calculators.  Data transfers;
were minimized, and all calculations were verified by a second person.  The report was reviewed and
approved by the Technical Director prior to transrnittal. In general, all accepted quality control standards
and practices recommended by the reference methods were followed.

       Stack Gas Volumetric Flow. The stack was measured with a certified tape to an accuracy of 0.1
inch. The velocity and sampling traverse points were marked on the probe with heat resistant glass fiber
tape.

       The S-type pilot tubes used to measure the velocity pressures were visually inspected for damage
prior to the test and damage was not  indicated.  The pyrometer used  to measure the stack gas
temperature and  all thermocouples for intermediate measurements were calibrated with respect to
standard thermometers. At the completion of the test, all equipment was visually inspected, and damage
was riot indicated.

       Stack Gas Molecular Weight.  Quality control on oxygen analyses by EPA Reference Method 3
involved the analysis of ambient air before sampling. If the measured concentration was less than 20.0
percent, the Orsat chemicals were changed before proceeding.

       If the measured concentration was greater than  20.0 percent but  less than 20.6 percent, the
sample data were corrected for the low measurement.  If the measured concentration was 20.6 percent
or greater, no correction was made. The Method 3 train was leak-checked with a KNF Pump capable of
producing a vacuum greater than 10 inches Hg.

       WESTON also participated satisfactorily in the most recent EPA Audit  Sample for Reference
Method 3.  Those data are on file at WESTON.

       Moisture Content.  Quality control of the moisture analysis involved the accurate measurement
of the  gas flow and the accurate  determination of the moisture condensed in  the sampling train.  A
calibrated triple beam balance was used to weigh the volume of water in each impinger before and after
sampling.  The silica gel was weighed, before and after its use, with the same balance to the neatest 0.1
gram. The sum of all differences in weights before and after sampling was considered to be the moisture
collected.

       Paniculate and Lead Concentration. The dry gas meter used to measure the sample volume
collected was calibrated before and after sampling. The calibration  obtained was within the required
                                             123

-------
specifications each time.  All thermocouples and other items used to calculate the mass emission rate
were calibrated on a routine schedule.

        WESTON participated satisfactorily in the most recent dry gas meter audit supplied by EPA.
Those data are on file at WESTON.                                               :

        WESTON used Class S 1-, 10-, 50-, and 100-gram weights to verify the accuracy!of the balance
for each used.  The weight is weighed when the filters are tared and also at the final weighings.  Any
significant difference in weight indicates a problem with the balance, and the balance is repaired before
proceeding.                                                                    :

        The rate of sample collection was determined to be within 10 percent of the isokinetic rate. Those
data indicate the validity of sample collection.

        Trie paniculate and lead analyses were performed according to the general procedures described
in EPA Reference Methods  5 and  12. The paniculate analysis was performed first on the probe wash
samples by drying the samples in tared 250 ml beakers and weighing to a constant weight. The filter was
then added to the beaker, dried, and weighed to a constant weight.

        The lead analysis was performed using the same samples.  The fitter was removed, and the nitric
acid impinger catch and probe wash made with 0.1N HNO3 were added to the beaker, covered with a
watch glass and taken to dryness.  Next, the filter was cut into strips, added to the beaker, and digested
according to EPA Method 12 Section 5.3.3. The final solution was filtered using a mixed cellulose ester
membrane filter and diluted to 100 ml. The lead concentration was determined using a Perkin-Elnier 2380
Atomic Absorption Spectrophotometer.

        Quality control procedures for this project consisted of duplicate analyses by separate analysts,
analysis of spiked filter samples, spiked liquid samples and blank samples, and use of the method of
standard addition.  The following tables summarize the results of the quality control analyses performed
for this project and shows that all guidelines have been met. The method of standard addition was
performed by spiking a known volume of standard solution with an aliquot of Sample No. CE 800. The
results showed an insignificant difference of two percent between the measured mass and the calculated
mass of lead  measured.

       Two blank filters were analyzed along with a reagent blank which included the acetone from the
paniculate analysis, the 0.1N HNO3, and deionized water.  All blank samples were found to be below the
detection limit of 20 ug.

       Carbon  Monoxide. Samples were  collected and conditioned in the field for laboratory analysis
of carbon monoxide by NDIR techniques.  Integrated samples were collected concurrently with and for
the same sampling time as  the paniculate and lead sampling. At the time of collection, moisture was
removed by a condensate trap. The samples were stored in Tedlar Bags prior to analysis. NBS traceable
gases were used to calibrate the NDIR analyzer before and after analysis.
                                             124

-------
Quality Assurance Objectives

       Table B-4 summarizes the QA objectives for measuring data in terms of precision, accuracy and
completeness. Representativeness will be defined primarily by the sampling locations and correlation of
sampling operations with process conditions.

Sampling Procedures

       Preliminary test data will be obtained at the stack sampling locations. The sampling port locations
and traverse points are selected in accordance with EPA Methods 1  and 2 as shown on Figures B-1 and
B-2, respectively.  Geometric  stack measurements  will  be recorded and  traverse point  distances}
calculated  Prior to formal testing, a check for the presence or absence of cyclonic flow will be conducted
at the stack test location. The cyclonic flow check must be negative to verify the test location is suitable
for  obtaining representative  samples.   Based upon the present  ductwork and  stack arrangement,
WESTON does not expect to  detect the presence of cyclonic flow at the stack sampling location.
However, if cyclonic  flow is detected, WESTON proposed to construct an  'egg-crate1 type  flow
straightening device and to lower it into position in the stack to alleviate the cyclonic flow condition.  This
device will be positioned such that the sampling port locations will not be affected.

       A series of three test  runs  (on each type of oil) for each parameter will be conducted
simultaneously at full load process conditions. The sampling ports illustrated on Figure B-1 will be used
during testing.   Two  main ports will be used for Method 5 traversing.  One smaller port, located
approximately six inches above the plane of the  main ports, will be used for the Method 10 stationary
sampling probe. Each test run will be two hours long.The sampling condition for each specific sampling
method will be followed to ensure that a quantitative recovery of each analyte is obtained from the source.
Outlines of the various sampling procedures are detailed in the following text. The oil will be transported
to Ft. Rucker by tanker truck, and fed to the boiler directly from the truck. This procedure will be used
to avoid oil contamination if the reclaimed oil was pumped into Ft. Pucker's oil tanks.

       The stack gas volumetric flow rate, molecular weight, and moisture content will be determined
simultaneously by using EPA Methods 1-4 with the EPA Method 5 runs.  The velocity of the gas stream
will be determined by reading the instantaneous velocity head using an inclined manometer at each point
with a calibrated S-type pitot tube. Stack gas temperatures will be determined by means of thermocouple
and a calibrated pyrometer. Carbon dioxide and oxygen concentrations will be determined using a Fisher
Brand Orsat with a 50 percent capacity. An integrated sample will be collected for the duration of  each
of the Method 1 -5 Runs and subjected to analysis by the Orsat Method. The moisture content of the gas
stream will be determined by weighing the impingers before and after each Method 1-5 Run.

       The paniculate and lead concentrations of the gas stream will be determined using EPA Methods
5 and 12 in the Method 5 sampling train. The filter will be of a low lead background, and the first two
impingers will be changed with 100 ml of 0.1 N HNO3 prior to each test run. Three test runs on each type
of oil (each run being at least two hours long) will  be performed.  The gas stream will be sampled
isokinetically at each sampling point by adjusting the sample flow rate to correspond to the measured
velocity at each point.  Following each run, the probe and nozzle will washed with acetone to remove
                                             125

-------
  TABLE B-4. SUMMARY OF DUPLICATE ANALYSES.
Lab No.
*
CE794
CE795
CE796
CE797
CE798
CE799
CE800
Run No.
1
2
3
4
5
6
7
%STD
2
5
1
3
4
3
1
% Accuracy
2
7
2
4
4
4
1
TABLE B-Su  SUMMARY OF SPIKED SAMPLE RESULTS.
ID
Measured
Mass (ug)
Actual
Mass (ug)
%Diff
A Filter
B Filter
A Liquid
B Liquid
100
144
9.97*
4.94*
100
150
9.90*
4.98*
0
4
1
1
Concentration in ug/ml.
                     126

-------
the adhering paniculate matter. The filter will be removed from the holder and stored in a petri dish until
analyzed. The filter holder will then be rinsed with acetone, and this rinse will be added to the probe
rinse. The first three impingers will be rinsed with 0.1 N HNO3, and this rinse will be transferred to the
same sample container with the impinger contents. The containers will be sealed and labeled and liquid
levels marked for transport to the laboratory.

       The mean temperature of the stack gas and the dry gas meter will be used in calculating the final
data The mean isokinetic sampling rate and the stack gas velocity (volumetric flow) will be calculated
from the mean of the square roots of the velocity pressure measured at each traverse point during the
paniculate! sampling.

       The concentration of carbon monoxide in the stack gas will be determined continuously by means;
of nondispersive infrared (NDIR) spectroscopy, in accordance with the analytical procedures identified in
Method 10.
                                             127

-------

-------