United States E PA - 600 /R- 98- 034
Environmental Protection
April 1998
<&EPA Research and
Development
PERSONAL COMPUTER MONITORS:
A SCREENING EVALUATION OF
VOLATILE ORGANIC EMISSIONS FROM
EXISTING PRINTED CIRCUIT BOARD LAMINATES
AND POTENTIAL POLLUTION PREVENTION ALTERNATIVES
Prepared for
Office of Prevention, Pesticides, and
Toxic Substances
Prepared by
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
-------
FOREWORD
The U. S. Environmental Protection Agency is charged by Congress with pro-
tecting the Nation* s land, air. and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions lead-
ing to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate. EPA's research
program is providing data and technical support for solving environmental pro-
blems today and building a science knowledge base necessary to manage our eco-
logical resources wisely, understand how pollutants affect our health, and pre-
vent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risks
from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air.
land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is-to catalyze
development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy decisions; and provide technical support and infor-
mation transfer to ensure effective implementation of environmental regulations
and strategies.
This publication has been produced as part of the Laboratory's strategic long-
term research plan. It is published and made available by EPA's Office of Re-
search and Development to assist the user community and to link researchers
with their clients.
E. Timothy Oppelt. Director
National Risk Management Research Laboratory
EPA REVIEW NOTICE
Thisreport has been peer and administratively reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
TWs document is available to the public through the National Te
-------
EP.A-600/R-98-034
April 1998
Personal Computer Monitors: A Screening
Evaluation of Volatile Organic Emissions
from Existing Printed Circuit Board Laminates
and Potential Pollution Prevention
Alternatives
Prepared by:
Dean R. Cornstubble and Donald A. Whitaker
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
EPA Cooperative Agreement No.: CR-822025-01
EPA Project Officer: Kelly W. Leovic
National Risk Management Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
Abstract
A vital operating component in many electronic products today is the printed circuit board.
In particular, printed circuit boards can be found in office products such as personal computers
(PCs), telephones, fax machines, and photocopiers. Many of these products are placed in
environments where indoor air quality and human exposure to volatile chemicals are major
concerns. Volatile chemicals emitted from office equipment during on-time use have been
attributed to adverse health effects in humans.
Offgassing from office products is most prominent during the initial break-in period when
electrical heating occurs. This is especially true in the case of PC monitors, the internal operating
temperatures of which can range from 60° to 70°C at full power.
In this evaluation, printed circuit board laminates, without circuitry, commonly found in
PC monitors were tested to determine if an alternative laminate would be less emitting than
conventional laminates. Test results qualitatively showed that the alternative, a glass/lignin-
containing epoxy resin laminate, emits fewer volatile compounds than paper/phenolic resin-based
laminates. The data also suggest that, if these laminates were used as a replacement for
paper/phenol circuit board laminates in PC monitors, reductions in emissions from PC monitors
could be achieved in indoor environments where they are used.
This report was submitted in fulfillment of EPA Cooperative Agreement No. CR-822025-01
by Research Triangle Institute under the sponsorship of the U.S. Environmental Protection Agency.
This report covers a period from October 1995 to October 1997, and work was completed as of
October 199 7.
-------
TABLE OF CONTENTS
Abstract ii
Figures v
Tables ...; v
Section 1.0 Project Description 1
1.1 Introduction 1
1.2 Project Background 2
1.3 Objective 3
1.4 Laminate Selection 3
1.5 Report Overview 4
Section 2.0 Materials and Methods 5
2.1 Materials 5
2.2 Testing Equipment 5
2.3 Measurements and Evaluations 6
2.4 Sampling Procedures 8
2.4.1 VOCs 9
2.4.2 Aldehydes and Ketones 9
2.4.3 Phenol/Cresol Samples 9
2.5 Analytical Procedures 9
2.5.1 VOCs (Including Phenol and Methylphenols [Cresols]) .-. 9
2.5.2 Aldehydes and Ketones 12
2.6 Data Objectives: Sampling and Analytical Objectives 13
Section 3.0 Results 14
3.1 Data Summary 14
3.1.1 Aldehydes and Ketones 14
3.1.2 VOCs 15
3.2 Aldehyde and Ketone Results 16
3.2.1 Formaldehyde Results - Figure 3-3 16
3.2.2 2-Butanone Results - Figure 3-4 16
3.2.3 Benzaldehyde Results - Figure 3-5 16
3.3 VOC Results 17
3.3.1 Methyl Ether - Figure 3-6 17
3.3.2 Toluene - Figure 3-7 17
3.3.3 Ethylbenzene - Figure 3-8 17
3.3.4 Salicylaldehyde - Figure 3-9 17
3.3.5 Phenol and m,p-Cresols - Figures 3-10 and 3-11 18
Section 4.0 Conclusions 30
Section 5.0 Data Quality 31
5.1 Quality Assurance Project Plan 31
5.2 Data Quality Indicator Coals 31
5.2.1 Precision 31
iii
-------
5.2.2 Accuracy 32
5.2.3 Completeness 33
5.2.4 Comparability 33
5.2.5 Representativeness 34
5.3 Quality Control Program 34
5.4 RTI Internal Technical System Audit (ISA) Results 34
5.5 Data Reduction, Validation, and Reporting 34
5.5.1 Chemical Data Reduction 34
5.5.2 Chemical Data Validation 36
Section 6.0 References 37
Appendix A Internal Operating Temperature of a Color Monitor A-1
Appendix B Industry Background B-l
Appendix C Concentration and Emission Factor Data: Aldehydes and Ketones C-1
Appendix D Concentration and Emission Factor Data: VOCs (Including Phenol/Cresols) ... D-1
Appendix E Internal QA Audit E-1
IV
-------
FIGURES
2-1 Chamber and sampling set-up 7
2-2 Sample can arrangement in chamber oven 7
3-1 Sum of measured concentrations at time t - 0 h and t - 336 h of atl reported aldehydes
for each laminate tested 19
3-2 Sum of measured concentrations at time t - 0 h and t - 336 h for all reported VOCs for
each laminate tested 20
3-3 Formaldehyde emission factors for paper/phenolic resin-based laminates tested 21
3-4 2-Butanone emission factors for the paper/phenolic resin-based laminates tested 22
3-5 Benzaldehyde emission factors for the paper/phenolic resin-based laminates tested 23
3-6 VOC emission factors (methyl ether) for each glass/lignin laminate tested 24
3-7 VOC emission factors (toluene) for the paper/phenolic resin-based laminates tested ... 25
3-8 VOC emission factors (ethylbenzene) for the paper/phenolic resin-based laminates tested
26
3-9 VOC emission factors (salicylaldehyde) for the paper/phenolic resin-based laminates tested
27
3-10 VOC emission factors (phenol) for the paper/phenolic resin-based laminates tested 28
3-11 VOC emission factors (m,p-cresols) for the paper/phenolic resin-based laminates tested
29
TABLES
1-1 VOCs Emitted from PC Monitors / VDTs (Brooks 1993) 1
2-1 Conditions for Steel Can Testing 6
2-2 Test Matrix 6
2-3 GC/MS Operating Conditions for Analysis of VOCs -. 10
2-4 HPLC Operating Conditions for Analysis of Aldehyde Emissions 12
5-1 Summary of Data Quality Indicator Coals 32
5-2 Concentration Data and Levels of Detection 35
-------
Section 1.0
Project Description
1.1 Introduction
A vital operating component in many electronic products today is the printed circuit board.
Printed circuit boards can be found in electronic products in markets such as communications,
consumer electronics, computers, industrial electronics, and instrumentation. Many of these
products exist in residential or office environments in which indoor air quality and human
exposure to volatile chemicals are major concerns. In particular, volatile chemicals emitted from
electronic equipment as a whole have been attributed to such adverse health effects as cancer,
chronic obstructive lung disease, allergy, tmmunologic disorders, mucous membrane irritation,
and odor-related phenomena (Brooks 1993). Pollutants emitted include volatile organic
compounds (VOCs), most of which are classified as hazardous air pollutants (HAPs) under the
Clean Air Act.
Offgassing from electronic products is most prominent during the initial break-in period
when electrical heating occurs. This is especially true in the case of personal computer (PC)s
monitors/video display terminals (VDTs), the internal operating temperatures of which can range
from 60° to 70°C at full power (see Appendix A). Studies have shown that total volatile emission
rates from new PC monitors / VDTs take approximately 144 to 360 h of on-time operation to
decay to insignificant levels (-10 ^g/h) (Brooks 1993). Typical VOCs emitted from PC monitors /
VDTs are shown in Table 1-1.
The implications of these emissions can be particularly significant in indoor air
environments such as schools, computer rooms, or offices that may contain numerous pieces of
new electronic equipment.
Table 1-1. VOCs Emitted from PC Monitors / VDTs (Brooks 1993)
Classified Classified
VOCs as a HAP VOCs as a HAP
n-Butanol (n-butyl alcohol) Diisooctyl phthalate
2-butanone (methyl ethyl ketone) X 2-ethoxyethyl acetate
2-butoxyethanol Ethylbenzene X
Butyl-2-methylpropyl phthalate Phenol X
Caprolactam X Toluene X
Cresol X Xylene X
Cyclosiloxane
Thus, demonstrating the potential reduction of volatile emissions through the use of low-
emitting alternative construction materials, such as printed circuit board laminates, could result in
-------
extensive application in the electronics industry effecting an improvement in air quality in indoor
environments (i.e., residences, schools, and offices) where computer electronic equipment is used.
1.2 Project Background
In March 1994, the U.S. Environmental Protection Agency's (EPA's) National Risk
Management Research Laboratory's Indoor Environment Management Branch met with a group of
industry technical advisors to discuss research plans for an existing cooperative agreement with
EPA entitled "The Application of Pollution Prevention to Reduce Indoor Air Emissions from Office
Equipment." Under Section 6602(b) of the Pollution Prevention Act of 1990 (Habicht, 1992),
Congress established a national policy that
Pollution should be prevented or reduced at the source whenever feasible
Pollution that cannot be prevented should be recycled in an environmentally safe manner
whenever feasible
Pollution that cannot be prevented or recycled should be treated in an environmentally
safe manner whenever feasible
Disposal or other release into the environment should be employed only as a last resort
and should be conducted in an environmentally safe manner.
Pollution prevention means "source reduction," as defined under the Pollution Prevention
Act, and other practices that reduce or eliminate the creation of pollutants through
Increased efficiency in the use of raw materials, energy, water, or other resources
Protection of natural resources by conservation.
The Pollution Prevention Act defines "source reduction" to mean any practice that
Reduces the amount of any hazardous substance, pollutant, or contaminant entering any
waste stream or otherwise released into the environment (including fugitive emissions)
prior to recycling, treatment, or disposal
Reduces the hazards to public health and the environment associated with the release of
such substances, pollutants, or contaminants.
The term includes: equipment or technology modifications, process or procedure
modifications, reformulation or redesign of products, substitution of raw materials, and
improvements in housekeeping, maintenance, training, or inventory control.
In discussions at the March 1994 technical advisors meeting, EPA, RTI, and the office
equipment industry participants identified several areas where pollution prevention could be
applied to improve indoor air quality. One such area was the evaluation of emissions from printed
circuit board laminates and the development of a low-polluting, cost-effective alternative. The
industry technical advisors agreed that such a replacement could have broad-based application in
numerous electronic products.
One company, under a grant from the Defense Sciences Office of the Advanced Research
Projects Agency, had already researched and was developing a new type of printed circuit board
laminate. This company was developing a circuit board using a resin composed of wood lignin
-------
and epoxy. Lignin is the noncarbohydrate portion of extractive-free wood, accounting for 20 to 30
percent of the weight of the wood. In the manufacture of highly purified pulps for papermaking,
all of the lignin is removed from the wood. Historically, the lignin fraction produced during
pulping has been considered either a waste product or an energy source.
Preliminary research by this company indicated that part of this waste lignin could be used
as a component in the manufacture of electronic circuit boards. The lignin-containing epoxy resin
laminate is produced by replacing the volatile component - the reactant - with a nonvolatile
component, the wood lignin. To date, this company has manufactured a lignin-containing epoxy
resin circuit board that has the required structural and performance characteristics for use in
electronic equipment at or below the cost of manufacturing glass/epoxy circuit boards. Thus, the
lignin-containing epoxy resin laminate could address indoor air quality problems by reducing
emissions from electronic equipment containing these laminates. In addition, use of the lignin-
containing epoxy resin laminates would address a portion of the solid waste problem in the
papermaking industry through the use of waste lignin.
Other EPA-funded projects related to the printed circuit board industry include EPA's
Design for the Environment case studies on the Printed Wiring Board industry in pollution
prevention work practices, onsite etchant regeneration, opportunities for acid recovery and
management, and plasma desmear (U.S. EPA, 1990,1995a, 1995b, 1995c, and 1996).
These projects, however, focus only on the reduction of emissions from manufacturing processes
in developing printed wiring boards and not on emissions from printed circuit board laminates
during their service life.
1.3 Objective
The primary objective of this project was to conduct a screening test to measure volatile
emissions from printed circuit board laminates made from glass/lignin-containing epoxy resin and
other laminate types to determine if these new laminates would be lower-emitting than
conventionally used laminates. Because laminates used in PC monitors are subjected to some of
the highest operating temperatures in an indoor air environment, they were selected as test
laminates.
1.4 Laminate Selection
Under a cooperative agreement with EPA, Research Triangle Institute (RTI) conducted a
screening evaluation to measure emissions from select laminate materials. These laminates were
chosen based on discussions with the industry as to what materials were the most common in PC
monitors/VDTs. Laminates selected for evaluation were based on the following laminate
base/resin binders:
Giass/lignin-containing epoxy
Glass/epoxy
Paper/phenol
Paper/reformulated phenolic resin.
Glass/lignin-containing epoxy resin laminates are relatively new to the printed circuit board
laminate marketplace. The resin portion, the epoxy resin, in these circuit board laminates can be
-------
manufactured in different ways. The most common methods are to manufacture the laminate with
100 percent epoxy resin or to use a two-component resin system where -50 weight percent epoxy
resin is combined with -50 weight percent of a curing agent. The curing agent is typically based
on a phenolic compound. These laminates have not seen widespread use in the industry.
Class/epoxy laminates are the most widely used material in the printed circuit board
industry because they have properties that satisfy the electrical and mechanical needs of most
applications. These laminates are constructed on multiple plies of epoxy-resin-impregnated woven
glass cloth. Their electrical, physical, and thermal properties make them an excellent material for
high-technology applications. They are typically used in aerospace, communications, computers
and peripherals, industrial controls, and automotive applications.
Paper/phenol and paper/reformulated phenol laminates are one of the cheapest printed
circuit board laminates manufactured in the printed circuit board industry. These laminates are
constructed on multiple plies of phenol-resin over a paper matte. Because they are inexpensive,
both the paper/phenol and the reformulated phenol laminates are primarily manufactured overseas
and are typically installed in toys, games, calculators, and PC monitors and VDT displays.
See Appendix B for a more detailed discussion of the background of the printed circuit
board laminate industry and the manufacturing process of these laminates.
Results from the screening test demonstrate that the laminates made from glass/ligntn-
containing epoxy resin are lower-emitting than the paper/phenol and the reformulated phenolic
resin laminates. Use of these laminates as substitutes for laminates used in the manufacture and
production of printed circuit boards for PC monitors could reduce the overall air emissions from
electronic office equipment. The glass/epoxy laminates were lower-emitting overall. However,
these laminates are not predominantly found in PC monitors.
1.5 Report Overview
The remainder of this report is divided into the following sections:
Section 2.0, Materials and Methods
Section 3.0, Results
Section 4.0, Conclusions
Section 5.0, Data Quality
Section 6.0, References.
Appendixes A through E contain pertinent referenced materials and the raw data for the screening
evaluation.
-------
Section 2.0
Materials and Methods
2.1 Materials
This section describes the materials and methods used to conduct a screening evaluation of
volatile emissions from the following base/resin printed circuit board laminates:
Class/lignin-containingepoxy
Glass/epoxy
Paper/phenol
Paper/reformulated phenolic.
The screening evaluation was conducted to determine if the glass/lignin-containing epoxy resin
and the reformulated phenolic laminates would be less emitting than conventional laminates
(paper/phenol). Glass/epoxy laminates were included in the evaluation because they exist
primarily in CPU units.
The glass/lignin-containing epoxy and glass/epoxy laminates were acquired from U.S.
manufacturers. The paper/phenol laminate was acquired from an overseas manufacturer because
almost all phenol-based laminates are produced overseas. The reformulated phenolic laminate is
also produced overseas and has been in use in Europe within the past 5 years.
Manufacturers participating in the collection of the test laminates were sent "sampling kits"
contain ing complete sampling instructions, labels, documentation (chain-of-custody forms), and
sample cans. These kits were sent to a designated person at each manufacturing facility who was
responsible for collecting the laminates. Each designated person was instructed to do the
following:
Take a laminate sample directly from the production line.
Cut the laminate to a sample size of 0.15 m wide by 0.25 m long.
Label the can containing the laminate with a unique sample code.
Immediately seal the laminate in the labeled, precleaned, air-tight, 7.85-L steel can
provided by RTI.
Each can was shipped by overnight delivery to RTI. Proper documentation of product type,
manufacturing information (i.e., date, time), and shipping times and dates accompanied the
laminates.
Each sealed can was inspected upon arrival at RTI. Each can was then stored at -10°C for
about 4 months prior to testing.
2.2 Testing Equipment
Emissions testing was conducted using 7.85-L (0.00785-m3) friction-sealed, flow-through
steel cans. The conditions for emissions testing are shown in Table 2-1.
-------
Table 2-1. Conditions for Steel Can Testing
Condition
Parameters
Can size
Oven and supply air temperature
Relative humidity (RH) of supply air
Air exchange rate (ACH)
Source area (A)'
Loading (L)b
ACH/L
0.00785 mj
65±3°C
50 ±5 percent
1.02/h
0.075 mj
9.43 mVm1
0.108 mVh-m2
' Source area calculated from both sides of laminate exposed (0.15 m width by 0.25 m length).
b Loading is equal to the source area of 0.075 m2 divided by the can size of 0.00785 m1.
Figures 2-2 and 2-3 illustrate the equipment used in this test.
2.3 Measurements and Evaluations
The intent of this evaluation was to measure emissions from the four different types of
laminates under controlled environmental conditions that might be typical for printed circuit board
laminates during on-time use in PC monitors. Table 2-2 illustrates the design of the test.
Table 2-2. Test Matrix
Laminate '
Air Sample Type:
VOCC
VCX duplicate
Aldehyde/ketone
Number of
sampling intervals
Number of air
samples taken
Background *
1
1
1
3
G/L1A
1
1
1
9
27
G/L1B
1
1
1
9
27
AirS
G/L2
1
1
1
9
27
amples
G/E1
1
1
1
9
27
G/E2
1
1
1
9
27
P/P
1
1
1
9
27
P/RPA
1
1
1
9
27
P/RPB
1
1
1
9
27
G/E - glass/epoxy
G/L - glass/lignin-containing epoxy
P/P - paper/phenol
P/RP - paper/reformulated phenolic
VOC - Volatile organic compound
* Structural base / binding resin. Laminates supplied without circuitry; copper coated on one side. Sample G/L 2 was
copper coated on both sides.
* A background air sample was taken from two separate test cans prior to the start of the test
c Includes phenol and cresols.
-------
Figure 2-1. Chamber and sampling set-up.
Figure 2-2. Sample can arrangement in chamber oven.
7
-------
Immediately prior to testing, the sealed storage cans containing the laminates were
removed from the freezer. The storage cans were opened and the laminate samples transferred to
individual clean steel cans for testing. The test cans were sealed with friction-seal lids. The lid of
each test can was fitted with Teflon inlet and outlet tubes that were attached to the supply air
manifold and sampling ports. This allowed continuous regulated air flow through the chambers
during the test period. Fine metering valves were used to adjust and regulate the airflow to each
individual test can.
The test cans were then placed in a temperature-controlled oven maintained at 65 ± 3 °C.
Temperature of the oven and humidity of the supply air (50 ± 5 percent RH at 23 °Q was
monitored and recorded using temperature/humidity probes. The humidity of the supply air at
23°C was 50 percent RH; however, as the air was warmed to 65°C, the percent RH dropped to
approximately 6 percent.
Total air flow to the system was controlled, monitored, and recorded using mass flow
controllers. Just prior to placing the test cans containing the laminate samples in the oven, each
can was purged for several minutes with clean air to flush the cans of laboratory air trapped during
the laminate transfer process. Collection of air samples from the test cans began immediately
(within 10 minutes of placing the test samples in the cans). Flows were measured and adjusted
immediately after the test cans were placed in the oven and during the middle and again at the end
of testing.
Background air samples from two test cans were collected prior to placing laminates in
them to provide information on the system background. For each laminate sample, air samples
were collected during times 0-2, 4-6, 11-13, 23-25,47-49, 95-97, 143-145, 215-217, and 335-337
hours. Mid point times for these sampling periods were 1, 5,12, 24,48, 96,144, 216, and 336
hours.
An initial set of air samples from each laminate sample was analyzed to identify specific
chemicals emitted from each laminate sample. Chemical identification consisted of analyzing
these air samples for the following compounds: VOCs, including phenol and cresols, and
aldehydes and ketones. A list of target compounds was generated from these analyses to perform
quantitative analyses on the remaining air samples for each laminate sample.
A list of target compounds was identified from an initial set of air samples taken for each
laminate sample. Quantitative analysis of these target compounds was then conducted for all
subsequent air samples taken for each laminate sample.
2.4 Sampling Procedures
Air samples were collected from each steel can containing test laminates using the
following methods:
Multisorbent cartridges for VOCs including phenol and methylphenol (cresol) analyses
DNPH-coated silica gel cartridges for aldehyde and ketone analyses
NaOH impinger for phenol and cresols.
8
-------
VOC samples were collected in duplicate; however, not all the duplicate samples were analyzed.
Some were analyzed as screening samples and some as quality control samples. The remaining
samples were stored at -10°C as backup samples in case they were needed until analysis was
complete. Single samples of aldehydes/ ketones were collected. The VOC and aldehyde/ketone
samples were collected simultaneously, followed by collection of the P/C samples. A brief
description of each of these collection methods follows.
2.4.7 VOCs
VOC, including phenol and methylphenols (cresols), emissions from each can's air space
were collected by passing can air through multisorbent cartridges containing Tenax TA and
Carboxen 1000 (200 mm x 6 mm outside diameter). Air samples were collected at a flow rate of
approximately 25 ml/min over a 2-h period to give nominal sampling volumes of 3 L.
2.4.2 Aldehydes and Ketones
Air samples for aldehydes and ketones were collected on DNPH-coated silica gel
cartridges. General procedures for use are outlined in EPA Reference Method TO-11. Air samples
were collected at a flow rate of approximately 30 ml/min for a 2-h period to provide nominal
sampling volumes of 3.6 L.
2.4.3 Phenol/Cresol Samples
Air samples were collected from each steel can containing test laminates for phenol/cresols
using an impinger method containing NaOH for phenol and cresols analyses. These samples were
collected using midget impingers containing NaOH. General procedures for this method can be
found in EPA Method TO-8. Air samples were collected at a flow rate of approximately 50 ml/min
for a 2-h period to provide nominal sampling volumes of 6 L. The phenol/cresol samples were
collected immediately following the collection of the VOC and aldehyde/ketone samples.
Collection in this manner was necessary since the total air flow through each can was only
approximately 100 mL/min. Collection of all samples at the same time would require a flow
exceeding each can's air flow thereby diluting pollutant concentration of each can's air space.
Quantitative analysis for phenol and cresols was not performed using EPA Method TO-8
because of a lack of project resources and because Method TO-8 has not been rigorously
evaluated by the scientific community. However, these compounds were included in the VOC
analysis as discussed in Section 2.5.1.
2.5 Analytical Procedures
2.5.7 VOCs (Including Phenol and Methylphenols [Cresols])
VOCs on exposed cartridges were thermally desorbed then analyzed by gas
chromatography/mass spectromefry (GC/MS). The operating conditions for the GC/MS are given
in Table 2-3.
Unknown sample constituents from the initial air samples for each laminate sample were
identified using an electronic search of the NIH/EPA/MSDC Mass Spectral Data Base (NBS library)
-------
and the Registry of Mass Spectral Library (Wiley library). Data were reviewed manually to verify
computer identifications and to identify compounds not found
Table 2-3. GC/MS Operating Conditions for Analysis of VOCs
Parameter
Setting
THERMAL DESORPTION
Trap type
Tube raised ambient
Initial carrier flow
Tube chamber heat time
Tube chamber temperature (maximum)
Secondary carrier flow
Trap 1 heat (maximum)
Trap 2 heat (maximum)
Trap-to-trap transfer time
Trap-to-column transfer time
CAS CHROMATOCRAPH
Instrument
Column
Temperature program
Carrier gas flow rate
MASS SPECTROMETER
Instrument
lonization mode
Emission current
Source temperature
Electron multiplier
1 - multisorbent or Tenax, 2 - multisorbent or Tenax
Off
1 min
6 min
320°C
2 min
270°C
310°C
2 min
20 min
Hewlett-Packard, Model 5890
DB-1 or DB-624 widebore fused silica capillary column
35°C (5 min) to 200°C (1 min) at 5°C/min
I.BmL/min
Hewlett-Packard, Model 5988A
Electron ionization scan 35-350 m/z
0.3mA
200°C
2,000 voltsa
* Typical value.
using the computer library search. Results of these analyses were used to select target VOCs for
quantitative emissions tests in subsequent analyses of air samples for each laminate sample.
Prior to air sample analysis, a set of standard cartridges were analyzed to show proper mass
calibration for the GC/MS system, to establish windows for CC retention time for selected VOCs,
and to generate instrumental response factors for target VOC quantitation. Standard cartridges
were spiked with known amounts of toluene and aliphatic hydrocarbons ranging in volatility from
10
-------
n-hexane to n-tetradecane. Two external standards, perfluorotoluene and
bromopentafluorobenzene (BFB), were added to each standard cartridge. Perfluorotoluene was
used to monitor instrumental tune (mass resolution and ion abundance) and BFB was used as an
external quantitation standard. For each day of sample analysis, an additional standard cartridge
was analyzed to demonstrate ongoing instrument performance.
During quantitative emissions tests, identification of target analytes was based on
chromatographic retention times relative to standards and the relative abundances of extracted ion
fragments selected for quantitation. Quantitation was accomplished using chromatographic peak
areas derived from extracted ion profiles. Calibration standards containing the target analytes were
prepared on Tenax TA cartridges at masses ranging from 10 to 500 ng/cartridge. Each calibration
standard and sample contained a known mass of the quantitation standard,
bromopentafluorobenzene. Standards were loaded by injecting a methanol solution with the
compounds into a flash evaporation system that vaporizes the compounds and flashes them into a
Tenax cartridge. Tenax, instead of multisorbent, cartridges were used for standards because the
methanol is not easily flushed off the multisorbent cartridge during the loading process and the
methanol interferes with the GC/MS analysis.
Relative response factors (RRFT) were calculated using the following equation:
A, x M0,
= -I - 2£ (2-1)
x MT
where
AT - peak area of the target analyte
MQS - mass of quantitation standard (ng/cartridge)
AQS - peak area of the quantitation standard (ng/cartridge)
MT - mass of target analyte (ng/cartridge).
Mean values and standard deviations of the RRFs were calculated for each target VOC.
The calibration curve was considered acceptable if the standard deviation for each response factor
was less than 30 percent.
During each day of sample analysis, an additional standard was analyzed. If the RRF
values for this standard were within ±25 percent of the RRFs obtained during instrument
calibration, the GC/MS system was considered "in control.* The RRF from the calibration was
used to calculate the mass of the target VOC on sample cartridges (MT). The ^ was calculated
using the following equation:
A, x M0,
M = I - 2L (2-2)
AQS x RRF7
11
-------
The target VOC was calculated from the reconstructed ion chromatogram (RIC). The total
area of the RIC was integrated for the retention time window from n-hexane through n-tetradecane.
The target VOC concentration was calculated based on the average total ion response factor
generated for toluene.
2.5.2 Aldehydes and Ketones
Quantitative analysis for aldehydes and ketones was generated from the initial samples for
each type of laminate. These values were then used in evaluating emissions from each type of
laminate. DNPH/aldehyde derivatives on sample cartridges were extracted by eluting each
cartridge with 5 mL of HPLC-grade acetonitrile into 5-mL volumetric flasks. The final volume was
adjusted to 5.0 ml and the samples aliquoted for analysis. Blank cartridges were eluted with each
sample set to identify background contaminants. Additional blank cartridges were spiked with
known amounts of DNPH/aldehyde standards as a method of assessing recovery.
DNPH/aldehyde derivatives in sample extracts were analyzed by HPLC with ultraviolet
(UV) detection. HPLC operating conditions for the analysis of aldehyde emissions are shown in
Table 2-4.
Table 2-4. HPLC Operating Conditions for Analysis of Aldehyde Emissions
Parameter . Setting
Instrument Waters Series 510
Column NOVA-PAKC18, 3.9x150 mm
Solvent system A: Water/Acetonitrite/Tetrahydrofuran 60/30/10 v/v
B: Acetonitrile/Water 40/60 v/v
Gradient 100% A for 3 min; then a linear gradient to 100% B in 10 min;
hold 15 min at 100% B
Mobile phase flow rate 1.5 mL/min
Injection size 20 //I
Ultraviolet wavelength 360 nm __
Purified and certified DNPH derivatives of the target aldehydes were used for the
preparation of calibration solutions. Target aldehydes were identified by comparison of their
chromatographic retention times with those of the purified standards. Quantitation of the target
compounds was accomplished by the external standard method using calibration standards
prepared in the range of 0.02 to 15 ng/,uL of the DNPH/aldehyde derivatives. Standards were
analyzed singly for the DNPH/aldehyde derivatives and a calibration curve calculated by linear
regression of the concentration and chromatographic response data. Calibration curves for all
target analytes must have an r2 * 0.998.
To demonstrate ongoing instrument performance, a calibration standard was analyzed each
day prior to the analysis of any samples. The calibration was considered "in control* if the
12
-------
measured concentration of the DNPH/aldehyde derivatives in the standard was 85 to 115 percent
of the prepared concentration.
The concentration of each target analyte in each air sample was calculated as follows:
Cy x V x Df
c* = ~r? (2-3)
where
C, - Concentration of aldehyde in the sample (^g/m3)
Cy - Concentration of DNPH/analyte derivative in the sample extract (ng/,uL)
Vy - Total volume of sample extract (i.e., 5,000 /A)
DF - Molecular weight of analyte + molecular weight of DNPH/ana!yte derivative
Vs - Sample volume in L.
2.6 Data Objectives: Sampling and Analytical Objectives
Duplicate air samples collected during emissions tests at time intervals specified in
Table 2-2 were stored and analyzed approximately 2 months later. Analytical evaluations of air
samples collected from the steel cans was performed using standard analytical techniques.
For a detailed discussion on data goals and results from this screening evaluation, see
Sections 3.0 and 5.0, respectively.
13
-------
Section 3.0
Results
Each laminate sample tested in this evaluation project was collected from separate printed
circuit board laminate manufacturers and placed in separate stainless steel, friction sealed, flow-
through cans as prescribed in Section 2.1 . Each can was placed in an incubator, heated to 65 °C,
and held at this temperature for 2 weeks (336 h), simulating the approximate operating
temperature of a full-powered PC monitor over time.
The temperature of the oven and the supply air percent RH were monitored and recorded
continuously for the duration of the test. The oven temperature averaged 65 °C and RH averaged
48 percent.
Over the 2-week test period, air samples were collected and analyzed to identify chemicals
emitted from each laminate tested, specifically aldehydes, ketones, and VOCs. The decay of
emissions over time was then plotted for each laminate tested.
Results of this test are described in three subsections: data summary, aldehyde and ketone
results, and VOC results. Figures corresponding to select emissions data are compiled in
respective order beginning on page 1 7. Only select aldehydes, ketones, and VOCs are presented
and discussed in this section.
For more detailed information on air sample concentration data results, including all
reported aldehydes, ketones, and VOCs, see Appendixes C and D. the tables in these appendixes
were used to generate the figures presented in this section.
3.1 Data Summary
The data presented in Figures 3-1 and 3-2 summarize the results of the total identified
aldehydes, ketones, and VOCs emitted from each printed circuit board laminate tested. The data
are expressed in terms of concentration U/g/m3) and illustrate the sum of measured concentrations
for all identified compounds emitted from each printed circuit board laminate at t - 0 h and
t - 336 h.
3.1.1 Aldehydes and Ketones
Figure 3-1 shows that, at time t - 0 h, the sum of measured aldehydes and ketones
concentrations from the paper/reformulated phenolic and the paper/phenol laminates ranged from
3,900 to 6,400 A*g/m3. However, for the same time point, the sum of measured aldehydes and
ketones concentrations from the glass/lignin laminates ranged from 200 to 270 Mg/m3 and from the
glass/epoxy laminates ranged from 33 to 200
After 336 h, of incubation at 65 "C, the sum of measured aldehydes and ketones
concentrations from the paper/reformulated phenolic and the paper/phenol laminates dropped to a
level ranging from 650 to 870 Atg/m3. For the same time interval, the sum of measured aldehydes
14
-------
and ketones concentrations from the glass/lignin laminates ranged from 42 to 53 /ug/m3 and the
glass/epoxy laminates ranged from 1 7 to 25
These numbers indicate that the paper/phenolic resin-based laminates emit more volatiles
in the first few days of simulated on-time operation than the glass/lignin or glass/epoxy laminates.
In addition, Figure 3-1 illustrates evidence that offgassing of volatile compounds would continue
beyond the 336 h of this screening evaluation.
Two other characteristics among the printed circuit board laminates are observed in
Figure 3-1 : The measured concentration values shown for the glass/epoxy sample 2 laminate are
greater than -those for the glass/epoxy laminate. This difference could have been due to the fact
that the glass/epoxy sample 2 laminate was manufactured by a separate company than the
glass/epoxy laminate. Other than this difference, it is uncertain why a large discrepancy in volatile
emissions resulted between the glass/epoxy sample 2 laminate and the glass/epoxy laminate.
All three glass/lignin laminate samples and the glass/epoxy laminates show an average of
emissions of 95 percent less than the paper/reformulated phenolic resin-based laminates.
3.7.2 VOCs
Figure 3-2 shows that, at time t - 5 h, the sum of measured VOC concentrations,
excluding aldehydes and ketones, from the paper/reformulated phenolic and the paper/phenol
laminates ranged from 9,600 to 23,000 ^g/rnV The sum of measured VOC concentrations,
excluding aldehydes and ketones> from the glass/lignin laminates ranged from less than 1 ,300 to
1,700 Mg/m3 and the glass/epoxy laminates ranged from less than 1 to 15 ^g/m3 at t - 5 h.
After 336 h of exposure at 65 °C, the sum of measured VOC concentrations from the
paper/reformulated phenolic and the paper/phenol laminates ranged from 3,900 to 6,200 ,ug/m3.
For the same time interval, the sum of measured VOC concentrations for the glass/lignin and the
glass/epoxy laminates ranged from nearly 0 to 1 00 ^g/m3.
Three observations can be made from the results shown in Figure 3-2. First, the sum of
measured VOC concentrations at t -5 h from the paper/reformulated phenolic laminates is less
than that for the paper/phenol, whereas the aldehyde and ketone concentrations were greater
(Figure 3-1). However, at t - 336 h, the converse is true. These results indicate that the
paper/reformulated phenolic laminates contain a smaller percentage of volatile matter in their resin
formulation than the paper/phenol laminate. In addition, these laminates appear to be less likely
to emit volatile matter at 65 °C than the paper/phenol laminate.
Second, as was the case for the aldehydes and ketones in Figure 3-1, all three glass/lignin
laminate samples and the glass/epoxy laminates show an average sum of measured concentrations
of VOCs to be 95 percent less than the paper/reformulated phenolic resin-based laminates.
1 Since data were not available for t - 0 h for the paper/phenol and paper/reformulated phenolic Sample A laminates,
data for t - 5 h were used in Figure 3-2.
15
-------
Third, VOC concentrations from both glass/epoxy laminates are virtually negligible. This is
a good indication, on a screening basis only, that VOC concentrations from glass/epoxy laminates
would probably not be a concern for indoor air quality issues.
3.2 Aldehyde and Ketone Results
Figures 3-3 through 3-5 illustrate the results of select identified aldehydes and ketones
analyzed for each laminate sample tested for the nine time periods. The results are shown as
emission factors expressed in ^g/h-m2. Aldehyde and ketone compounds selected for illustration
are formaldehyde in Figure 3-3, 2-butanone (commonly known as methyl ethyl ketone) in
Figure 3-4, and benzaldehyde in Figure 3-5.
3.2.1 Formaldehyde Results Figure 3-3
Figure 3-3 shows formaldehyde emission factors for the paper/reformulated phenolic resin-
based laminates over the 2-week test period. Emission factors for each of the laminates were
approximately 120 ^g/h-m2 at t - 0 h to less than 20 A^g/h-m2 at t - 336 h. This figure shows that,
in this screening evaluation, over 83 percent of formaldehyde emissions occur within the 2-week
testing period at an elevated temperature of 65°C. Also, very little difference can be observed in
emission factors between the different types of paper/reformulated phenolic resin-based laminates
tested. However, this figure shows clearly that formaldehyde emissions from these laminates
would continue beyond the 336 h of elevated temperature exposure.
3.2.2 2-Butanone Results - Figure 3-4
Figure 3-4 shows 2-butanone emission factors for the paper/reformulated phenolic resin-
based laminates over the 2-week test period. Emission factors for each of the laminates were
approximately 500 ^g/h-m2 at t - 0 h to less than 100 Mg/h-m2 at t - 336 h. This figure shows
that, in this screening evaluation, over 80 percent of 2-butanone emissions occur within the 2-
week testing period at an elevated temperature of 65°C. Also, very little difference can be
observed in emission factors between the different types of paper/reformulated phenolic resin-
based laminates tested. However, this figure clearly shows that 2-butanone emissions from these
laminates would continue beyond the 336 h of elevated temperature exposure.
3.2.3 Benzaldehyde Results Figure 3-5
Figure 3-5 shows benzaldehyde emission factors for the paper/reformulated phenolic resin-
based laminates over the 2-week test period. Emission factors for each of the laminates ranged
between 100 and 300 jug/h-m2 at t - 0 h to less than 25 Aig/h-m2 at t - 336 h. This shows that, in
this screening evaluation, 25 to over 92 percent of benzaidehyde emissions occur within the 2-
week testing period at an elevated temperature of 65°C. Unlike the formaldehyde and 2-butanone
emissions in Figures 3-3 and 3-4, the paper/phenol laminate had a lower emission factor than the
other paper/reformulated phenol samples A and B. However, again, benzaldehyde emissions from
these laminates would continue beyond the 336 h of elevated temperature exposure.
An interesting observation in Figure 3-5 is that benzaldehyde emissions from paper/phenol
laminate are approximately one-half those from the paper/reformulated phenolic laminates.
16
-------
3.3 VOC Results
Figures 3-6 through 3-11 illustrate the results of emission factors for selected VOCs
analyzed for each laminate sample tested. Those compounds selected for lustration were methyl
ether, toluene, ethylbenzene, salicylaldehyde, phenol, and m,p-cresols. Methyl ether was the
most predominant compound among the glass/lignin laminates. The other compounds listed are
exclusive to the paper/phenolic resin-based laminates. Toluene and ethylbenzene are shown
because they are known HAPs. Salicylaldehyde was chosen for illustration because this
compound had the greatest measured emission concentrations among all of the VOCs identified
for the paper/reformulated phenolic laminates.
3.3.I Methyl Ether - Figure 3-6
Figure 3-6 shows methyl ether emission factors for the glass/lignin laminates over the 2-
week test period. Emission factors for each of the laminates ranged from 200 to 275 Mg/h-m2 at
t - 0 h to less than 20 Mg/h-m2 at t - 336 h. This figure shows that, in this screening evaluation,
over 93 percent of methyl ether emissions occur within the 2-week testing period at an elevated
temperature of 65°C. Also, very little difference can be observed in emission factors between the
different samples of glass/lignin laminates tested.
3.3.2 Toluene -Figure3-7
Figure 3-7 shows toluene emission factors for the paper/reformulated phenolic resin-based
laminates over the 2-week test period. As shown in the figure, data were not available for the
paper/reformulated phenolic Sample A laminate at sampling times t - 0,12, 24, 48,144, and
216 h. Data were not available for the paper/phenol laminate at t - 0 h.
Data results show that there was a problem in the analysis for toluene for each of the
paper/reformulated phenolic resin-based laminates. Emission factors began at a low level and
cycled up and down over the 336 h of exposure for each laminate type. The only observation that
can be made from this figure is that the paper/phenol laminate appears to have a larger quantity of
toluene in its resin formulation than the paper/reformulated phenolic resin-based laminate
samples.
3.3.3 Ethylbenzene - Figure 3-8
Figure 3-8 shows ethylbenzene emission factors for the paper/reformulated phenolic resin-
based laminates over the 2-week test period. As shown in the figure, data were not available for
the paper/reformulated phenolic Sample A and paper/phenol laminates at sampling times t - 0 h.
In addition to showing that emissions consistently decay over time, Figure 3-8 shows that the
paper/phenol laminate, once again, appears to have a larger quantity of ethylbenzene in its resin
formulation than the paper/reformulated phenolic resin-based laminate samples.
3.3.4 Salicylaldehyde -Figure 3-9
Figure 3-9 shows salicylaldehyde emission factors for the paper/reformulated phenolic
resin-based laminates over the 2-week test period. As shown in the figure, data were not available
for the paper/reformulated phenolic Sample A and paper/phenol laminates at sampling times t - 0
17
-------
h. Figure 3-9 shows that salicylaldehyde did not decay over time as much was seen in the other
figures. In addition, it appears that salicylaldehyde emissions from these laminate types began to
slowly decay at 336 h and would continue to do so if the test were extended beyond the 336 h.
3.3.5 Phenol and m,p-Cresols - Figures 3-10 and 3-11
Figures 3-10 and 3-11 show emission factors for phenol and m,p-cresols, respectively, for
the paper/reformulated phenolic resin-based laminates over the 2-week test period. As shown in
these figures, data were not available for the paper/reformulated phenolic Sample A and
paper/phenol laminates at sampling time t - 0 h. Data results show that there were some data lost
for both phenols and m,p-cresols for each of the paper/reformulated phenolic resin-based
laminates.
Also, the data in Figures 3-10 and 3-11 show that emissions do not appear to steadily
decline over time as compared to those VOC compounds in Figures 3-6 to 3-9. The reason for this
discrepancy could have been due to a slower emission rate of these VOCs over time or collection
of the data could have been in error. However, without further investigation, the real cause, or
causes, to these discrepancies remains unknown.
18
-------
7,000
6,000
5,000
"I
f 4,000
3,000
2,000
1,000
Paper/reformulated Paper/reformulated
phenolic Sample B phenolic Sample A
Paper/phenol Class/lignin Sample Glass/lignin Sample Class/lignin Sample Glass/epoxy Sample Glass/epoxy Sample
1A 2 IB 2 1
[Ot = Oh » = 336 h |
Figure 3-1. Sum of measured concentrations at time t = 0 h and t = 336 h of all reported aldehydes for each laminate tested.
-------
25,000
Paper/phenol Paper/reformulated Paper/reformulated Glass/lignin Sample Glass/lignin Sample Glass/lignin Sample Glass/epoxy Sample Glass/epoxy Sample
phenolic Sample A phenolic Sample B 1A IB 2 1 2
lit« data poml for Uminjln papn/phfiiol, papf r/ rttomtulalrd phroolk Sampln A and * it
D« - OorSh ! - 336 h
I . 5 h. Oat IKK »«ibblr for I . 0 h
Figure 3-2. Sum of measured concentrations at time t = 0 or 5 h and t = 336 h for all reported VOCs for each laminate tested, excluding
aldehydes and ketones.
-------
140
120
100
BO
§
?
E
5 60
1
1
1
1
(2
144
24 48 96
Time, hours
ID Paper/phenol Q Paper/reformulated phenolic Sample A O Paper/reformulated phenolic Sample B |
216
336
Figure 3-3. Formaldehyde emission factors for the paper/phenolic resin-based laminates tested.
-------
K)
700
601)
500
400
300
200
100
S
I
1
1
II
53=.
^
1
53=,
24
48
Time, hour*
-------
U
350
300
250
200
i 150
100
r,0
1
1
i
S
r§i
12
'J
4H
Time, hours
>(.
144
216
336
| DPaper/phenol O Raper/reformulated phenolic Sample A O Paper/reformulated phenolic Sample B |
Figure 3-5. Benzaldehyde emission factors for the paper/phenolic resin-based laminates tested.
-------
300
250
200
f
100
1
I
48
Time, hours
144
216
336
DGIass/lignin Sample 1A
QGlast/lignin Sample IB
HCIass/ligoin Sample 2
Figure 3-6. VOC emission factors (methyl ether) for each glass/lignin laminate tested.
-------
»J
450
400
350
300
250
200
150
100
a b f=\
F^H
0
1 1 1 1
J 1 |b|| 1 |b| | |b| i yj |-]b I |bB | 1
5 12 24 48 96 144 216 336
* Data lost for paper/phenol.
ne' * Data lost for paper/ reformulated phenolic Sample A.
D Paper/phenol O Paper/reformulated phenolic Sample A Q Paper/reformulated phenolic Sample B |
Figure 3-7. VOC emission factors (toluene) for the paper/phenolic resin-based laminates tested.
-------
IJ
ON
30
25
Emission Factor, ug
* * 1
3 ut :
S
a b
0 \ 1
0
^^
^^
^^^
^^^
"
_
^-^
^^^
=
-
.
I
y///y/////////
|
i
5
MM
I
B
:v
12
1 1 rt
k^W V^
|l §=, sj
fc fe ffe nsfc,
V.X »v T^1 wx x^ i i K.X wx i
24 48 96 144 216 336
Time, hours " Data lost for paper/phenol.
b Data lost for paper/ reformulated phenolic Sample A.
D Paper/phenol Q Paper/reformulated phenolic Sample A O Paper/reformulated phenolic Sample Bj
Figure 3-8. VOC emission factors (ethylbenzene) for the paper/phenolic resin-based laminates tested.
-------
1200
1000
800
600
400
200
a b
1
1
r
i
12
24
48
Time, hours
96 144 216 336
* Data lost for paper/phenol.
b Data lost for paper/ reformulated phenolic Sample A.
ID Paper/phenol D Paper/reformulated phenolic Sample A D Paper/reformulated phenolic Sample B |
Figure 3-9. VOC emission factors (salicylaldehyde) for the paper/phenolic resin-based laminates tested.
-------
1400
120O
1000
800
600
400
200
0
a b =
0
[tr^TfcT^
5 12 24 48 96 144 216 336
Time, hours " Data lost for P»P«f/Phenol.
b Data lost for paper/ reformulated phenolic Sample A.
D Paper/phenol D Paper/reformulated phenolic Sample A H Paper/reformulated phenolic Sample
Figure 3-10. VOC. emission factors (phenol) for the paper/phenolic resin-based laminates tested.
-------
200
180
160
140
120
100
BO
60
40
a b
>
I
n
an
Trine, hours
144 216 336
* Data loat for paper/phenol.
* Data lot! for paper/reformulated phenolic Sample A.
ID Paper/phenol Q Paper/reformulated phenolic Sample A H Paper/reformulated phenolic Sample B
Figure 3-11. VOC emission factors (m,p-cresols) for the paper/phenolic resin-based laminates tested.
-------
Section 4.0
Conclusions
Several conclusions can be drawn from the screening test data results presented in
Section 3.0. They are as follows:
Glass/lignin laminates emit fewer volatile compounds than paper/phenolic resin-based
laminates. Although this test was conducted on only eight laminate samples of four
different laminate types, the results presented in Section 3.0 do show that, for the samples
tested, glass/lignin-containing epoxy resin laminates emit less volatile compounds than the
paper/phenolic resin-based laminates during on-time operation at 65°C. As is evident in
Figures 3-1 and 3-2, the results qualitatively show that the glass/lignin laminates do emit
less than the paper/phenolic resin-based laminates. The data also suggest that, if these
laminates were used as pollution prevention alternatives for paper/phenol circuit board
laminates in PC monitors, reductions in VOC emissions from PC monitors could be
achieved in indoor environments where they are used. However, an initial exposure
period at an elevated temperature would be beneficial in reducing volatile emissions prior
to human operation.
Volatile emissions from glass/epoxy laminates are relatively small compared to glass/lignin
and paper/phenol laminates. The data results clearly show that there are minimal VOC
compounds emitted from glass/epoxy laminates. In fact, the results show that the
glass/epoxy laminates emitted fewer aldehydes and ketones and VOCs than the gtass/lignm
laminates.
Although these printed circuit board laminate types appear to be good substitutes for
paper/phenol laminates, they are not predominantly used in PC monitors. This is because
glass/epoxy laminates are designed for high-speed applications and data processing
whereas PC monitors do not perform the same operating functions nor experience the
same operating conditions as central processing units of PCs.
Paper/reformulated phenolic resin laminates emit more aldehydes and ketones but fewer
volatile compounds than the paper/phenol laminate. In this screening evaluation, the
objective was to determine if emissions from the paper/reformulated phenolic resin-based
laminate would be less than those from the paper/phenol laminate. Results in Section 3.0
show that this is true for reported aldehydes and ketones but not true for the VOCs.
Therefore, reformulating the base resin appears to have both a positive and a negative
impact on indoor air emissions of volatile compounds.
Degree of decay of emissions over time. Plotting emission factors over time gave a good
indication as to how quickly or how slowly volatile emissions would decay over time for
each of the laminate types tested. The results in Section 3.0 show that a majority of the
compounds identified and reported did in fact decay to low levels {-50 Mgm-m ) by 336 h
of exposure to 65°C. However, the data clearly show that the identified volatile
compounds would likely have continued to emit from the laminates beyond the 336 h of
the elevated temperature exposure test.
30
-------
Section 5.0
Data Quality
Quality assurance (QA) activities were an integral part of this screening evaluation. QA
activities were carried out following the guidelines and procedures detailed in the QA Project
Plan (QAPP) and included
Development of data quality indicator goals for study data
Preparation of a QAPP
Monitoring quality control procedures and results
Systems audits of major study components.
5.1 Quality Assurance Project Plan
A revised QAPP, An Evaluation of Applied Pollution Prevention Technologies that Reduce
Indoor Air Emissions from Printed Circuit Board Laminates, was prepared by the RTI project team
in November 1996 and submitted to EPA for approval prior to the start of the proposed testing.
5.2 Data Quality Indicator Coals
Emissions data collected from printed circuit board laminates made from paper/phenol,
glass/epoxy, and glass/lignin-containing epoxy were required to have specific goals for accuracy,
precision, and completeness. The data quality indicator goals presented in the QAPP are
summarized in Table 5-1. However, Table 5-1 has been modified to reflect the incorporation of
phenol/cresols identification and analysis in the VOC analysis.
Data quality goals for precision an.d accuracy for each analysis were based on (1) the
analytical methods used, (2) previous experience with these analyte/matrix samples, and (3) the
objectives of this project. Each indicator goal in Table 5-1 - accuracy, precision, completeness,
comparability, and representativeness - is discussed below.
5.2.7 Precision
The precision measurement is expressed as the percent relative standard deviation (% RSD)
between replicate samples. Percent RSD was calculated as:
% RSD = -I x 100 (5_D
where
S - standard deviation
Y - mean replicate of samples.
31
-------
Table 5-1. Summary of Data Quality Indicator Goals
Precision Accuracy Completeness
Parameter (% RSD) (% recovery) (%)
Air concentrations in steel cans:
VOCs (including phenol and cresols} <;25 *75
Aldehydes/ketones <.2S 2 75
Oven temperature (thermocouple) ±1.0°C NIST 100
traceable
RH of supply air (capacitive detector) *50 NIST 100
traceable
Emission factors
ng/h-m2 <;30 *70 *95
NIST - National Institute of Standards and Technology.
RH - Relative humidity.
RSD - Relative standard deviation.
VOCs - Volatile organic compounds.
The % RSDs are a function of not only the matrix but also the analyte concentration in the
matrix and reproducibility of emission test conditions. Since the concentration of specific target
compounds in samples of a particular matrix may vary, the % RSDs may vary accordingly with
higher %RSD for concentrations near the method quantitation limit. Precision of the analytical
methods was evaluated by collecting and analyzing duplicate air samples. Precision of the
emissions tests was evaluated by conducting duplicate emissions tests on the same sample.
Duplicate air samples were analyzed for each type of laminate evaluated.
The overall precision for five valid pairs of duplicate VOC (and phenol) air concentration
samples was 11.4 percent RSD. Two of the 20 measurements exceeded the data quality indicator
goal of *25 percent RSD.
There were no duplicate samples collected or analyzed for aldehydes due to airflow
constraints through each can. There was insufficient airflow to collect duplicate DNPH samples in
addition to the other samples.
5.2.2 Accuracy
Table 5-1 indicates the accuracy objectives for the analytical data. However, as shown in
the table, there were two types of data generated: chamber air concentrations and emission rates.
For VOCs, only sem(quantitative data were generated for chamber air concentrations and the
accuracy of these measurements was not evaluated.
The accuracy parameter in Table 5-1 is expressed as the recovery from fortified (spiked)
samples. Accuracy of the analytical methods was evaluated by analyzing spiked sample cartridges.
32
-------
For the chamber air concentrations of aldehydes and ketones, DNPH cartridges were spiked with
the DNPH derivatives of the target compounds at a known level. For measurements where matrix
spikes are used, the percent recovery was calculated as follows:
% R = 100% x
A-B
(5-2)
where
A - measured concentration in spiked aliquot
B - measured concentration in unspiked aliquot
Cu - actual concentration of spike added.
The data quality indicator goal of *75 percent recovery was met for the analysis of
aldehydes. The mean recovery was 95.5 percent for 10 aldehydes and ketones spiked onto three
control cartridges. There were no spiked controls for VOCs and phenols.
5.2.3 Completeness
Completeness goals for testing, as shown in Table 5-1, indicate the estimated amount of
valid analytical data expected relative to the amount of data proposed. It does not include "not
detected" analytical results. The % completeness was calculated as follows:
M
% C = 100 % x - (5-3)
n
where
V .- number of measurements judged valid
n - total number of measurements planned.
Completeness for aldehyde measurements was 100 percent. For VOC and phenol/cresol
measurements, completeness was 96 percent. Completeness for temperature and RH
measurements was 100 percent. All completeness goals were met.
5.2.4 Comparability
Comparability in this evaluation focused on VOC emission levels from different types of
printed circuit board laminates, namely
Glass/lignin-containing epoxy
Glass/epoxy
Paper/phenol
Paper/reformulated phenolic laminates.
33
-------
Because each type of laminate had essentially the same physical characteristics and was tested
using the same dimensions, a direct comparison of emissions could be made between each type of
laminate. However, the glass/epoxy sample 2 was copper cladded on both sides of the laminate.
It is uncertain as to how this cladding affected the results of the study. In comparison with the
emissions from the glass/epoxy laminate, the results seem to indicate that the additional cladding
did not affect the rate of volatile emissions.
5.2.5 Representativeness
Representativeness refers to how accurately and precisely the data represent a
characteristic of interest. In the context of the proposed experiment, inferences were limited to the
specific laminates used in the experiment. With respect to the hypotheses of interest, the major
issue was whether each type of laminate from each manufacturer represented that "typical"
laminate used in the printed circuit board industry. The probability of representativeness being
compromised was limited by obtaining laminates from major suppliers to the printed circuit board
manufacturing industry.
5.3 Quality Control Program
Method controls and laboratory controls were prepared to evaluate recovery. Duplicate
samples were collected and duplicate tests run for evaluation of precision. In addition, chamber
blanks and laboratory blanks were analyzed to monitor background and accidental contamination.
Calibration curves were prepared prior to analysis of samples, and check standards were analyzed
at regular intervals to ensure that the calibration remained valid. All data were generated when the
analytical systems were operating within the control criteria.
5.4 RTI Internal Technical System Audit (TSA) Results
The RTI/ACS QA Officer conducted unit-wide inspections and audits during the course of
this program. Laboratory (systems) inspections are conducted every 6 months and laboratory
notebooks and instrument log notebooks are inspected every 6 months. Training files are inspected
yearly and standard operating procedures and good laboratory practices are reviewed yearly.
5.5 Data Reduction, Validation, and Reporting
5.5.1 Chemical Data Reduction -
Data flow for the temperature, humidity, and air exchange rate measurements, VOCs, and
aldehydes was monitored. Quantitative measurement results were reported as steel can air
concentration of target analytes. All reported values contain two significant figures.
Primary results from detailed emissions testing were VOC (including phenol/cresols) and
aldehyde/ketone concentrations in sample air. Emissions factors were calculated for each target
VOC and aldehyde/ketone identified.
Levels of detection limits and background concentrations for identified compounds in this
screening evaluation are shown in Table 5-2. Background concentration data are also shown in
34
-------
Table 5-2. Concentration Data and Levels of Detection
Compound
Aldehydes/Ketones "
Formaldehyde
Acetaldehyde
2-Butanone
Benzaldehyde
VOCs: c
Acetone
Toluene
Hexanol
Ethylbenzene
m,p-Xylene
o-Xylene
Salicylaldehyde
Phenol
o-Cresol
m,p-Cresol
2,4-Dimethylphenol
m-Anisaldehyde
2-Hydroxybenzyl alcohol
4-Hydroxybenzaldehyde
C8H1002d
QH1202 "
Methyl ether
1 -Methoxy-2-propanol
Cartridge blank
concentration, Mg/m1
11
no
6.5
15
Total system background
concentration, Mg/m>
7.2
114.8
7.4
8.0
P/RP sample A Chamber
13.01
0.11
0.39
0.02
0.11
0.00
11.15
26.43
2.20
16.87
2.53
6.22
129.23
6.87
0.00
0.00
NC
NC
Limit of detection,
Mg/m"
13
12
15
12
5.4
7.67
7.67
9.83
11.8
9.17
9.77
27.9
5.73
13
9.1
9.43
51.7
9.1
8.33
8.33
57.3
21.5
Level of detection based on lowest calibration standard
b Average concentration in chambers #1 and #8.
e Background concentration in one chamber.
d Isomer unknown.
NC - Negligible concentration.
P/RP - Paper/reformulated phenolic.
35
-------
Appendixes C and D. None of the figures and data presented in this report have been corrected
for background concentrations.
5.5.2 Chemical Data Validation
Data validation began after air samples were analyzed and during the period of data reduction
and review. Data validation was performed by the sample analysis supervisor. The QA Officer
verified that all validation steps were complete. See Appendix E for a copy of the Internal Audit
report that details what the QA Officer did in auditing this screening evaluation.
36
-------
Section 6.0
References
Brooks, B. O., etal. 1993. Chemical Emissions from Electronic Products. International
Symposium on Electronics and the Environment, Arlington, VA, May 10-12.
Coombs, Jr., C. F. 1988. Printed Circuits Handbook, McGraw-Hill, 3rd edition, New York, NY.
Flatt, M. 1992. Printed Circuit Board Basics. Miller Freeman, Inc. San Francisco, CA.
Habicht, F. H. II. 1992. Memorandum from the Office of the Administrator. Subject: EPA
Definition of Pollution Prevention. U.S. Environmental Protection Agency, Washington,
DC, May 28.
U.S. Bureau of the Census. 1994. Current Industrial Report - Semiconductors, Printed Circuit
Boards, and Related Equipment. MA36Q(94)-1.
U.S. EPA (Environmental Protection Agency). 1990. Guide to Pollution Prevention, The Printed
Circuit Board Manufacturing Industry. EPA/625/7-90/007 (NTIS PB 90-256413),
Cincinnati, OH.
U.S. EPA. 1995a. EPA's Design for the Environment Cooperative Case Studies with the Printed
Wiring Board Manufacturing Industry: Case Study 1, Pollution Prevention Work Practices.
EPA/744-F-95-004. Cincinnati, OH.
U.S. EPA. 1995b. EPA's Design for the Environment Cooperative Case Studies with the Printed
Wiring Board Manufacturing Industry: Case Study 2, On-site Etchant Regeneration.
EPA/744-F-95-005. Cincinnati, OH.
U.S. EPA. 1995c. f PA's Design for the £nv/ronment Cooperative Case Studies \v/th the Printed
Wiring Board Manufacturing Industry: Case Study, 3 Opportunities for Add Recovery and
Management. EPA/744-F-9 5-009. Cincinnati, OH.
U.S. EPA. 1996. EPA's Design for the Environment Cooperative Case Studies with the Printed
Wiring Board Manufacturing Industry: Case Study 4, Plasma Desmear. EPA/744-F-96-003.
Cincinnati, OH.
37
-------
APPENDIX A
Internal Operating Temperature of a Color Monitor
Explanation: The internal temperature of a Compaq 151FS color monitor was measured to
determine the actual temperatures during various modes of operation (full power,
suspended, screen saver, and energy saver). This information was useful in
determining a testing temperature for the printed circuit board emissions screening
evaluation.
Experimental: The outer case of a Compaq 151FS color monitor (approximately 6 months old)
was removed and a wire thermocouple was taped to a wire between the tube and
the heat sink and approximately 3.81 cm above the printed circuit board. The
cover was replaced and the monitor powered up. The thermocouple was fed
through the ventilation grid of the monitor cover and attached to a recording
readout instrument.
Temperatures were monitored at four stages of operation as shown in Table A-1,
Table A-1. Power Consumption for Various Operating Modes of a Color Monitor
Operating mode Power consumption, Watts
Full power - 70
Screen saver U
Suspended power < 30
Energy saver -^8
U - Unknown.
Ambient temperature of the room was 24.6 to 25.3°C. The monitor was powered up and allowed
to stabilize to consistent temperature readings. Temperatures were recorded at 1-min intervals.
The results are shown in Table A-2.
Table A-2. Resulting Temperature Ranges and Averages for Various Operating Modes of a Color
Monitor
Operating mode
Full power
Screen saver
Suspended power
Energy saver
Temperature range, °C
67.4 - 72.2
63.0-70.3
37.8-33.1
28.1 -28.2
Average, "C ± SD '
69.9 ±1.2
67.5±2.2
32.4 ±0.4
28.1 ±0.05
Average of last 15 time intervals.
A-1
-------
APPENDIX B
Industry Background1
Printed circuit boards are found in virtually all manufactured electronic equipment. They
are custom designed and often the most expensive single component of that equipment. Some
printed circuit boards may cost more than a thousand dollars a piece and consist of additional
expensive components in the form of circuitry, or components of an electrical circuit.
Printed circuit boards were not used until after World War II. Dr. Paul Eisler, an Austrian
scientist working in England at the time, is usually credited with making the first printed circuit
board. His concept of the printed circuit board was to replace radio tube wiring with something
less bulky.
In the 1950s and 1960s, printed circuit boards with circuitry on one side, called single-
sided, were the dominant variety. Still in use today, single-sided boards are the simplest variety of
printed circuit boards. Logic follows that they are manufactured in high volume, most often for
consumer electronics, and are the least expensive to produce.
During the late 1960s and early 1970s, processes were developed for plating copper on the
walls of the drilled holes in circuit boards, allowing top and bottom circuitry to be electrically
interconnected. This gave way to the concept and production of double-sided boards, which
quickly became the industry standard. Used in more sophisticated consumer products and
extensively in computer peripheral equipment, double-sided boards are more expensive than
single-sided boards.
As the densities and complexities of electronic components increased, the multilayer
board - a process of sandwiching several circuitry layers together - was developed. By the mid-
1980s, multilayers accounted for the majority of U.S. output. Today's computers, aerospace
equipment, and instrumentation and telecommunications gear all contain multilayer boards.
Multilayers are the most expensive type of printed circuit board to produce.
B.1 Printed Circuit Board Laminates and Their End-Use
Printed circuit board laminates, in general, are dielectric, or nonconducting, substrates
with metallic circuitry photochemically formed upon the substrate. There are three major
classifications of printed circuit boards: single-sided, double-sided, and multilayered.
Single-sided boards are dielectric substrates with circuitry on one side of the board only.
There may or may not be holes for circuit components. Double-sided boards are dielectric
substrates with circuitry on both sides of the board. To electronically connect both sides, holes are
drilled through the dielectric substrate and a thin layer of copper is plated, usually by
electrodeposition, through the holes. The end product of applying copper in this manner is
usually referred to as a copper-clad laminate. Multilayered boards are constructed by stacking and
1 This summary is adapted from Printed Circuit Board Basics by Michael Flatt, 1992, San Francisco, CA: Miller
Freeman, Inc.
B-1
-------
bonding together two or more dielectric materials already having circuitry formed on them.
Electrical connections are established from one side to the other and to the inner layer circuitry by
drilled holes with a thin layer of copper plated through them.
Table B-1 summarizes the most common copper-clad laminates used in the printed circuit
board industry.
Table B-1. Most Common Copper-Clad Printed Circuit Board Laminates'
NEMA
grade
XXXPC
FR-2
FR-3
CEM-1
CEM-3
FR-6
GOO
FR-4
c-11
FR-5
Cl
(military)
Laminate description
Commercial electric and electronic applications.
Stable electric properties in high humidity and
temperature conditions.
Fire resistant.
Fire resistant. High insulation resistance.
Composite with electrical properties, moisture
resistance and thermal stress equal to G-10.
Composite with electrical properties, moisture
resistance and thermal stress equal to G-10.
Fire resistant. Designed for low-capacitance or
high-impact applications.
Superior mechanical strength, high electric
properties over wide humidity and temperature
ranges. Excellent chemical resistance, low
warpage, and predictable machinability.
Dimensional ly stable.
Fire-resistant version of G-10.
Similar to G-10. Retention of electrical and
mechanical properties in heated environments.
Fire-resistant version of G-1 1 .
Superior performance and excellent electrical
stability at very high temperatures and humidity.
Excellent resistance to hostile processing
conditions, including heat and chemicals. Very
high mechanical strength and very iow
dimensional change.
Most stable performance at highest temperatures.
Unaffected by processing chemicals.
Resin
system
phenolic
phenolic
epoxy
epoxy
epoxy
polyester
epoxy
epoxy
epoxy
epoxy
triazine
polyimide
Base
paper
paper
paper
paper-
fiberglass
fiberglass
matte
fiberglass
matte
fiberglass
fiberglass
fiberglass
fiberglass
fiberglass
fiberglass
Color
opaque
brown
opaque
brown
opaque
cream
opaque tan
translucent
opaque
white '
translucent
translucent
translucent
translucent
translucent
(yellow-
green)
translucent
(dark
amber)
Operating
T,°C
125
105
105
130
130
105
130
130
170
170
220
260
' Coombs, Jr., C. F. 1988. Printed Circuits Handbook, McGraw-Hill, 3rd edition, New York,
NEMA - National Electrical Manufacturers Association.
NY.
B-2
-------
The 12 laminates, described in Table B-1, are discussed below.
NEMA grade XXXPC (Phenolic/Paper). Phenolic/paper substrates are one of the cheapest
laminates manufactured in the printed circuit board industry. However, the use of phenolic resin
over paper is limited in the United States because of greater demands for enhanced chemical
resistance and overall reliability in electronic components.
NEMA grade FR-2 (Phenolic/Paper). The FR-2 laminates are composed of multiple plies of
paper that have been impregnated with a flame-retardant phenolic resin. Major advantages
include low cost and good electrical and hole punching qualities. They are typically used in
applications where tight dimensional stability is not required, such as in radios, calculators, toys,
and television games.
NEMA grade FR-3 (Epoxy/Paper). These laminates are made of multiple plies of paper
impregnated with an epoxy-resin binder. They have higher electrical and physical properties than
the FR-2 but lower than epoxy laminates that have woven glass cloth as a reinforcement. FR-3
laminates are used to manufacture printed circuits used in consumer products, computers,
television sets, and communication equipment.
NEMA grade CEM-1 (Epoxy/Fiberglass/Paper Composites). These types of substrates are
constructed using a paper core irnpregnated with epoxy resin. The two surfaces of this paper core
are covered with a woven glass cloth impregnated with the same resin. This construction allows
the material to have punching properties similar to those of FR-2 and FR-3, with electrical and
physical properties approaching those of FR-4. CEM-1 boards are used in smoke detectors,
television sets, calculators, and automobiles as well as in industrial electronics.
NEMA grade CEM-3 (Epoxy/Fiberglass). A majority of printed circuits used in home
computers, automobiles, and home entertainment products are manufactured using this type of
laminate. CEM-3 is a composite of an epoxy-resin-impregnated non woven fiberglass core with
epoxy-resin-impregnated woven glass cloth surface sheets. It is higher in cost than CEM-1, but it is
more suitable for plated-through holes.
NEMA grade FR-6 (Polyester/Fiberglass). This is a fairly cheap laminate with good
moisture absorption and electrical properties. However, it is inferior to epoxy substrates,
especially in mechanical strength and heat resistance. It is finding wider acceptance for circuits
used in extremely high volume such as electronic game cartridges and computer disk drives.
NEMA grade FR-4/G-10 (Epoxy/Fiberglass). These laminates are constructed on multiple
plies of epoxy-resin-impregnated woven glass cloth. It is the most widely used material in the
printed circuit board industry because its properties satisfy the electrical and mechanical needs of
most applications. Its excellent electrical, physical, and thermal properties make it an excellent
material for high-technology applications. It is used in aerospace, communications, computers
and peripherals, industrial controls, and automotive applications.
NEMA grade FR-5/C-11 (Epoxy/Fiberglass). FR-5 types are laminated using multiple plies
of woven glass cloth impregnated with mostly polyfunctional epoxy resin. They are used where
higher heat resistance is needed than is attainable with FR-4 but not where the very high thermal
properties of Cl type materials are needed.
B-3
-------
NEMA grade Triazine/Fiberglass. Laminates made of triazine/fiberglass composites
demonstrate exceptional performance and electrical stability at very high temperatures and
humidity.
NEMA grade Cl (Polyimide/Fiberglass). Cl type materials are composed of multiple plies
of woven glass cloth impregnated with a polyimide resin. These laminates have excellent
electrical and mechanical properties when operated at elevated temperatures. Polyimide is very
hard and makes machining of the laminate difficult. A major application is for burn-in boards or
boards on which electrical components are tested for reliability. It is expensive and is sometimes
substituted with epoxy/fiberglass laminates. Polyimide is also used for multilayer construction to
enhance mechanical properties of the board. It is commonly used in high-reliability applications,
usually for the military.
B.2 Printed Circuit Board Laminate Manufacturing Process
In a printed circuit board laminate manufacturing process, the base raw material - paper,
fiberglass matte, woven glass cloth - is impregnated or coated with either a phenolic, epoxy,
polyester, polyimide, or triazine resin. This resin is then partially polymerized suitable for
subsequent storage and further processing. A machine called a "treater" or "coater" is used for
treating the material. First, the material passes through a dip pan of resin, where it is impregnated,
and then through a set of metering rollers (squeeze rollers) and a drying oven. The oven is air-
circulating or infrared and can be up to 37 m long. Most VOCs, such as solvents in the resin, are
driven off in the oven, and the resin is partially polymerized to a condition known as a "B-stage."
In the B-stage, the resin can be subsequently heated and caused to flow, thereby allowing final
curing to a desired shape. This semicured state is also known as "prepreg." The prepreg is dry and
nontacky.
Rigid process control is applied during treatment so that the ratio of resin to base material,
the final thickness of the prepreg, and the degree of resin polymerization can be monitored. Beta-
ray gauges may compare the raw material with the final semicured product and automatically
adjust the metering rollers above the resin dip pan so that the proper ratio of resin to base material
is maintained. The degree of polymerization of the resin is controlled by the treater air
temperature, air velocity, and speed at which the material passes through the treater.
The prepreg material is usually stored in an area where the temperature is controlled below
21 °C and below 35 percent humidity until the time of pressing operation.
Besides the base material and resin, the other principal component of printed circuit board
laminates is a conductive material for cladding the laminate. The conductor is most commonly
copper, although aluminum, chrome, nickel, and other metals have been used. Almost all metal
cladding is electrodeposited using an electroless metal. This electroless metal is a thin layer of
conductive material deposited on the laminate surface from an autocatalytic plating solution
without the application of an electrical current. Other techniques of affixing metal to base
laminates include adhesives, pressure/heat bonding, and sometimes screws.
Subsequent electronic component processing is usually performed at a separate printed
circuit board assembly plant.
B-4
-------
B.3 Today's Economics of Printed Circuit Boards in PC Monitors / VDTs
In terms of economics, the United States controls approximately 30 percent of the global
electronics interconnection market. According to the Institute for Interconnecting and Packaging
Electronic Circuit's Technology Marketing Research Council, the computer and computer
electronics industry shipped a total of $1.7 billion, or 35 percent of the total of all printed circuit
board sales for independent producers of printed circuit boards in 1995. These numbers are
exclusive to printed circuit boards made from fiberglass (glass)/epoxy (base/resin) laminates.
Printed circuit boards made from paper/phenol laminates are manufactured in and imported to the
United States primarily from Taiwan and Mainland China, Korea, and Japan. Table B-2 shows the
distribution of these paper base laminates imported to the United States as well as import
information for color computer monitors. Imports for color computer monitors are included in
Table B-2 because almost all color monitors manufactured outside of the United States contain
paper/phenol printed circuit board laminates.
Table B-2. General Import Information for Paper-based Printed Circuit Board Laminates and
Color Computer Monitors'
General Imports for Consumption - Paper Laminates
1993 1994
Country
Mainland China
Korea
Taiwan
Japan
All others
Total
Quantity
497,948
439,679
663,626
421,363
2,022,616
Value
$403,418
-
$249,858
$4,798,509
$411,601
$5,863,386
Quantity
410,620
2,470
945,664
397,434
301,672
2,057,860
Value
$337,674
$14,820
$463,507
$2,845,387
$514,768
$4,176,156
Total Imports for Color
Monitors from all countries 15,313,690 $4,112,549,382 16,418,262 $4,470,581,738
' U.S. Bureau of the Census, Current Industrial Report - Semiconductors, Printed Circuit Boards, and Related
Equipment- 1994, MA36Q(94)-1.
B-5
-------
APPENDIX C
Concentration and Emission Factor Data:
Aldehydes and Ketones
Table C-1 illustrates an example of how emission factors were calculated for each
aldehyde/ketone identified for each air sample of each type of laminate tested. The loading factor
used in calculating emission factors in each table presented in this Appendix are based on printed
circuit board laminate dimensions of 0.15 m wide by 0.25 m long with both sides of the laminate
exposed.
Table C-1. Example (formaldehyde) of Emission Factor Calculation for Each Aldehyde/Ketone
Identified for Each Type of Laminate Tested
Sample interval,
h
0-2
4-6
11-13
23-25
47-49
95-97
143-145
215-217
335-337
Time point,
h
1
5
12
24
48
96
144
216
336
Meas. cone.,
uR/mJ
0.0s
714.5
871.1
562.B
421.2
296.5
203.9
172.7
153.2
116.2
Derivative,
dOdt
714.5
39.1
-44.0
-11.8
-5.2
-1.9
-0.7
-0.3
-0.3
Avg derivative,
dC/dt
376.8
-2.4
-27.9
-8.5
-3.6
-1.3
-0,5
-0.3
-0.3
Emission factor,
uE/h-m2
117
94
58
45
32
22
19
17
13
1 Initial concentration at time zero. Also, a value is required for the derivative calculation.
The following equations were used to generate the data in this table:
dC/dtt + dCldl.
Avg. (dC/dt) = ±- (C-2)
Avg. (dPcfl,, * Ct) N
where:
C-1
-------
dC/dt - derivative of concentration (j/g/m3) with respect to time (h)
C,, - mean sample air concentration fc/g/m3) at time interval t,
CIO - mean sample air concentration Gwg/m3} at time interval t,,
(dC/dt),, - derivative of concentration fc/g/m3) with respect to time t,
(dC/dt)u - derivative of concentration fo/g/m3) with respect to time 12
EF - emission factor (j/g/h-m2)
N - air exchange rate in each steel can (1.02 h'1)
L - product loading factor (9.43 m2/m3).
Similar data tables were generated for all laminate sample types for each aldehyde/ketone
identified but are not shown here for paper reduction purposes. Emission factors calculated in
those tables were used to generate Tables C-2 through G9. These tables were used to generate
Figures 3-3 through 3-5 in Section 3.0, Results. Table C-10 was used to generate the
aldehyde/ketone summation figure, Figure 3-1, in Section 3.0. Table C-11 represents the raw
concentration data at each time interval for all laminate types analyzed in the screening test.
C-2
-------
Table C-2. Emission Factors of Selected Aldehydes and Ketones for Paper/Phenol Laminate
Time point, Formaldehyde, Acetaldehyde,
h ug/h-m2
0 116
5 93
12 57
24 44
48 31
96 21
144 18
216 16
336 12
Table C-3. Emission Factors of Selected
Phenolic Sample A Laminate
Ug/h-m2
10
14
9
6
3
4
3
2
5
Aldehydes and
Time point, Formaldehyde, Acetaldehyde,
h - ug/h-m2
0 126
5 76
12 51
24 38
48 28
96 20
144 17
216 14
336 11
Table C-4. Emission Factors of Selected
Phenolic Sample B. Laminate
Ug/h-m2
7.9
4.3
7.4
6.1
5.3
3.9
3.0
2.7
1.7
Aldehydes and
Time point, Formaldehyde, Acetaldehyde,
h ug/h-mj
o in
5 60
12 39
24 30
48 21
96 16
144 13
216 11
336 9
Ug/h-m2
7.1
5.8
2.7
4.4
3.4
1.2
0.9
1.8
2.1
2-Butanone,
Ug/h-m2
409
360
232
174
121
81
69
57
44
Benzaldehyde,
Ug/h-m2
115
102
76
50
35
23
18
15
10
Ketones for Paper/Reformulated
2-Butanone,
Ug/h-m2
534
406
281
218
152
105
89
68
58
Benzaldehyde,
Ug/h-m2
225
213
150
112
77
49
38
29
24
Ketones for Paper/Reformulated
2-Butanone,
Ug/h-m2
604
342
220
202
119
86
72
62
51
Benzaldehyde,
Ug/h-m2
295
183
117
88
57
38
32
24
18
C-3
-------
Table C-5. Emission Factors of Selected Aldehydes and Ketones for Class/Lignin Sample 1A
Laminate
Time point, Formaldehyde,
h ug/h-m2
0 9
5 4
12 2
24 2
48 3
96 1
144 1
216 1
336 1
- - Value below the limit of detection.
Acetaldehyde,
Ug/h-ffl2
12.0
5.7
2.6
3.0
3.7
0.4
2.3
4.5
1.9
2-Butanone,
UR/h-m2
9
3
3
1
2
1
2
2
1
Benzaldehyde,
Ug/h-m2
2
0
1
0
-
-
-
0
0
Table C-6. Emission Factors of Selected Aldehydes and Ketones for Class/Lignin Sample IB
Laminate
Time point, Formaldehyde,
h ug/h-m2
0 11
5 4
12 3
24 8
48 2
96 2
>44 2
216 2
336 1
Acetaldehyde,
Ug/h-m2
16.7
8.5
4.5
5.3
3.8
4.2
3.9
4.5
2.0
2-Butanone,
Ug/h-m1
12
3
4
3
2
2
1
3
2
Benzaldehyde,
Ug/h-m2
2
i
3
a
1
1
1
a
0
- - Value below the limit of detection.
' Data not available.
C-4
-------
Table C-7. Emission Factors of Selected Aldehydes and Ketones for Class/Lignin Sample 2
Laminate
Time point,
h
0
5
12
24
48
96
144
216
336
Formaldehyde,
Hg/h-m2
11
4
3
2
2
2
1
1
1
Acetaldehyde,
Hg/h-m2
14.7
9.6
5.1
5.0
2.5
2.4
1.6
2.5
3.3
2-Butanone,
Hg/h-m2
12
4
3
2
2
2
2
2
-
Benzaldehyde,
2
1
0
1
1
-
-
a
1
- - Value below the limit of detection.
" Data not available.
Table C-8. Emission
Factors of Selected Aldehydes and
Ketones for Class/Epoxy
Laminate
Time point,
h
0
5
12
24
48
96
144
216
336
Formaldehyde,
Hg/h-m2
2
1
0
0
0
1
0
0
-
Acetaldehyde,
Hg/h-m2
1.3
-
-
-
2.6
0.5
-
-
0.8
2-Butanone,
Hg/h-m2
1
1
1
a
0
2
1
1
1
Benzaldehyde,
Hg/h-m2
1
1
0
0
1
0
0
-
1
- - Value below the limit of detection.
C-5
-------
Table G9. Emission Factors of Selected Aldehydes and Ketones for Class/Epoxy
Sample 2 Laminate
Time point,
h
0
5
12
24
48
96
144
216
336
Formaldehyde,
Hg/h-m2
4
2
1
1
1
1
1
1
0
Acetaldehyde,
|ig/h-m2
12.3
9.5
5.7
3.5
2.6
2.6
1.0
2.4
-
2-Butanone,
UK/h-m2
13
7
4
3 .
0
3
1
2
1
Benzaldehyde,
Hg/h-m2
3
a
1
2
0
2
-
a
0
- - Value below the limit of detection.
" Data not available.
C-6
-------
Table C-10. Sum of Measured Concentrations at t=0 h and t = 336 h for All Reported
Aldehydes and Ketones for Each Laminate Tested
Sum of aldehyde/ketone Sum of aldehyde/ketone
concentrations in ng/m3, concentrations in ng/m3,
Laminate type t = 0h t«336h
Paper/phenol 3930 653
Paper/reformulated phenolic Sample A 5456 869
Paper/reformulated phenolic Sample B 6372 734
Class/lignin Sample 1A 268 47
Class/lignin Sample IB 203 53
Class/lignin Sample 2 250 42
Glass/epoxy 33 25
Class/eooxv Samole 2 197 17
C-7
-------
Table C-11. Raw Concentration, mg/rn3, Data for Identified Aldehydes and Ketones
n
00
Sample/
HIM poini ronneroenyoa AC
Paper/phenol
1
5
12
24
4B
96
144
216
336
714.49
871.064
562.8
421. t57
296.495
203.948
172.675
153.161
116.174
tttalctohyde
171.778
239.438
201.923
169.615
145.401
153.206
139.82
136.895
157.007
Acetone Proplonaldehyde
282.929
SOI .395
336.302
366.498
279.082
300.26
262.28
276.945
460.944
2.273
19.805
19.289
12.52
2.089
15.221
9.285
2-Butanone Butraldehyde
2476.245
3311.824
2254.875 4.646
1653.351
1143.971
764.615
645.755
534.704 1.298
417.221
Benialdehyde
704.909
933.28
733.323
482.59
335.858
225.507
173.854
146.112
99.999
Vatoraldehyde
32.058
55.14
24.29
32.532
48.408
27.071
5.29
73.511
37.391
Paper/reformulated phenoNc Sample A
1
5
12
24
48
98
144
216
336
Papor/retornuMiea
1
5
12
24
48
96
144
216
336 .
795.969
731.673
504.776
367.589
287.5 fl
195.287
163.057
135.660
105.695
ilmiifife^ Tnmnln fl
708.99
592.92
382.541
287.17
207.811
157
123.615
109.878
86.275
164.411
154.26
181.774
172.069
163.658
151.091
142.62
140.187
130.69
157.69
168.622
141.415
155.056
146.271
126
122.914
131.014
133.934
396.814
298.909
267.616
284.581
377.73
310.613
265.759
314.31
218.416
372.994
379.667
177.214
215.217
278.997
336.873
260.383
309.87
355.95
14.923
2.256
11.284
14.566
0.573
4.O95
8.545
21.929
4.026
4,08
9.132
12.598
3276.691 6.342
3772.158
2708.741
2059.792
1433.862 1.665
988.332 1.955
635.086
640.557
542.936
3797.045
3310635 2.679
2136.607
1903.982
1126.495
805.872
674.257
583.058
479.223
1356.246
1940.373
1440.558
1068.309
729.177
467.279
362.785
279.178
226.96
1847.63
1750.188
1148.356
840.482
542.481
357.816
302.186
233.067
171.565
34.677
44.299
47.55
27.792
23.711
10.618
25.119
2.41
71.742
48.676
19.109
32.347
0.09
7.518
m-TohuMehyde Hexinal
38.244
39.62
17.176
2.182
9.626
6.925
19.362
9.545
56.64
54.74
32.392
13.774 13.939
20.588
1.023
13.458
18.792
17.32
57.265
50.771
33.499
7.421
7.69
12.83
(Continued)
-------
Table C-11. Raw Concentration, mg/m3, Data for Identified Aldehydes and Ketones (con.)
9
Sample/
lima point Formaldehyde
Glass/llgntn Sample 1A
1
5
12
24
48
96
144
216
336
Glass/flgnln Sample
1
5
12
24
48
96
144
216
336
63.305
43.735
25.353
23.669
34.359
15.366
20.573
18.736
15.077
IB
78.327
47.669
37.872
84.684
26.131
26.349
24.02
25.158
20.236
Glass/ltgnin Sample 2
1 78.88
5
12
24
48
96
144
216
336
49.851
34.386
26.122
27.25
23.129
18.607
19.659
20.502
Acataldehyde
190.953
172.166
140.985
142.077
148.874
118.995
135.729
156.585
132.594
220.232
196.744
159.398
163.65
150.633
153.455
151.241
156.535
133.32
208.101
206.634
164.642
161.937
138.79
137.331
129.661
137.742
144.975
Acetone Proplonaldehyde
352.557
192.797
117.349
240.371
240
297.945
250.784
584.007
207.212
200.982
441.094
104.864
223.045
290.556
253.104
153.891
334.539
240.185
337.132
258.383
176.185
267.767
118.945
117.161
131.782
149.051
288.966
19.224
4.587
17.587
2.738
6.226
16.004
1.917
25.846
14.76
7.818
15.386
6.278
18.923
32.569
18.352
11.012
2.343
7.308
13.056
2-Butanone Butraldehyde
68.176
39.048
38.39
19.616
26.694
13.392
25.698
24.742
17.735 3.912
89.071
38.636
41.424
35.269
24.871
26.892
16.68
38.94
27.03
84.313
52.642
38.016
27.22
25.09
27.734
24.922
26.1
0.35 4.639
Bemaldehyde
18.211
8.524
18.33
9.134
1.177
3.635
5.212
9.247
11.237
18.009
31.098
13.986
17.756
13.598
9.755
17.746
18.884
10.419
13.943
12.767
6.983
3.3
13.564
Vatoraldehyde
12.473
23.011
2.793
6.456
4.595
12.575
16.494
18.616
12.934
8.902
1.767
30.694
15.122
17.481
45.395
18.636
0.733
67.128
m-Tolualdehyde Hexanal
12.704
9.646
17.93
5.542
13.492 13.902
11.216
62.067 15.693
44.382 18.692
17.03 13.615
11.79 9.156
13.684
5.042
22.207
9.72 17.032
8.955
12.245 10.789
38.293
42.823
2.287
6.99
11.716 9.559
8.409
(Continued)
-------
Table Oil. Raw Concentration, mg/m3, Data for Identified Aldehydes and Ketones (con.)
0
o
Simpta/
Dm* point FonrnkMiydt AntahMiyd*
1
5
12
24
48
96
144
216
336
to1
19.231
12.788
11.473
10.47S
8.624
16.565
8.701
9.902
6.683
123.93
113.859
111.34
100.738
138.543
119.632
105.256
111.18
122.399
Action* Proplonahtohvcto
318.219
186.006
152.926
126.6
270.787
144.612
132.098
188.879
187.599
6.707
15.266
1.669
7.979
2-ButtnotM Butraktohyd*
10.827
14.026
17.785
9.404
26.681
16.389
17.819
17.25 2.7
Bmuldehycta
16.744
16.567
12.237
10.359
14.073
11.079
8.637
6.023
16.558
Vatoralctehydt
5.513
3.593
4.774
5.624
15.863
24.328
Gtasi/Epaxy Swnpto 2
1
5
12
24
48
98
144
216
336
CART BLANKS
3
CONTROLS
3
CHMBBKOS
CHMB 1 BKO
CHMB8BKQ
30.372
25.207
17.906
13.381
20.671
19.953
16.773
13.818
9.277
35
ss
99
36.011
760
775
789.082
10.048
4.378
189.936
203.075
170.316
148.151
138.773
139.053
124.488
136.777
114.416
465
410
436.706
1145
1170
1181.835
122.544
107.064
162.082
303.493
226.746
191.349
139.089
430.259
288.746
367.564
141.549
1005
490
357.102
1265
1120
1446.234
338.741
113.001
47.211
39.6
21.941
12.686
4.268
3.994
5.393
27.413
85
755
760
844.246
15.404
8.189
92.707
71.887
47.81
32.36
7.675
32.376
14.324
28.214
18.471
30 15
25
22.536
665 720
690 765
699.551 800.467
7.368
21.353
A
re.848
23.88
9.444
28.09
7.554
12.429
105
35
36.579
775
745
760.863
11.385
4.627
15.503
27.671
5
248.814
590
615
602.119
99.619
m-Toliiildehyde Hnrnnl
31.891
37.594 9.72
10.607
31.529
1 1.635
7.604
15.295
19.691
10.728
13.552
11.985 12.024
1.846 4.943
5.102
7.679
5.561
15.608
45
90 30
60.794
685 710
715 705
701.118
34.15 8.82
4.038
{Continued)
-------
Table C-11. Raw Concentration, mg/m3, Data for Identified Aldehydes and Ketones (con.)
Sampta/
flint) point
0.5X RF
060497-2
060497-8
060497-14
060497-22
060497-29
060497-36
060497-43
060597-2
060597-8
060597-16
060597-23
060597-32
060597-39
060697-2
060697-8
060697-14
060697-20
Formaldehyde
(noW
0.219
0.205
BAD INJECTION
0.212
0.178
0.212
0.214
0.226
BAD INJECTION
0.217
0.218
0.22
BAD INJECTION
0.218
0.22
0.221
0223
Acetaldehyde
0.233
0.211
0.223
0.189
0.224
0.233
0.237
0.228
0.23
0.233
0.224
0.228
0.231
0231
Acetone
0.222
0.242
0.256
0.218
0.258
0.244
0.247
0.257
0.249
0.247
0.253
0.255
0.238
0.28
Proplonaldehyde
0.221
0.215
0.239
0.19
0.232
0.232
0.235
0.234
0.23
0.231
0.241
0.242
0.228
0.242
2-Butanone
0.228
0.213
0.229
0.174
0.222
0.225
0.238
0.223
0.22
0.233
0.222
0.219
0.218
0.217
Bulraldehyde
0.237
0.214
0.245
0.2
0.238
0.25
0.244
0.255
0.249
0.248
0.249
0.248
0.247
0.248
Benzaldehyde
0.205
0.196
0.214
0.181
0.215
0.211
0.214
0.216
0.235
0.234
0.216
0.224
0.218
0.225
Valeraldehyde
0.224
0.196
0.212
0.177
0.229
0.23
0.261
0.222
0.223
0.252
0.227
0.228
0.225
0.217
m-Tolualdehyde
0.208
0.213
0.205
0.174
0.209
0.238
0.265
0.207
0.214
0.218
0.214
0.208
0.215
0.209
Hexanal
0.213
0.211
0.243
0.178
0.218
0.227
0.228
0.208
0.223
0.209
0.224
0.227
0.207
0.216
AcroNn. crotanakfehyde, and methacroteln are not reported due to Instability of compounds during storage.
-------
APPENDIX D
Concentration and Emission Rate Data:
VOCs (Including PhenoI/Cresols)
Table D-l illustrates an example of how emission factors were calculated for each VOC
identified for each air sample of each type of laminate tested. The loading factor used in
calculating emission factors in each table presented in this appendix is based on printed circuit
board faminate dimensions of 0.15 m wide by 0.25 m long with both sides of the laminate
exposed.
Table D-1. Example (for Methyl Ether) of Emission Factor Calculation for Each VOC Identified
for Each Type of Laminate Tested
Sample interval,
h
0-2
4-6
11-13
23-25
47-49
95-97
143-145
143-145
215-217
335-337
Time point,
h'
1
5
12
24
48
96
144
144
216
336
Meas. cone.,
#R/m3
0.0"
1528.0
1144.5
725.6
509.2
305.2
158.1
129.7
125.8
97.5
64.6
Derivative,
dC/dt
1528.0
-95.9
-59.8
-18.0
-8.5
-3.1
-0.6
-0.7
-0.4
-0.3
Avg derivative,
dC/dt
716.1
-77.9
-38.9
-13.3
-5.8
-1.8
-0.6
-0.5
-0.3
-0.3
Emission factor,
wn/h-mj
241.3
115.6
74.4
53.7
32.4
16.9
14.0
13.6
10-5
7.0
* Duplicate air sample taken at 144 h.
* Initial concentration at time zero. Also, a value is required for the derivative ca/culation.
The following equations were used to generate the data in this table:
.
M ~
dCldl. +
Avg. (dC/cft), -- ^-r - - (D-2)
Avg. (dCfdt), « C. N
± * (D-3)
where
dC/dt - derivative of concentration fc/g/m3) with respect to time (h)
D-1
-------
C,, - mean sample air concentration (j/g/m3) at time interval t,
Q, - mean sample air concentration (j/g/m3) at time interval t<,
dC/dt,, - derivative of concentration Ot/g/m3) with respect to time t,
dCd^ - derivative of concentration U/g/m3) with respect to time t2
EF - emission factor ty/g/h-m*)
N - air exchange rate in each steel can (1.02 h"')
L - product loading factor (9.43 mVm3).
Similar data tables were generated for all laminate sample types for each VOC identified
but are not shown here for paper reduction purposes. Emission factors calculated in those tables
were used to generate Tables D-2 through D-9. These tables were used to generate Figures 3-6
through 3-11 in Section 3.0.
D-2
-------
Table D-2. Emission Factors, in itg/m1, of Selected VOCs for Paper/Phenol Laminate
X
I
p
0
5
12
12A
24
48
96
144
216
336
Acetone 1
1
3
2
1
1
1
0
0
0
0
,
399
72
93
61
125
80
24
52
65
X
a
5
2
2
1
1
0
1
1
0
£
a
33
23
21
18
10
7
4
3
2
!
a
25
19
17
14
7
5
4
3
2
f
a
9
7
7
6
2
2
2
1
1
Salicylaldehyde J
a
1138
451
424
427
190
276
227
153
123
f
a
1573
269
344
432
247
382
365
95
109
|
a
5
13
8
17
1
6
6
1
2
1
E"
a
121
43
55
117
10
60
67
6
9
2,4-Oimethylphenol 1
a
2
1
2
5
0
2
2
-
0
m-Anisaldehyde y
a
4
7
5
13
0
5
4
-
0
1
1
o
X
X
rt
'
47
.
102
94
79
44
50
-
63
4-Hydroxybenzaldehyde |
a
39
-
3
3
4
3
7
-
20
C8H10O2 9
a
111
64
62
123
15
67
79
13
21
C9H12O2 |
a
39
106
70
113
8
53
47
7
9
- Value below the limit of detection.
1 Data not available.
-------
Table D-3. Imlwlon Facton, In jig/m*, of Selected VOCs for Paper/Reformulated Phenolic Sample A Laminate
I
i.
o
5
12
24
48
96
144
216
336
a
3
1
1
1
0
0
0
0
a
13
0
20
0
8
a
B
5
3
2
2
1
1
1
a
25
14
10
6
4
3
3
2
a
20
10
7
4
3
2
2
1
a
B
3
2
a
535
508
412
232
264
216
206
177
a
391
391
352
241
366
266
366
319
a
33
10
9
2
8
4
6
4
a
179
48
62
20
112
38
94
59
a
13
1
2
0
5
1
3
2
a
68
11
II
2
13
4
6
5
a
228
212
107
85
66
44
15
6
a
12
17
5
7
3
2
1
4
a
32
28
27
9
41
17
38
24
a
290
149
125
25
102
53
63
51
- - Value below the limit of detection.
' Data not available.
-------
Table D-4. Emission Factors, in (ig/m1, of Selected VOCs for Paper/Reformulated Phenolic Sample B Laminate
f.
*
0
5
12
24
48
96
144
216
336
3
1
3
1
1
1
0
0
0
0
?
20
112
98
70
66
46
6
13
4
I
6
9
4
3
2
1
1
1
1
I
t
Ul
10
10
7
5
3
2
2
2
1
|
E
8
7
6
4
3
2
1
1
1
i
01
3
2
2
1
1
1
1
0
0
(aldehyde |
fr
^
(/>
319
349
307
275
187
250
197
194
167
a.
175
392
370
286
135
331
393
442
382
j
20
1
3
3
1
5
6
6
5
|
E
155
29
52
23
16
65
119
125
94
imethylphenol |
O
fi
9
1
2
1
0
2
4
4
4
saldehyde B
I
40
-
-
4
1
9
7
6
6
Iroxybenzyl alcohol
^
!
30
113
116
108
45
35
19
10
6
Iroxybenzaldehyde 1
f
-
8
3
9
5
4
2
-
-
§
1
17
13
15
11
7
30
41
-
0
(S
8
»-
Ot
u
163
42
40
56
17
84
66
-
0
- - Value below the limit of detection.
m
-------
Table D-5. Emission Factors of Selected VOCs for Glass/Lignin Sample 1A Laminate
Time interval,
h
Methyl ether
emission factor,
iig/h-m2
1 -Methoxy-2-propanol
emission factor,
txg/h-m2
0
5
12
24
48
96
144
144A
216
336
241.3
115.5
74.4
53.7
32.4
16.9
14.0
13.6
10.5
6.9
5.9
2.1
1.3
0.9
-
-
0.0
0.0
0.3
-
- - Value below the limit of detection.
Table D-6. Emission Factors of Selected VOCs for Glass/Lignin Sample 1B Laminate
Time interval,
h
0
5
12
24
24A
48
96
144
216
336
Methyl ether
emission factor,
tig/h-m2
197.4
93.2
88.9
60.0
58.5
37.5
20.4
16.6
11.3
6.9
1 -Methoxy-2-propanol
emission factor,
jig/h-m2
5.2
2.5
2.1
1.4
1.2
0.5
0.5
0.3
-
0.2
- - Value below the limit of detection.
Table D-7. Emission Factors of Selected VOCs for Glass/Lignin Sample 2 Laminate
Time interval,
h
0
5
12
24
48
96
144
216
336
Methyl ether
emission factor,
UK/h-m2
265.5
120.3
83.5
31.2
46.2
22.4
15.5
12.2
9.6
1-Methoxy-2-propanol
emission factor,
UR/h-m2
0.1
0.1
0.0
0.8
1.0
0.0
0.5
0.4
0.2
- - Value below the limit of detection.
D-6
-------
Table D-8. Sum of Measured Concentrations at t-0 h and t- 336 h for All Reported
VOCs for Each Laminate Tested
SumofVOC SumofVOC
concentrations in Hg/mJ, concentrations in fig/m3,
Laminate type * t«0h t«336h
Paper/phenott
Paper/reformulated phenolic Sample A
Paper/reformulated phenolic Sample B
Glass/lignin Sample 1A
Glass/lignin Sample IB
Glass/lignin Sample 2
Glass/epoxy
Glass/epoxy Sample 2
23,014
15,809
9,602
1,566
1,285
1,688
1
15
3,933
6,145
6,214
64
66
93
0
1
* First data point for giass/lignin samples 1A, 16, and glass/epoxy sample at t-5 h.
D-7
-------
Table D-9. Raw Concentration, |ig/m3, Data for Identified VOCs
00
Paper/phenof
Compound
Acetone
Toluene
Hexanal
Ethylbenzene
m.p-Xytene
o-Xytene
Sallcytaldehyde
Phenol
o-Cresol
m.p-Cresot
2,4-Dimethylphenol
m-Anisaldehyde
2-Hydroxybenzyl alcohol
4-Hydroxybenzaldehyde
C8H10P2
C9H1202
Compound
Acetone
ToluGno
Hexanal
Ethylbenzene
m,p-Xylene
V«J!AMA
o-Ayiene
Salicytaldehyde
Phannl
r IHJIIwl
0*Cf6SOl
m.p-Cresol
2,4-Dimethylphenol
m-Anisafdehyde
2-Hydroxybenzyl alcohol
4-Hydroxybenzaldehyde
T5-t
34.30
2615.03
35.16
204.31
152.51
56.79
7294.99
10353.64
29.31
798.23
17.86
25.29
461.11
273.82
693.71
183.57
T5-1
42.67
114.46
72.49
212.12
170.07
68.75
4500.48
3300.12
289.79
1589.00
117.78
608.48
2047.01
107.53
271.12
2511.80
T12-1
31.90
879.93
16.98
213.94
174.67
65.90
4554.56
3384.70
114.12
454.10
12.57
68.98
602.38
905.78
T12-1
26.05
46.91
139.88
98.51
34.30
4725.27
3628.74
104.28
530.63
22.51
143.97
2119.09
168.30
262.27
1457.36
T12-10
27.07
874.91
15.43
197.48
163.06
64.12
3968.16
3182.22
78.35
495.20
20.77
55.07
1005.75
29.89
550.17
652.61
21.96
29.39
93.09
66.13
22.80
3888.65
3315.71
85.56
597.23
20.40
106.72
1166.49
60.92
25000
1189.69
T24
19.03
567.34
11.94
166.59
131.67
52 34
4002.85
4021.81
156.62
1091.56
52.71
125.20
1001.27
36.71
1136.36
1047.81
19.27
22.06
56.00
39.38
12.52
2183.80
2262.30
17.17
198.34
5.91
23.34
923.71
73.49
79.32
241.05
T48
24.87
1147.18
13.12
89.84
69.66
22.79
1805.64
2333.17
16.49
122.27
4.78
9.94
864.77
48.25
155.15
91.69
14.79
187.02
15.21
37.86
27.87
10.50
2449.93
3407.38
74.74
1053.88
48.17
123.36
742.48
36.14
380.90
937.31
T96
17.15
747.00
4.58
62.34
50.38
21.38
2555.61
3548.42
60.62
565.12
19.41
49.87
538.01
30.19
613.07
489.05
14.65
14.08
27.51
20.30
7.92
2016.05
2488.75
34.80
368.75
14.33
46.20
536.63
29.51
161.77
496.94
T144
16.18
221.31
8.21
41.74
34.55
14.86
2117.44
3420.22
61.96
637.75
21.43
46.44
596.72
73.98
736.33
436.67
13.56
12.23
23.84
17.71
6.51
1915.96
3406.20
58.40
883.70
32.78
57.17
266.20
18.26
348.18
584.55
T216
15.87
480.14
7.88
29.09
24.07
9.51
1429.85
919.03
9.40
72.52
1.43
3.70
121.94
4.53'
124.66
63.36
13.05
69.52
10.85
17.84
13.50
5.06
1645.70
2980.92
42.20
562.49
24.78
54.25
189.05
41.50
221.17
468.71
T336
15.41
603.05
3.71
17.86
14.96
5.97
1152.82
1037.21
18.05
96.48
3.00
8.92
705.08
186.57
197.29
81.36
-------
Table D-9. Raw Concentration, u,g/m3, Data for Identified VOCs (con.)
Paper/reformulated phenolic Sample B
vo
Compound
Acetone
Toluene
Hexanal
Ethylbenzene
m,p-Xylene
o-Xytene
Salicytaldenyde
Phenol
o-Cresol
m,p-Cresol
2,4-Dlmethylphenol
m-Anisaldehyde
2-Hydroxybenzyl alcohol
4-Hydroxybenzaldehyde
C«H1002
C9H1202
TO-1
19.80
51.33
31.28
58.71
49.38
19.74
1888.27
913.04
132.78
1032.73
64.91
277.17
248.98
107.04
1062.68
T5-1D
39.84
924.32
84.83
90.49
68.12
21.23
3099.50
3338.96
23.01
354.66
17.46
36.25
1060.69
71.11
117.23
469.26
T12-1
25.82
915.82
37.14
67.47
52.80
16.89
2874.22
3468.46
27.62
494.06
19.69
1191.22
34.39
141.75
370.37
T24
21.07
655.33
29.48
44.93
35.67
11.61
2578.70
2730.23
25.78
237.04
8.23
42.88
1139.20
86.19
100.00
523.00
T48
17.81
609.40
22.27
30.97
24.63
7.66
1751.84
1282.01
13.12
160.69
5.53
15.96
553.69
55.72
6342
156.04
T96
15.46
430.58
12.40
21.83
18,01
6.54
2318.13
3065.05
51.79
607.33
23.00
89.54
453.67
40.33
276.29
775.95
T144
15.79
61.82
9.01
15.64
12.81
4.84
1835.95
3652.25
57.95
1109.80
41.35
74.72
303.54
22.10
377.08
619.79
T216
15.49
115.62
10.02
15.17
12.45
4.62
1807.93
4111.38
58.33
1169.49
40.17
65.25
226.08
4.62
T336
15.44
40.15
9.31
10.75
9.14
3.47
1557.87
3560.50
47.23
890.91
40.67
62.53
180.78
BKGB1
13.01
0.11
0.39
0.02
0.11
11.15
26.43
2.20
16.87
2.53
6.22
129.23
6.87
(Continued)
-------
Table D-9. Raw Concentration, fig/m1, Data for Identified VOCs (con.)
Glasaffignfn Sample 1A
9
o
Compound
Methyl ether
1 -Methoxy-2-propanol
Glass/ligntn Sample 1B
Compound
Methyl ether
1 -Methoxy~2*propano1
Glass/lignln Sample 2
Compound
Methyl ether
1 *Methoxv-2*DfODanol
fcompouno
1 Methoxy-2-propanol
2-ethvl-1 hexanol
QjaMJegox^Samgje^^ mmfm
Compound
4 .lulathrtyu.9.nmnflnol
TO-1
1528.03
3B.01
TO-t
1252.65
3278
TO-1
1687.07
098
15.41
H^^BHMMBBgMgg*""''
TO-1
T5-1
1144.45
21.84
T5-1
908.84
2478
T5-1
1196.44
095
7.79
^M
T5-1
0.64
T12-1
725.63
13.15
T12-1
838.03
1967
T12-1
819.75
5.15
tmmm^mm**
T12-1
T24
509.19
8.82
T24
566.96
13 33
T24
306.84
6.98
4.57
^MMM^BB
T24
0.01
T4B
305.19
0.02
T24-1D
545.32
11 65
T48
427.34
9.00
2.62
I^BBHB
T48
T96
158.08
T48
352.63
5.09
T96
210.45
0.01
2.51
T96
T144
129.69
T96
190.84
4.29
T144
144.45
4.64
1.51
T144
T144-0
125.79
T144
154.02
3.05
T216
113.43
3.38
1.38
T144-1D
T218
97.47
2.64
T216
105.23
T336
91.15
2.25
1.24
T216
T336
64.59
T336 BKG-1
64.71 0.24
1.63
0.55
T336
-------
APPENDIX E
ACS QA Data Audit Report Form
Audit Information:
Document or Data: VOC/aldehyde data
Study Director: L. Sheldon/J. Keever
QA Auditor: D. J. Smith
Response Provided:
Project: 5784-004
Audit Date: 7/7, 7/8, 7/9/97
Study Director/Date
All findings have been corrected or resolved:
QA Officer/Date
Documents/Sources included in Audit:
VOC data
Aldehyde data
Draft data sheets (GC/MS) Draft data sheets (summary data)
Raw data notebook (includes Chain of Custody Product testing records
sheets) Notebook #7691
HP 5988 #2 Logbook #5, #7068 Raw data notebook
Specific Audit Items or Phases:
Data audit
Review of documentation
Adherence to QA Plan, SOPs and Protocols
Attachments
Summary and Comments
Additional Distribution
D. Whitaker
J. Keever
E-1
RTI/ACS-QA-92-6R1
-------
Summary and Comments
Study/Document: VOC data, 5783-004
I. VOC Data:
Summary:
The VOC data set is complete. The record is well organized, complete, and for the most part
clearly annotated.
Overall completeness will meet completeness goal of 95% (72 samples, excluding
duplicates and chamber blanks) if 3 samples are declared invalid. If 4 samples are invalid,
completeness drops to 94%.
One of the 3 chamber blanks on the custody sheets was not run and is not accounted for.
This is currently being investigated.
Overall, the data set appears to be sound.
The issue of valid data near the lowest calibration point for secondary quadratic
calibrations is an issue in this study. The data below the lowest calibration, and some data
above this point appear to be invalid. This is being investigated. Strategies for evaluating
second-order calibrations, similar to those used for linear calibrations, need to be put in
place.
In addition, strategies for evaluating the quality of data at and above the highest calibration
point(s) need to be evaluated as well.
I have some concerns about the acetonitrile data. The RF is almost unbelievably high.
QC
Very little QC is specified in the QA Plan.
QC procedures documented include daily RF checks and review of the quant reports and
data reports.
Spiked controls were specified in the QA Plan, but apparently not implemented.
Concerns
Calibration is a technical, not a QA concern. However, I do have some calibration
concerns. First, second-order calibrations should be used when technical reasons dictate,
not just to get a better r-value. Second, 4 points are not sufficient to describe second-order
curves. Third, some of the plots are equivocal. Take acetone, for example. Saturation of
the detector would suggest a convex shape, or flattening out of the top. This second-order
plot is concave, suggesting to me that one of the points is an outlier.
Comments:
Item
Comment
Check results vs. samples taken (completeness). All samples analyzed, reported.
One chamber blank was not analyzed. There appear to have been 2 blanks taken
on the same day (different volumes), but with the same ID.
Calibration. Multilevel calibration was performed. Range is reported correctly in the
data sheets.
Data sheets consist of 3 pages: direct transcription from quant report; quant ion
summary (ng); concentration (ng/L).
- 2 RTIMCS-QA-93-04
-------
Item
Comment
RF checks. lOOx cartridge run each day; reports are complete in the notebook.
However, no information available on acceptance criteria; no documentation that
results were checked and "passed". A second cart, was run in a number of cases; an
explanation was included only once.
Checked representative data for correct, complete transcription to data sheets. No
transcription errors were found. There were some concerns, most of which have been
resolved.
There seems to be a problem with toluene in the paper/reformulated phenolic
sample A series. Toluene has been recalculated for most, but not all of the samples,
but the recalculation has not been included in the data sheet along with the other
manual recalculations. Resolved: the data reported are the correct data; manual
data was one-point calibration ca/c. that was not used
There seems to be a problem with phenol in the glass/epoxy sample 2 duplicate
pair T48/T48-D. Please look at phenol. It was manually "zeroed" in one sample,
but not the other. The manual calculation was not transcribed. Phenol needs to be
reviewed for this whole series. Resolved: data in question will be below the
reporting level.
There seems to be something wrong with some of the calibrations. 2,4-
dimethylphenol, for example. In glass/epoxy sample 2, 63 area counts result in 32
ng. This is clearly wrong, since the *25X" cartridge (27 ng) has area counts of
10894. Resolved:.result of second order calibration. Lower limit will be reviewed.
Are the recalculated values for salicylaldehyde, phenol and m,p-cresol a different
issue from the toluene recalculations (paper/phenol)?
II Aldehyde Data
Comments:
1. Overall, the record is complete and well-organized. Raw data files appear to be complete;.
reported results are complete. Completeness - 100%. The organization and completeness of
the record are quite good and this was very helpful for reviewing the data and supporting
confidence in the results.
2. The data are reported accurately. Minor transcription errors probably have little to no effect on
the data.
3. Data Quality Indicator Coals (QA Plan)
Completeness - met
Accuracy (recovery of spiked samples). Met, but (see below).
Precision (RSD for duplicate air samples and duplicate tests). Duplicate air samples were
not collected; duplicate tests not yet evaluated. For QA/QC purposes, this should be done.
Oven temperature - checked with NIST-traceable device. Don't know if this was done.
RH of supply air checked with NIST-traceable device. Don't know if this was done.
Emission factors. Not evaluated.
4. QA Plan included little QC information/criteria for aldehydes.
Linear calibration with r2 * 0.998 required. Linear calibration was not used. Need to
address guidelines and criteria for other calibration schemes.
Check standards within 85-115% of prepared cone. Some standards are out of control
E-3
RTI/ACS-QA-93-04
-------
(high) in several RFs.
There are data validation steps in section 6.2 of the QA Plan that need to be reviewed.
Some of these have been done, but others have not (?). These need to be completed so that
OAO can verify that this has been completed.
Item
1
2
3
4
5
6
7
8
9
10
Comment
Checked results for completeness (vs. product testing logs). No problems noted.
Checked results (summary data sheets) vs. raw data. No problems or errors noted. A
few minor comments (~ 18 of 72).
Class/epoxy. Value for propfonald. not transcribed. Fairly small amount.
Assume that data will be "trimmed" so that values below some point (lowest
calibration, or whatever) will not be reported.
There are a few values for 2-butanone that are above the highest calibration
point. Would it be useful to tag them in the report?
Checked calibration. Documentation is complete.
COC sheets included in raw data notebook. Documentation is complete.
0.5x RF run at regular intervals - after every 5-6 samples. Did not check to verify
that ail results were accounted for, but the record appears to be complete.
The summary included with sample results is extremely helpful and useful.
The documentation doesn't show that the results were reviewed/approved.
Several appear to be high (060497,43,1 and 060597,1 6,1 , for example), while
one run, 060497,29,1 appears to be just inside acceptance limits (low),
It is not entirely clear what "bad injection* means for 3 of the 0.5x RF runs. Did
this occur only with the RF and not samples?
Checked volume used for calculation (Multichrom vs. protocol sheets). No problems
noted (5 of 31).
Only checked samples run 6/4/97. Could not find analysis list for 6/5/97, so
couldn't verify volumes.
Checked volumes on protocol sheets vs. raw form (14 of 72 checked).
transcription error from raw data - final flow glass/epoxy. Small change in final
volume.
very trivial transcription error - .01 in final time glass/epoxy-T5.
The flows were very similar, so it is doubtful that small errors would affect the data.
As an internal QC check, it would be a good idea for someone to independently
check transcription and calculation of some of the highest and lowest flow rates.
Did not check temp, humidity log to verify that chamber conditions were met.
Notebook record is OK, and includes helpful information.
It would be helpful to include source for die DNPH standards.
Controls. QA Plan *75% recovery. It appears that this goal is met for all
compounds; however, this includes subtract! nR background which is substantial for
acetaldehyde and acetone. I'm not sure that results with background this high can
K considered valid. Presumably the background will then be subtracted from the
sample data as well.
E-4
RTI/ACS-QA-93-04
-------
TECHNICAL REPORT DATA
f Please read Instructions on the re»ene before completing)
. REPORT NO.
EPA-600/R-98-034
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Personal Computer Monitors: A Screening Evaluation
of Volatile Organic Emissions from Existing Printed
Circuit Board Laminates and Potential Pollution*
5, REPORT DATE
April 1998
6, PERFORMING ORGANIZATION CODE
. AOTHOR(S)
Dean H.
Cornstubble and Donald A. Whitaker
8. PERFORMING ORGANIZATION REPORT NO.
92U- 5783-003
'. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P. O. Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR 822025-01
IX SPONSORING AGENCY NAME AND ADDRESS
13,.
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
JIOO COVERED
14. SPONSORING AGENCY CODE
EPA/600/13
«. SUPPLEMENTARY NOTES APPCD project officer is Kelly W. Leovic, Mail Drop 54, 919 /
541-7717. (*) Prevention Alternatives
is. ABSTRACT
repOrt gives results of a screening evaluation of volatile organic emis-
sions from printed circuit board laminates and potential pollution prevention alterna-
tives. In the evaluation, printed circuit board laminates, without circuitry, com-
monyl found in personal computer (PC) monitors, were tested to determine if an al-
ternative laminate would be less emitting than conventional laminates. Test results
qualitatively showed that the alternative, a glass/lignin- containing epoxy resin lam-
inate, emits fewer volatile compounds than paper/phenolic resin-based laminates.
The data also suggest that, if these laminates were used as a replacement for paper/
phenol circuit board laminates in PC monitors, emissions from PC monitors could
be reduced in indoor environments where they are used.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Pollution
Computers
Monitors
Laminates
Printed Circuits
Organic Compounds
Volatility
Emission
Pollution Prevention
Stationary Sources
Personal Computers
Volatile Organic Com-
pounds (VOCs)
13 B
09B
14G
11D
09A
07C
20 M
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report}
Unclassified
21. NO. OF PAGES
74
2O. SECURITY CLASS fThii page)
Unclassified
22. PRICE
EPA Form 222O-1 JS-73)
E-5
------- |