United States
                      Environmental Protection
                      Agency
 Risk Reduction
 Engineering Laboratory
 Cincinnati, OH 45268
                      Research and Development
 EPA/600/SR-93/114    August 1993
vxEPA       Project  Summary

                      Mobile  Onsite  Recycling of
                      Metalworking  Fluids
                      Arun R. Gavaskar, Robert F. Olfenbuttel, and Jody A. Jones
                        Product quality, waste reduction, and
                      economic issues were evaluated for a
                      technology designed to recycle metal-
                      working fluids. Emulsion-type fluids
                      were tested at two sites and a  syn-
                      thetic fluid was tested at a third  site.
                      The specific recycling unit being evalu-
                      ated is based on the technology of fil-
                      tration,  pasteurization, and centrifuga-
                      tion. This recycling unit is mounted on
                      a truck that goes from site to site, per-
                      forming  the  recycling  at  each
                      customer's location. The customer is
                      charged a fixed fee for the service. Met-
                      alworking fluid recycling was found to
                      have good potential as a way to reduce
                      waste and  save money. The product
                      quality achieved by this unit was ac-
                      ceptable for the applications studied.
                      Product quality was evaluated by  con-
                      ducting  performance tests and by
                      chemical characterization of the spent,
                      recycled, and virgin fluids. Performance
                      tests included tests for  corrosion re-
                      sistance, emulsion  stability, foaming
                      resistance, lubricity, and biological re-
                      sistance.
                        This Project Summary was developed
                      by EPA's Risk Reduction Engineering
                      Laboratory, Cincinnati, OH, to announce
                      key findings of the research project
                      that is fully documented in a separate
                      report of the same title (see Project
                      Report ordering information at back).

                      Introduction
                       The objective of  the Waste Reduction
                      Innovative Technology Evaluation (WRITE)
                      Program conducted by the U.S. Environ-
                      mental Protection  Agency (EPA)  is  to
evaluate, in a typical workplace environ-
ment, examples of prototype or innovative
commercial technologies that have poten-
tial for reducing wastes. The goal of the
metalworking fluid recycling study was to
evaluate (a)  the quality of the  recycled
coolant, (b) the waste reduction  potential
of the technology, and (c) the economic
feasibility of the technology.
  The mobile metalworking fluid recycling
unit is operated by  Safety-Kleen Corp.*
Elgin, IL. Safety-Kleen provides fluid re-
covery services to a variety of businesses,
primarily those that generate  relatively
small quantities of fluid waste. The mobile
service  performs the  recycling on  the
generator's property, thus eliminating the
need to transport potentially hazardous
wastes.  Each mobile truck-mounted  unit,
operating off its own power, is capable of
processing fluid at a maximum rate of 300
gal/hr.
  The recycling process (Figure 1) con-
sists  of filtering, pasteurizing, and centri-
fuging the  spent fluid. The fluid is first
sent through a 100-u, filter to remove any
large  particulates. It  is then  pumped
through a preheater and then a heat ex-
changer to kill bacteria and fungi, as well
as to reduce  fluid viscosity. Centrifuging,
where tramp oil and other debris are sepa-
rated from the usable fluid, is next. Addi-
tives are then incorporated into the fluid to
restore performance. In the final step, the
fluid flows through a 1-p. filter to remove
any remaining particulates. The fluid is
* Mentionof tradenamesorcommercial products does
 not constitute endorsement or recommendation for
 use.


          ^ZZ> Printed on Recycled Paper

-------
    Vacuum
      Hose
150 Micron.
                                                        Main
                                                        Filter
                                                      Chamber
                      • Remdte Filter' '
                     Chamber'* Caddy
   Customer
   Recyctabte
     fluid
    Storage
   Container
                                                                                     Recirculate
Clean
Fluid
Holding
Tank
i
f
                                                                                   Pump
                             Customer
                              Chan
                              Ftuftf
                             Storage
                             Container
Fluid Test
Pass
Fail
                     Customer
                       Waste
                     Container
                      Outside
                        Van

To
Centrifuge
Bypass
Centrifuge
Figure 1. Metalworking fluids recycling system flowchart.
then returned to the client's clean holding
tank for reuse.
  The technology was evaluated at three
different small- to medium-sized machine
shops (sites) in the Philadelphia, PA, vi-
cinity. The three sites were chosen from
among Safety-Kleen's customer base. Two
of the sites (called E1 and E2) used emul-
sion-type metalworking  fluids. The  third
site (called S1) used a synthetic fluid. At
each  site, one sample each of the spent,
recycled, and virgin fluids (at their  use
concentrations) was collected and  sub-
jected to the same series of tests.  The
comparison  between the  spent  and  re-
cycled fluids indicates the improvement
                      achieved by recycling, and a comparison
                      between the recycled and virgin fluids in-
                      dicates how closely the recycled product
                      approximates the virgin product. A limited
                      number of  samples was taken at each
                      site because the objective was to conduct
                      a broad spectrum of analytical tests on a
                      few samples, rather than a statistical com-
                      parison based on a large number  of
                      samples.

                      Product Quality Evaluation
                        The main purpose of metalworking flu-
                      ids  in machining operations is to provide
                      lubricity and cooling without causing cor-
                      rosion  or other problems. Through use,
the fluids lose some measure of these
functions because of the accumulation of
contaminants and microbes. The recycling
process attempts to restore these func-
tions.
  Degree of removal of nondissolved and
dissolved particulates during recycling is
shown in Table 1. High concentrations of
these particulates affect tool life, surface
finish, and chemical breakdown. Particles
also provide substrates for microbial
growth. At  all three sites, the  results
showed considerably lower concentrations
of nondissolved particulates in the recycled
fluids (E1-R, E2-R, and  S1-R) as com-
pared with concentrations in the spent flu-

-------
 Table 1. Analysis of Non-Dissolved and Dissolved Solids
                    ; Non-Dissolved Solids Concentration
                      	(mg/100 mL.)	
 Sample No.
Total
               Inorganic
Dissolved Solids
 (Conductivity)

 (umhos/cm2)
E1-S"
E1-R
E1-V
E2-S»
E2-R
E2-V
S1-S
S1-R
S1-V
79.10
22.55
3.55
12.55
5.60
4.50
33.80
17.00
5.18
27.25
1.45
2.50
0.50 c
3.00
2.00
14.50
1.95
0.78
2,400
1,810
700
1,820
1,750
810
1,450
1,460
1,930
   ByASTM D 2276. Particulates smaller than 8 microns.
   Analyzed after skimming off and discarding the floating tramp oil. E1-S = spent emulsion, site 1;
   E1-R = recycled emulsion, site 1; E1-V= virgin emulsion, site 1; etc.
   Possible inhomogeneity giving a low value.
ids (E1-S, E2-S, S1-S). The accumulation
of very small participates over  time  and
use could limit the number  of  times  a
given  batch  of fluid  could  be  recycled.
Conductivity of the samples was measured
as  an indicator of the  dissolved  solids
levels  in the fluids. Dissolved solids levels
remained  approximately the  same  after
recycling, which indicated the effect of con-
taminant precipitation  and fresh additive
introduction.
  Users of metalworking fluid  often moni-
tor the pH as an easily measured indica-
tor  of  fluid quality. A change  in  pH may
indicate chemical  degradation or degra-
dation  due to microbial growth. The recy-
cling process seeks to  restore pH  to  a
range  of 8.5 to 9.5. This alkaline pH im-
proves emulsion stability and corrosion
resistance characteristics of the  fluid. At
the three  sites tested, the pH of the re-
cycled  fluids was returned to this range.
                     Corrosion characteristics (Table  2) are
                   important parameters for water-based met-
                   alworking fluids because of their effect on
                   workpiece quality and tool life. The results
                   of the iron  chip corrosion test (ASTM  D
                   4627) on the virgin samples  (E1-V, E2-V,
                   and S1-V)  showed  that E1-V and S1-V
                   generated no rust at the use concentra-
                   tion (approximately 5% solution of the con-
                   centrate in  tap water).  In this test, the
                   lower the concentration of the fluid  in wa-
                   ter at which there is no rust, the better the
                   corrosion resistance. S1-V showed stron-
                   ger corrosion inhibition since there were
                   no rust  stains even at 30% of  the  use
                   concentration. E2-V showed rust stains at
                   the use concentration itself, indicating that
                   this virgin fluid had  lower strength  corro-
                   sion inhibition  properties  compared with
                   the other two. Recycled  sample E1-R
                   showed  considerable  improvement over
                   the spent sample (E1-S),  indicating  that
Table 2. Corrosion Test Results of the Metalworking Fluids
Sample No.
E1-SC
E1-R
E1-V
E2-SC
E2-R
E2-V
S1-S
S1-R
S1-V
Iron Chip
Corrosion Breakpoint '
Rust at use concentration
No rust at 50% of use concentration
No rust at use concentration
Rust at use concentration
Rust at use concentration
Rust at use concentration
Rust at use concentration
Rust at use concentration
No rust at 30% of use concentration
Copper
Corrosion "
1A
1A
1A
1B
1B
1A
1A
1A
1B
   Analyzed by ASTM D 4627.  Breakpoint is the lowest concentration tested that left no rust stains
   on filter paper.
   Analyzed by ASTM D 130.  The rating scale is from 1 to 4, where 1 indicates slight tarnish and 4
   indicates corrosion. 1A indicates a light orange color (almost the same as the freshly polished
   strip) and 1B indicates a dark orange color.
   Analyzed after skimming off and discarding the floating tramp oil.
 its corrosion inhibition properties had been
 restored. E2-R and S1-R showed some
 rust at the use  concentration,  indicating
 that  stronger  iron  corrosion  resistance
 properties need  to be  imparted  to these
 fluids. All of the collected samples (Table
 2) fared virtually the same in the copper
 corrosion test  (ASTM  D 130) with a  high
 rating of 1A or 1B, indicating that none of
 the samples have much effect on copper.
   Tramp oil is the nonemulsified floating
 oil that  builds up in  metalworking  fluid
 sumps from sources such as leaking equip-
 ment  seals  (hydraulic  oils, gear oils)  or
 from  the workpiece itself. These oils can
 contaminate the workpiece  or generate
 smoke from the heat of machining. Tramp
 oils  are also the biggest  contributors  to
 fluid  rancidity  and odor. Table 3  shows
 the results of the tramp oil analysis. Spent
 samples  E1-S and E2-S contained ap-
 proximately 6% and 2% (by volume) re-
 spectively of tramp oil.  No phase separa-
 tion  was  noticed  in any of the  recycled
 samples, indicating the tramp oil had been
 removed.  Virgin  sample  E1-V  showed
 some  phase separation, but this was at-
 tributed to some unemulsified concentrate
 in the fluid.
   The results of  emulsion stability testing
 at elevated temperature (Table 3) showed
 small  amounts of phase separation  in
 spent  samples E1-S and  E2-S. The re-
 cycled samples  remained as  a  single
 phase even after  96  hr,  indicating  that
 emulsion stability had been restored  dur-
 ing recycling.
   Foaming  can  reduce  effective  film
 strength, reduce  heat transfer, and inter-
 fere with the settling of metal fines. Ten-
 dency of the fluids to foam was tested by
 ASTM D 892-89.  Foam volume in  the
 recycled samples (E1-R, E2-R, and S1-R)
 was significantly  higher than  that  in the
 spent  or virgin samples. This can be at-
 tributed  to  introducing  fresh  emulsifier
 (surfactant)  during recycling. A correction
 can be made for this effect by adding an
 antifoam agent during  recycling.  Safety-
 Kleen, however, does not typically add an
 antifoam agent unless  the user  specifi-
 cally reports a foaming problem.
   At all three sites, the recycled and vir-
 gin fluid viscosities were very close (Table
 4); this indicated that the recycling  pro-
 cess  had restored this parameter.  The
 viscosity measurements also indicate  that
the recycling  process succeeded  in re-
turning the fluids to the  required use con-
centration (oihwater ratio). The concentra-
tion of the recycled fluid is adjusted during
the recycling process by taking refracto-
 meter readings. More virgin fluid is added

-------
Table 3. Tramp Oil Separation and Emulsion Stability

                 Tramp Oil Separation
                 (Room Temperature)
             Emulsion Stability'
            (Temperature = 8S°C)
Sample
No.
E1-S
E1-R
E1-V
E2-S
E2-R
E2-V
S1-S
S1-R
S1-V
Total
Initial
Volume of
The Fluid
Samples (mL)
898
850
882
846
850
850
850
850
850
Volume
(mL) of
Upper Layer
Separating
Out
After 4 Hr
51
0
22 c
13
0
0
0
0
0
Upper Layer Volume
(mL)
Total
Initial
Volume (mL)"
100
100
100
100
100
100
NA
NA
NA
After
48 Hr
1
0
0
1.5
0
0.7
NA
NA
NA
After
96 Hr
1
0
0
1"
0
0"
NA
NA
NA
   By ASTM D 3707. An NA indicates not analyzed.
   After discarding the upper layer formed at room temperature.
   Unemulsified concentrate.
   Upper layer that formed after 48 hr reduced or disappeared after 96 hr.
to the recycled batch if needed to restore
the use concentration.
  Lubricity and wear preventive charac-
teristics  of  a metalworking  fluid  affect
workplace quality and tool life.  Lubricity
and wear characteristics were measured
by the standard  "four-ball test" (ASTM D
445) (Table  4). For Site E1, the  recycled
sample caused a much lower average scar
diameter than did the spent sample, but
not as low as the virgin sample. This indi-
cated that the recycled and virgin  samples
had better lubricity and wear characteris-
tics than the spent fluid and that the virgin
sample  was slightly better than  was the
recycled. The Site E2 samples showed no
noticeable differences in performance, al-
though the  recycled and  virgin  samples
performed about the same. The presence
of some emulsified tramp oil could have
improved the lubricity results of the spent
sample E2-S.
  A major factor in metalworking fluid spoil-
age (rancidity) is  microbial growth. In the
recycling process, existing microbes are
killed  during the  pasteurization step, the
dead biomass is removed during the cen-
trifugation step, and a measured quantity
of biocide is added to control future micro-
bial growth. ASTM E 686-85  evaluates
the effectiveness  of biocides at use con-
centrations. No microbial growth was ob-
served in the recycled samples  even after
6wk.
  The performance tests conducted in this
evaluation  (viscosity,  lubricity  and wear,
iron  corrosion,  copper  corrosion,
bioresistance,  foaming  tendency,  and
emulsion stability) are a measure of the
integral effect of both the contaminant lev-
els  as  well as the level of additives and
Table 4. Lubricity and Wear, and Viscosity Characteristics of the Metalworking Fluids
Sample No.
E1-S"
E1-R
E1-V
E2-S*
E2-R
E2-V
S1-S
S1-R
S1-V
Viscosity '
(cs)
0.77
0.85
0.81
0.69
0.81
0.77
0.77
0.75
0.75
Average Wear Scar Diameter b
(mm)
1.26
0.83
0.64
0.97
1.18
1.17
NA
NA
NA
   By ASTM D 445. An NA indicates not analyzed.
   Analyzed after skimming off and discarding the floating tramp oil.
other fluid components. The levels of par-
ticular contaminants that can be tolerated
in the recycled fluids are difficult to judge
in isolation and are often affected by the
properties of other fluid components and
additives. The recycling  process brings
about considerable  improvement in fluid
quality, to  make  recycling  a technically
feasible option. The recycled fluid showed
some tendency toward foaming and iron
corrosion when compared with the virgin
fluid;  but these could possibly be adjusted
by appropriate additives. Some solubilized
contaminants  (such  as calcium, magne-
sium,  etc.)  remain in  the recycled fluid
because the smallest  filter (1  |i) in  the
recycling unit does not remove them. The
levels of these contaminants in the fluids
at the three sites evaluated did not, how-
ever,  appear to affect  their performance.
Retention of solubilized constituents in re-
cycled fluids also has the potential for old
and  new additives to clash if  they  are
mismatched.
   Currently, there are  no published stan-
dards for recycled fluids. Each user has to
evaluate his/her own requirement based
on the same factors used in selecting a
virgin fluid  brand. At the  three test sites
evaluated in this study, recycled fluids ap-
peared to  satisfy  the  functional  require-
ments of the users.

Waste Reduction  Potential
   The waste volume  reduction  potential
of this technology involves the amount of
spent metalworking fluid kept from  being
disposed into the  environment  (either by
landfilling or by onsite wastewater treat-
ment  and sewer  disposal).  On  an aver-
age,  Safety-Kleen visits each user once
every 10 wk and recycles an average of
250 gal of spent fluid per visit. Thus, there
is potential for an annual reduction of 1,250
gal from a typical small user.  Approxi-
mately 4 gal  of  tramp oil  per  visit  are
generated during, recycling. This tramp oil
is hauled away at a competitive fee by
Safety-Kleen for use as supplemental fuel.
Residue generated on the filters (mostly
metal chips)  is transferred to the user's
waste metal bin and  later reclaimed for its
metal value.
   According to a  1991  study by the Inde-
pendent  Lubricant Manufacturer's Asso-
ciation, the volume of metalworking fluids
(concentrate) manufactured  in the United
States, has increased steadily from 67 mil
gal in 1985 to 92 mil gal in  1990. By
extending the life of metalworking  fluids
through  onsite recovery, considerable
amounts of fluid  can be  prevented from
going to waste. The actual total volume of
fluids  going to waste, in some cases, may

-------
be as much as 20 times higher than the
manufacturer volumes, since many types
of fluids are diluted into 3% to 5% solu-
tions in water.

Economic Evaluation
  The  economic  evaluation compared
costs for recycling versus costs for  dis-
posal. Recycling costs included the onsite
service charge for the customer arid tramp
oil disposal cost. Disposal costs included
spent fluid disposal cost and hazard analy-
sis costs. The annual savings for a typical
small user, who recycles 1,250 gal/yr of
metalworking  fluid was approximately
$1,600, if the spent fluid was nonhazard-
ous, and $7,800, if the spent fluid was
hazardous (by the Toxicity Characteristic
Leaching Procedure).

Conclusions
  This evaluation found that  recycling of
metalworking fluids is a good option for
small-  to medium-sized plants with ma-
chining operations. In the absence of pub-
lished standards for recycled fluids quality
and performance, the user has to evalu-
ate the recycled product by the same cri-
teria used to select a virgin brand. In addi-
tion to the testing performed in this evalu-
ation,  shop-floor testing of the recycled
fluids  over an  extended period of time to
determine the  effect on workpiece quality
and tool life would be desirable.
  The full report was submitted in fulfill-
ment  of Contract No.  68-CO-0003  by
Battelle Memorial Institute under the spon-
sorship of the  U.S. Environmental Protec-
tion Agency.
                                                                     •&U.S. GOVERNMENT PRINTING OFFICE: 1993 - 750-071/80049

-------

-------