June 2002
DTSC R-02-03
EPA 600/R-02/043
Environmental Technology
Verification Report
ABB Inc.
BIOTEMP® Vegetable Oil-
Based Insulating Dielectric
Fluid
Prepared by
California Environmental Protection Agency
Department of Toxic Substances Control
Under a cooperative agreement with
U.S. Environmental Protection Agency
ET1/ET1/ET1/
-------
June 2002
Environmental Technology Verification
Report
ABB Inc.
BIOTEMP®
Vegetable Oil-Based Insulating
Dielectric Fluid
By
California Environmental Protection Agency
Department of Toxic Substances Control
Office of Pollution Prevention and Technology Development
Sacramento, California 95812-0806
-------
Notice
The information in this document has been funded in part by the U.S. Environmental Protection
Agency (EPA) under a Cooperative Agreement number CR 824433-01-0 with the California
Environmental Protection Agency (CalEPA), Department of Toxic Substances Control (DTSC).
The Pollution Prevention and Waste Treatment Technology Center under the U.S. EPA
Environmental Technology Verification (ETV) Program supported this verification effort. This
document has been peer reviewed by the EPA and recommended for public release. Mention of
trade names or commercial products does not constitute endorsement or recommendation by the
EPA or the Department of Toxic Substances Control (DTSC) for use.
This verification is limited to the use of the ABB BIOTEMP® Vegetable Oil-Based Insulating
Dielectric Fluid for use in pole-mounted, small distribution and small power transformer units as
an alternative to mineral oil-based dielectric fluids or those containing PCBs. EPA and DTSC
make no express or implied warranties as to the performance of the ABB BIOTEMP® Vegetable
Oil-Based Insulating Dielectric Fluid technology. Nor does EPA and DTSC warrant that the
ABB BIOTEMP® Vegetable Oil-Based Insulating Dielectric Fluid is free from any defects in
workmanship or materials caused by negligence, misuse, accident or other causes.
June 2002 ii
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's air, water, and land resources. Under a mandate of national environmental laws, the
EPA strives to formulate and implement actions leading to a compatible balance between human
activities and the ability of natural systems to support and nurture life. To meet this mandate, the
EPA's Office of Research and Development (ORD) provides data and science support that can
be used to solve environmental problems and to build the scientific knowledge base needed to
manage our ecological resources wisely, to understand how pollutants affect our health, and to
prevent or reduce environmental risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA, to
verify the performance characteristics of innovative environmental technologies across all media
and to report this objective information to the permitters, buyers, and users of the technology,
thus substantially accelerating the entrance of new environmental technologies into the
marketplace. Verification Organizations oversee and report verification activities based on
testing and Quality Assurance protocols developed with input from major stakeholders and
customer groups associated with the technology area. There are now six ETV technology centers,
which include the original twelve ETV technology areas. Information about each of the
environmental technology centers covered by ETV can be found on the Internet at
http://www.epa.gov/etv.htm
Effective verifications of pollution prevention and treatment technologies for hazardous waste
are needed to improve environmental quality and to supply cost and performance data to select
the most appropriate technology. Through a competitive cooperative agreement, the California
Department of Toxic Substances Control (DTSC) was awarded EPA funding and support to plan,
coordinate, and conduct such verification tests, for "Pollution Prevention and Waste Treatment
Technologies" and report the results to the community at large. Information concerning this
specific environmental technology area can be found on the Internet at
http:\|www.epa.gov/etv/03/03_main.htm.
The following report reviews the performance of the ABB BIOTEMPR Vegetable Oil-Based
Insulating Dielectric Fluid. BIOTEMP® is used as an insulating dielectric fluid for pole-
mounted, small distribution, and small power units as an alternative to mineral oil-based
dielectric fluids or those containing PCBs.
June 2002 iii
-------
Acknowledgment
DISC wishes to acknowledge the support of all those who helped plan and implement the
verification activities, and prepare this report. In particular, a special thanks to Ms. Norma
Lewis, Project Manager, and Ms. Lauren Drees, Quality Assurance Manager, of EPA's National
Risk Management Research Laboratory in Cincinnati, Ohio.
DTSC would also like to thank Mr. Ron West of Pacific Gas and Electric for their support and
for providing the facility and necessary resources to conduct the verification field test.
Additionally DTSC would like to thank Mr. Jim Baker, Mr. Phillip Collins, and Mr. Gerry
Schepers of ABB Inc. for their participation in this Environmental Technology Verification Pilot
Project.
June 2002 iv
-------
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
PROGRAM
EPA OCal/EPAl
California Environmental Protection Agency I
U.S. Environmental Protection Agency ^^^^^^^^^^^^^^^^^^^^^^J
ETV JOINT VERIFICATION STATEMENT
TECHNOLOGY TYPE: VEGETABLE OIL-BASED INSULATING DIELECTRIC
FLUID
APPLICATION: VEGETABLE OIL-BASED INSULATING DIELECTRIC
FLUID FOR USE IN 3-PHASE TRANSFORMERS, UP TO
20MVA
TECHNOLOGY NAME: BIOTEMP® INSULATING DIELECTRIC FLUID
COMPANY: ABB INC.
ADDRESS: 2135 PHILPOTT ROAD PHONE: (540)688-4929
SOUTH BOSTON, VIRGINIA 24592 FAX: (540) 688-3844
WEB SITE http://www.abb.com/us/
EMAIL: don.cherry@us.abb.com
The U.S. Environmental Protection Agency has created the Environmental Technology Verification
(ETV) Program to facilitate the deployment of innovative or improved environmental technologies
through performance verification and information dissemination. The goal of the ETV Program is to
further environmental protection by substantially accelerating the acceptance and use of innovative,
improved, and more cost-effective technologies. The ETV Program is intended to assist and inform those
individuals in need of credible data for the design, distribution, permitting, and purchase of
environmental technologies.
ETV works in partnership with recognized testing organizations to objectively and systematically
document the performance of commercial ready environmental technologies. Together, with the full
participation of the technology developer, they develop plans, conduct tests, collect and analyze data, and
report findings. Verifications are conducted according to an established workplan with protocols for
quality assurance. Where existing data are used, the data must have been collected by independent
sources using similar quality assurance protocols.
June 2002 v VS-R-02-03
-------
EPA's ETV Program, through the National Risk Management Research Laboratory (NRMRL), has
partnered with the California Department of Toxic Substances Control (DTSC) under an ETV Pilot
Project to verify pollution prevention, recycling, and waste treatment technologies. This verification
statement provides a summary of performance results for the ABB Inc. BIOTEMP®Vegetable Oil-
Based Insulating Dielectric Fluid.
TECHNOLOGY DESCRIPTION
ABB Inc. (ABB) has developed a dielectric insulating fluid called BIOTEMPR which is comprised of
>98.5% vegetable oil and <1.5% antioxidants. BIOTEMP® is used in liquid-filled electrical transformers
to act as an electrical insulating medium, and to transport heat generated in the transformer around the
windings, core and connected circuits to cooling surfaces. BIOTEMP* is currently used in pole-mounted,
distribution, network, and small power transformers with a voltage rating < 69 kV and a maximum kVA
rating of 20 MVA. Approximately 250 transformers supplied with BIOTEMPR fluid are presently in-
service.
EVALUATION DESCRIPTION
The evaluation consisted of:
Developing a Technology Evaluation Workplan by DTSC to independently evaluate the technology
with respect to the identified performance objectives for general performance, aquatic
biodegradability, flammability, acute toxicity, chemical composition, and worker health and safety;
Implementing the Technology Evaluation Workplan by DTSC and ABB at their manufacturing
facility in South Boston, Virginia and at Pacific Gas and Electric's (PG&E) in-service transformers in
San Francisco, California. The field sampling included collection of 12 samples from three different
unused (virgin) product lots, and four samples from four different in-service transformers (one
sample per in-service transformer);
Analyzing virgin product samples for general performance parameters (fire and flash point, dielectric
breakdown, dissipation factor, oxidation stability, viscosity, pour point, water content), aquatic
biodegradation, aquatic toxicity using the California sample preparation method, fatty acid content,
phenolic antioxidants, SVOCs, and metals. In-service transformer sample analyses included general
performance parameters (fire and flash point, dissipation factor, water content, conductivity), fatty
acid content, phenolic antioxidants, SVOCs, and metals;
Reviewing supporting documentation on BIOTEMP® including ASTM data, an acute toxicity report,
aquatic biodegradability data, and material safety data sheets (MSDSs).
VERIFICATION OF PERFORMANCE
Performance results of ABB Inc. BIOTEMP® Vegetable Oil-Based Insulating Dielectric Fluid are as
follows:
• General Performance. The average sample results for the each virgin product lot and the overall
average for all three lots are presented in Table 1. BIOTEMP® met the ASTM and ABB
performance specifications for dielectric breakdown (minimum and gap), oxidation stability at 72
hours (sludge generation and neutralization number), and oxidation stability for 164 hours
(sludge generation only) for all three lots. Only two lots had values that met the ASTM D3487
and ABB performance specifications for dissipation factor at 25 °C. All three BIOTEMP® lots
met the ABB performance specifications for dielectric breakdown (impulse), pour point, water
content and viscosity at 0°C, 40°C, and 100°C while only two lots met the ABB specification for
June 2002 vi VS-R-02-03
-------
dissipation factor at 100°C. However, the data consistently exceeded the neutralization number
listed for all three specifications for the oxidation stability at 164 hours. The data also did not
meet the oxidation stability criteria for the rotating bomb method for ABB and ASTM D3487
specifications.
Table 1. Summary of Virgin Product Sampling Results
Performance Parameters
Specification Standards
ABB
ASTM D3487
ASTM D5222
Average Sample Results
Lot 2000-216
Lot 2000-224
Composite Lot*
Average
Dielectric properties
Dielectric breakdown (kV)
Minimum
gap
Impulse**
Dissipation Factor (%)
@ 25 °C
@ 100°C
>30
>28
>100
>30
>28
> 145
>42
>30
-
<0.05
<2.0
<0.05
<0.3
<0.01
<0.3
46 ±4
37 ±3
177 ±83
0.160 ±0.184
2.95 ± 1.15
51 ±6
37 ±5
200 ±68
0.022 ±0.011
0.837 ±0.307
55
39
173
0.028
0.931
50 ±3
37 ±2
185 ± 32
0.075 ±0.054
1. 665 ±0.762
Chemical Properties
Oxidation Stability
Percent Sludge (%)
after 72 hours
after 164 hours
Neutralization No. (mgKOH/g)
after 72 hours
after 164 hours
Rotary Bomb (minutes)
Water Content (ppm)
<0.2
<0.2
<0.1
<0.2
<0.2
<0.5
<200
<150
<0.3
<0.4
<195
N/A
-
800-1,000
N/A
0.02 ±0.01
0.03 ±0.04
0.19 ±0.04
21.13± 1.31
118±4
75 ±21
0.02 ±0.015
0.02 ±0.02
0.16 ±0.02
18.41 ±3.66
116±5
72 ±37
0.02
0.02
0.16
16.02
116
102
0.02 ±0.00
0.02 ±0.01
0.17 ±0.02
19.02 ± 1.85
117±2
79 ±14
Physical Properties
Pour Point (°Q
Viscosity (cSt)
@ 100°C***
@ 40°C
@ 0°C***
-15 to -25
N/A
N/A
<10
<45
<300
N/A
N/A
N/A
N/A
N/A
N/A
-18 ±6
8.61
40. 73 ±0.51
276.27
-17 ±5
8.57
40. 75 ±0.3 8
274.7
-18
8.55
40.45
275.84
-17±2
8.59 ±0.05
40.68 ±0.1 9
275.77 ±1.19
Note: Bold values met the ABB, ASTM D3487, and ASTM D5222 specification values. Underlined values met the ABB and ASTM D3487
specification values. Italicized values met the ABB specification values. Data variability was calculated at 95% confidence using a
two-tailed T-test and assuming a normal distribution.
*The values listed are based on the results for two samples except for the viscosity at 100°C and 0°C where only one sample was analyzed.
**Due to large variations between sample results analyzed at different points in time for the same lot, the lower impulse voltages
(averaging around 133 kV) were assumed to be correct as a conservative assumption.
***These values are based on the results for two samples except for the composite lot values where only one sample was analyzed.
Acronyms and Abbreviations:
— = No value provided in the specification for this parameter
ABB = Virgin product specification for BIOTEMP® developed by ABB, Inc.
ASTM D3487 = American Society for Testing and Materials (ASTM) standard specification for mineral insulating oil used in electrical
apparatus.
ASTM D5222 = ASTM standard specification for high fire-point electrical insulating oil.
cSt = centistokes
kV = kilovolt
mgKOH/g = milligrams of potassium hydroxide per gram
N/A = Not applicable due to the differences in physical and chemical characteristics between BIOTEMP® and mineral oil and high temperature
hydrocarbon oil.
ppm = parts per million
June 2002
Vll
VS-R-02-03
-------
Although the oxidation stability test method states there is no correlation between the fluid's
performance in the test and its performance in service, the test is used to evaluate oxidation
inhibitors and to check the consistency of oxidation stability for a particular fluid.
The in-service transformer sample results are presented in Table 2. All four in-service transformer
samples had dissipation factors and water contents below the maximum value listed for the IEC
1203 specification. All four in-service transformer samples had conductivity values higher than the
minimum ABB specified value. The higher results listed for sample INS-07 relative to the other
samples may be due to the extreme operating conditions (e.g., overloads) the transformer was
subjected to as part of ABB's ongoing research project.
Table 2. Summary of In-service Transformer Sampling Results
Performance Parameters
Dissipation Factor @ 25°C (%)
Water Content (ppm)
Conductivity @ 25 °C (pS/m)
Specification Standards
ABB
<0.05
<150
<2.0
IEC 1203
<0.8
<400
-
Sampling Results
INS-01
0.13
15
16.17
INS-02
0.088
12
11.5
INS-03
0.082
16
8.51
INS-07
0.252
78
24.65
Note: Underlined values met both ABB and IEC 1203 specification values. Italicized values met either IEC 1203 or ABB
specifications.
1 . Samples INS-0 1 , INS-02, and INS-03 collected from transformers owned by PG&E
2. Sample INS-07 collected from a transformer owned by ABB which is used for testing BIOTEMP®
conditions.
Acronyms and Abbreviations:
ABB = Virgin product specification for BIOTEMP® developed by ABB, Inc.
under extreme operating
IEC 1203 = International Electrochemical Commission (IEC) specification for Synthetic Organic Esters for Electrical
Purposes - Guide for Maintenance of Transformer Esters in Equipment.
ppm = parts per million
pS/m = picosiemens per meter
Aquatic Biodegradabilitv. The average biodegradability of BIOTEMP® was 99% ± 3% after 21
days. The average biodegradation rates for BIOTEMP® and mineral oil based on literature data
are presented in Table 3.
Table 3. Aquatic Biodegradation Results
Compound
BIOTEMP®
Mineral oil
Biodegradation Rates
ABB ETV1
99% ± 3% after 21 days
—
Universite de Liege2
—
70% after 40 days
CONCAWE3
—
28% after 28 days
USAGE45
—
42-49% after 28 days
'U.S. EPA, Environmental Technology Verification Report ABB Inc. BIOTEMP* Vegetable Oil-Based Insulating
Dielectric Fluid, 2001.
Cloesen,C. & Kabuya, A, Research RWN° 2174 Physical and chemical properties of environment friendly
lubricants, no date.
Conservation of Clean Air and Water-Europe (CONCAWE), Lubricating Oil Basestocks, pp. 20-22, June 1997.
4U.S. Army Corps of Engineers (USAGE), Engineering and Design Environmentally Acceptable Lubricating Oils,
Greases, and Hydraulic Fluids, April 1997.
5USACE, Engineering and Design Environmentally Acceptable Lubricating Oils, Greases, and Hydraulic Fluids,
February 1999.
June 2002
Vlll
VS-R-02-03
-------
Based on the information above, the virgin BIOTEMP® fluid appears to biodegrade more readily
than mineral oil. Although BIOTEMP® readily biodegrades per this test, releases to water should
be prevented. The product's ability to degrade in the environment is dependent on site-specific
factors such as climate, geology, moisture, pH, temperature, oxygen concentration, dispersal of
oil, presence of other chemicals, soil characteristics, nutrient quantities, and populations of
various microorganisms at the location.
Flammability. The flash and fire point for the virgin and in-service fluid were consistently above
the minimum values listed in the ASTM D3487, D5222, and ABB performance specification
presented in Table 4. The fire point results obtained also agreed with values reported by the
Factory Mutual Research Center (FMRC) and Underwriters Laboratories (UL). The flash point
results agreed with the values reported by FMRC but were higher than the values reported by the
UL due to the different ASTM method used.
Table 4. Flash and Fire Point Results for Virgin and In-Service Samples
Product Lot No./
Transformer SN
Flash Point (°C)
Specification Criteria
ABB
ASTMD3487
ETV
Result
Fire Point (°C)
Specification Criteria
ABB
ASTMD5222
ETV
Result
Virgin Product
2000-216
2000-224
composite
Average
>300
>300
>300
>300
>145
>145
>145
>145
329 ±4
331 ±5
337
331 ±3
>300
>300
>300
>300
304-310
304-310
304-310
304-310
361 ±3
360 ±3
360
360 ±1
In-service Transformer Fluid
ISFR3-01
ISFR3-02
ISFR3-03
ISFR3-06
>300
>300
>300
>300
>145
>145
>145
>145
330
334
334
328
>300
>300
>300
>300
304-310
304-310
304-310
304-310
362
364
362
362
Note: Data variability was calculated at 95% confidence using a two-tailed T-test assuming a normal
distribution.
SN = Sample Number
• Acute Toxicity. The average LC^n for virgin BIOTEMP® was less than 250 mg/L. This low LC50
value is thought to reflect the physical impacts on fish due to oil coating the gills and preventing
oxygen exchange. The average LC50 indicates the spent (or waste) BIOTEMP* fluid may exhibit
a hazardous characteristic when tested under California regulations (California Code of
Regulations, Title 22, Section 66261.24(a)(6)). This determination is based on a limited set of
data for the virgin product and may not apply in states other than California where hazardous
waste criteria and test methods may differ. End-users should characterize their spent
BIOTEMP® fluid at the time of disposal since changes to the oil may occur due to use, storage,
or age. End-users should also consult their appropriate local, state, or federal regulatory
authority on applicable waste characteristic definitions and available disposal options.
• Chemical Composition. Virgin BIOTEMP® samples contained 80.1% ± 0.3% oleic acid, 10.5%
± 0.1% diunsaturated fatty acids, 0.3% ± 0.0% triunsaturated fatty acids, and 9.2% ± 0.2%
saturated fatty acids which agree closely with the formulation. The in-service transformer
samples contained 79.5% to 84.4% oleic acid, 5.3% to 10.7% diunsaturated fatty acids, 0.2% to
0.3 % triunsaturated fatty acids, and 9.5% to 10.0% saturated fatty acids. Other tentatively
identified compounds were TBHQ, 2-isopropyl-l,4-benzenediol,
2,3-dihydro-2-methyl-5-phenyl-benzofuran, 2-isopropyl-l,4-benzoquinone,
June 2002 ix VS-R-02-03
-------
p,p'-dioctyldiphenylamine, beta-sitosterol, squalene, and vitamin E. Metals were not detected in
the in-service transformer samples except for one sample, which had a zinc concentration of 2.3
mg/kg. For the virgin samples, copper ranged from non-detect to 4.13 mg/kg, barium ranged
from non-detect to 0.32 mg/kg and zinc ranged from non-detect to 2.02 mg/kg.
The phenolic antioxidant content was between 3,207 mg/kg ±103 mg/kg for the virgin
BIOTEMP® fluid and between 2,990 and 3,600 mg/kg for the in-service transformer samples.
Variations observed in the antioxidant content may be due to the varying quantities of
antioxidant added by ABB's off-site blender.
• Worker Health and Safety. Based on the MSDS information from the Vermont Safety
Information Resources, Inc. (SIR!) MSDS archive, BIOTEMP* appears to have personal
protective equipment (PPE) requirements similar to select mineral oil-based transformer fluids
but less stringent when compared to select silicone oil-based transformer fluids. BIOTEMP® has
a slightly higher nuisance particulate permissible exposure level (PEL) than mineral oil based on
the OSHA PEL for an 8-hour TWA exposure. In California, the nuisance particulate PEL is 10
mg/m3. BIOTEMP® also contains no IARC confirmed carcinogens. Some mineral oil-based
transformer fluids contain a light naphthenic petroleum distillate that has been identified by the
IARC as a confirmed carcinogen. Although BIOTEMP® appears to contain ingredients that
cause less serious health effects, the end-user must comply with all applicable worker health and
safety regulations for use of this product.
• Cost Comparison. The initial purchase cost of a new transformer unit containing BIOTEMP®
costs approximately 1.25-1.30 times more than that of a comparable mineral oil transformer.
When comparing the price per gallon of BIOTEMP® to mineral oil, the difference may be
between $4 to $9 depending on the volume purchased. Based on historical accelerated aging test
results, the estimated life expectancy of a BIOTEMP® transformer is estimated to be 20 years,
which is a comparable to mineral oil-based transformers.
Results for this verification/certification show that the ABB Inc. BIOTEMP® Vegetable Oil-Based
Dielectric Fluid is a biodegradable, vegetable oil-based dielectric fluid with a flash and fire point above
300°C. The product has dielectric breakdown voltages comparable to mineral oil and high temperature
hydrocarbon oil. The product may have varying amounts of antioxidants based on past and current
oxidation stability results. BIOTEMP® samples from in-service transformers had flash and fire points
above 300°C, and only one sample showed signs of oil degradation due to extreme operating conditions.
LC50 results indicate the spent BIOTEMP® may exhibit a hazardous characteristic per California's
hazardous characteristic definition but this is based on limited data for the virgin product. The end-user
should characterize the spent BIOTEMP® at the time of disposal since changes may occur to the oil due
to use, storage, or age.
Although BIOTEMP® is a vegetable oil-based product, end-users are still subject to the federal oil
pollution prevention regulations under 40CFR112. End-users should contact their appropriate local,
state, or federal regulatory authority regarding the management of BIOTEMP* (virgin and spent), and
BIOTEMP® spills.
June 2002 x VS-R-02-03
-------
Original signed by: Original signed by:
E. Timothy Oppelt 6/10/02 Kim F. Wilhelm 6/5/02
E. Timothy Oppelt Date Kim F. Wilhelm, Acting Chief Date
Director Office of Pollution Prevention
National Risk Management Research Laboratory and Technology Development
Office of Research and Development Department of Toxic Substances Control
United States Environmental California Environmental Protection Agency
Protection Agency
NOTICE: Verifications are based on an evaluation of technology performance under specific,
predetermined criteria and the appropriate quality assurance procedures. EPA and Cal/EPA make no
expressed or implied warranties as to the performance of the technology. The end-user is solely
responsible for complying with any and all applicable federal, state, and local requirements. Mention of
commercial product names does not imply endorsement.
June 2002 xi VS-R-02-03
-------
Availability of Verification Statement and Report
Copies of the public Verification Statement and
Verification Report are available from the following:
1 U.S. EPA
Web site: http://www.epa.gov/etv/library.htm (electronic copy)
2. Department of Toxic Substances Control
Office of Pollution Prevention and Technology Development
P.O. Box 806
Sacramento, California 95812-0806
Web site: http://www.dtsc.ca.gov/sciencetechnology/etvpilot.html
http: //www. dtsc. ca.gov/sciencetechnology/techce rt_index .html
or http://www.epa.gov/etv (click on partners)
(Note: Appendices are not included in the Verification Report
and are available from DTSC upon request.)
June 2002 xii VS-R-02-03
-------
TABLE OF CONTENTS
Notice ii
Foreword iii
Acknowledgment iv
Verification Statement v
Glossary of Terms xviii
Section 1. Introduction 1
Section 2. Description of Technology 4
Section 3. Field Sampling Verification Objectives 6
Section 4. Verification Activities and Results 7
4.1 Verification Activities 7
4.2 Results: Objective 1, General Performance 11
4.3 Results: Objective 2, Aquatic Biodegradability 19
4.4 Results: Objective 3, Flammability 21
4.5 Results: Objective 4, Acute Toxicity 23
4.6 Results: Other Verification/Certification Objectives 25
Section 5. Regulatory Considerations 31
5.1 Regulation of Virgin BIOTEMP® Dielectric Fluid 31
5.2 Waste Characterization/Disposal Requirements 32
5.3 Spill Management 34
Section 6. Conclusions 36
6.1 Objective 1, General Performance 36
6.2 Objective 2, Aquatic Biodegradability 36
6.3 Objective 3, Flammability 36
6.4 Objective 4, Acute Toxicity 37
6.5 Other Verification/Certification Objectives 37
Section 7. Vendor's Comment Section 39
References 40
June 2002
xill
-------
TABLES
Table 1. Summary of 1992 PCS Waste Generation - Electric Utility 2
Table 2. BIOTEMP® Samples and Analyses 8
Table 3. Equipment Information on Sampled Transformers 10
Table 4. Performance Results for Virgin BIOTEMP® 12
Table 5. Performance Results for In-Service BIOTEMP® Samples 17
Table 6. Aquatic Biodegradability Results 19
Table 7. Flash Points for Virgin and In-service BIOTEMPR Samples 21
Table 8. Fire Points for Virgin and In-service BIOTEMP® Samples 22
Table 9. Fish Bioassav Results for Virgin BIOTEMP® Samples 23
Table 10. AOAC Results for Virgin BIOTEMP® Samples 25
Table 11. AOAC Results for In-service BIOTEMP® Samples 26
FIGURES
Figure 1. Transformer Cross Section 4
Figure 2. Transformer Core 4
Figure 3. Drum Sampling 9
Figure 4. Tank Sampling 9
Figure 5. Flushing Sampling Port 10
Figure 6. Transformer Sampling 10
Figure 7. Trends for In-Service Transformer Parameters 18
June 2002
xiv
-------
APPENDICES
Appendix A: ABB Field Test Results
Appendix A-l: Select ASTM Test Methods: Objective 1
Appendix A-2: Aquatic Biodegradability Test Method: Objective 2
Appendix A-3: Flammability ASTM Test Method: Objective 3
Appendix A-4: Acute Toxicity Test Method: Objective 4
Appendix A-5: Worker Health and Safety Assessment: Other Verification/Certification
Objectives
Appendix A-6: AOAC Test Methods, EPA Test Method 8270 (SVOCs), and EPA Test
Appendix B: ABB Field Test Plan
Technology Evaluation WorkPlan (ABB), April 2, 2001; Department of Toxic
Substances Control, Office of Pollution Prevention and Technology Development.
Note: Appendices are not included in the Verification Report and
are available upon written request to DTSC at the following address:
Department of Toxic Substances Control
Office of Pollution Prevention and
Technology Development
P.O. Box 806
Sacramento, California 95812-0806
June 2002 xv
-------
List of Abbreviations and Acronyms
Qcm ohm-centimeter
ABB ABB,Inc.
ANSI American National Standards Institute
AOAC Association of Analytical Chemists
ASTM American Society of Testing and Materials
BHA butylated hydroxy anisole
BHT 3,5-di-tert-butyl-4-hydroxytoluene
EC degrees Celsius
C-H carbon-hydrogen bond
CAA Clean Air Act
CAS Chemical Abstracts Service
CCR California Code of Regulations
CEC Coordinating European Council
CFR Code of Federal Regulations
CH2 ethyl
CH3 methyl
CC>2 carbon dioxide
CONC AWE Conservation of Clean Air and Water-Europe
cSt centistokes (millimeter squared per second or mm2/s)
CWA Clean Water Act
DI deionized
DL detection limit
DISC California Department of Toxic Substances Control
EPA United States Environmental Protection Agency
EPCRA Emergency Planning and Community Right-to-Know Act
ETV Environmental Technology Verification
FDA Food and Drug Administration
FMRC Factory Mutual Research Center
FRP facility response plan
g gram
HML Hazardous Materials Laboratory
HSC Health and Safety Code
HTH high temperature hydrocarbons
IARC International Agency for Research on Cancer
LEG International Electrochemical Commission
IEEE Institute of Electric and Electronic Engineers
IR infrared spectroscopy
KOH potassium hydroxide
kPa kilopascals
KV or kV kilovolts
kVA kilovolt amperes
lethal concentration for 50% of the test population
50 lethal dose for 50% of the test population
June 2002 xvi
LD
-------
mg/kg milligrams per kilogram
mg KOH/g milligrams of potassium hydroxide per gram
mg/L milligrams per liter
ml milliliter
mmHg millimeters of mercury
MSDS material safety data sheet
MVA megavolt amperes
NIOSH National Institute for Occupational Safety and Health
NRMRL National Risk Management Research Laboratory
OPPTD Office of Pollution Prevention and Technology Development
ORD U.S. EPA Office of Research and Development
OSHA Occupational Safety and Health Administration
PCBs polychlorinated biphenyls
PEL permissible exposure limit
PG&E Pacific Gas and Electric
PPE personal protective equipment
ppm parts per million
psi pounds per square inch
pS/m picosiemens per meter
RCRA Resource Conservation and Recovery Act
SIRI Safety Information Resources, Inc.
SOP standard operating procedure
SPCC Spill Prevention, Control, and Countermeasures
SVOCs semivolatile organic compounds
TBHQ mono-di-tert-butyl hydroquinone
TCLP toxicity characteristic leaching procedure
TSCA Toxic Substances Control Act
TWA time weighted average
UL Underwriter Laboratories
USAGE U.S. Army Corps of Engineers
U.S. EPA United States Environmental Protection Agency
WET waste extraction test
June 2002
xvn
-------
Dielectric breakdown
(gap)
Glossary of Terms
The dielectric breakdown voltage indicates the fluid's ability to
resist electrical breakdown at a power frequency of 60 Hz and is
measured as the minimum voltage required to cause arcing
between two submerged electrodes in the fluid spaced 1.0 mm or
2.0 mm apart. This method is considered more sensitive to the
adverse effects of moisture in the oil in insulating systems.
Dielectric breakdown
(impulse)
The impulse dielectric breakdown voltage indicates the fluid's
ability to resist electrical breakdown under transient voltage
stresses such as lightning and power surges and is measured as the
voltage required to cause arcing between a submerged point and
sphere electrode.
Dielectric breakdown
(minimum)
The dielectric breakdown voltage at a 60 Hz test voltage indicates
the fluid's ability to resist electrical breakdown and is measured as
the minimum voltage required to cause arcing between two
submerged electrodes in the fluid. This test is recommended for
acceptance tests on virgin product.
Dissipation Factor
(maximum)
The fluid's dissipation factor is a measure of the dielectric losses in
the fluid. A low dissipation factor indicates low dielectric losses
and a low concentration of soluble, polar contaminants.
Diunsaturated fatty
acids
Fatty acids consisting of several carbons with 2 double carbon
bonds (e.g., C18:2).
Flash point
The lowest temperature corrected to a barometric pressure of 101.3
kPa (760 mmHg) at which application of a test flame causes the
vapor of a specimen to ignite.
Fire point
The lowest temperature at which the fluid will sustain burning for
5 seconds.
Kinematic viscosity
The measure of the time for a volume of liquid to flow under
gravity through a calibrated glass capillary viscometer.
Linoleic acid
A diunsaturated acid found as a triglyceride in high oleic oils.
has 18 carbons and 2 carbon-carbon double bonds (C18:2).
It
June 2002
xvill
-------
Linolenic acid
A triunsaturated acid found as a triglyceride in high oleic oils. It
has 18 carbons and 3 carbon-carbon double bonds (C18:3).
Monounsaturated
fatty acids
Fatty acids consisting of several carbons with 1 carbon-carbon
double bond (e.g., CIS: 1).
Neutralization
number
This number is a measure of the acidic or basic substances in the
oil and is used as a quality control indicator. An increase in the
value of the neutralization number may indicate degradation of the
oil due to increased water content. This value is measured by
dissolving the oil sample in a mixture of toluene, isopropyl
alcohol, and a little water. A color indicator, />-naphtholbenzein, is
added to this mixture and then titrated with potassium hydroxide
until an orange (acid) or green-brown (base) color change occurs.
Oleic acid
A monounsaturated acid found as a triglyceride in many natural
oils such as sunflower, olive, and safflower oil. This compound
has 18 carbons with one carbon-carbon double bond (C18:l).
Oxidation stability
This value measures the amount of sludge and acid products
formed by the oil under accelerated aging conditions. The oil
sample is oxidized in a 110°C bath containing a copper catalyst
coil. Oxygen is bubbled through duplicate specimens for 72 and
164 hours, respectively. At the end of each period, the amount of
sludge and acid formed are measured. The sludge results are
expressed as a percentage, which is calculated by dividing the
weight of sludge formed by the weight of the oil sample. The acid
content is determined by titrating the filtered solution containingp-
naphtholbenzein (a color indicator) with potassium hydroxide
(KOH). The acid content is expressed in the units of milligrams of
KOH per grams of oil (mg KOH/g).
Polar contaminant
A polar contaminant in a dielectric fluid ionizes and imparts
electrical conductivity to the solution. Examples of polar
contaminants in dielectric fluids include water, dirt, and metals.
June 2002
xix
-------
Polyunsaturated fatty
acids
Fatty acids consisting of diunsaturated and triunsaturated fatty
acids (i.e., several carbons with 2 or more carbon-carbon double
bonds, respectively such as C18:2, C18:3)).
Pour Point
The lowest temperature at which the movement of the oil is
observed. An average electrical power distribution application will
require a dielectric fluid to have a pour point below -20°C.
Rotary Bomb
Oxidation Stability
The time measured for the oil to react with a given volume of
oxygen. The oil is placed in a copper vessel (bomb) with a glass
sample container and exposed to oxygen at an initial pressure of 90
psi. The bomb is placed in a 140°C bath and agitated until a
specific pressure drop occurs in the bomb. The time that elapses
between the start of the experiment and the pressure drop is
measured and recorded. This method is designed to evaluate the
oxidation stability of new mineral oil containing 2,6-tertiary-butyl-
para-cresol or 2,6-ditertiary-butyl phenol from shipment to
shipment. According to the method, the method's applicability for
inhibited insulating oils with a viscosity greater than 12 centistokes
(cSt) at 40°C has not been determined.
Stearic acid
A saturated acid found as a triglyceride in high oleic oils. It has 18
carbons and no double carbon bonds (C18:0).
Triunsaturated fatty
acids
A triunsaturated acid found as a triglyceride in high oleic oils
consisting of several carbons with 3 carbon-carbon double bonds
(i.e., C18:3).
Water content
The measure of the presence of water in oil expressed as a
concentration (ppm). Water in the insulating oil will increase the
breakdown rate of fatty acid esters in the vegetable oil base and
leads to the formation of polar contaminants. This breakdown rate
is proportional to the amount of water available for the reaction.
An indicator of such reactions is a significant increase in the value
of the neutralization number due to the increased acidity of the
fluid. Compared to conventional mineral oils, vegetable oils have
a much higher water content saturation point, typically well over
500 ppm at room temperature. Five to 10% of the saturation level
(25 to 50 ppm) is the recommended range for vegetable oil after
processing.
June 2002
xx
-------
Section 1. Introduction
Background
Electric power utilities use electrical transformers for a variety of applications, including power
distribution. The transformers generate significant amounts of heat, and must contain
cooling/insulating (dielectric) media to prevent gas formation, electrical shorts, fire or explosion,
and transformer damage. Most transformers currently use some type of mineral oil as the cooling
fluid; however high temperature hydrocarbons (HTHs) and synthetics (less-flammable fluids) are
used in transformers that must operate in safety-related applications (near or inside buildings).
Recently, transformer fluid vendors have developed vegetable seed oil-based dielectric fluids.
These fluids have been certified as meeting "less-flammable" safety-related requirements by
organizations such as Underwriters Laboratories or Factory Mutual Research Corporation.
Typically, liquid-containing distribution class transformers store from 30 to 1,000 gallons of oil.
Spills from transformers are potentially an environmental concern because even small amounts
of oil can contaminate bodies of water, possibly deplete oxygen, coat plant and animal life, be
toxic or form toxic products, affect breeding, produce rancid odors, or foul shorelines or other
habitats. Effects on soils are not as well characterized.
Polychlorinated Biphenyls (PCBs) are still in use but no longer produced because of their high
toxicity - they are regulated under the federal Toxic Substances Control Act (TSCA). According
to Title 40 Code of Federal Regulations Section 261.8 (40CFR 261.8), dielectric fluids and
electric equipment with dielectric fluids regulated under TSCA are not regulated under the
federal Resource Conservation and Recovery Act (RCRA). Non-PCB transformer fluids do not
meet the requirements for regulation as hazardous waste under RCRA; however, mineral oils that
have been in service for approximately 10 years have exceeded California's acute toxicity levels
for copper due to leaching from the transformer coils.
Facility owners and operators that handle, store, or transport oils (e.g., petroleum oils, vegetable
oils, animal fats, etc.) are required to report an oil spill, which "may be harmful to the public
health or welfare, or environment". A reportable oil spill is defined as one that either (1) violates
water quality standards, (2) causes a sheen or discoloration on the surface of a body of water, or
(3) causes a sludge or emulsion to be deposited beneath the surface of the water or on adjoining
shorelines. The oil spill must be contained, cleaned up, and reported to the National Response
Center, the federal point of contact for all chemical and oil spills.
Table 1 illustrates the types and amounts of waste oil change-outs, spills, and associated clean-up
costs that a small to medium-sized electrical utility transmission system monitoring and
maintenance facility experienced in 1992. This facility, which is only one of several operated by
the electrical utility, generated 155 tons of spilled oil and contaminated soil, most of which was
caused by accidents involving utility poles and transformers.
June 2002
-------
Table 1. Summary of 1992 PCB Waste Generation - Electric Utility
Waste Generated Annual Quantity Annual
Generated (tons) Costs ($)
Oil Spill and Leak
Residue 155 46,000
Source of Waste: Primarily damage to transformers
Waste Oil from
Electrical Transformers 126 100,000
Source of Waste: Draining of oil prior to reconditioning or decommissioning transformers
Wastes Containing PCB 28 50,000
Source of Waste: Primarily damage to transformers and PCB recovery
Source: U.S. EPA, Risk Reduction Engineering Laboratory, EPA/600/S-92/063 - October 1992
BIOTEMP® Dielectric Insulating Fluid
ABB Inc. (ABB) has developed a dielectric insulating fluid called BIOTEMPR which is
comprised of >98.5% vegetable oil and <1.5% antioxidants and optional color additives.
BIOTEMP® is used in liquid-filled electrical transformers to act as an electrical insulating
medium, and to transport heat generated in the transformer around the windings, core and
connected circuits to cooling surfaces. BIOTEMP® is currently used in pole-mounted,
distribution, network, and small power transformers with a voltage rating <69 kV and a
maximum kVA rating of 20 MVA. Approximately 250 transformers supplied with BIOTEMP®
fluid are in-service. Customers that use this product include Pacific Gas & Electric, Boston
Edison, Seattle City Light, Montana Power, American Electric Power, Empire District Electric,
Southern Company Services, Carolina Power & Light, Arco Alaska, Hawaiian Electric, Cone
Mills and US Gypsum.
Evaluation Approach
The BIOTEMP® evaluation was designed to provide the data necessary to draw conclusions on
the fluid's performance, chemical composition, toxicity, and safety. The evaluation included a
review of supporting documents, information, and laboratory data submitted by ABB, and field
sampling to provide independent data on the technology's performance, chemical composition,
and toxicity.
The field sampling was conducted at ABB's manufacturing facility in South Boston, Virginia and
at Pacific Gas and Electric's (PG&E) in-service transformers in San Francisco, California.
PG&E is an ABB customer and agreed to provide staff and access to three in-service
transformers as part of the field sampling activities. Prior to the field sampling, the Department
June 2002 9
-------
of Toxic Substances Control staff (DISC) prepared a Technology Evaluation Workplan
(Workplan) to identify specific field objectives, data quality objectives, testing procedures, and
roles and responsibilities. ABB assumed overall responsibility for obtaining access to all
locations where field sampling was conducted. DTSC staff provided independent oversight and
was present to observe all field sampling activities. The agreed-upon Workplan specified that
DTSC would maintain a record of all samples collected, and record all measurements and
observations made during sampling.
The oldest transformer in-service using BIOTEMP® as the dielectric insulating fluid is 2.5 years
old. Since the technology is still new, no data was available to assess the long-term transformer
performance and waste characteristics of BIOTEMP® fluid at the end of its service life.
According to ABB, the service life is expected to be in the range of 20 years.
June 2002
-------
Section 2. Description of Technology
BIOTEMPR, developed by ABB Inc., is a vegetable oil-based dielectric fluid comprised of
greater than (>) 98.5% vegetable oil and less than (<) 1.5% antioxidants. The product may use
up to three different antioxidants to prevent unsaturated bonds in the oil from polymerizing with
oxygen. The vegetable oil used in BIOTEMPR is manufactured off-site in a four-step process:
crushing and refining, bleaching, deodorizing, and winterizing. The oil is extracted from crushed
seeds using a solvent such as hexane. As part of the bleaching process, the oil is subject to a clay
treatment to remove polar contaminants. Next, the oil is deodorized using steam distillation to
remove unwanted volatile compounds. The last step, winterizing, involves chilling the oil to
remove excessive saturates. In the past, the vegetable oil and antioxidants were blended at a
contract blending facility per ABB's product specifications. ABB is currently using blending
equipment at ABB's South Boston, Virginia facility to oversee and control this portion of the
process.
BIOTEMP® is used in liquid-filled electrical transformers as an electrical insulating medium. An
example of a 3-phase transformer is presented in Figure 1. The main parts of a transformer are
the core, the windings, the tank containing the core and windings, and the cooling system. The
core is made of thin steel sheet laminates, which are coated, with an oxide film to insulate the
sheets from each other. Two distinct sets of coils called windings are placed upon the core at a
suitable distance from each other. These windings consist of wire insulated with a paper
covering. An example of a three-phase transformer core is presented in Figure 2. When the
transformer is in-service, the oil and core expands and contracts as the heat generated by the
transformer windings varies with the load. As the oil becomes heated, the hot oil rises to the top
of the transformer where heat is dissipated to the outside, and then moves along the case to the
bottom. Fins are sometimes attached to deflect moving air against the case and to increase the
cooling area. Overheating the core can lead to damage and overheating the windings can cause
the insulation to deteriorate, which reduces the life of the transformer.
Figure 1. Transformer Cross Section
Figure 2. Transformer Core
June 2002
-------
For large power transformers, the tops of the tanks are designed to have a nitrogen gas seal to
prevent the oil from oxidizing with the air. The expansion of the oil reduces the volume of the
nitrogen gas causing the gas pressure to be greater during power load periods. Large
transformers may also use radiators, fans, circulating pumps or cooling water to increase heat
exchange.
June 2002
-------
Section 3. Verification Objectives
The verification/certification objectives were to verify the applicant's technology performance
claims listed below.
Verification/Certification Claim #1 - General Performance
• In the following composition ratio (98.5% vegetable oil, 1.5% additives), BIOTEMP®
meets criteria for oxidative, thermal, and chemical stability, as measured by Oil
Qualification Tests - ASTM D3487 (Mineral Oil) and ASTM D5222 (High Temperature
Hydrocarbons).
Verification/Certification Claim #2 - Aquatic Biodegradability
• BIOTEMP® biodegrades 97% in 21 days, based on the average of several performance
tests as measured by the Coordinating European Council (CEC) Test Method CEC-L-33-
A-93.
Verification/Certification Claim #3 - Flammability
• BIOTEMP® has a Flash Point of at least 300°C, and a minimum Fire Point of 300°C,
based on the average of several performance tests as measured by ASTM D92 (Cleveland
Open Cup).
Verification/Certification Claim #4 - Acute Toxicity
• The virgin BIOTEMP® product passes the aquatic toxicity characteristic criterion
specified in the Code of California Regulations, Title 22, Section 66261.24(a)(6) based
on U.S. EPA/600/4-90/027F Test for Acute Toxicity of Effluents and Receiving Waters
to Freshwater and Marine Organisms.
Other Verification/Certification Tests:
• Verify that BIOTEMP® consists of >98.5 % vegetable oil and <1.5% antioxidant and
color additives, and that the formulator is meeting selected ABB purchase specifications.
• Establish a baseline for measuring potential metals leaching and oil degradation of
BIOTEMP® under electrical loading over time.
• Evaluate the worker health and safety aspects of BIOTEMP®.
• Estimate expected lifetime costs of BIOTEMP® as compared to mineral oil.
June 2002
-------
Section 4. Verification Activities and Results
4.1 Verification Activities
4.1.1 Field Sampling
Prior to sampling, DISC developed a technology evaluation workplan, which described the
sample collection procedures and analyses to be performed. A copy of the technology evaluation
plan is included in Appendix B. To ensure independent and representative samples were
collected, DTSC personnel oversaw sample collection in the field of virgin product and in-
service transformers. Samples were assigned a field sample identification number, which was
determined prior to sampling. Table 2 lists the samples collected and the analysis performed as
part of this verification/certification. Proper chain of custody and storage procedures were
followed.
Virgin Product
Samples of virgin fluid were collected at ABB's manufacturing facility in South Boston,
Virginia. Three different lots were sampled by an ABB representative with DTSC oversight. A
total of 12 samples (four samples per lot) were collected. Initially, three samples from each lot
were analyzed for SVOCs, metals, acute toxicity, aquatic biodegradation, and select AOAC and
ASTM methods. One duplicate was analyzed for SVOCs, metals, and select AOAC and ASTM
methods. Two matrix spikes and an equipment blank were analyzed for SVOCs and metals. A
field blank was analyzed for metals only. Six additional samples, consisting of two samples from
Lot 2000-216, three samples from Lot 2000-224, and one sample from the composite lot, were
analyzed by the ASTM methods listed in Table 2 to verify performance results reported by Doble
Engineering.
Samples from Lots 2000-216 and 2000-224 were collected from 55-gallon drums. Samples were
also collected from a 250-gallon holding tank used to store residual unused fluid from several
different lots (the composite lot). Barrel samples were collected using a glass Coliwasa. A new
glass Coliwasa was used at each new barrel sampled to reduce the potential of cross
contamination. The tank samples were collected at a sampling spigot located at the bottom of the
tank. The tank contents were not mixed prior to sampling. Approximately one pint of oil was
drained from the tank via the spigot prior to sampling. Sampling activities are presented in
Figures 3 and 4.
Virgin product samples collected as part of this verification/certification were from lots produced
by ABB's off-site blender. Since BIOTEMP® was blended off-site, ABB was not able to
continuously monitor the blending of antioxidants into the oil and make adjustments based on
atmospheric conditions such as humidity. Lots blended at ABB's South Boston facility were not
available for this sampling event since ABB was completing installation and testing of their on-
site blending equipment.
June 2002
-------
t®
Table 2. BIOTEMP Samples and Analyses
Sample ID
BIO-01
BIO-02
BIO-03
BIO-04
BIO-05
BIO-06
BIO-07
BIO-08
BIO-09
BIO-10
BIO- 11
BIO-12
BIO- 13
BIO- 14
INS-1
INS-2
INS-3
INS-4
INS-5
INS-6
INS-7
Lot No.
2000-216
2000-216
2000-216
2000-216
2000-224
2000-224
2000-224
2000-224
composite
composite
composite
composite
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
SVOCs
a
a
a
a
a
a
c
c
a
a
a
c
c
c
a
Metals
b
b
b
b
b
b
b
b
b
b
b
b
b
Acute
Toxicity
e
e
e
Aquatic
Biodegradation
d
d
d
AOAC
Methods
f
f
f
f
f
f
f
f
ASTM
Methods
g,h,i,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,i,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,i,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,j,k,l,m,
n,p,q,r
g,h,i,k,l,m,
n,p,q,r
g,o,r,s
g,o,r,s
g,o,r,s
g,o,r,s
Comments
Duplicate sample. Analyzed for ASTM
methods. Collected from same barrel as
BIO-01.
Duplicate sample analyzed for methods
marked.
Duplicate sample. Analyzed for ASTM
methods. Collected from same barrel as
BIO-03.
Matrix spike for metals and SVOC.
Analyzed for ASTM methods.
Duplicate sample. Analyzed for ASTM
methods. Collected from same barrel as
BIO-05.
Duplicate sample. Analyzed for ASTM
methods. Collected from same barrel as
BIO-07.
Matrix spike for metals and SVOC.
Analyzed for ASTM methods.
Duplicate sample not analyzed
Duplicate sample not analyzed
Field blank
Equipment blank
Field blank
Equipment blank
Equipment blank
The letter assigned to each sample corresponds to the analysis performed:
a - U.S. EPA Method, 8270 (SVOC screening) and prepared per U.S. EPA Method 3580
b - U.S. EPA Method 6010 (metals screening) and prepared per U.S. EPA Method 5030
c - U.S. EPA Method, 8270 (SVOC screening) and prepared per U.S. EPA Method 305 1
d - CEC-L-33-A-93, Biodegradability of Two-Stroke Cycle Outboard Engine Oils in Water
e -U.S. EPA Method 600/4-90/027F, Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and
Marine Organisms and prepared per the requirements in California Regulations, Title 22, Section 66261.24(a)(6), Static Acute
Bioassay Procedures for Hazardous Waste Samples.
f - AOAC Method 981.11, Oils andFats, AOAC Method 972.28, Total Fatty Acids in Oils andFats, AOAC Method 963 .22, Methyl
Esters of Fatty Acids in Oils andFats, AOAC Method 983.15, Phenolic Antioxidants in Oils, Fats, andButter, and AOAC Method
977.17, Polymers and Oxidation Products of Vegetable Oil.
g - ASTM Method D92, Hash and fire point n - ASTM Method D924, dissipation factor (25°C & 100°C)
h - ASTM Method D97, pour point o - ASTM Method D924, dissipation factor (25°C)
i - ASTM Method D445, kinematic viscosity (0, 40, & 100°C) p - ASTM Method D2440, oxidation stability
j - ASTM Method D445, kinematic viscosity (40°C) q - ASTM Method D21 12, oxidation stability
k - ASTM Method D877, dielectric breakdown (minimum) (rotary bomb)
1 - ASTM Method D1816, dielectric breakdown (gap) r - ASTM Method D1533, water content
m - ASTM Method D3300, dielectric breakdown (impulse) s - ASTM Method D4308, conductivity
June 2002
-------
Figure 3. Drum Sampling
Figure 4. Tank Sampling
In-Service Transformer
Samples of in-service fluid were taken from transformers that have been in use for at least one
year and part of a regular sampling/testing environment. Samples from the PG&E transformers
were collected by PG&E and ABB representatives under DTSC oversight and in conjunction
with PG&E's on-going sampling program. The sample from the ABB transformer was collected
by an ABB representative under DTSC oversight. Only one sample per transformer was
collected to minimize the amount of fluid removed from each transformer and the impact to the
ongoing test program. New Tygon tubing connectors were used at each transformer fluid
sampling port to reduce the potential of cross contamination.
The transformer pressure valve is checked to confirm the unit is under positive pressure prior to
sampling. A sampling syringe with Tygon tubing and a T-shaped sampling valve are attached to
the sampling port. The T-shaped sampling valve is set to allow oil to flow through a purge line,
which bypasses the sampling syringe. The sampling port valve is cracked open and oil is purged
through the Tygon tubing, sampling valve, and purge line. The sample bottles are filled after a
pint of oil has been purged through the line.
Four transformers were sampled; three owned by PG&E in San Francisco, California and one
owned by ABB in South Boston, Virginia. Two of the PG&E transformers were located in
underground vaults on Mission Street between First and Second Street. The other PG&E
transformer was located in an underground vault on Howard Street between Fremont Street and
Beale Street. The three PG&E transformers were in normal service. The ABB transformer was
used for testing BIOTEMP® under extreme operating conditions. Table 3 lists information on the
transformer type, size, and condition at the time of sampling. Transformer sampling activities
are presented in Figures 5 and 6.
June 2002
-------
Table 3. Equipment Information on Sampled Transformers
Owner
PG&E
PG&E
PG&E
ABB
Transformer Information
Type
3 -phase vault network
transformer
3 -phase vault network
transformer
3 -phase vault network
transformer
3 -phase RSL insulated
unit substation
Serial
Number
NAB4424-003T
NAB4424-004T
NAB4424-005T
PAO79 14-001
kVA
Rating
(kVA)
1,000
1,000
1,000
1,000
Primary
Voltage
(kV)
12,000
12,000
12,000
—
Secondary
Voltage
(kV)
480
480
480
—
Temp.
Rise
(°C)
65
65
65
65
Initial
In-Service
Date
March 2000
March 2000
March 2000
June 2000
Figure 5. Flushing Sampling Port
Figure 6. Transformer Sampling
4.1.2 Historical Data
DTSC staff also reviewed internal product development testing data provided by ABB. These
data were collected as part of ongoing testing for internal use by ABB prior to entry into the
verification/certification agreement. Historical data collected by independent testing facilities
under contract with ABB were also reviewed. These data provided background information on
the technology performance for past virgin lots and indicated trends on the fluid's performance in
tested transformers for select ASTM parameters.
June 2002
10
-------
4.2 Results: Objective 1, General Performance
For this verification/certification, BIOTEMPR was tested for select physical (e.g., pour point,
viscosity), chemical (e.g., water content, oxidation stability), thermal (e.g., flash and fire point)
and dielectric (e.g., dielectric breakdown, dissipation factor) properties to verify general
performance claims listed in ABB's product specifications. Since no standard suite of general
performance tests exist for vegetable oil-based dielectric fluids, two ASTM specifications
developed for mineral oils (ASTM D3487) and high temperature hydrocarbons (HTH) (ASTM
D5222) were used. These ASTM standards were selected because ABB claimed the dielectric
and oxidation properties for BIOTEMPR were similar to those for mineral oil and HTH fluid.
For the in-service transformer samples, results were compared to the International
Electrochemical Commission (IEC) 1203 specification for in-service synthetic organic esters
since BIOTEMP® has similar fluid characteristics to synthetic esters when in use. Doble
Engineering (Doble), an independent testing laboratory, tested virgin and in-service samples for
physical, chemical, thermal, and dielectric properties using the ASTM methods listed in Table 2.
The results for the thermal properties are discussed in Section 4.4. Results for the other
properties are discussed below.
4.2.1 Virgin Product Performance Results
Dielectric Properties (or Dielectric Strength)
Dielectric breakdown is the basic property used to evaluate a dielectric fluid's performance. The
dissipation factor is also used to evaluate a dielectric fluid's performance but this property may
vary between various dielectric fluids due to the chemical properties. Table 4 lists the test
results, and dielectric breakdown and dissipation factor specification standards for ASTM
D3487, D5222, and ABB that were used to evaluate BIOTEMP®'s electrical performance.
Dielectric Breakdown
Both the minimum and gap dielectric breakdown tests measure the minimum voltage
required to cause arcing between two submerged electrodes in a dielectric fluid. A low
dielectric breakdown value may indicate the presence of water, dirt, or other electrically
conductive particles in the oil, which may cause damage to the transformer core or
windings due to arcing. The dielectric breakdown values for the virgin BIOTEMPR
samples were higher than the lowest value specified for the minimum and 1.0 mm gap
dielectric breakdown voltages for all three specifications. Precision criteria are not
specified in ASTM Method D877 (minimum breakdown voltage) and ASTM Method
D1816 (gap breakdown voltage). Since BIOTEMPR' s dielectric breakdown values were
higher than the ABB and ASTM specifications, the fluid met these performance criteria
and would not likely cause damage to the transformer core or windings due to arcing.
June 2002 11
-------
Table 4. Performance Results for Virgin BIOTEMP11
Performance Parameters
Specification Standards
ABB
ASTM D3487
ASTM D5222
Sampling Results
Lot 2000-216
BIO-01
BIO-02
BIO-03
BIO-04
Average*
Lot 2000-224
BIO-05
BIO-06
BIO-07
BIO-08
Average*
Composite Lot
BIO-09
BIO-10
Average
Dielectric Properties
Dielectric breakdown (kV)
minimum
gap
impulse
>30
>28
>100
>30
>28
> 145
>42
>30
NA
Dissipation Factor (%)
@ 25 °C
(2j 100°C
<0.05
<2.0
<0.05
<0.3
<0.01
<0.3
48
34
134
45
39
220
48
37
130
43
38
224
46 ±4
37 ±3
177 ± 83
0.128
2.6
0.242
3.3
0.098
2.42
0.141
3.12
0.160 ±0.184
2.95 ±1.15
49
34
226
52
40
220
56
34
136
48
38
216
51 ±6
37 ±5
200 ± 68
0.015
0.74
0.017
1.08
0.029
0.636
0.025
0.89
0.022 ± 0.011
0.837 ± 0.307
55
36
214
54
41
132
0.031
1.06
0.025
0.801
55
39
173
0.028
0.931
Chemical Properties
Oxidation Stability
Percent Sludge (%)
after 72 hours
after 164 hours
<0.2
<0.2
<0.1
<0.2
Neutralization No. (mgKOH/g)
after 72 hours
after 164 hours
Rotary Bomb (minutes)
Water Content (ppm)
<0.2
<0.5
>200
<150
<0.3
<0.4
>195
<35
NA
NA
NA
NA
800-1,000
<25
0.02
0.06
0.01
0.01
0.02
0.04
0.01
0.01
0.02 ± 0.01
0.03 ± 0.04
0.17
21.68
120
93
0.23
20.51
117
62
0.19
21.98
120
76
0.18
20.34
115
70
0.19 ±0.04
21.13 ±1.31
118 ±4
75 ±21
0.02
0.01
0.01
<0.01
0.03
0.04
0.01
0.02
0.02 ± 0.015
0.02 ± 0.02
0.17
18.99
117
55
0.16
21.43
120
51
0.14
16.61
115
80
0.17
16.63
112
101
0.16 ±0.02
18.41 ± 3.66
116 ±5
72 ±37
0.02
0.01
0.01
0.03
0.16
17.61
117
98
0.16
14.44
115
106
0.02
0.02
0.16
16.02
116
102
Physical Properties
Pour Point (°Q
Viscosity (cSt)
@100°C
@ 40 °C
(gj 0°C
-15 to -25
-40
-24
<10
<45
<300
<3
<12
<76
11.5-14.5
100-140
1,800-2,200
-21
-15
-21 | -15
-18 ±6
—
40.55
276.42
—
41.04
-
8.61
40.38
276.1
—
40.96
-
8.61
40.73 ± 0.51
276.27
-15
-15
-21
-15
-17 ±5
—
41.08
-
—
40.71
-
8.57
40.5
274.7
—
40.71
-
8.57
40.75 ± 0.38
274.7
-15
-21
—
40.50
-
8.55
40.39
275.8
-18
8.55
40.45
275.84
Note: Due to the differences in the physical and chemical properties (e.g., water content) of BIOTEMP® versus mineral oil and high fire-point hydrocarbon insulating oils, the values listed under these headings are compared
the ABB specification values only. The ASTM specification values are provided as a reference to the reader.
*Data variability was calculated at 95% confidence using a two-tailed T-test assuming a normal distribution.
4$ronyms and Abbreviations:
-= g ample not tested for this parameter
BB = Virgin product specification for BIOTEMP® developed by ABB, Inc.
STM D3487 = American Society for Testing and Materials (ASTM) standard specification for mineral insulating oil used in electrical apparatus.
STM D5222 = ASTM standard specification for high fire-point electrical insulating oil.
t = centistokes
V = kilovolt
gKOH/g = milligrams of potassium hydroxide per gram
/A = Not available
m ....
= parts per million
ppm
June 2002
12
-------
The impulse dielectric breakdown test is designed to determine the minimum voltage to
cause arcing in the fluid under lightning or power surge conditions. The minimum
breakdown voltages the oil must exceed for use are listed for each specification in Table
4. Of the ten samples analyzed, six samples had voltages higher than the minimum
voltage listed under ASTM D3487. All ten samples exceeded the ABB minimum voltage
specification.
The ten samples listed in Table 4 were analyzed at two different points in time. Initially,
Doble analyzed samples BIO-01, BIO-03, BIO-07, and BIO-10. ABB later requested
Doble to analyze samples BIO-02, BIO-04, BIO-05, BIO-06, BIO-08, and BIO-09 to
verify results for certain parameters. The confidence interval for data obtained for the
two sets of samples were calculated separately. The two sets of data met the repeatability
criteria in ASTM Method D3300 of ± 13KV with a 95% confidence at ± 4.1KV and ±
4.8KV, respectively. When the data were combined into one data set, the 50%
confidence interval was ±10.0KV, which did not meet the method criteria of 5KV at
50% confidence for any two series of samples tested.
The percent difference between sample results collected from the same barrel and the
same lot but analyzed at different points in time was between 48% and 54% (i.e., BIO-01
and BIO-02, BIO-03 and BIO-04, BIO-07 and BIO-08). The percent difference for
samples BIO-05 and BIO-06, which were from the same barrel, the same lot and analyzed
at the same point in time, was 3%. ASTM D3300 does not require the test instrument to
be calibrated before and after testing using a calibration solution. Instead, the instrument
calibration is based on the five successive voltage readings obtained for the test fluid.
These large variations in the sample results from the same drum and lot suggest inherent
inaccuracies within the method and possible quality issues associated with Doble's
analyses.
A low impulse voltage might be due to a high contaminant or water content in the oil. A
high contaminant or water content might also cause a higher dissipation factor or lower
gap voltage. A direct correlation between these three values was not apparent from the
data in Table 4. Since the impulse voltages varied greatly for samples taken from the
same barrel and the same lot, it wasn't clear which results for BIOTEMP® were correct.
Assuming conservatively that the lower impulse voltages were correct, BIOTEMP® met
ABB's performance specification for the impulse dielectric breakdown voltage, but not
the ASTM D3487 specification.
Dissipation Factor
The dissipation factor is used to measure the dielectric losses to an insulating dielectric
fluid (such as oil) when it is exposed to an alternating electric field. For ASTM Method
D924, the dissipation factor is determined by passing an alternating electric current
through a test cell filled with dielectric fluid and measuring the capacitance with an
electronic bridge circuit. This value is used to control the product quality, and to
determine changes in the fluid due to contamination or degradation during use. A low
June 2002 13
-------
dissipation factor indicates a low dielectric loss and a low contaminant concentration
(e.g., dirt, water, or metals).
The dissipation factor for the four samples from Lot 2000-216 were much higher than the
six samples taken from Lot 2000-224 and the composite lot. Results indicate the four
samples from Lot 2000-216 did not meet any specification values at 25°C and 100°C.
None of the ten samples were found to meet the ASTM D5222 specification values.
BIOTEMP® met the ABB specification for the dissipation factor at 25°C and 100°C for
two out of three lots sampled. The same two lots also met the ASTM D3487
specification for the dissipation factor at 25°C. The high dissipation factors for Lot 2000-
216 indicate a higher contaminant concentration in this lot compared to the other lots.
This may be due to contaminants introduced while storing empty barrels or during the
barrel filling process. The higher dissipation factors at 100°C may also be due to the
higher water content of BIOTEMP® versus mineral oil and HTH.
Chemical Properties
Oxidation Stability
Oxidation stability was originally designed to assess the amount of sludge and acid
products formed in mineral transformer oils under specific test conditions. Good
oxidation stability minimizes the formation of sludge and acid in order to maximize the
service life of the oil. Oils that met the requirements specified for ASTM Method D2440
tend to minimize electrical conduction, ensure acceptable heat transfer, and preserve
system life. According to ASTM Method 2440, there is no proven correlation between
performance in this test and performance in service, since the test does not model the
whole insulation system (oil, paper, enamel, and wire). However, the test can be used as
a control to evaluate oxidation inhibitors and to check the consistency of the oxidation
stability of production oils.
The first oxidation stability tests on BIOTEMP® were performed per ASTM Method
D2440 over a 72-hour period (the 72 hour test). The percentages of sludge generated and
the neutralization numbers after 72 hours met both ABB and ASTM D3487
specifications. Data were within the precision criteria listed in ASTM Method D2440.
The difference between results at the 95% confidence level did not exceed 0.017% for the
generated sludge and 0.093 mg KOH/g for the neutralization number.
Oxidation stability tests were also performed on BIOTEMP® per ASTM Method D2440
over a 164-hour period (the 164 hour test). The percentage of sludge generated after 164
hours met the ABB and ASTM D3487 specifications for all samples. Sludge results were
within the precision criteria listed in ASTM Method D2440 and did not exceed 0.026% at
95% confidence. However, the neutralization number after 164 hours exceeded the
maximum value for the ABB and ASTM D3487 specifications for all samples. Doble's
chemist verified the results were correct and no unusual material formation was observed.
The confidence interval for the neutralization numbers at 95% confidence were
± 1.85 mg KOH/g that did not meet the precision criteria of ± 1.027 mg KOH/g.
June 2002 14
-------
Past oxidation stability tests performed by Doble Engineering reported the neutralization
number after 164 hours at 0.25 mg KOH/g. A yellow deposit formed at the top of the test
tube, which had a pH of 4.6 when dissolved in water. ABB had observed these deposits
during product testing but noted that they did not form on a consistent basis. The yellow
deposit was thought to be composed of acid crystals produced when the fluid degraded
due to heat and oxidation. ABB believed the deposit might be volatile acids associated
with the antioxidants used in BIOTEMP® (Lewand, 2001). The high neutralization
numbers reported for the samples listed in Table 4 may be due to the formation of these
acids in solution.
The oxidation stability of BIOTEMP® was also tested using ASTM Method 2112,
oxidation stability by rotating bomb (the rotary bomb test). The rotary bomb test was
developed as a rapid method for evaluating the consistency of the oxidation stability for a
new mineral oil between shipments. Results ranged between 112 and 120 minutes, which
were below the minimum ABB, ASTM D3487 and D5222 specification values. The
confidence limit at 95% for the data was ± 2.0 minutes, which met the precision criteria
in ASTM D2112 of ± 23 minutes at 95% confidence. Past testing performed by Doble
reported a rotary bomb test results of 162 minutes, which was less than ABB's
specification (Lewand, 2001) and D3487's specification.
BIOTEMP® met ABB's specifications and was comparable to mineral oil for the 72-hour
oxidation stability test, and for the percent of sludge generated using the 164-hour
oxidation stability test. The fluid did not meet ABB's oxidation stability specification for
the neutralization number using the 164-hour test or the rotary bomb test. BIOTEMPR
was not comparable to the HTH fluids per the rotary bomb test. As stated earlier, there is
no proven correlation between performance in this test and performance in service, since
the test does not model the whole insulation system. However, these tests indicate
possible inconsistencies in the addition and/or blending of antioxidants used in
BIOTEMP® due to the low results for the rotary bomb tests and the high neutralization
number for the 164-hour test.
Water Content
Water content is used by industry to monitor a dielectric fluid's quality and as an
indicator of possible oil deterioration, which could adversely affect the oil's electrical
properties such as dielectric breakdown. This value is based on the relative saturation of
the water in the dielectric fluid. The relative saturation is based on the amount of water
dissolved in the oil divided by the total amount of water the oil could hold at that
temperature. The dielectric strength of oil starts to fall when saturation reaches about
50%. For petroleum based dielectric oils, 50% saturation at room temperature is 30-35
mg/kg. Synthetic esters and vegetable oil contain about 500-600 mg/kg of water at room
temperature and 50% saturation. A water content at or near 50% saturation may indicate
the oil has deteriorated and may cause a lower dielectric breakdown voltage, which can
damage the transformer core and windings.
June 2002 15
-------
Water content was tested by Doble using ASTM Method D1533, water in insulating
liquids. The water contents for all samples were below the maximum value listed for the
ABB specification of 150 ppm. No precision criteria were available for results greater
than 50 ppm. The water content varied between barrels from the same lot by
approximately 20% to 40% in Lot 2000-216, 23% to 66% for Lot 2000-224, and 7% for
the composite tank. For Lot 2000-216 and 2000-224, two samples each were collected
from two separate barrels for each lot. When sample results for the same barrel and same
lot were compared, the water content varied 40% for samples BIO-01 and BIO-02, 8% for
BIO-03 and BIO-04, 8% for BIO-05 and BIO-06, and 23% for samples BIO-07 and BIO-
08. This may be due to variability in the analytical method, atmospheric conditions at the
time of testing, or sample storage conditions. Although BIOTEMP®'s water content is
higher compared to mineral oil and HTH, the water content did not adversely affect the
dielectric strength since BIOTEMP® met the dielectric breakdown specifications for
ABB, ASTM D3487, and ASTM D5222. Presently, ABB is using their own blending
equipment at their South Boston, Virginia facility to ensure future consistency of their
product.
Physical Properties
Pour Point
The pour point indicates the lowest temperature an oil can be used. Initially, BIO-01,
BIO-03, BIO-07, and BIO-10 were analyzed and the pour point was measured at -21°C
for all four samples. The other six samples were analyzed at a later date with pour points
all measured at -15°C. These combined data exceeded the repeatability criteria of 3°C
between readings per ASTM Method D97. This may be due to a different operator
conducting the tests. The pour points for all samples were within the ABB specification
range. BIOTEMP® did not meet the values listed for the ASTM D3487 and D5222
specifications and was not expected to meet these values since they were based on the
physical properties of mineral oils and HTH.
Viscosity
The dielectric fluid's viscosity is used by transformer designers to confirm that the fluid is
appropriate for the unit under certain operating conditions. The viscosity of BIOTEMP®
was determined at 0°C, 40°C, and 100°C. The viscosities at 0°C, 40°C, and 100°C
varied slightly between samples and were below the ABB maximum specification values
at these temperatures. The fluctuations in the measured viscosity at 40°C and 100°C
were not within the precision criteria listed in ASTM Method D445 of < 0.35% of the
sample mean. This may be due to different operators testing the material at two different
points in time. No precision criteria were listed for viscosities measured at 0°C.
BIOTEMP® was not expected to meet and did not meet ASTM D3487 and D5222
specifications for viscosity. These ASTM specifications were developed for mineral oils
and HTH that have different physical properties and were provided as a reference only.
June 2002 16
-------
4.2.2 In-service Transformer Fluid Results
For in-service transformer samples, monitoring results for the past year are presented in Figure 7.
The sample results generated as part of this verification/certification are presented in Table 5.
These samples were tested for flash and fire point, dissipation factor, water content, and
conductivity. Flash and fire point results are presented in Section 4.4. In-service transformer
results are compared to the IEC 1203 performance specification, which was developed to
evaluate the quality of in-service synthetic esters (IEC 1203). The performance of BIOTEMP®
in service is similar to that of synthetic esters. The performance specifications for the dissipation
factor, water content, and conductivity listed under ASTM D3487, D5222, and ABB are for
virgin product and are used to determine if the oil has degraded.
®
Table 5. Performance Results for In-Service BIOTEMP Samples
Performance Parameters
Dissipation Factor @ 25°C (%)
Water Content (ppm)
Conductivity @ 25 °C (pS/m)
Specification Standards
ABB
<0.05
<150
>2.0
ASTMD3487
<0.05
<35
--
ASTMD5222
<0.01
<25
--
IEC 1203
0.8
400
--
Sampling Results
INS-01
0.13
15
16.17
INS-02
0.088
19
11.5
INS-03
0.082
16
8.51
INS-07
0.252
78
24.65
Note:
1. Samples INS-01, INS-02, and INS-03 collected from transformers owned by PG&E.
2. Sample INS-07 collected from a transformer owned by ABB which is used for testing BIOTEMP® under extreme
operating conditions.
3. Sample results for the dissipation factor are compared only to IEC 1203. The values listed for ABB, ASTM D3487,
and D5222 are for virgin product.
4. Water content values are compared to the ABB and IEC 1203 specification values.
Acronyms and Abbreviations:
— = Specification did not list a value for this parameter
ABB = Virgin product specification for BIOTEMP® developed by ABB, Inc.
IEC 1203 = International Electrochemical Commission (IEC) specification for Synthetic Organic Esters for
Electrical Purposes - Guide for Maintenance of Transformer Esters in Equipment.
ppm = parts per million
pS/m = picosiemens per meter
The dissipation factor for all four transformer samples were below the IEC 1203 maximum
value. All but two historical data points associated with the ABB transformer monitoring
program were below the IEC 1203 maximum values. The ABB transformer sample (INS-07)
had a higher value than the PG&E transformer samples and was noted to have an amber-orange
color. The PG&E samples were described as light yellow. According to ABB, the ABB
transformer was used to test BIOTEMP® under extreme operating conditions such as overload
scenarios. Historical results for the ABB transformer showed a steady rise in the dissipation
factor, which corresponded to the overload scenarios. When the in-service sample results were
compared to the ABB virgin product specification, the in-service sample results ranged from
64% to 404%, indicating the oil may have a higher contaminant content due to use. The color
June 2002
17
-------
and higher dissipation factor for the ABB transformer might indicate thermal decomposition of
the fluid or possible oxidation.
The water content for the in-service transformer samples were all below the maximum value
listed for LEG 1203 and ABB. Most of the historical water content data for the ABB transformer
(TNS-07) were below the ABB maximum value. When compared to the other transformer
sample results, the ABB transformer sample (TNS-07) had the highest water content. The higher
water content observed in INS-07 corresponds to the overload tests conducted by ABB.
The conductivity values for all four samples are greater than the minimum value specified by
ABB specifications. IEC 1203 did not specify a conductivity value but does specify a minimum
volume resistivity value of 6.00 x 1011 Qcm. The conductivity values are the inverse of
resistivity and can be converted. The calculated volume resistivity values for samples INS-01,
INS-02, INS-03, and INS-07 are 6.2 xlO12 Qcm, 8.7 x 1012 Qcm, 1.2 xlO13 Qcm, and 4.0 x 1012
Qcm, respectively. These values were greater than the minimum volume resistivity specified in
IEC 1203. It should be noted that the ABB transformer sample had the higher conductivity value
compared to PG&E transformer samples. Again, the higher conductivity value for INS-07
corresponds to overload tests and was probably the result of extreme operating conditions.
Figure 7. Trends for In-Service Transformer Parameters
0.900 -
?
o 0.600 -
0
s.
o
«
Q. 0.300 -
en
4
b 1
0.000 -
c
Dissipation Factor Results
x.x
x
x
A X
X
'x xX x xx x •
+
D O O O O ->•
* INS-3 (NAB4424003-T)
• INS-2 (NAB4424004-T)
A INS-1 (NAB4424005-T)
X INS-7 (OODV065)
Vdue(McKirnjiri)
o t\) s* ay co o t\)
Years in Service
Conductivity Results
90
I
s
Years in Service
Water Content Results
Water Content (ppm)
x
xxxxx
INS-2 NAB4424004-T)
INS-1 NAB4424005-T)
X INS-7 OODV065)
Vdu3(McKirruTJ>
0 0 0 0 0 ->• -*
o k) ^ o> co o k)
Years in Service
June 2002
18
-------
4.3 Results: Objective 2, Aquatic Biodegradability
Three virgin BIOTEMP® samples were analyzed by the Coordinating European Council (CEC)
test method CEC-L-33-A-93. This method was originally intended to measure the
biodegradability of hydrocarbons, specifically two-stroke motor oils, in water. This method
compares the biodegradation potential of BIOTEMP® against the standard oil specified in the test
method. BIOTEMP® and the standard oil are placed in separate flasks containing an
inoculum/mineral substrate mixture. Two separate poisoned flasks containing BIOTEMP® and
the standard oil are also prepared with 1 ml of mercuric chloride and no inoculum. The extract
solutions from these flasks are collected on the zero-day and after the 21 -day incubation period.
The extract solution is analyzed by infrared spectroscopy (TR) measuring the maximum
absorption of the stretch between carbon and hydrogen (C-H) at the ethyl -methyl (CH^-CHa)
bond. This is conducted at a wavelength of 2930 cm"1 ±10cm~1. The biodegradability is
expressed as a percent difference in the residual oil contents between the poisoned flasks and the
respective test flasks. Powertech, an independent testing laboratory, performed these tests.
Table 6 presents the results for the three virgin product samples sent to Powertech. Also
presented are historical results for virgin product analyzed by another independent laboratory,
Parametrix, using the same test method. The average biodegradability of BIOTEMP® was 99%
after 21 days. An earlier study by ABB showed 90% biodegradation after 21 days. The results
from Powertech met the method's repeatability criteria of less than 14.8% at 95% confidence.
The biodegradability result for the method reference oil, RL130, at 88% was compared to an
inteiiaboratory program value of 89.6% and met the reproducibility criteria of less than 25.2% at
95% confidence. Biodegradability results reported by Powertech and Parametrix also met the
reproducibility criteria.
Table 6. Aquatic Biodegradability Results
Sample ID
BIO-01
BIO-07
BIO-10
Average
Historical Data1
Biodegradability (%)
100
98
100
99 ±3
89 ±8
Note: Data variability is calculated at 95% confidence.
lrThis value is the average of six test results reported in an internal
ABB document dated March 1997.
While mineral oil was not tested as part of this study, literature data were available on
biodegradability using the same CEC method, a U.S. EPA method, and an Organization of
Economic Cooperation and Development (OECD) method. The Universite de Liege study
reported the biodegradability of mineral oil over 70% after 40 days using test method
CEC-L-33-T-82 (Cloesen, C. & Kabuya, A., no date). This method has been replaced by
method CEC-L-33-A-93.
June 2002
19
-------
Biodegradation rates for conventional mineral oil ranged from 42-49% after 28 days using U.S.
EPA Method 560/6/-82-003, Aerobic Aquatic Biodegradability (USAGE, 1997, 1999). Another
study by CONCAWE reported a ready biodegradation rate for a light naphthenic distillate
mineral oil of 28% after 28 days when analyzed by OECD 30 IB, Sturm Test (CONCAWE,
1997). Both methods estimated the degree of biodegradability by the amount of carbon dioxide
(CO2) produced and expressed this result as a percentage of the theoretical CO2, which can be
produced. These methods are not considered equivalent to CEC-L-A-33-93 but the data does
indicate that mineral oil is not readily biodegraded.
Based on a comparison to the reported biodegradation rates for mineral oil, the BIOTEMPR fluid
appears to biodegrade more readily. Although BIOTEMP® readily biodegrades per this test, the
product's ability to degrade in the environment is dependent on site-specific factors such as
climate, geology, moisture, pH, temperature, oxygen concentration, dispersal of oil, presence of
other chemicals, soil characteristics, nutrient quantities, and populations of various
microorganisms at the location (U.S.EPA 1997).
June 2002 20
-------
4.4 Results: Objective 3, Flammability
The flash point and fire point for virgin and in-service BIOTEMP® fluid were determined using
ASTM Method D92, Cleveland Open Cup test. The flash point was measured to assess the
overall flammability of the fluid and determine the presence of volatile or flammable material at
elevated temperatures. The fire point was measured to determine the temperature at which the
fluid would support combustion. These values were compared to ABB's specifications for
BIOTEMP®. They were also compared to ASTM D3487 for flash point and ASTM D5222 for
fire point, which are designed for virgin mineral oil and HTH oil, respectively. Both ASTM
D3487 and ASTM D5222 specify ASTM D92 (Cleveland Open Cup) to determine flash and/or
fire point. Results are presented in Tables 7 and 8. The individual and average flash and fire
point values for both the virgin and in-service fluid met the ABB and ASTM specifications. The
deviation in the flash and fire point values for the virgin product were within the precision
margin of ± 8°C at 95% confidence specified in ASTM Method D92. Since the in-service fluid
samples were collected from different transformers and a duplicate was not collected, the results
were not compared to the precision criteria. After being in operation for over one year, the flash
and fire points for the in-service transformer fluids were well above the minimum ABB, ASTM
Method D3487, and ASTM D5222 specifications.
®
Table 7. Flash Points for Virgin and In-service BIOTEMP Samples
Sample Numbers
Virgin Lot No./
Transformer SN
Specification criteria (°C)
BIOTEMP®
ASTM D3487
ASTM D5222
Flash Point
(°Q
Virgin Product
BIO-01
BIO-02
BIO-03
BIO-04
Average*
BIO-05
BIO-06
BIO-07
BIO-08
Average*
BIO-09
BIO-10
Average
Overall Average*
2000-216
2000-224
composite
NA
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>145
>145
>145
>145
>145
>145
>145
>145
>145
>145
>145
>145
>145
>145
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
328
328
332
326
329 ±4
328
328
332
334
331±5
334
340
337
331±3
In-service Transformer Fluid
INS-01
INS-02
INS-03
INS-07
Average
NAB4424-005T
NAB4424-004T
NAB4424-003T
PAO7914-001
NA
>300
>300
>300
>300
>300
>145
>145
>145
>145
>300
NA
NA
NA
NA
NA
330
334
334
328
332 ±5
NA = Not Applicable 'Calculated at a 95% confidence interval
June 2002
21
-------
,®,
Table 8. Fire Points for Virgin and In-service BIOTEMP Samples
Sample Numbers
Virgin Lot No./
Transformer SN
Specification criteria (°C)
BIOTEMP8
ASTM D3487
ASTM D5222
Fire Point
(°C)
Virgin Product
BIO-01
BIO-02
BIO-03
BIO-04
Average*
BIO-05
BIO-06
BIO-07
BIO-08
Average*
BIO-09
BIO-10
Average
Overall Average*
2000-216
2000-224
composite
NA
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
>300
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
304-310
362
360
362
358
361 ±3
360
360
362
358
360 ±3
360
360
360
360 ±1
In-service Transformer Fluid
INS-01
ms-02
INS-OS
INS-07
Average*
NAB4424-005T
NAB4424-004T
NAB4424-003T
PAO7914-001
NA
>300
>300
>300
>300
>300
NA
NA
NA
NA
NA
304-310
304-310
304-310
304-310
304-310
362
364
362
362
363 ±2
NA = Not Applicable 'Calculated at a 95% confidence interval
The fire point results agreed with those obtained by Underwriters Laboratory (UL) and the
Factory Mutual Research Center (FMRC). UL and FMRC evaluated this product using ASTM
Method D92 (Cleveland Open Cup) and reported fire points of 354°C and 360°C, respectively.
UL determined the flash point of 243 °C using ASTM Method D93 (Pensky-Martens closed-cup)
while FMRC determined a flash point of 330°C using ASTM Method D92. The lower flash
point reported by UL was due to their use of a different test method.
BIOTEMP® is one of five products that UL has classified as a dielectric medium with a fire
hazard rating of 4 to 5, and is less of a fire hazard than paraffin oil (UL, 2001). The UL fire
rating system uses the flash point determined by Pensky-Martens closed-cup to rate the material's
flammability. The material's flammability is rated and classified using the following scale
arranged from flammable to nonflammable: ether rated at 100, gasoline from 90 to 100, ethyl
alcohol from 60 to 70, kerosene from 30 to 40, paraffin oil from 10 to 20, and water at 0.
FMRC classified this product as a less flammable transformer fluid. FMRC also identified
BIOTEMP® as an alternative to high fire point hydrocarbons, silicone fluids, and synthetic esters
or hydrocarbons where fire resistance, improved high temperature operation, and improved
cooling are desired (FMRC, 1999).
June 2002
22
-------
4.5 Results: Objective 4, Acute Toxicity
Three virgin BIOTEMP® samples, one from each lot, were analyzed by U.S. EPA method,
Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and
Marine Organisms, EPA/600/4-90/027F, August 1993. Tests were performed by Associated
Laboratories, a California certified laboratory, which performed the work under contract with
DTSC. Based on the fish bioassay results provided by the client per this method, the screening
test was not conducted and instead three test chambers were prepared containing 750 mg/1, 500
mg/1, and 250 mg/1 of BIOTEMP®. Duplicate testing was performed in parallel with the test
samples. The tests used juvenile pimephales promelas (fathead minnow) instead of juvenile
oncorhynchus mykiss (rainbow trout) as stated in the test plan. Samples were prepared in
accordance with the "Static Acute Bioassay Procedures for Hazardous Waste Samples"
developed by the California Department of Fish and Game, Water Pollution Control Laboratory
and specified in the Code of California Regulations, Title 22, Section 66261.24(a)(6). This
procedure requires shaking the sample for six hours using a wrist-action or similar type of shaker
to dissolve the oil in 200 ml of water before the sample is added to the aquatic bioassay fish tank.
Dissolved oxygen (DO) content, pH, and temperature were monitored and maintained at 6.0-7.0
mg/1, 7.0-7.5, and 20°C, respectively as required by the method.
Earlier tests performed by Parametrix, an independent laboratory under contract with ABB, were
conducted per U.S. EPA method, Methods for Measuring the Acute Toxicity of Effluents and
Receiving Waters to Freshwater and Marine Organisms, EPA/600/4-90/027F, August 1993. The
test species used was juvenile rainbow trout. DO content, pH, and temperature were monitored
and maintained at 9.0-1 1.0 mg/1, 7.5 - 8.0, and 12°C, respectively as required by the method.
Results are presented in Table 9 and compared to the hazardous waste toxicity characteristic
criterion listed in the Code of California Regulations, Title 22, Section 66261.24(a)(6). A waste
is considered to exhibit a toxic characteristic if the LCso is less than 500 milligrams per liter
when measured in soft water (total hardness 40 to 48 milligrams per liter of calcium carbonate).
®
Table 9. Fish Bioassay Results for Virgin BIOTEMP Samples
Sample Numbers
BIO-01
BIO-07
BIO-10
Average
Historic Data2
California Toxicity Criteria1
(mg/1)
<500
<500
<500
<500
<500
Sample Results
(mg/1)
<250
<250
<250
<250
776
lrThe virgin oil is considered to exhibit a toxic characteristic if the LC50 is less
than 500 mg/1 when measured in soft water.
2The result is for a single sample collected by ABB in November 1998.
The 95% confidence interval was 668 mg/L to 901mg/L.
June 2002
23
-------
A DISC toxicologist reviewed the reports prepared by Associated Laboratories and Parametrix
to identify the differences, which could lead to such conflicting results. As part of the review, the
toxicologist also reviewed the test methods, and material safety data sheets for BIOTEMP® and
its additives. The tank water was not analyzed for breakdown products associated with degraded
vegetable oil. The main difference between the two sets of tests was the sample preparation
method used. Associated Laboratories used a wrist-action shaker per the method specified.
Parametrix prepared their samples using a carrier solvent, which is listed in U.S. EPA method,
Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and
Marine Organisms, to make the oil miscible in water. Oil samples prepared using the wrist
action method are thought to stratify, with the oil at the top of the tank. Fish swimming through
this upper layer of the tank will become coated with the product and gill exchange will be
impaired. Oil samples prepared using the wrist shaker method are thought to provide more
realistic results for conditions, which may occur during an environmental release. Samples
prepared using the carrier solvent provided results that reflect systemic (chemical) impacts on
fish.
In California, insoluble, viscous waste samples are prepared using the wrist-shaker method and
ultrasonic method, and sometimes the solvent carrier method as part of the fish bioassay
screening tests for hazardous waste characterization. The preparation method yielding the most
conservative LCso result is then used to perform the definitive tests. This methodology is
required by DTSC Waste Evaluation Unit and overseen by the Department of Health Services
Environmental Laboratory Accreditation Program's Aquatic Toxicity Bioassay Section who
certifies laboratories performing aquatic toxicity tests for DTSC. ABB disagrees with DTSC's
methodology (see vendor's comment section for ABB's opinion). The reader should note that
this methodology is used to characterize the hazardous characteristics for waste. Any statement
concerning the hazardous characteristic of the BIOTEMP® fluid applies to the spent (waste)
fluid only and is not intended to classify the virgin product.
The lower LCso results and physical effects described above are similar to those presented by the
U.S. EPA in their responses to comments on the rule for Oil Pollution Prevention at Non-
Transportation Related Onshore Facilities (40 CFR Part 112). The physical effects observed in
the toxicity tests performed by Associated Laboratories have been observed in vegetable oils, and
oils in general, and were therefore expected. These results, which are based on virgin product
and a relatively small number of samples, suggest that spent BIOTEMP® may be classified as
hazardous waste and need to be managed accordingly. The end-user should characterize the
spent BIOTEMP® at the time of disposal since changes may occur to the oil due to use. The end-
user should also consult their appropriate regulatory authority on the appropriate waste
characterization and disposal method for their state.
June 2002 24
-------
4.6 Results: Other Verification/Certification Objectives
Chemical Composition
The chemical composition of the virgin and in-service fluids was analyzed for semivolatile
organics (SVOCs) and metals to verify chemical composition. In addition, the samples were
analyzed by various Association of Analytical Chemist (AOAC) methods to create a chemical
"fingerprint" Krueger Food Laboratories analyzed samples per the AOAC methods while the
Hazardous Materials Laboratory (HML) analyzed the SVOC and metals samples. Appendix B
contains a list of the AOAC methods used.
According to ABB, BIOTEMP® is composed >98.5% vegetable oil and <1.5% additives (e.g.,
antioxidants and color). The vegetable oil is comprised of at least 75% oleic acid, less than 10%
diunsaturated fatty acids, less than 3% triunsaturated fatty acids, and less than 8% saturated fatty
acids. According to the manufacturer, the antioxidants may consist of combination of
antioxidants, which include butylated hydroxyl anisole (BHA), mono-tertiary butyl hydroquinone
(TBHQ), 3,5-di-tert-butyl-4-hydroxytoluene (BHT or DBPC), or Vitamin E.
Tables 10 and 1 1 present the sample results for virgin and in-service BIOTEMP® fluid. Analytes
detected at percentages greater than 5% in virgin sample results meet the repeatability criteria
listed in AOAC Method 963.22 with a relative percent difference between results of < 3% and an
absolute percent difference of < 1%. Results for the in-service samples were not compared to the
precision criteria. The in-service samples were collected from different transformers and a
duplicate sample was not collected to minimize impacts on the transformer and the on-going
sampling program.
®
Table 10. AOAC Results for Virgin BIOTEMP Samples
Analyte
Total Fatty Acids
Hexadecanoic (Palmitic)
Octadecanoic (Stearic)
Octadecenoic (Oleic)
Octadecadienoic (Linoleic)
Octadecatrienoic (Linolenic)
Eicosanoic (Arachidic)
Docosanoic (Behenic)
Tetracosanoic (Lignoceric)
Phenolic Antioxidants (mg/kg)
16:0
18:0
18:1
18:2
18:3
20:0
22:0
24:0
Polymers and Oxidation Products (g/1 OOg)
Sample Number
BIO-01
3.67%
3.54%
79.92%
10.52%
0.27%
0.30%
1.46%
0.31%
3,139
2.80
BIO-03
3.64%
3.40%
80.17%
10.41%
0.26%
0.30%
1.51%
0.30%
3,187
2.29
BIO-07
3.63%
3.38%
80.23%
10.41%
0.24%
0.29%
1.50%
0.31%
3,206
1.81
BIO-10
3.69%
3.50%
79.91%
10.47%
0.26%
0.30%
1.56%
0.30%
3,294
2.00
Average
3. 66% ±0.04%
3. 46% ±0.12%
80.06% ± 0.26%
10.45% ± 0.08%
0.26% ± 0.02%
0.30% ±0.01%
1.51% ±0.07%
0.31% ±0.01%
3,207 ± 103
2.23 ±0.69
Note: Data variability calculated at 95% confidence using a two-tailed T-test assuming normal
distribution.
Results are presented for the individual fatty acids along with their number of carbons and the
number of double bonds (i.e., 18:1 represents 18 carbons and one double carbon bond). The
June 2002
25
-------
percentage of monounsaturated, diunsaturated, and triunsaturated fatty acids are determined by
adding the fatty acids with one, two or three double carbon bonds together, respectively. For
example, the percentage of diunsaturated fatty acids would consist of fatty acids with two double
carbon bonds or octadecadienoic acid (18:2). The percentage of saturated fatty acids is
determined by summing the results for fatty acids with no double carbon bonds such as
hexadecanoic (16:0), octadecanoic (18:0), eicosanoic (20:0), docosanoic (22:0), and
tetracosanoic (24:0). The virgin BIOTEMP R samples had oleic acid ranging from 80. 1% ± 0.3%,
diunsaturated fatty acids ranging from 10.5% ± 0.1%, triunsaturated fatty acids ranging from
0.2% ± 0.0%, and saturated fatty acids ranging from 9.2% ± 0.2%, which agree closely with the
formulation listed above. The in-service BIOTEMP® samples had oleic acid ranging from 79.5%
to 84.4%, diunsaturated fatty acids ranging from 5.3% to 10.7%, triunsaturated fatty acids
ranging from 0.2% to 0.3%, and saturated fatty acids ranging from 9.5% to 10.0%. The in-
service samples are similar to the formulation above except that three samples had a low
diunsaturated content compared with the virgin BIOTEMP® samples. This may be due to
variations in formulation associated with the basestock oil.
®
Table 11. AOAC Results for In-service BIOTEMP Samples
Analyte
Sample Number
INS-01
INS-02
INS-03
Total Fatty Acids
Hexadecanoic (Palmitic)
Octadecanoic (Stearic)
Octadecenoic (Oleic)
Octadecadienoic (Linoleic)
Octadecatrienoic (Linolenic)
Eicosanoic (Arachidic)
Docosanoic (Behenic)
Tetracosanoic (Lignoceric)
Phenolic Antioxidants (mg/kg)
16:0
18:0
18:1
18:2
18:3
20:0
22:0
24:0
Polymers and Oxidation Products (g/1 OOg)
3.85%
3.79%
84.41%
5.38%
0.22%
0.42%
1.62%
0.30%
3,586
1.78
3.84%
3.85%
84.39%
5.41%
0.25%
0.46%
1.46%
0.33%
3,022
1.84
3.83%
3.77%
84.41%
5.40%
0.21%
0.46%
1.54%
0.39%
3,196
1.39
INS-07
3.97%
3.38%
79.55%
10.68%
0.27%
0.31%
1.53%
0.31%
2,993
2.40
AOAC Method 983.15, Phenolic Antioxidants in Oils, Fats, and Butter Oil, was used to
determine the concentrations of 7 commonly used antioxidants in food grade oils and fats. The
results for the virgin and in-service transformer samples are presented in Tables 10 and 11.
Phenolic antioxidants were detected in the virgin product between 3,207 mg/kg ±103 mg/kg.
The in-service transformer samples had antioxidant concentrations between 2,990 and 3,600
mg/kg.
The polymers and oxidation products listed in Tables 10 and 11 above are simple indicators used
in the food industry to assess the quality of vegetable oil after exposure to heat. If lower values
are reported for an oil as it is reheated, the difference is assumed to show an increase in non-
elution material (compounds not dissolved using a solvent) that indicates the polar compounds in
the oil such as unsaturated fatty acids are degrading. This method does not list a precision
criteria for the data. Compared to the average virgin product value of 2.2% ± 0.7%, the in-
service fluid samples appear to have degraded slightly due to use.
June 2002 26
-------
Virgin and in-service samples were screened for 65 standard SVOC compounds using U.S. EPA
Method 8270/3580. Virgin samples and one in-service sample (INS-07) were extracted outside
the 7 day extraction period, which deviated from the holding time requirements listed in the test
evaluation plan. HML noted the recovery of pyrene in the matrix spikes could not be reliably
calculated due to matrix interference, and the recovery of two surrogate compounds (2,4,6-
tribromophenol and terphenyl-d!4) may have also been affected due to difficulty in separating
the oil samples using U.S. EPA Method 8270. Due to this difficulty and extraction times
exceeding those listed in the test plan, the reported SVOC results should be regarded as
approximations and not be used in lieu of actual waste characterization data.
For the 65 standard SVOC compounds analyzed by the HML lab, only n-nitrodiphenylamine was
detected around the detection limit of 20 mg/L for the virgin and in-service transformer samples.
This may be a component of one of the antioxidants used in the fluid. For the in-service fluid,
bis-(2-ethylhexyl)phthalate was also detected. This compound, a widely used plasticizer, was
also detected in the equipment and field blanks collected. Other tentatively identified
compounds were TBHQ, 2-isopropyl-l,4-benzenediol, 2,3-dihydro-2-methyl-5-phenyl-
benzofuran, 2-isopropyl-l,4-benzoquinone, p,p'-dioctyldiphenylamine, beta-sitosterol, squalene,
and vitamin E. Due to the deviations discussed above, the SVOC data should be considered a
qualitative measurement but does not change the assessment that BIOTEMP® consists primarily
of vegetable oil with a small percentage of antioxidants.
Virgin and in-service samples were analyzed by U.S. EPA Method 6010/5030. Other than the
sample preparation method used, the laboratory noted no other deviations to the final test
evaluation plan. Metals were not detected in the in-service transformer samples except for INS-
2, which had a zinc concentration of 2.3 mg/kg. For the virgin samples, copper was detected at
4.13 mg/kg in sample BIO-01. Barium was detected at 0.31 mg/kg in samples BIO-05, 0.32
mg/kg in BIO-07, and 0.32 mg/kg in BIO-10. Zinc was detected at 2.02 mg/kg in sample BIO-
10. The detection limit was 2.50 mg/kg for copper, 0.25 mg/kg for barium, and 2.00 mg/kg for
zinc. No metals were detected in the equipment blank.
Worker Health and Safety Aspects
This section presents some of the potential hazards associated with BIOTEMPR and compares
them to those for select mineral oil-based and silicone oil-based transformer fluids. This is not
considered a comprehensive review where all potential hazards associated with BIOTEMP® have
been identified. End-users should review all applicable worker health and safety regulations for
this product.
BIOTEMP® is a dielectric insulating fluid used to cool the core and coils within a transformer.
The fluid is held in a tank inside the transformer where the tank headspace is filled with nitrogen
to prevent the oil from oxidizing with the ambient air. A pressure relief valve installed on the
tank releases gases in the headspace to the ambient air. Transformers that use mineral oil or
other types of insulating fluid are also equipped with pressure relief valves. BIOTEMP® is
designed for use in transformers where higher fire protection is required such as in or adjacent to
buildings, or in underground vaults.
June 2002 27
-------
The BIOTEMP material safety data sheets (MSDS) lists the components as >98.5% vegetable
oil and <1.5% additives (e.g., antioxidants and color). The antioxidants used in this product are
not listed as hazardous materials (see Section 5.1 for reference). Two of the antioxidants have
been cleared by the Food and Drug Administration (FDA) for use as an indirect food additive in
food packaging (Ciba-Geigy, 1996) while the third antioxidant is identified as a food grade
antioxidant (Eastman, 1996). Although the BIOTEMP® components may be food grade, this
product should not be used as a food product.
According to the BIOTEMP® MSDS, this product is also not considered a hazardous substance
as defined under Title 8, California Code of Regulations, Section 5194, Hazard Communications.
However, this does not relieve the end-user who uses this product from providing workers with
information and training necessary to handle BIOTEMP® safely. Workers should review the
MSDS and be familiar with the information concerning first aid procedures, physical properties,
personal protective equipment (PPE), respiratory protection, and slip hazards. Workers should
wash skin that has contacted the product with soap and water. For eye contact, the eyes should
be flushed with water. The primary physical property workers should be aware of is the
product's flash point of greater than 300°C. In the case of a BIOTEMP® spills, employees
should be aware of the increased slip hazard in the affected area due to the product.
Before working with BIOTEMP®, employees should ensure the work area has adequate
ventilation, and the appropriate respiratory protection and protective clothing are selected. When
working with hot BIOTEMP®, workers should don neoprene gloves, rubber boots and aprons.
Respiratory protection should only be worn if oil mists or dusts contaminated with oil are
detected at concentrations equal to or exceeding the permissible exposure limit (PEL). The
Occupational Safety and Health Administration (OSHA) has set the permissible exposure limit
(PEL) for vegetable oil mist as a nuisance particulate at 15 mg/m3 and 5 mg/m3 for respiratory
protection for an 8-hour time-weighted average (TWA) exposure. In California, the nuisance
parti culate PEL is 10 mg/m3. The end-user should consult the appropriate regulatory authority
about applicable nuisance particulate PELs used in their area.
If the transformer is located in a poorly ventilated area, then workers should use appropriate
engineering controls to ventilate the area. Based on the MSDS information on BIOTEMP®'s
antioxidants, BIOTEMP® may produce carbon monoxide, carbon dioxide, nitrogen oxides, and
other toxic compounds when the antioxidants thermally decompose. Mineral oil-based and
silicone oil-based transformer fluids may also thermally decompose and produce fumes, smoke,
carbon monoxide, aldehydes and other products. For some mineral oil-based transformer fluids,
sulfur oxides are also listed as a possible decomposition product while silicon dioxide is listed
for some silicone oil-based fluids. No data are available on the composition of emissions from
transformers in general.
When comparing the PPE requirements for handling BIOTEMPR to select mineral oil-based
transformer fluids, the requirements were found to be similar. This comparison is based on
MSDS information for select mineral-oil-based transformer fluids obtained from the Vermont
Safety Information Resources, Inc. (SIRI) MSDS archive. However, respiratory protection for the
mineral oil-based transformer fluids is required when the mineral oil mist concentration equals or
exceeds the OSHA PEL set at 5 mg/m3 for an 8-hour TWA exposure. For select silicone oil-
June 2002 28
-------
based transformer fluids found in the Vermont SIRIMSDS archive, workers are advised to don
impervious gloves and chemical goggles when handling the fluid.
Occupational exposure to transformer fluid is limited and associated to infrequent activities such
as filling, draining, or sampling of transformers. These activities are not likely to generate a mist
or aerosol at concentrations approaching the PEL. Potential hazards associated with filling or
draining the transformer include slipping on work surfaces where the product was spilled, or
splashing of the material into the eyes or onto the skin. Potential hazards associated with
sampling the transformer include coming in contact with extremely hot oil, potential electrical
arcing from the transformer, or slipping hazards due to spilled BIOTEMP® on the floor.
MSDS information for three silicone transformer fluids identified as less-flammable transformer
oils by UL and FMRC were reviewed along with several mineral oil-based transformer fluids
listed in the Vermont SIRI MSDS Archive. Health and safety information on the components
listed on the MSDSs was compared to information listed in Sax's Dangerous Properties of
Industrial Materials. The primary component of the mineral oil-based transformer fluid was a
hydrotreated light naphthenic petroleum distillate (CAS No 64742-53-6) ranging from 30-100%
which was identified as an International Agency for Research on Cancer (IARC) confirmed
carcinogen based on experimental data for animals (Lewis, 2000). The primary ingredient of the
silicone oil-based transformer fluids was dimethyl polysiloxane (CAS No. 63148-62-9) listed at
100% and identified as a combustible liquid, a teratogen, and the cause of reproductive effects
based on experimental data on animals (Lewis, 2000).
Estimated Cost of Using BIOTEMP® versus Mineral Oil
An average transformer life of 20 years was used to compare the costs of BIOTEMP® versus
mineral oil based on historical life testing results performed by ABB per ANSI/IEEE C57.100-
1986, the accelerated life test. The ANSI/IEEE accelerated life tests performed on transformers
using BIOTEMPR passed with an operational equivalence of 100 years, which is five times the
normal transformer. If the initial purchase cost of a new transformer unit containing BIOTEMP®
is compared to a mineral oil transformer, the BIOTEMP® transformer unit costs approximately
1.25-1.30 times more. The price of the BIOTEMPR fluid ranges from $7 to $11 per gallon
depending on the volume purchased and is based on estimates provided by ABB. The fluid is
available in 5 gallon containers, 55 gallon drums, 200 gallon totes, 6,000 gallon tanker trucks, or
by the rail car. Prices for mineral oil typically range from $2 to $3 per gallon (Cooper, 2001).
Monitoring costs will vary depending on the maintenance program the purchaser has in place.
The waste characterization cost for a transformer using BIOTEMP® or mineral oil are anticipated
to be approximately the same except for mineral oil suspected to contain PCBs where the costs
will be higher. The disposal cost for mineral oil and BIOTEMP® are assumed to comparable
since data are not available on the waste characteristics of BIOTEMP® after 20 years of use.
For a retrofilled transformer, no additional costs due to modifications of the transformer unit are
incurred for using BIOTEMP®. The costs associated with draining and disposing of the used oil
are expected to be the same for both mineral oil and BIOTEMP®. Costs associated with flushing
and filling a retrofilled transformer with BIOTEMP® versus mineral oil are also anticipated to be
June 2002 29
-------
higher since BIOTEMPR costs between $4 to $9 per gallon more than mineral oil depending on
the volume purchased.
June 2002 30
-------
Section 5. Regulatory Considerations
A review of Federal and California regulations was conducted to identify applicable regulations
for virgin and spent (used) BIOTEMP®. The regulations listed below are based on the limited
data available on this product. This review is not considered to be a comprehensive review of
existing regulations. The reader should consult their local environmental regulatory agency
concerning other applicable local and State regulations and the status of the regulations cited
below. The regulations cited below may have been updated or superceded since this review was
conducted.
Virgin (or unused) BIOTEMP® fluid is a vegetable oil-based dielectric fluid consisting of
>98.5% food-grade vegetable oil and < 1.5% additives such as antioxidants and color. The
product has a flash point of 243°C by ASTM Method D93 and an average fire point of 331 °C by
ASTM Method D92. The product has a neutral pH (pH = 7.0) and is not reactive with other
chemicals at room temperature but is incompatible with strong oxidizers. The virgin
BIOTEMP® fluid has a reported aquatic LCso value of less than 250 mg/L based on test results
reported in Section 4.5 of this report and 776 mg/L based on historical results provided by ABB.
The difference between the results was thought to be due to the sample preparation method used.
The lower LCso value was thought to reflect the physical impacts and the higher LCso the
systemic (chemical) impacts of an oil spill to fish.
5.1 Regulation of Virgin BIOTEMP® Dielectric Fluid
Information on new product and materials introduced for commercial use are submitted to the
U.S. EPA for review under the Toxic Substances Control Act unless the new product is a mixture
of listed materials. The components of BIOTEMP® are listed under the Toxic Substances
Control Act (TSCA) as Chemicals in Commerce. None of the components are listed as an
imminently hazardous chemical substance or mixture which the EPA Administrator has "taken
action under" Section 7. BIOTEMP® and its components are not listed as hazardous substances
under Section 3001 of Resource Conservation and Recovery Act (RCRA), and Section 112 of the
Clean Air Act (CAA). The product is included under Section 311 of the Clean Water Act, which
addresses oil and hazardous substance releases to water. The product is shipped as a non-
hazardous material per Department of Transportation regulations.
The components of BIOTEMP® are not listed in the Consolidated List of Chemicals Subject to
Emergency Planning and Community Right-To-Know Act (EPCRA) and Section 112(r) of the
CAA and therefore, are not reportable under Section 313. However, a material safety data sheet
(MSDS) is required as part of the EPCRA under Section 311. California facilities should consult
Health and Safety Code (HSC) Chapter 6.8 and determine if business plans need to be modified
in the areas of emergency preparedness and response, and water quality if BIOTEMP® is used at
their facilities.
June 2002 31
-------
5.2 Waste Characterization/Disposal Requirements
5.2.1 Waste Characterization and Disposal of Virgin BIOTEMP®
Under the RCRA definition of a hazardous waste, a waste is considered hazardous if it is a listed
waste under Section 261.2 or exhibits a hazardous characteristic as defined in 40CFR261.20
through 40CFR261.24. A hazardous characteristic is defined as either having a flash point less
than 60°C (ignitability), has a pH < 2.5 or pH > 12.5 (corrosivity), is reactive, or contains a
contaminant equal to or greater than the regulatory value listed in 40CFR 261.24 (toxicity) per
the Toxicity Characteristic Leaching Procedure (TCLP). The virgin BIOTEMP® is not a listed
RCRA waste nor does it meet the definition of a hazardous waste per 40CFR261.20. Virgin
BIOTEMP® fluid which is off-specification or has exceeded its shelf life is not listed as a
hazardous waste per 40CFR 261.33 and may be returned to the manufacturer or disposed of as a
non-hazardous material.
In California, a waste is considered hazardous if it is a RCRA listed waste or exhibits a
hazardous characteristic per California Code of Regulations (CCR), Title 22, Division 4.5,
Chapter 11, Article 3, Section 66261.20 (22CCR66261.20). The ignitability, corrosivity, and
reactivity criteria listed under 22CCR66261.20 are the same as those listed for 40CFR261.20
above. The toxicity characteristic defined under 22CCR261.24 lists several criteria which are as
follows: (1) the waste meets the criteria per 40CFR261.24, (2) the waste contains a substance
listed in 22CCR66261.24 as determined by the Waste Extraction Test (WET), (3) the waste has
an acute oral lethal dose (LDso) of less than 5,000 mg/kg, (4) the waste has an acute dermal LDso
of 4,300 mg/kg, (5) the waste has an acute inhalation lethal concentration (LCso) of less than
10,000 ppm as a gas or vapor, (6) the waste has a acute aquatic 96-hour LCso of less than 500
mg/L, or the waste contains any of the substances listed in 22CCR66261.24(a)(7). Since LCso
results reported under Section 4.5 of this report indicate that spent BIOTEMP® may exhibit a
hazardous characteristic, off-specification material may be subject to hazardous waste
management regulation. Off-specification material may be considered a retrograde material if it
meets the criteria per HSC 25121.5 and may be returned to the manufacturer without a manifest.
5.2.2 Waste Characterization of Spent BIOTEMP®
Spent BIOTEMP® fluid should be characterized by the end-user per 40CFR261.20 or per the
applicable State regulation prior to disposal. To date, the longest continuous use of BIOTEMP®
in a transformer has been approximately 2.5 years. The average service life of a transformer is
approximately 20 years. Since changes to the oil may occur due to use, the spent BIOTEMPR
must be characterized by the end-user prior to disposal. As part of the waste characterization for
transformers that exclusively used BIOTEMP®, the end-user should determine the metals
concentration per EPA Method 1311 and the TCLP. For retrofilled transformers, the spent
BIOTEMP® must also be tested for PCBs per EPA Method 8082 if the transformer was known
or suspected to have contained PCBs prior to using BIOTEMP®. If the spent BIOTEMP® fluid is
characterized as hazardous per 40CFR261.20, then the fluid must be managed as a hazardous
waste.
For spent BIOTEMP® generated in California, the Waste Extraction Test (WET) should also be
performed as defined in 22CCR66261.24 (a)(l) and 66261.24 (a)(2), respectively, in addition to
June 2002 32
-------
EPA Method 1311. The spent oil should also be characterized for acute aquatic toxicity per
22CFR66261.24(a)(6) in addition to the TCLP. If the spent BIOTEMP® fluid is characterized as
hazardous per 40CFR261.20, then the fluid must be managed as a hazardous waste. If the spent
BIOTEMP® fluid is characterized as hazardous per 22CCR66261.20 but not by 40CFR261.20,
then the fluid must be managed as a used oil per 22CCR66279.1.
Characterization results for BIOTEMP® for a specific transformer model may be used for others
if the transformer has only used BIOTEMPR and has not been retrofilled with a different
dielectric fluid during its service life. Depending on the results of the waste characterization, the
spent BIOTEMP® fluid may be sent to a waste oil recycler or fat Tenderer for end-users located
outside California. End-users outside of California should consult their appropriate regulatory
authority about certified waste oil recyclers or fat Tenderers in their area and the recyclers'
acceptance criteria for used vegetable oil. In California, the spent BIOTEMP® may only be sent
to a licensed waste oil recycler if the waste characterization results show the fluid to exhibit a
hazardous characteristic per 22CCR66261.20 and not by 40CFR261.20.
5.2.3 Disposal of Spent BIOTEMP®
Under the federal Used Oil Management Program, spent BIOTEMP® is not included under the
definition of used oil. The U.S. EPA defines used oil as being "refined from crude oil or any
synthetic oil, that has been used and as a result of such use is contaminated by physical or
chemical impurities" (40CFR279.1). The U.S. EPA has stated that animal and vegetable oils are
excluded from the federal used oil definition even when used as a lubricant (U.S. EPA, 1996).
However, spent BIOTEMP® may be subject to hazardous waste management under RCRA if the
spent oil meets the federal hazardous waste characteristics listed in 40CFR261.20 or contains a
listed RCRA hazardous waste. End-users outside California should contact their appropriate
regulatory authority about applicable used oil management regulations for their area.
In California, spent BIOTEMP® may be included in the Used Oil Program under the definition of
a synthetic oil per 22CCR66279.1(d). As part of the synthetic oil definition, "vegetable or
animal oil used as a lubricant, hydraulic fluid, heat transfer fluid or for other similar industrial
purposes shall be managed as used oil if it is identified as a non-RCRA hazardous waste. Used
vegetable or animal oil identified as RCRA hazardous waste is not used oil"(22CCR66279.1(d))
and must be managed as a hazardous waste. A non-RCRA hazardous waste is one that does not
contain a RCRA listed waste, does not exhibit a federal hazardous waste characteristic per
40CFR261.20 through 40CFR261.24 but does exhibit a hazardous waste characteristic per
22CCR66261.20. If the spent BIOTEMP® meets the synthetic oil definition but contains more
than 5 ppm of PCBs or has a total halogen content of greater than 1,000 ppm, then it cannot be
included in the Used Oil Program and must be managed as a hazardous waste.
Used oil (e.g., mineral oils, synthetic oils) managed under the California program must be
managed as a hazardous waste unless it is shown to meet one of the specifications for recycled
oil in Health and Safety Code (HSC) Section 25250. l(b) or qualifies for a recycling exclusion
under HSC 25143.2. Used oil generators are required to meet all used oil generator requirements
except householders who perform their own oil changes. DTSC issues an EPA Identification
Number for each site where used oil is stored except for generators of 100 kilograms or less of
June 2002 33
-------
hazardous waste per month (including used oil) who ship used oil under a modified manifest.
Above-ground storage tanks and containers accumulating used oil, and fill pipes used to transfer
used oil to underground storage tanks must be labeled with the words "USED OIL _
HAZARDOUS WASTE" and the initial date of accumulation. Used oil must be sent to an
authorized used oil storage or treatment facility by a registered hazardous waste transporter.
However, spent BIOTEMP® fluid may be exempt from the California used oil regulations if the
oil is removed from a transformer, filtered, and then reused on-site in electrical equipment as a
dielectric fluid (HSC 25250.4(b)). This exemption does not apply to transformer fluid that has
been removed, filtered, and then sent off-site for reuse. Facilities should contact their local
environmental agency on applicable recycling regulations.
5.2.4 Disposal of Waste the Clean-up of Virgin and Spent BIOTEMP® Spills
In the event of a spill, responders should consult the MSDS and their spill prevention, control,
and countermeasures (SPCC) plan or facility response plan (FRP), if applicable, for the
appropriate clean-up measures. Facilities should consult with their local environmental
regulatory agency on clean-up levels and disposal options for waste generated from these spills.
Since virgin BIOTEMP® may exhibit a hazardous characteristic per California's hazardous waste
definition, the waste generated from spill clean-ups in California should be presumed hazardous
until the waste has been characterized.
5.3 Spill Management
The spill management regulations listed in this section apply to both virgin and spent
BIOTEMP®. Facilities should contact their appropriate regulatory authority on other local or
State regulations pertaining to oil spill management.
Oil Discharge
Under 40CFR 110, Discharge of Oil Regulation, facility owners and operators that handle, store,
or transport oils are required to report an oil discharge which "may be harmful to the public
health or welfare, or the environment". A reportable spill is defined as one that either; (1)
violates water quality standards, (2) causes a sheen or discoloration on the surface of a body of
water, or (3) causes a sludge or emulsion to be deposited beneath the surface of the water or on
adjoining shorelines. The term "oil" applies to petroleum based oil products and non-petroleum
based oil products, which include animal fats, vegetable seed-based oils, and synthetic oils.
Adding dispersants or emulsifiers to the oil prior to discharge is prohibited under Section 40CFR
110.4.
Oil discharged into or upon the navigable waters of the United States must be reported to the
National Response Center, contained, and cleaned up. Depending on the discharge volume,
extent and proximity to sensitive areas (e.g., wildlife areas), coordination and involvement of
local emergency response agencies and the National Response Center may be required for the
clean up effort. These reporting requirements apply to mineral oils and synthetic oils, as well as
vegetable oils.
June 2002 34
-------
Oil Pollution Prevention
Under 40 CFR Part 112.1 through 112.7 of the Oil Pollution Prevention; Non-Transportation
Related Onshore Facilities, facilities "that could be expected to discharge oil into or upon the
navigable waters of the United States or adjoining shorelines, and that have (1) a total
underground buried storage capacity of > 42,000 gallons, (2) a total aboveground oil storage
capacity of > 1,320 gallons, or (3) an aboveground oil storage capacity in a single container of >
660 gallons" are required to prepare and submit a SPCC plan. Some facilities may not be
regulated if, due to their location, they could not reasonably be expected to discharge oil into
navigable waters of the U.S. or adjoining shorelines.
Under the 40 CFR Part 112, facilities are required to prepare and submit a facility response plan
(FRP) if they transfer > 42,000 gallons of oil over water to a vessel or have a storage capacity >
1,000,000 gallons and meet at least one of these four criteria; inadequate secondary containment,
proximity to environmentally sensitive areas, proximity to public drinking water intakes, or
occurrence of a 10,000 gallon or more oil spill in the last 5 years. The FRP includes response for
worst-case discharges, estimates of planned resources, emergency response plans, and training
drills/exercises. Under this regulation, the requirements for animal fats and vegetable oils are
similar to those for petroleum oils, but involve new specific methodology for planning response
actions for vegetable oils and animal fats.
The U.S. EPA's analysis of the impacts of the SPCC program indicated that a majority of electric
utility substations and transformer installations would meet the aboveground storage capacity
thresholds. Facilities such as schools and small business complexes are not anticipated to meet
the SPCC or FRP program requirements. Typically, these facilities have several pad-mounted
transformers with an average oil tank capacity of 40 gallons. For compliance, the facility owner
is required to determine if oil storage capacity at a given site meets the criteria listed in the SPCC
and FRP.
June 2002 35
-------
Section 6. Conclusions
6.1 Objective 1, General Performance
The general performance specifications are useful for end users to determine whether the product
will meet their specific needs. Verification testing confirmed that BIOTEMPR meets or exceeds
the manufacturer's product specifications for dielectric breakdown (minimum, gap, and impulse),
pour point, viscosity, water content, and oxidation stability at 72 hours. Two of the three lots
tested met the manufacturer's specifications for dissipation factor (25°C and 100°C).
BIOTEMP® did not meet the manufacturer's product specifications for oxidation stability at 164
hours or using the rotary bomb method. This may be due to possible inconsistencies in the
addition and/or blending of antioxidants used in BIOTEMPR. When compared to the ASTM
specifications, BIOTEMP® met some but not all of the specifications listed. It met ASTM
D3487 and D5222 specifications for dielectric breakdown (minimum and gap). It met ASTM
D3487 specifications for oxidation stability at 72 hours, while two of the three lots met the
dissipation factor at 25°C. It did not meet the oxidation stability at 164 hours or by the rotating
bomb method, nor did it meet the dissipation factor at 100°C for ASTM D3487 and D5222.
BIOTEMP® also did not meet the pour point, water content, and viscosity standards per ASTM
D3487 and D5222, but was not expected to meet these standards since the physical properties of
mineral oils and HTH are different.
For in-service transformer fluid samples, the dissipation factor and water content values were
below the maximum allowable value listed for in-service synthetic esters per IEC 1203. The
conductivity values were all above the minimum performance value specified by ABB. The
higher dissipation factor, water content, and conductivity values for INS-07 relative to the other
transformers is likely due to the extreme operating conditions (e.g., overloads) the transformer
was subjected to as part of ABB's ongoing research project.
6.2 Objective 2, Aquatic Biodegradability
The average biodegradability of BIOTEMP® was 99% ± 3% after 21 days as measured by CEC-
L-33-T-82. Based on these results, the virgin BIOTEMP® fluid appears to biodegrade more
readily than mineral oil. Although BIOTEMP® readily biodegrades per this test, releases to
water should be prevented. The product's ability to degrade in the environment is dependent on
factors such as climate, geology, moisture, pH, temperature, oxygen concentration, dispersal of
oil, the presence of other chemicals, soil characteristics, nutrient quantities, and populations of
various microorganisms at the location (U.S.EPA 1997).
6.3 Objective 3, Flammability
The flash and fire point for the virgin and in-service fluids were consistently above the minimum
values listed in the ASTM D3487, D5222, and ABB performance specifications. The fire point
results obtained also agreed with values reported by the FMRC and UL. The flash point results
agreed with the values reported by FMRC.
June 2002 36
-------
6.4 Objective 4, Acute Toxicity
The average LCso for virgin BIOTEMP® was less than 250 mg/L, which indicates that spent
®
BIOTEMP may exhibit a hazardous characteristic per 22CCR 66261. 24(a)(6) based on limited
data for virgin product. The end-user should characterize their spent BIOTEMP® at the time of
disposal since changes to the oil may occur due to use, storage, or age. End-users should also
consult their appropriate regulatory authority about the applicable waste characterization
definitions and available disposal options for their State.
6.5 Other Verification/Certification Objectives
Chemical Composition
Verification test results for the virgin BIOTEMP® samples showed the fluid consisted of oleic
acid ranging from 80.1% ± 0.3%, diunsaturated fatty acids ranging from 10.5% ±0.1%,
triunsaturated fatty acids ranging from 0.3% ± 0.0%, and saturated fatty acids ranging from 9.2%
± 0.2%. The in-service transformer fluid had oleic acid ranging from 79.5% to 84.4%,
diunsaturated fatty acids ranging from 5.3% to 10.7%, triunsaturated fatty acids ranging from
0.2% to 0.3%, and saturated fatty acids ranging from 9.5% to 10.0%. These results are consistent
with the formulation provided by ABB.
The virgin BIOTEMP® fluid contained phenolic antioxidants ranging from 3,207 mg/kg ±103
mg/kg while the in-service transformer samples had concentrations between 2,990 mg/kg and
3,600 mg/kg. These concentrations were similar to the formulation provided by ABB.
For the 65 standard SVOC compounds analyzed by the HML lab, only n-nitrodiphenylamine was
detected around the detection limit of 20 mg/L for the virgin and in-service transformer samples.
Other tentatively identified compounds were TBHQ, 2-isopropyl-l,4-benzenediol,
2,3-dihydro-2-methyl-5-phenyl-benzofuran, 2-isopropyl-l,4-benzoquinone,
p,p'-dioctyldiphenylamine, beta-sitosterol, squalene, and vitamin E.
Metals were not detected in the in-service transformer samples except for INS-2, which had a
zinc concentration of 2.3 mg/kg. For the virgin samples, copper was detected at 4. 13 mg/kg in
sample BIO-01. Barium was detected at 0.31 mg/kg in samples BIO-05, 0.32 mg/kg in BIO-07,
and 0.32 mg/kg in BIO-10. Zinc was detected at 2.02 mg/kg in sample BIO-10.
Worker Health and Safety
Based on the MSDS information, BIOTEMP® and mineral oil-based transformer fluids appear to
have similar PPE requirements for material handling. When the PPE requirements for silicone
oil-based transformer fluids are compared to BIOTEMP®, BIOTEMP® has less stringent PPE
requirements. BIOTEMP® also has a slightly higher nuisance particulate OSHA PEL than
mineral oil. The end-user must comply with all applicable worker health and safety regulations
concerning this product.
June 2002 37
-------
The ingredients for BIOTEMPR appear to pose less of a health risk than those listed for the select
mineral oil-based and silicone oil-based transformer fluids reviewed as part of this
verification/certification. These select mineral oil-based transformer fluids listed a hydrotreated
light naphthenic petroleum distillate, which is an IARC confirmed carcinogen, ranging from 30-
100%. The silicone oil-based transformer fluids listed dimethyl polysiloxane as the primary
ingredient at 100%, which is a teratogen in animals.
Estimated Cost of Using BIOTEMPR versus a Mineral Oil
The initial purchase cost of a new transformer unit containing BIOTEMP® costs approximately
1.25 tol.30 times more than a comparable mineral oil-based transformer. When comparing the
price per gallon of BIOTEMP® to mineral oil, the difference may be between $4 to $9 more
depending on the volume purchased. Based on historical accelerated aging test results, the
estimated life expectancy of a BIOTEMPR transformer is estimated to be 20 years which similar
to a comparable mineral oil-based transformer.
June 2002 38
-------
Section 7. Vendor's Comment Section
The following information was provided by ABB Inc. The purpose is to provide the vendor with
the opportunity to share their comments regarding their environmental technology verification
report. This information does not reflect agreement or approval by U.S. EPA and Cal/EPA.
Vendor's Comment:
ABB is concerned that the aquatic toxicity of BIOTEMP as determined by the California
Environmental Protection Agency, Department of Toxic Substances Control, may be
misrepresented by the method of evaluation chosen as the preferred one. ABB had previously
determined the LCso for fathead minnows was 776 mg/L wherein the US EPA method, Methods
for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine
Organisms, was used. California EPA using a different method has determined that the LCso
values were less than 250 mg/L. By California EPA definition, a waste is considered to exhibit a
toxic characteristic if the LCso is less than 500 mg/L.
The US EPA method utilizes samples prepared with a carrier solvent to make the oil miscible in
water. The California EPA used a system of sample preparation referred to as the wrist-action
shaker method. The wrist-action shaker method can reasonably be expected to involve both a
systemic effect (toxic to the minnows' system) as well as a physical effect (mixing the oil with
the water in a fashion such that the gills and the skin of the minnows can be coated). The US
EPA method used by ABB in their evaluation uses a carrier solvent which is chosen to be less
toxic that the material being investigated. This causes a shift to a matrix that is more directly
restrictive to the systemic effect on the minnow.
California EPA maintains that the wrist-action shaker method may be more applicable to real
world spill incidents, but there is no real evidence to suggest that this is the case. The wrist-
action shaker method can be more appropriate when materials such as powders or other difficult
to dissolve substances are being tested. It is likely that any vegetable oil subjected to this method
will suffer from enhanced toxicity by including the physical (mixing) as well as systemic (actual
toxicity) effects together.
ABB does not wish to minimize either the hazards or the toxicity of their fluid. The physical
hazards are well known and are stated in the Materials Safely Data Sheet. We are principally
concerned that the aquatic toxicity is at best exaggerated by the wrist-action shaker method and
that an enhanced level of alarm may be construed from this.
June 2002 39
-------
REFERENCES
Boatelle Riera, J., et.al., Recycled Cooking Oils: Assessment of Risks for Public Health.,
September 2000.
Ciba-Geigy Corporation, Material Safety Data Sheet for antioxidant (proprietary information),
1996.
Cloesen, C. & Kabuya, A., Research RWN° 2174 Physical and chemical proper ties of
environment friendly lubricants, no date.
CONCAWE, Lubricating OilBasestocks, pp. 20-22, June 1997.
Cooper Power Systems, Personal communication with Suzanne Davis, September 2001.
Department of Toxic Substances Control (DTSC), Used Oil and Oil Filter Management, Fact
Sheet, April 2001.
Eastman Chemical Company, Material Safety Data Sheet for antioxidant (proprietary
information), 1996.
Factory Mutual Research Corporation, Approval Report, BIOTEMP* Less-Flammable
Transformer Fluid, March 1999.
Franklin, A.C., and Franklin, D.P., The J & P Transformer Book, A Practical Technology of The
Power Transformer, llth Edition, 1983, pp.1-34, 391-420.
International Programme on Chemical Safety (IPCS), INCHEM, Environmental Health Criteria
20, Selected Petroleum Products, 1982.
Lewand, Lance R., Laboratory Evaluation of Several Synthetic and Agricultural-Based
Dielectric Liquids, 2001 Doble Client Conference, March 2001.
Lewis, Richard J., Rowley's Condensed Chemical Dictionary, Thirteenth Edition, 1997.
Lewis, Sr., Richard J., Sax's Dangerous Properties of Industrial Materials, 2000.
Oommen, Thottathil V., Claiborne, C. Clair, Electrical Transformers Containing Electrical
Insulation Fluids Comprising High Oleic AcidOilCompositions, U.S. Patent Number 5,949,017,
September 7, 1999.
Polisini, Jim, Evaluation of Aquatic Toxicity Test Results for Soybean Oil-Based Transformer
Fluid (PCA 40088, Site 720045 and PCA 40088, Site 720044 WP72), July 16, 2001.
June 2002 40
-------
Underwriter Laboratories, File MH2689, Project 98NK3283, Report on Dielectric Mediums
Under the classification Program ABB Power T&D Co. Inc., June 1998.
U.S. Army Corps of Engineers (USAGE), Engineering and Design Environmentally Acceptable
Lubricating Oils, Greases, and Hydraulic Fluids., April 1997.
USAGE, Engineering and Design Environmentally Acceptable Lubricating Oils, Greases, and
Hydraulic Fluids, February 1999.
U.S. EPA, Analysis of the Applicability of EPA 's SPCC Program to the Electric Utility Industry,
July 1996.
U.S. EPA, Managing Used Oil-Advice to Small Businesses, November 1996.
U.S. EPA, Oil Pollution Prevention; Non-Transportation Related Onshore Facilities; Rule,
October 20, 1997, Volume 62, Number 202, pp. 54507-54543.
U.S. EPA, Oil Pollution Prevention Response; Non-Transportation-Related Facilities, June 30,
2000, Volume 65, Number 127, pp. 40775-40817.
U.S. EPA, SPCC Requirements and Pollution Prevention Practices for Electrical Utilities,
February 1998.
Western Area Power Administration, Electric Power Training Center Transformers Lesson XII,
no date.
June 2002 41
------- |