Developing a Reference Material for
Formaldehyde Emissions Testing
Final Report
Submitted to:
Xiaoyu Liu, Ph.D.
U.S. Environmental Protection Agency (E305-03)
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Indoor Environment Management Branch
Research Triangle Park, NC 27711, USA
Submitted by:
John Little, Zhe Liu, Xiaomin Zhao, and Steven Cox
Department of Civil and Environmental Engineering
Virginia Tech, Blacksburg, VA USA
Date: April 16, 2013
-------
DISCLAIMER
The work reported in this document was funded by the United States Environmental
Protection Agency (EPA) under Contract No. EP11C000135 and has been subjected to
the Agency's peer and administrative reviews and has been approved for publication as
an EPA report. Any opinions expressed in this report are those of the authors and do not
necessarily reflect the official positions and policies of the EPA. Any mention of
products or trade names does not constitute recommendation for use by the EPA.
-------
TABLE OF CONTENTS
DISCLAIMER i
TABLE OF CONTENTS ii
LIST OF FIGURES iv
LIST OF TABLES v
EXECUTIVE SUMMARY 1
SYMBOLS AND ABBREVIATIONS 3
ACKNOWLEDGEMENTS 4
1 INTRODUCTION AND OBJECTIVES 5
1.1 Introduction 5
1.2 Objectives 6
2 MATERIALS AND METHODS 7
2.1 Selecting polymer substrates 7
2.2 Generating gas-phase formaldehyde 7
2.3 Determining mass-transfer properties of selected polymers 8
2.4 Loading the identified polymer substrate with formaldehyde 9
2.5 Measuring formaldehyde emissions from pre-loaded films in small chambers 10
2.6 Predicting formaldehyde emissions from pre-loaded films 11
2.7 Task allocation 13
3 RESULTS AND DISCUSSION 16
3.1 Generating gas-phase formaldehyde 16
3.2 Evaluating mass-transfer properties of selected polymer films 17
3.3 Validating overall approach by small-scale chamber testing 21
3.3.1 Preliminary approach by small-scale chamber testing 21
3.3.2 Inter-laboratory study 22
3.3.3 Mass balance analysis 24
3.4 Measuring effectiveness of foil packaging and storage materials and methods 24
3.4.1 Packaging material evaluation 25
3.4.2 Shelf-life evaluation 25
3.5 Measuring emissions at different relative humidity (RH) levels 28
3.6 Statistical evaluation of measured and model-predicted concentrations 29
4 QUALITY ASSURANCE AND QUALITY CONTROL 31
4.1 VT quality assurance and quality control 31
4.2 EPA quality assurance and quality control 31
5 CONCLUSIONS 32
6 RECOMMENDATIONS 34
7 REFERENCES 35
APPENDIX A 38
APPENDIX B 42
APPENDIX C 45
APPENDIX D 55
-------
APPENDIX E 56
APPENDIX F 60
APPENDIX G 61
111
-------
LIST OF FIGURES
Figure 1. Strategy to develop a reference material for formaldehyde emissions testing....6
Figure 2. Microbalance and loading vessel system 10
Figure 3. Schematic representation of a formaldehyde-containing source in a test
chamber showing mechanisms controlling the emission rate 12
Figure 4. Measured weight decrease of diffusion vials overtime 16
Figure 5. Comparison of directly measured formaldehyde concentration with that
calculated from diffusion vial weight change 17
Figure 6. Sorption/desorption data and analysis for the 0.025 cm thick PMP 18
Figure 7. Sorption/desorption data and analysis for the 0.025 cm thick PC 19
Figure 8. Sorption/desorption data and analysis for the 0.051 cm thick PC 20
Figure 9. Comparison of duplicated analysis for 0.025 cm PC 20
Figure 10. Comparison of the measured and predicted formaldehyde emission profiles. 22
Figure 11. Comparison of measured and model predicted emission profiles in the ILS. 23
Figure 12. Comparison of the measured formaldehyde emission profiles in chamber tests
using films with and without foil wrapping 25
Figure 13. Comparison of the measured formaldehyde emission profiles in chamber tests
of different shelf-life and the model prediction 26
Figure 14. Comparison of model-predicted and measured formaldehyde emission
profiles from films stored for different durations 27
Figure 15. Peak gas phase concentration values of different shelf-life chamber tests 27
Figure 16. Emission profiles at different RH levels 28
Figure Al. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.48 g/m3,
D: 1.6x10'13m2/s, K: 210) 38
Figure A2. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 1 g/m3, D:
2.1xlO'13m2/s, K:270) 38
Figure A3. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.48 g/m ,
D: 1.5xlO'13m2/s, K: 190) 39
Figure A4. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.96 g/m3,
D:2.2xlO'13m2/s, K: 280) 39
Figure A5. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.86 g/m3,
D:2.2xlO'13m2/s, K: 210) 40
Figure A6. Sorption/desorption data and analysis for a 0.051 cm thick PC (y;n: 0.86 g/m3,
D:4.0xlO'13m2/s, K: 155) 40
Figure A7. Sorption/desorption data and analysis for a 0.051 cm thick PC (y;n: 0.97 g/m3,
D: 3.7x10'13m2/s, K: 180) 41
Figure A8. Sorption/desorption data and analysis for a 0.025 cm thick PMP (y;n: 1.73
g/m3, D: 3.5xlO'14m2/s, K: 40) 41
Figure Bl. Sorption data of loading process for VT 10 (Batch 1) 42
Figure B2. Sorption data of loading process for VT 11 (Batch 2) 43
Figure B3. Sorption data of loading process for VT 12 (Batch 3) 43
IV
-------
LIST OF TABLES
Table 1. Generating gas-phase formaldehyde 8
Table 2. Determining mass-transfer properties of selected polymer by microbalance
sorption/desorption test 9
Table 3. Loading the identified polymer substrate with formaldehyde 10
Table 4. Measuring formaldehyde emissions from pre-loaded films in small chambers. 11
Table 5. Predicting formaldehyde emissions from pre-loaded films 13
Table 6. Task allocation summary 14
Table 7. Formaldehyde mass balance analysis 24
Table 8. Statistical analysis summary 29
Table Cl. Preliminary small-scale chamber raw data for Batch 1 — film B1F1 45
Table C2. Preliminary small-scale chamber raw data for Batch 1 — film B1F2 45
Table C3. Preliminary small-scale chamber raw data for Batch 1 — film B1F3 46
Table C4. Small-scale chamber raw data for Batch 2 — film B2FA1 46
Table C5. Small-scale chamber raw data for Batch 2 — film B2FA2 47
Table C6. Small-scale chamber raw data for Batch 2 — film B2FA4 47
Table C7. Small-scale chamber raw data for Batch 2 — filmB2FA5 48
Table C8. Small-scale chamber raw data for Batch 2 — film B2FA6 48
Table C9. Small-scale chamber raw data for Batch 2 — film B2FA7 49
Table CIO. Small-scale chamber raw data for Batch 2 — film B2FB1 49
Table Cll. Small-scale chamber raw data for Batch 2 — filmB2FB2 50
Table C12. Small-scale chamber raw data for Batch 2 — film B2FB3 50
Table CIS. Small-scale chamber raw data for Batch 2 — film B2FB4 51
Table C14. Small-scale chamber raw data for Batch 3 — film B3FA3 51
Table CIS. Small-scale chamber raw data for Batch 3 — film B3FA2 52
Table C16. Small-scale chamber raw data for Batch 3 — filmB3FCl 52
Table C17. Small-scale chamber raw data for Batch 3 — film B3FC2 53
Table CIS. Small-scale chamber raw data for Batch 3 —filmBSFAl 53
Table C19. Small-scale chamber raw data for Lab A 54
Table Dl. Total formaldehyde in the film as measured using the microbalance 55
Table D2. Formaldehyde mass-balance calculation using chamber test data 55
-------
EXECUTIVE SUMMARY
EXECUTIVE SUMMARY
Exposure to formaldehyde has been shown to produce broad and potentially severe
adverse human health effects. With ubiquitous formaldehyde sources in the indoor
environment, formaldehyde concentrations in indoor air are usually higher than outdoors,
ranging from 10 to 4000 ug/m3. As a result, industry and government are taking actions
to minimize formaldehyde exposure in the indoor environment. A critical step toward
mitigating formaldehyde exposure in the indoor environment is assessing the potential of
building materials and indoor furnishings to emit formaldehyde. These assessments
usually involve emissions measurements obtained using environmental chambers.
However, some variability currently exists with respect to chamber testing results. A
formaldehyde emissions reference material could be used to identify and eliminate or
minimize the root causes of formaldehyde emissions measurement variability. The
objective of this research project was to create and evaluate such a reference material.
The formaldehyde emissions reference material development progressed through the
following major steps: (1) identifying a suitable polymer film for use as a reference
material and characterizing its mass-transfer properties; (2) loading formaldehyde into the
selected polymer film; (3) predicting formaldehyde emissions from the pre-loaded
polymer films through the use of a fundamental emission model; and, (4) measuring
formaldehyde emissions from the pre-loaded films in small-scale environmental
chambers. The reference material was evaluated by (1) comparing actual formaldehyde
emission profiles measured using small-scale emission chambers to model predictions;
(2) evaluating the effect of storage duration; (3) evaluating the effect of the packaging
material; and, (4) investigating how reference material formaldehyde emissions were
affected by humidity.
Polycarbonate film was selected as a suitable reference material substrate due to its
relatively large partition coefficient with respect to formaldehyde. Using measured mass-
transfer coefficients and chamber parameters, the emissions model was used to predict
formaldehyde concentration profiles during small-scale chamber testing. Although
measured chamber concentrations tended to be lower than model predictions in early time
periods and higher than model predictions in later time periods, the measured emission
profiles were overall quite similar to model predictions.
Subsequent chamber tests were conducted to evaluate the effect of reference material
storage duration. Results suggested that some formaldehyde was lost from the reference
materials over time but also showed good agreement with model predictions over the
course of 144-hour tests when samples had been kept in storage for periods of less than 2
weeks.
The last two evaluations included testing the effect of the foil packaging material and the
effect of humidity on formaldehyde emissions. Results of the packaging material test
indicate that a tight aluminum foil wrap reduces but does not eliminate formaldehyde loss
from the reference material. Although the results of the humidity evaluation were not
consistent, the data suggests that formaldehyde emissions from the reference material are
1
-------
EXECUTIVE SUMMARY
possibly affected by humidity. This could be due to water molecules plasticizing the
polymer film or possibly facilitating formaldehyde chemical reactions.
Recommendations for future work include further investigation of formaldehyde
chemistry and reaction pathways and mass-transfer behavior, packaging materials and
storage methods, evaluation of additional substrates for use as emissions reference
materials, scale-up for use in large chambers, and additional testing for overall method
evaluation.
-------
SYMBOLS AND ABBREVIATIONS
SYMBOLS AND ABBREVIATIONS
SYMBOLS
A exposed surface area of the material (m2)
C material-phase formaldehyde concentration (g/m3)
Co initial uniform material-phase formaldehyde concentration (g/m3)
D diffusion coefficient of the VOC in the material (m2/s)
K partition coefficient of the VOC between the material and air (dimensionless)
L thickness of the material (m)
Q volumetric air flow rate (m3/h)
r correlation coefficient
t time (s)
V well-mixed chamber volume (m3)
x distance from the base of the material (m)
y gas-phase formaldehyde concentration in the bulk chamber air (g/m3)
y;n gas-phase formaldehyde concentration in the influent air (g/m3)
ABBREVIATIONS
ASTM American Society for Testing and Materials
ATSDR Agency for Toxic Substances and Disease Registry
BDL below detection limit
DAD diode array detector
DCC daily calibration check
DNPH 2,4-dinitrophenylhydrazine
EPA United States Environmental Protection Agency
HPLC high performance liquid chromatography
IAP Internal Audit Program
IARC International Agency for Research on Cancer
ILS inter-laboratory study
NIST National Institute of Standards and Technology
NMSE normalized mean square error
PC polycarbonate
PMP polymethylpentene
QAPP quality assurance project plan
QAQC quality assurance and quality control
RH relative humidity
TSCA Toxic Substances Control Act
UF urea-formaldehyde
UV ultraviolet
VOCs volatile organic compounds
VT Virginia Tech
-------
ACKNOWLEDGEMENTS
ACKNOWLEDGEMENTS
The authors thank Nancy F. Roache, Corey A. Mocka, and Robert H. Pope of ARCADIS
for conducting all the EPA small chamber tests and Dr. Chip Frazier of Virginia Tech for
comments on the draft report.
-------
INTRODUCTION AND OBJECTIVES
1 INTRODUCTION AND OBJECTIVES
1.1 Introduction
Formaldehyde (H2C=O), the simplest member of the aldehyde family, is a flammable,
colorless gas with a pungent odor at room temperature. Since the 1880s, formaldehyde
has been produced commercially and in recent years, annual global industrial production
of formaldehyde is estimated at more than 21 million tonnes (Bizzari, 2009).
One of the primary uses of formaldehyde is for producing synthetic resins, such as urea-
formaldehyde (UF), phenol-formaldehyde, melamine-formaldehyde and polyacetal
resins, which are used as adhesives and impregnating resins in wood products, curable
moulding products, and textile, leather, rubber and cement industries (IARC, 2006).
These products may emit formaldehyde during the use phase although the emission rate
may vary greatly (Kelly et al, 1999; Meyer and Boehme, 1997; Weigl et al, 2009).
Due to the ubiquitous presence of formaldehyde emission sources indoors as well as the
slow removal rate in the indoor environment (outdoor formaldehyde is readily removed
by photolysis and reaction with hydroxyl radicals in the presence of sunlight to produce
carbon dioxide), formaldehyde concentration in indoor air (10 to 4000 ug/m3) is usually
much higher than outdoors (3 to 70 ug/m3) (ATSDR, 2008; IARC, 2006; Salthammer et
al., 2010; WHO, 1989).
Due to the potential health risks associated with indoor formaldehyde exposure, various
guidelines, standards, and recommendations have been established around the world
(Salthammer et al., 2010). The Formaldehyde Standards for Composite Wood Products
Act, enacted as Title VI of the Toxic Substances Control Act (TSCA), was signed into
law by President Barack Obama in July 2010. TSCA Title VI requires formaldehyde
emissions testing in chambers to demonstrate compliance with these standards.
Inter-laboratory studies are often used to evaluate chamber testing performance.
However, these studies can be costly and time-consuming and may lead to inconclusive
results. The creation of a well characterized reference material for formaldehyde
emissions testing is therefore critical for validating and calibrating emissions testing
procedures.
In collaboration with the National Institute of Standards and Technology (NIST),
researchers at Virginia Tech (VT) have developed a prototype reference material for
VOCs emissions testing (Cox et al., 2010; Howard-Reed et al., 2011). It consists of a thin
polymethylpentene (PMP) film that is loaded with toluene to equilibrium using a carrier
gas stream containing a known gas-phase toluene concentration. Extensive chamber tests
at NIST and other emissions testing laboratories have shown that the emissions behavior
of the reference material resembles actual homogenous building materials with emission
profiles that can be accurately predicted by a mechanistic model. The model predicted
emission profiles therefore serve as true reference values and can be compared to
emissions testing results of individual laboratories. This is of considerable benefit
because the model not only provides reference emission values for validating individual
-------
INTRODUCTION AND OBJECTIVES
laboratories' performance, but also can provide insight into the likely causes of
variability and experimental errors. VT's recent work shows that the same approach
using PMP is applicable to n-butanol, a more polar compound than toluene.
In this project, a similar procedure was employed to create a reference material for
formaldehyde following the same basic steps (Figure 1): (1) identifying a suitable
polymer substrate and determining its mass-transfer properties; (2) loading formaldehyde
into the polymer film; (3) predicting formaldehyde emissions from the pre-loaded
polymer films employing a fundamental emission model; (4) measuring formaldehyde
emissions from the pre-loaded films in small-scale environmental chambers; and, (5)
comparing the predicted emission profiles to the measured results.
/-~
Identify a suitable substrate
Select a candidate
h
\.
ymer substrate
• Uniform
• Stable
• Pure
• Unreactive
^^ 1 Determine its mass-
+c
isfer properties
~~~~\
• Have sufficient
solubility of
formaldehyde
• Material-phase
diffusion of
formaldehyde
is Field an in
nature
_^-^
Load the suitable
polymer substrate
Figure 1. Strategy to develop a reference material for formaldehyde emissions testing.
1.2 Objectives
The specific objectives of this research project were to:
• develop a method to reliably generate a dry gas stream containing a controllable
concentration of formaldehyde;
• evaluate different polymer films for use as a formaldehyde source and select the
most suitable material for use as a reference material substrate;
• measure the diffusion coefficient, D, and partition coefficient, K, of formaldehyde
for the selected polymer film;
• validate the overall approach in (1) small-scale chamber testing, (2) formaldehyde
mass balance analysis, and (3) comparison of measurements with model
predictions;
• investigate the effectiveness of packaging and storage methods; and
• investigate the effect of temperature and humidity on formaldehyde emission
characteristics.
-------
MATERIALS AND METHODS
2 MATERIALS AND METHODS
2.1 Selecting polymer substrates
As outlined in Figure 1, identifying a suitable polymer substrate was the first step in
creating a viable reference material. An ideal polymer should be uniform and stable so
that the material's properties do not change. It should neither react with formaldehyde
nor contain any reactive or volatile impurities (additives or contaminants) that may
confound mass-transfer of formaldehyde within the material. Furthermore, the success of
this method depends on two key criteria: first, formaldehyde needs to be sufficiently
soluble in the polymer substrate so that an adequate amount of formaldehyde can be
loaded into and then allowed to diffuse from the substrate; and second, the diffusion of
formaldehyde in the polymer substrate must be ideal or Fickian in nature.
Although the encouraging application of PMP for both toluene and n-butanol suggests
that diffusion of formaldehyde within PMP, a non-polar polymer, may be ideal, its
solubility is rather low given that formaldehyde is quite polar. It is expected that
formaldehyde has higher solubility in polar matrices. For example, the solubility in
polycarbonate (PC) is reported to be about 150 times higher than in polypropylene
(Hennebert, 1988). However, additional attention should be paid when selecting polar
polymers because formaldehyde may react with hydroxyl (-OH) or amine (-NEk) groups
in polymers by forming a methylol (-CH2OH) group with the active hydrogen (Walker,
1975). Such reactions have been observed in cellulose, paper, nylon, latex and polyester,
rendering formaldehyde's transport within these materials non-ideal (Hennebert, 1988).
Therefore, PMP material with a thickness of 0.025 cm and two polycarbonate materials
of thicknesses 0.025 and 0.051 cm, respectively, were evaluated for potential use as a
formaldehyde emissions reference material. These polymeric materials are commercially
available and were purchased directly from manufacturers in large sheets.
2.2 Generating gas-phase formaldehyde
As described later, a continuous gas stream with a constant formaldehyde concentration
was needed to characterize mass-transfer properties of candidate polymers and to load
formaldehyde into the polymer substrate. The formaldehyde gas generating system
consisted of a diffusion vial placed in a temperature-controlled calibration gas generator
(Dynacalibrator Model 190, VTCI Metronics Inc., Santa Clara, CA) with a purge clean air
flow regulated by a mass-flow controller (Model FC-280S, Tylan General, Carson, CA).
Solid paraformaldehyde (97%, Alfa Aesar, Ward Hill, MA) contained in the diffusion
vial depolymerized to monomeric formaldehyde gas at elevated temperatures in the oven
of the calibration gas generator (Rock et al, 2010), which then diffused into the purge
flow of dry and clean air (UN1002, Airgas Inc., Radnor, PA). While maintaining
constant purge gas flow rate, formaldehyde concentration in the generated gas stream was
varied by adjusting the oven temperature and using diffusion vials with different
diffusion path lengths. The formaldehyde gas generation parameters are summarized in
Table 1.
-------
MATERIALS AND METHODS
To determine the rate of formaldehyde emissions from the diffusion vial containing
paraformaldehyde, the vial was weighed using a mechanical balance (with a precision of
-10 jig) over appropriate time intervals. The linearity between the measured weight and
time can be examined to determine whether the formaldehyde release rate was constant.
The formaldehyde concentration in the generated gas stream can then be calculated by
dividing the formaldehyde release rate by the purge flow rate. The true gas flow rate was
measured using a bubble flowmeter (mini-Buck Calibrator, A.P. BUCK Inc., Orlando,
FL).
For comparison, the formaldehyde concentration in the gas stream was also directly
measured by visible absorption spectrometry, following NIOSH Analytical Method 3500
(NIOSH, 1994). Briefly, an appropriate volume of the gas stream was passed through
two impingers in series containing 20 mL 1% sodium bisulfite solution so that the gas-
phase formaldehyde was completely absorbed by the solution, forming HOCFLySOsNa.
The backup impinger was used to check collection efficiency. Then aliquots of the
impinger solution were transferred to a flask and mixed with 0.1 mL 1% chromotropic
acid and 6 mL concentrated sulphuric acid. After heating the sample solution at 95 °C for
15 minutes and maintaining it at room temperature for 2 hours so that the chromophore
was fully developed, the absorbance at 580 nm was measured using a spectrophotometer
(Spectronic 20D+, Thermo Scientific, West Palm Beach, FL). Meanwhile, a blank and
six calibration standard solutions prepared from a formaldehyde standard aqueous
solution (1000 |ig/mL, AccuStandard, New Haven, CT) were also treated with the
reagents and analyzed by the spectrophotometer for absorbance at 580 nm. A calibration
line (absorbance versus formaldehyde concentration of the calibration standard) was
constructed and the formaldehyde concentration in the tested solution sample was
obtained from the calibration line. Finally, formaldehyde concentration in the gas stream
was calculated using an appropriate aliquot factor and the gas sample volume.
Table 1. Generating gas-phase formaldehyde
Test ID
VT 1
VT2
VT3
VT4
VT5
VT6
Vial ID
Viall
Viall
Vial 2
Vial 3
Vial 3
Vial 3
Diffusion path
length
(mm)
76
76
38
25
25
25
Temp.
85
95
95
95
100
105
Test
duration
(h)
140
310
290
70
140
65
Avg. flow
rate
(mL/min)
250
250
250
250
250
250
Avg.
emission
rotp
(ug/min)
61
120
230
320
450
640
Note: VT 1 - VT 6 were conducted by VT.
2.3 Determining mass-transfer properties of selected polymers
The key VOC mass-transfer parameters of a given polymeric material include the
diffusion coefficient of the VOC in the material (D) and the partition coefficient of the
VOC between the material and air (K) (Cox et al, 2010; Xiong et al, 201 la; Xiong et al,
20lib). To determine D and K, a microbalance sorption/desorption method was
-------
MATERIALS AND METHODS
employed (Cox et al, 2001). As shown in Figure 2(a), the mass of a polymer film sample
was continuously measured using a high-resolution (0.1 jig) dynamic recording
microbalance (Thermo Cahn D-200, Thermo Fisher Scientific, Waltham, MA). During
the sorption test, as air containing a known concentration of formaldehyde was passed
across the film, formaldehyde sorbed into the material and the mass gain of the film was
recorded, generating a sorption curve. Once the film had reached sorption equilibrium
with the formaldehyde-containing gas stream, a desorption curve was created by passing
clean air across the film while again using the microbalance to monitor formaldehyde
mass loss. When Fickian diffusion controls the sorption and desorption process, D can be
determined by fitting a Fickian diffusion model to the sorption and desorption curves.
Under the experimental configuration, the mass change caused by Fickian diffusion of
formaldehyde inside the film is given by (Crank, 1975):
8
">
where Mt is the total formaldehyde mass that has entered or left the film via diffusion in
time t, Moo is the formaldehyde mass in the film when partition equilibrium is reached
between the film and the air, and H is the diffusion pass length. Furthermore, K can be
derived by dividing M^ by the volume of the film sample and the gas-phase
formaldehyde concentration. The selected films and parameters are summarized in Table
2.
Table 2. Determining mass-transfer properties of selected polymer by microbalance
sorption/desorption test.
Test
ID
VT7
VT8
VT9
Selected
polymer
PMP
PC
PC
Film dimensions
(cm x cm x cm)
3.6x3.6x0.025
3.6x3.6x0.051
3.6x3.6x0.025
Test
duration
(h)
840
410
340
Avg. flow
rate
(mL/min)
250
250
250
Avg. gas-phase
formaldehyde
concentration
(g/m3)
1.70
1.70
1.70
Avg.
emission
rate
(ug/min)
430
430
430
Note: VT 7 - VT 9 were conducted by VT.
2.4 Loading the identified polymer substrate with formaldehyde
After a suitable polymer substrate was identified, reference materials were created by
loading precisely-cut film samples with formaldehyde. As shown in Figure 2(b),
formaldehyde was infused into the films by passing air containing formaldehyde through
a loading vessel containing several films and allowing material-phase/gas-phase sorption
equilibrium to be reached. The effluent from the loading vessel was passed across an
additional film installed on the microbalance so that formaldehyde mass gain could be
continuously monitored during the loading process. Because the film on the
microbalance was subject to the same mass-transfer process as the films in the loading
vessel, the microbalance data were used to determine when material-phase/gas-phase
equilibrium was reached. Furthermore, the material-phase concentration in the loaded
-------
MATERIALS AND METHODS
films, Co, can be obtained from the microbalance data, by dividing the final measured
mass of formaldehyde infused into the film by the film sample volume.
Microbalance
Microbalance
Exhaust
Loading Vessel
i
Clean air/'airwith
forni al dehvde
Inlet
(a) (b)
Figure 2. (a) Microbalance sorption/desorption test: air with a constant formaldehyde
concentration is swept across the film for the sorption test and clean air is swept across
the sample for the desorption test; (b) the microbalance and loading vessel system.
Table 3. Loading the identified polymer substrate with formaldehyde.
Avg.
Test Loading batch Material Adsorption Avg. temp. RH Avg. flow rate gas-phase
ID ID (quantity) duration (day) (°C) (%) (mL/min) concentration
(g/m3)
VT 10
VT 11
VT 12
Batch 1
Batch 2
Batch 3
PC (3)
PC (11)
PC (9)
5
6
5
25
25
25
0
0
0
250
250
250
0.90
0.90
0.91
Note: VT 10 - VT 12 were conducted by VT.
2.5 Measuring formaldehyde emissions from pre-loaded films in small chambers
After material-phase/gas-phase absorption equilibrium had been reached in the loading
vessel, films were quickly removed from the loading vessel, wrapped in aluminium foil,
sealed in zip-loc bags, and placed in insulated containers that were then packed with dry
ice. The containers were shipped via overnight mail to EPA for emissions testing. Once
received, the films were retained in the original package and stored at -12 °C prior to
being tested in small chambers. Formaldehyde emissions were measured by the EPA at
23 °C using a 53-L stainless steel chamber with an air change rate of 1 h"1, following the
guidelines of ASTM International Standard Guide for Small-scale Environmental
Chamber Determinations of Organic Emissions from Indoor Materials/Products (ASTM
Standard D5116-2010) (ASTM 2010). Emissions measurement was also conducted by
another participating laboratory (Lab A) at 25 °C using a 21 L stainless steel chamber
with an air exchange rate of 2.9h"\ During chamber testing both sides of the film were
fully exposed to the chamber air using a custom fabricated sample holder. The chamber
10
-------
MATERIALS AND METHODS
air was sampled at appropriate time intervals to measure the gas-phase formaldehyde
concentration development in the chamber, according to EPA standard method
Determination of Formaldehyde and Other Aldehydes in Indoor Air Using a Solid
Adsorbent Cartridge (EPA method IP-6A) (US EPA, 1990). Briefly, the chamber air was
pulled through a cartridge containing silica gel coated with 2,4-dinitrophenylhydrazine
(DNPH) so that the gas-phase formaldehyde was collected on the cartridge by forming
hydrazones with DNPH. After sampling, the cartridge was eluted with acetonitrile to
extract the hydrazones, which were then analyzed by high performance liquid
chromatography (HPLC) and ultraviolet (UV) spectroscopy. Small chamber test
conditions are summarized in Table 4.
Table 4. Measuring formaldehyde emissions from pre-loaded films in small chambers.
Test ID
AT 1
AT 2
AT 3
AT 4
AT 5
ET1
ET2
ET3
ET4
ET5
ET6
ET7
ET8
ET9
ET10
ET11
ET12
ET13
ET14
ET15
ET16
ET17
ET18
Film dimension
(cm x cm x cm)
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
Test
duration
(h)
140
140
140
140
140
130
100
53
150
150
150
150
150
150
150
150
170
170
1700
150
150
150
150
Inlet RH
±STD
50
50
50
50
50
50±0.03
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
51±0.35
50±0.70
BDL
71±0.91
RH
±STD
50
50
50
50
50
47±1.2
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
50±0.34
49±0.65
BDL
69±1.4
Temp.
(°C)± ,
STD ^
25
25
25
25
25
24±0.08
24±0.04
24±0.09
25±0.03
25±0.02
25±0.03
25±0.02
25±0.17
25±0.22
25±0.02
25±0.02
25±0.02
25±0.03
25±0.17
25±0.03
25±0.03
25±0.04
25±0.05
ACH
h'1) ±STD
2.9
2.9
2.9
2.9
2.9
1.1±0.03
l.liO.Ol
l.liO.Ol
.0±0.007
.0±0.008
.0±0.005
.0±0.003
.0±0.007
.0±0.008
.OiO.OOl
.0±0.002
.0±0.002
.0±0.002
l.OiO.OO
.0±0.002
.0±0.003
.0±0.004
.0±0.005
Chamber
volume
(L)
21
21
21
21
21
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Notes: AT 1 - AT 5 were conducted by Lab A; ET 1 - ET 18 were conducted by EPA.
2.6 Predicting formaldehyde emissions from pre-loaded films
Figure 3 shows the mechanisms governing the emission of formaldehyde from a
homogeneous solid material slab in a test chamber. If it is assumed that the external
convective mass-transfer rate is much faster than internal diffusion, then a simple
fundamental model can predict the emission profile. The following derivation applies to
emissions from a single-sided source, however the solution can easily be adjusted and
applied to a double-sided source by considering the source to be two single-sided sources.
11
-------
MATERIALS AND METHODS
Yin= 0, Q
x = L
x=0
V
y(t)
y(t), Q
Figure 3. Schematic representation of a formaldehyde-containing source in a test
chamber showing mechanisms controlling the emission rate.
The transient diffusion equation in the material slab is given by Pick's second law:
ac a2c
where t is time, x is the distance from the base of the slab, and C is the material -phase
concentration of formaldehyde as a function of t and x. The initial condition assumes a
uniform material-phase concentration of formaldehyde in the slab, Co. The boundary
condition at the base of the slab assumes there is no mass flux through the bottom
surface. The boundary condition at the exposed surface is imposed via a mass balance on
formaldehyde in the chamber air, or
dy.v=Q.y. _
, ^< J in
dt
-,
dx
-Q-y
(3)
x=L
where y;n and y are the gas-phase formaldehyde concentration in the influent air and the
bulk chamber air respectively, Q is the volumetric air flow rate, V is the well-mixed
chamber volume, A is the exposed surface area of the slab, and L is the thickness of the
slab. A linear and instantaneously reversible equilibrium relationship is assumed
between the slab surface and the chamber air, or
K = C T/y
x=L/ J
(4)
Equation (4) implies that the convective mass-transfer resistance through the boundary
layer at the exposed surface is negligible compared to internal diffusion, which is
common for compounds with small values of D (Cox et al, 2010). Assuming y;n and the
initial chamber concentration are zero, an analytical solution to these equations was given
by Little et al. (1994):
(5)
(6)
where
h = Q/(ADK)
12
-------
MATERIALS AND METHODS
k = V/(AK) (7)
and the qns are the positive roots of
qntan(qnL) = h-kqJ (8)
When key model parameters, D, K and Co are determined as described above and other
parameters (V, Q, L, and A) are obtained from the chamber test configuration, C can be
obtained using Equation (5) and y can be simply calculated using Equation (4). In
addition, when the model is used to predict emissions from a pre-loaded film with both
sides exposed to the chamber air, L should be half of the film thickness and A should be
the total surface area of both sides. The Matlab code for emissions models are attached
in Appendix F, with model parameters summarized in Table 5.
Table 5. Predicting formaldehyde emissions from pre-loaded films.
Test ID
VT 13
VT 14
VT 15
VT 16
Co
(g/m3) ±STD
190±27
160±8.9
170±19
170±19
Film dimension
(cm x cm xcm)
10x10x0.025
10x10x0.025
10x10x0.025
8.5x8.5x0.025
Chamber
volume
(m3)
0.053
0.053
0.053
0.021
Avg. flow
rate (m3/h)
0.053
0.053
0.053
0.061
Duration
(h)
144
144
144
144
Note: VT 13 - VT 16 were conducted by VT.
2.7 Task allocation
The work described in this report was created through the combined efforts of three
organizations, EPA, Lab A, and VT. This specific work completed by each organization
is summarized in Table 6.
13
-------
MATERIALS AND METHODS
Table 6. Task allocation
_. . . Experiment RH
Organization ^ /0/x
& content (%)
EPA Small-scale 50
chamber 10
preliminary test 0
0-week shelf-life 0
test 0
2-week shelf-life 0
test 0
4-week shelf-life 0
test 0
6-week shelf-life 0
test 0
10-week shelf- 0
life test 0
Packaging test 0
50
ILS EPA
50
50
ILS Lab A 50
50
50
0
50
RHtest
70
VT 0
0
Generating gas- „
nhase
JJIICIO^ _
formaldehyde
0
Microbalance 0
sorption/ 0
Desorption test 0
,. r,1 0
Loading films „
for chamber test
. . 0
Emissions „
profiles model
prediction
Notes: aFilms B1F1, B1F2 and B1F3;
bFilms B2FA1, B2FA2, B2FA3,
B2FB3, B2FB4;
cFilms B3FA1, B3FA2, B3FB1,
Test ID
ET1
ET2
ET3
ET4
ET5
ET6
ET7
ET8
ET9
ET10
ET11
ET12
ET13
ET14
ET15
ET16
AT 1
AT 2
AT 3
AT 4
AT 5
ET17
ET15
ET16
ET18
VT1
VT2
VT3
VT4
VT5
VT6
VT7
VT8
VT9
VT 10
VT11
VT12
VT13e
VT14f
VT158
VT16h
summary.
Film ID
B1F1
B1F2
B1F3
B2FA1
B2FA2
B2FA4
B2FA5
B2FA6
B2FA7
B2FB1
B2FB2
B2FB3
B2FB4
B2FA3
B3FA1
B3FC2
B3FB1
B3FB2
B3FB3
B3FB4
B3FB5
B3FA2
B3FA1
B3FC2
B3FC1
None
None
None
None
None
None
None
None
None
Notea
Noteb
Notec
Note1
Note1
Notek
Note1
Film dimension
(cm x cm x cm)
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
8.5x8.5x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
10x10x0.025
3.6x3.6x0.025
3.6x3.6x0.051
3.6x3.6x0.025
10x10x0.025
10x10x0.025
10xlOx0.025d
10x10x0.025
10x10x0.025
10x10x0.025
8.5x8.5x0.025
B2FA4, B2FA5, B2FA6, B2FA7,B2FB1
B2FB2, B3FB3, B3FB4, B3FB5, B3FC1
Loading
batch
1
1
1
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
None
None
None
None
None
None
None
None
None
1
2
3
1
2
3
3
, B2FB2,
, B3FC2;
14
-------
MATERIALS AND METHODS
d!0 cm x 10 cm x 0.025 cm film size for EPA, 8.5 cm x 8.5 cm x 0.025 cm film
size for Lab A;
eVT 13 is the emission model prediction for films loading in Batch 1;
fVT 14 is the emission model prediction for films loading in Batch 2;
gVT 15 is the emission model prediction for films loading in Batch 3 with the size
of 10 cm x 10 cm x 0.025 cm (EPA);
hVT 15 is the emission model prediction for films loading in Batch 3 with the size
of 8.5 cm x 8.5 cm x 0.025 cm (Lab A);
'Films B1F1, B1F2 and B1F3;
gFilms B2FA1, B2FA2, B2FA3, B2FA4, B2FA5, B2FA6, B2FA7,B2FB1, B2FB2,
B2FB3, B2FB4;
kFilms B3FA1, B3FA2, B3FC1, B3FC2;
'Films B3FB1, B2FB2, B3FB3, B3FB4, B3FB5;
15
-------
RESULTS AND DISCUSSION
3 RESULTS AND DISCUSSION
3.1 Generating gas-phase formaldehyde
By weighing the diffusion vial containing paraformaldehyde at certain time intervals, the
formaldehyde release rate could be determined. Figure 4 shows the measured weight
change of diffusion vials (Vial 1, 2 and 3) containing paraformaldehyde maintained at
different temperatures (85, 95, 100, and 105 °C). It is found that the weight decrease in
all cases followed a linear pattern: when linear regression is performed between weight
and time in each case, coefficients of determination (R2) for all six cases are larger than
0.999. Therefore, the formaldehyde release rate in each case, derived from the slope of
the corresponding linear regression, was constant over time. The difference in diffusion
vials (Vial 1, 2, and 3) is that Vial 1 has the longest diffusion path, Vial 3 has the shortest
diffusion path, and Vial 2 lies somewhere between the two extremes. As shown in Figure
4, at a fixed temperature, the formaldehyde release rate and thus the concentration in the
generated gas stream increases when the diffusion path length decreases (from Vial 1 to
3). Moreover, Figure 4 also shows the formaldehyde release rate from a diffusion vial
increases with temperature, which is due to the faster depolymerization rate of solid
paraformaldehyde at higher temperatures.
0.0
3-0.5 -
ill (S5°C)
al 1 (95 °C)
a!2(95°C)
al 3 (95 °C)
a!3(100°C)
1ial3(105°C)
6 9
Time (day)
Figure 4. Measured weight decrease of diffusion vials over time: marker color indicates
temperature and marker shape indicates emission vials with different diffusion path
length.
16
-------
RESULTS AND DISCUSSION
y = 0.999x
R2 = 0.998
0
0 0.5 1 1.5 2 2.5 3
Concentration determined from
diffusion vial's weight change (g m3)
Figure 5. Comparison of directly measured formaldehyde concentration with that
calculated from diffusion vial weight change.
Figure 5 compares the directly measured formaldehyde concentration in the generated
gas stream using visible absorption spectrometry with that determined from the
corresponding diffusion vial's weight change. It is found that the gas-phase
formaldehyde concentrations obtained by these two approaches match well (a paired t-
test yields a P value of 0.66), suggesting that either approach is able to determine the
concentration accurately. Overall, the results in Figure 4 and 5 prove that a gas stream
with a constant formaldehyde concentration at different levels can be achieved using the
formaldehyde gas generating system.
3.2 Evaluating mass-transfer properties of selected polymer films
Considering the criteria in Figure 1, one PMP film with a thickness of 0.025 cm and two
different PC films with thicknesses of 0.025 and 0.051 cm, respectively, were chosen as
candidate substrates. To determine their mass-transfer properties, small film samples
were cut from the original large sheets for the microbalance sorption/desorption tests.
The measured mass gain of a 0.025 cm thick PMP sample (3.6 cm x 3.6 cm) during a
sorption/desorption test is shown as blue circles in Figure 6(a). The gas-phase
formaldehyde concentration for the sorption test was 1.70 g/m3.
17
-------
RESULTS AND DISCUSSION
0.07
0.07
200 400
Time (h)
(a)
600
800
Ma^s gain due to pojj*merization
model
100
200 300 400
Time(h)
(b)
500 600
Figure 6. Sorption/desorption data and analysis for the 0.025 cm thick PMP.
If simple Fickian diffusion governs the sorption process, the mass would level off when
reaching sorption equilibrium, as demonstrated for toluene/PMP and phenol/vinyl
flooring (Cox et al., 2001; Cox et al, 2010). However, in contrast to what would be
expected for simple Fickian diffusion, the mass of the PMP film continued to increase
indefinitely as shown in Figure 6(a). Meanwhile, the mass desorbed from the PMP film
during the desorption period (-0.025 mg) was less than the mass sorbed by the film
during the sorption period (-0.07 mg). A possible explanation is that formaldehyde
adsorption and polymerization or other chemical reaction occurred on the film surface,
with overall mass gain during the sorption cycle due to both Fickian diffusion inside the
film (absorption) and polymerized formaldehyde accumulating on the film surface. It has
been shown that surfaces such as glass and stainless steel adsorb formaldehyde, with the
amount being dependent on the nature of the surface, relative humidity, gas-phase
formaldehyde concentration, and exposure time (Braswell et al., 1970). As demonstrated
in the work of Braswell et al. (1970), surface polymerization may occur even at very low
humidity levels because trace amounts of water (and many other nucleophilic surface
contaminants) induce polymerization, building polyoxymethylene on surfaces (Walker,
1975). Moreover, the linear mass increase possibly due to polymerization implies that
the surface polymerization rate would be relatively constant throughout the sorption
period. As shown in Figure 6(b), if surface polymerization was occurring, the rate of
mass gain due to polymerization would be linear and could be estimated from the slope
of the overall mass gain (blue line) at later times (after 200 hours). Assuming that the
rate of surface polymerization is constant and began at t=0, the linear mass gain due to
polymerization (purple line) could be subtracted from the total mass gain (blue line),
yielding the net mass gain due to diffusion (green line) which can be described by the
Fickian diffusion model given in Equation (1). Assuming that polymerization at the film
surface and Fickian diffusion inside the film are independent and using the method
described above, D and K can be determined to be (3.5±0.2)xlO"14 m2/s and 40±5,
respectively. K of formaldehyde between PMP and air is much smaller than that of
toluene, which is 500±30 (Howard-Reed et al., 2011), indicating that formaldehyde has
rather low solubility in PMP. During the desorption period, the mass decrease should be
due to Fickian diffusion from within the film and possibly depolymerization of
18
-------
RESULTS AND DISCUSSION
polyoxymethylene from the film surface. However, as shown in Figure 6(a), it is found
that the Fickian diffusion model (Equation (1)), with D and K obtained from the sorption
test, predicts the overall mass decrease well.
to 0.08 -
i
E 0.06 H
u
£
0.04 -
0.02
o -:
-*/••'"
X
Linear
region
Sorption
^ Fickian diffusion
\\ __ model
Desorption
0.1
Total mass gain _
_^r-- — **
^ ~~
""Fickian diffusion model
0 50 100 150 200 250 300 350
Time (h)
(a)
Mass gam due to diffusion
50 100 150
Time (h)
(b)
200
Figure 7. Sorption/desorption data and analysis for the 0.025 cm thick PC.
Figure 7 and 8 show the microbalance sorption/desorption results for a 0.025 cm thick PC
sample (3.6 cm x 3.6 cm) and a 0.051 cm thick PC sample (3.6 cm x 3.6 cm). The gas-
phase formaldehyde concentration for these sorption tests was 0.86 g/m3. The mass
increase of these two films during the sorption period followed a trend similar to PMP, as
a result of the combined effect of polymerization or other chemical reactions at the film
surface and Fickian diffusion inside the film. Based on the net mass gain due to diffusion
during the sorption period (green lines in Figure 7(b) and 8(b)), D and K at 25 °C, 0% RH
for the two PC films can be determined. D and K were found to be (1.9±0.3)xlO"13 m2/s
and 230±40 respectively for the 0.025 cm thick PC, and (3.9±0.2)xlO"13 m2/s and 170±20
respectively for the 0.051 cm thick PC. Therefore, these two PC films are slightly
different in nature, but both of them have much greater formaldehyde solubility than
PMP. In addition, the desorption curves of these two film samples can also be well
predicted using the Fickian diffusion model with D and K obtained from the sorption
tests.
In addition to Figures 6-8, replicate sorption/desorption tests were also performed and are
summarized in Appendix A.
19
-------
RESULTS AND DISCUSSION
0.18
I 0.15
0.16
0 50 100 150 200 250 300 350
Time(h)
(a)
Mass gain due to diffusion
ass gain due to polymerization
0 30 60 90 120 150
Time (h)
(b)
Figure 8. Sorption/desorption data and analysis for the 0.051 cm thick PC.
In summary, the sorption/desorption tests of the polymeric materials suggest that the total
mass uptake during the sorption period was a combined result of constant-rate
polymerization or irreversible chemical reactions involving formaldehyde at the film
surface and Fickian diffusion inside the film. Because diffusion appears to dominate the
desorptive mass-transfer process, the mass of formaldehyde emitted from the film can be
predicted solely based on Fickian diffusion. Finally, the 0.025 cm thick PC was selected
for use as a formaldehyde emissions reference material for this project due to the higher
K value and the reduced time to reach gas-phase/solid phase equilibrium.
0. 14
Mass gain due to polymerization 1
Mass gain due to polymerization 2
60 90 120 150 180
Time (s)
Figure 9. Comparison of duplicated analysis for 0.025 cm PC.
Figure 9 compares duplicate sorption tests of 0.025 cm PC films under similar conditions
suggesting that the irreversibly-sorbed formaldehyde fractions are consistent.
20
-------
RESULTS AND DISCUSSION
3.3 Validating overall approach by small-scale chamber testing
3.3.1 Preliminary approach by small-scale chamber testing.
Three 10 cm x 10 cm films were cut from the 0.025 cm thick PC sheet and loaded in
Batch 1 using a gas stream containing -0.90 g/m3 formaldehyde (VT 10). The duration
of the loading process was five days, long enough for the films to reach absorption
equilibrium with the gas stream. Using the same analysis as in Figure 7(b), the net
uptake of formaldehyde into the films through diffusion can be obtained and the
formaldehyde concentration in the films was determined to be 190±27 g/m3. The loading
process data and Co calculations are summarized in Appendix B. The films were then
shipped to EPA and tested in small-scale chambers. Figure 10 shows the chamber test
results of the three pre-loaded films as well as the test conditions (shelf-life and humidity
level). The emission profiles are very similar, although the age effect and different
humidity conditions may explain some of the difference. The small-scale chamber data
are attached in Table Cl, Table C2 and Table C3 in Appendix C.
Because the desorbed (emitted) mass of formaldehyde from the PC films are primarily
due to diffusion, the emission model based on diffusion introduced earlier is applicable
for formaldehyde with D and K values determined to be (1.9±0.3)xlO"13 m2/s and 230±40
(25 °C, 0% RH) from the net mass gain due to diffusion during the sorption period. Co
has been determined based on the net uptake of formaldehyde into pre-loaded films
through diffusion, which is 190±27 g/m3. Therefore, Equation (4) and (5) can be used to
predict the formaldehyde concentration profile in the chamber air during the emission
tests.
To further estimate the uncertainties in model predicted concentrations associated with
the uncertainties of D, K and Co, the Monte Carlo method (Kim et al, 2004) was
employed. 10,000 repeated model simulations were carried out with D, K and Co
randomly sampled from their probability distributions, while the other parameters (L, A,
Q, and V) were fixed for each individual run. The results of the 10,000 model
predictions were then pooled to assess the expected variation in y as a function of time.
Figure 10 shows the model prediction, with the black solid line indicating the mean of the
transient gas-phase formaldehyde concentration in the chamber air and the shaded area
indicting the range of mean ± one standard deviation of the transient gas-phase
concentration. Compared with the measured results, the model overestimates emissions
during the first 20 hours. Possible reasons include: (1) formaldehyde escaped from the
pre-loaded films during packaging, shipping, and storage (shelf-life) period, especially
when they were removed from the loading vessel and were wrapped; and, (2) the
chamber tests were carried out at 24 °C while the D and K used in the model prediction
were obtained from sorption/desorption tests performed at 25 °C. Higher temperature
will tend to increase D and reduce K, thus accelerating emissions (Deng et al., 2009;
Zhang et al., 2007). The longer-term predicted concentrations nevertheless compare
reasonably well with the measured results.
21
-------
RESULTS AND DISCUSSION
1400
0>
o
C
o
o
0)
t/3
03
^
PL,
CO
£
O
1200-
1000-
800-
600-
400-
200-
o-
Model prediction
ET 1: 1 day old, 50% RH
ET 2: 9 days old, 10% RH
ET 3: 15 days old, 0% RH
40 60 80 100
OH
0
20 40 60 80 100
Time (h)
Figure 10. Comparison of the measured and predicted formaldehyde emission profiles.
3.3.2 Inter-laboratory study
To validate the formaldehyde reference material, an inter-laboratory study (ILS) was
conducted with EPA laboratories and Lab A. Six 8.5 cm x 8.5 cm PC films and six 10
cm x 10 cm PC films were loaded in Batch 3 for six days (VT 12). The loading process
data are summarized in Figure B3 in Appendix B. Based on microbalance data the
formaldehyde concentration in the films was determined to be 170±19 g/m3. Five of the
8.5 cm x 8.5 cm PC films were used by Lab A and two of the 10 cm x 10 cm PC films
were used by EPA for the ILS. The ILS chamber tests conducted by EPA and Lab A
were carried out at 50% RH and a temperature of 25 °C.
O
^
~5b
^
+j
03
£
O
C
O
0
0>
03
'En
1Z,UU
1000-
800-
600-
400-
200-
o-
!
Model prediction
Q ET 15
o ET 16
i
9 9i
\B
V^o
^~~~~—_ ° r>
0 20 40 60 80 100 120 140 160
Time (h)
(a)
22
-------
RESULTS AND DISCUSSION
^ 1200-
| 1000-
o 800-
g 600-
§ 400-
200-
o-
03
CO
03
<
Model
° AT 1
o AT 2
A AT 3
v AT 4
& 0 AT 5
i
I
v>
, ^%. £ «?
prediction
o
0 20 40 60 80 100 120 140 160
Time (h)
(b)
Figure 11. Comparison of measured and model predicted emission profiles in the ILS.
Direct comparison of the ILS test results generated by the two laboratories is not possible
because they employed different chamber characteristics and air flow rates (Table 4 and
Table 6) and have different emission profiles. However, the test results of each
laboratory can be compared to the emissions model predictions based on the appropriate
chamber characteristics. Figure 11 shows the measured formaldehyde concentrations and
model predicted concentration profiles for each laboratory (raw data are summarized in
Table C17-Table C19 in Appendix C), with Figure 11 (a) comparing the test results of
EPA to the model prediction and Figure ll(b) comparing the Lab A test results to the
model prediction. The measured data and the model prediction of both EPA and Lab A
show some deviation, although the deviations are similar in some respects. With respect
to both EPA and Lab A, measured formaldehyde concentrations during the first 10 hours
of testing are lower than model predictions, while measured formaldehyde concentrations
after that period tend to be higher than the model predictions. The longer-term predicted
concentrations nevertheless compare reasonably well with the measured results.
The lower concentrations during the early period could be explained by some loss of
formaldehyde from the pre-loaded films during packaging, shipping, and storage, while
higher concentrations during the later period might be explained by some formaldehyde
depolymerization during the chamber test. It is also possible that the chamber conditions
themselves do not correspond with the model assumptions. The model assumes that the
air in the chamber is well mixed, and these conditions may not have been obtained by
Lab A in reality. The overall deviation might also be explained by some impact of RH on
formaldehyde emission profiles. The ILS chamber tests were conducted at 50% RH
while the D and K values used in the model were obtained at 0% RH conditions. It is
well known that RH affects formaldehyde emissions from composite wood products
manufactured with UF (Parthasarathy, 2011). UF polymerization is a water-producing
condensation reaction, and ambient moisture will naturally displace the reaction
23
-------
RESULTS AND DISCUSSION
equilibrium with ambient moisture effectively reversing the polymerization reaction to
release formaldehyde. In contrast, the potential effect of RH on formaldehyde emissions
from PC films would be different. Sorbed water can catalyze formaldehyde
polymerization/depolymerization, but reaction rates will vary with many possible acidic
or basic species (Brown, 1967). Finally, while the test results for the two laboratories
differed with respect to the model predictions, the tests results of each laboratory are
quite consistent.
3.3.3 Mass balance analysis
To further validate the overall approach, a formaldehyde mass balance analysis was
conducted using data obtained during selected chamber tests. The formaldehyde
emission rate was estimated using the chamber flow rate and the measured gas-phase
concentration data. The emission rate was then integrated over the duration of each
chamber test using a simple trapezoidal method of numerical integration. The emissions
model was used to predict the mass emitted from each PC film during the same duration
of the respective chamber test. The full integration results are contained in Appendix D
and summarized in Table 7.
Table 7. Formaldehyde mass balance analysis.
Film ID
B2FA1
B2FA2
B3FA2
B3FC1
B3FC2
B3FA1
B3FB2
B3FB4
B3FB5
Model
prediction ID
VT 14
VT 14
VT 15
VT 15
VT 15
VT 15
VT 16
VT 16
VT 16
Measured
emissions (|o,g)
570
580
570
620
580
615
340
290
330
Predicted
emissions (|o,g)
410
410
430
430
430
430
310
310
310
Recovery1
139%
141%
133%
144%
135%
143%
110%
94%
110%
Recovery of adsorbed formaldehyde fraction only.
As shown in Table 7, measured formaldehyde mass emitted from each PC film is
generally greater than the mass of formaldehyde monomer estimated to have diffused into
the film during initial loading. This could be explained by polymerized formaldehyde on
the film surface depolymerizing during chamber testing. However, according to previous
tests by EPA, depolymerization rate is usually positively correlated with temperature.
Further study will be necessary to more accurately describe formaldehyde polymerization
and depolymerization chemistry.
3.4 Measuring effectiveness of foil packaging and storage materials and methods
To investigate the effectiveness of the packing material and storage methods, a series of
small-scale chamber emissions tests were conducted. Several 10 cm x 10 cm films were
cut from a 0.025 cm thick PC sheet and loaded in Batch 2 for five days using a gas stream
containing -0.91 g/m3 formaldehyde (VT 11). The loading process data are summarized
in Figure B2 in Appendix B. The formaldehyde concentration in the films was
24
-------
RESULTS AND DISCUSSION
determined to be 160±8.9 g/m
dry ice, and shipped to EPA.
All films were packaged in aluminium foil, packed in
3.4.1 Packaging material evaluation
To investigate the effectiveness of the packing material, small-scale chamber emissions
testing was conducted using a film tightly wrapped in the original aluminium foil
packaging. The test was conducted at 0% RH with a chamber temperature of 25 °C.
Figure 12 shows the measured formaldehyde emission profiles with the film placed in the
chamber, either with or without the aluminum foil wrapping. The gas-phase
formaldehyde concentration without the foil wrapping is obviously much higher than that
with the foil wrapping. What the results clearly show is that films wrapped in foil do lose
formaldehyde at room temperature, although the foil wrapping reduces the formaldehyde
loss rate.
800-
600-
o
1
-M
8 400
a
o
o
1 200
ji
O
ET 4, 5-. Emission with foil wrapping
ET 14: Emission without foil wrapping
400 800 1200
Time (h)
1600
Figure 12. Comparison of the measured formaldehyde emission profiles in chamber tests
using films with and without foil wrapping.
3.4.2 Shelf-life evaluation
To study formaldehyde emission profiles with respect to film storage duration, a series of
chamber tests were conducted using PC films that had been stored for time periods of
between 0 and 10 weeks. All films used for this evaluation were simultaneously loaded
with formaldehyde in Batch 2. All of the tests were conducted in duplicate.
25
-------
RESULTS AND DISCUSSION
concentration
c3
43
SFL|
£
O
i^uu -
1000-
800-
600-
400-
200-
o-
3 Week
Week
& 9 Week
& 9 Week
E» • Week
*K 1 DO
m* 1UU
s» \
!l m
^f 25_ ^
T Q ^^ :
. prediction
0: ET 4; ET 5
2: ET 6; ET 7
4: ET 8; ET 9
6: ET 10; ET 11
10: ET 12; ET 13
V <>5 20 40 60 80 100 120
0 20 40 60 80 100 120 140 160
Time (h)
Figure 13. Comparison of the measured formaldehyde emission profiles in chamber tests
of different shelf-life and the model prediction.
1200
Model prediction
• Week 0: ET 4; ET 5
1200
0 20 40 60 80 100 120 140 160
Time (h)
(a)
Model prediction
Week 2: ET 6; ET
20 40 60 80 100 120 140 160
Time (h)
(b)
1200
Model prediction
• Week 4: ET 8; ET 9
1200
0 20 40 60 80 100 120 140 160
Time (h)
(c)
1000-
Model prediction
Week 6: ET 10; ET 11
20 40 60 80 100 120 140 160
Time (h)
(d)
26
-------
RESULTS AND DISCUSSION
1000-
800-
600-
400-
200-
v> Week 10: ET 12; ET 13
>
20 40 60 80 100 120 140 160
Time (h)
(e)
Figure 14. Comparison of model-predicted and measured formaldehyde emission
profiles from films stored for different durations.
Figure 13 summarizes shelf-life test results compared to the model prediction, with the
points showing the average value of duplicate tests and the error bars showing the
deviation of duplicate tests (raw data is attached in Table C4-Table C13 in Appendix C).
All gas-phase formaldehyde concentration measurements fit the model well, however
emissions from films with shorter storage duration fit the model even better during the
first 20 hours of testing. The grey shaded area provides an expression of the uncertainty
of the mean at ± one standard deviation associated with estimates of D, K and Co using
the Monte Carlo method (Kim et al, 2004). D, K and C0are respectively (1.9±0.3)xlO"13
m2/s, 230±40 and 160±8.9 g/m3 (25 °C, 0% RH). Figure 14 shows the results of each
shelf-life test in a separate figure where the effect of storage duration can be seen more
clearly.
1100
.1
GO
o
900-
I 700-
o
03
8 500
O
-Q-Peak gas-phase concentration
0 2 4 6 8 10
Shelf-life (week)
Figure 15. Peak gas phase concentration values of different shelf-life chamber tests.
27
-------
RESULTS AND DISCUSSION
Figure 15 shows that the peak values of gas phase formaldehyde concentration decrease
with longer shelf life. The results suggest that some formaldehyde is lost from films even
during storage at -12 °C. The detailed view in Figure 13 also shows that the gas-phase
formaldehyde concentrations after the first 40 hours were somewhat higher than the
model predictions. A possible reason could be slow depolymerization of formaldehyde
from the film surface during chamber testing, or the underestimation of Co caused by the
deficiencies of the model. Despite these small deviations, the results presented in Figure
13 show very good agreement with the emissions model, particularly for films stored for
shorter durations.
3.5 Measuring emissions at different relative humidity (RH) levels
To further investigate the impact of humidity on formaldehyde emission profiles, four 10
cm x 10 cm films loaded in Batch 3 for six days (VT 12) were employed for an emission
test at different humidity levels (0% RH, 50% RH and 70% RH). Based on microbalance
data the formaldehyde concentration in the films was determined to be 170±19 g/m3.
Figure 16 shows the test results and model prediction using a D value of (1.9±0.3)xlO"13
m2/s and a K value of 230±40. All of the RH tests were conducted at a chamber
temperature of 25°C. The RH test raw data is attached in Table C15 and Table C16 in
Appendix C.
•O
"M
^i
C
•,§
Ja
g
O
O
w
03
'a
CO
03
O
l^UU
1000-
800-
600-
400-
200-
o-
1
, Mp..^
| 1V1U L
n ET
a. ° ET
W A ET
Vt T^T1
Mi, O ET
iia
M^
0,^
\
\Ski
^ fi^^^^in
n 9n AC\ «n «
lei
17
15
16
18
n
prediction
: 0% RH
: 50% RH
: 50% RH
: 70% RH
^S^V n ^\
mn 1 9n 1 AH 1^
Time (h)
Figure 16. Emission profiles at different RH levels.
As with earlier small-scale chamber studies that indicated that humidity had an effect on
some VOC emission characteristics for some materials (Wolkoff, 1998; Fang, 1999), this
study suggests that formaldehyde emission profiles from the PC films may also be
affected by humidity, as shown in Figure 16. The deviation increases somewhat as
humidity increases. The test results generated from duplicate 50% RH tests showed good
consistency, indicating that formaldehyde has the same emission profile at a specific RH
level. Humidity may impact either the D or K values through a "plasticizing" effect of
28
-------
RESULTS AND DISCUSSION
the PC film material or with water molecules potentially competing for sorption sites. A
recent study also shows that higher RH tends to increase both D and K values (Xu et al,
2012). A 35% increase in RH could increase formaldehyde emissions by a factor of 1.8 -
2.6 (Parthasarathy, 2011). The results could also be explained by a combination of these
factors.
3.6 Statistical evaluation of measured and model-predicted concentrations
Differences between chamber-measured and model-predicted gas-phase formaldehyde
concentrations result from 1) errors associated with the measurement procedure, 2) model
construction and use, and 3) inaccuracies in model parameters. Two statistical analyses
were used to assess the degree of agreement between model predictions and chamber
measurements for the small-scale chamber testing conducted during this project. In the
first analysis, the correlation coefficient, r, was calculated for each data set. The
correlation coefficient is a measure of the strength of the relationship between measured
and predicted concentrations. The second analysis, normalized mean square error
(NMSE), is a measure of the magnitude of the prediction error relative to measured and
predicted concentrations. ASTM D5157-97 advises that an r value of 0.9 or greater and a
NMSE of 0.25 or lower generally indicate acceptable agreement between model-
predicted and chamber-measured gas-phase concentrations (ASTM 2008). The results of
the analysis are summarized in Table 8.
Table 8. Statistical analysis summary.
Test ID
ET 1
ET2
ET3
ET4
ET5
ET6
ET7
ET8
ET9
ET 10
ET 11
ET 12
ET 13
ET 15
ET 16
ET 17
ET 18
AT 1
AT 2
AT 3
AT 4
ATS
r
0.99
0.90
0.85
0.97
0.97
0.91
0.92
0.92
0.92
0.89
0.88
0.91
0.92
0.98
0.96
0.90
0.99
0.96
0.96
0.96
0.96
0.92
NMSE
1.0
1.1
1.4
0.058
0.061
0.20
0.17
0.20
0.17
0.29
0.52
0.23
0.19
0.15
0.34
0.27
0.17
3.0
0.9
2.1
0.93
0.76
29
-------
RESULTS AND DISCUSSION
The statistical analysis shows that the agreement between the model and chamber-
measured concentrations represented by r, is relatively strong for all tests indicating that
the model predictions compare well to the experimental observations. The prediction
error represented by NSME is within an acceptable range for most of the chamber tests
conducted in accordance with ASTM 5116-10. The root cause of deviation between
model-predicted and chamber-measured concentrations could be formaldehyde reactivity.
Polymerization, depolymerization and other formaldehyde reaction pathways affect
measurements of the model parameters D and K, as well as measurements of gas-phase
formaldehyde concentrations during chamber testing. The relatively large prediction
error associated with Tests ET 1 - ET 3 could be due to the length of time films from
Batch 1 were exposed to air, and consequential loss of formaldehyde, during the
packaging process at VT while preparing films for shipment. Employing experience
gained from packaging Batch 1 films, the exposure time between removal from the
loading chamber and packaging was reduced for Batch 2 and Batch 3 films. The NSME
for Tests ATI-ATS could be due to incomplete mixing in the chambers during the
measurement process, while model condition is that the chambers are well-mixed.
30
-------
QUALITY ASSURANCE AND QUALITY CONTROL
4 QUALITY ASSURANCE AND QUALITY CONTROL
4.1 VT quality assurance and quality control
Work on this project was performed in accordance with Quality Assurance Project Plan
- Developing a Reference Material for Formaldehyde Emissions Testing attached in
Appendix F. Six microbalance data sets were used to obtain estimates of the mean D and
K for formaldehyde in the 0.025 cm PC films. Each microbalance data set used to
determine D, K, and Co consisted of a minimum of 600 data points. All the PMP/PC film
samples were obtained from a single roll of additive-free film purchased directly from the
manufacturer. Flow rates were verified using a flow meter calibrated to a NIST-traceable
primary standard. The microbalance was calibrated before each test using standard
weights that had been verified using mass standards whose calibrations are NIST-
traceable.
4.2 EPA quality assurance and quality control
Quality assurance (QA) and quality control (QC) procedures were implemented in this
project by following guidelines and procedures detailed in the approved Category III
Quality Assurance Project Plan (QAPP), Indoor Source Emissions and Sink Effect Study
of Formaldehyde, Addendum 4 - Procedure for Evaluation of Reference Material for
Formaldehyde Emissions Testing using Small Environmental Chambers (Appendix G).
The Agilent 1200 High Performance Liquid Chromatography (HPLC) with a Diode
Array and Multiple Wavelength Detector (DAD) was calibrated using an external
standard method with formaldehyde in the range of 0.04-15 ug/mL. The Internal Audit
Program (IAP) was implemented to minimize the systematic errors. Chamber
environmental system components such as Gilibrator, temperature sensors, relative
humidity sensors, and mass flow controllers were calibrated annually in EPA's
Metrology Laboratory. Quality control samples, including background, field blank, and
duplicate, all met the data quality indicator goals for critical measurements listed in the
QAPP. On each day of analysis, at least one daily calibration check (DCC) sample was
analyzed to document the performance of the instrument. The recoveries met the
criterion of 85 to 115% recovery for acceptable performance of the HPLC instrument.
31
-------
CONCLUSIONS
5 CONCLUSIONS
This project investigated the feasibility of creating a reference material for use in
formaldehyde emissions testing by loading formaldehyde into a suitable polymer
substrate and predicting the emission rate from pre-loaded films.
Gas-phase formaldehyde can be controllably generated through depolymerization of
paraformaldehyde at elevated temperatures. Consistent carrier gas flow rate and
formaldehyde gas generator temperature produced a consistent gas-phase formaldehyde
concentration in the carrier gas. The gas-phase formaldehyde concentration could be
regulated by adjusting the gas generator temperature, the length of the diffusion vial, or
by adjusting the carrier gas flow rate.
During sorption/desorption testing of a PMP material and two types of PC material, it
was found that the formaldehyde sorption process is complicated due to simultaneous
Fickian diffusion inside the polymer and possible formaldehyde polymerization or other
irreversible chemical reactions on the polymer surface. However, Fickian diffusion
appears to dominate desorption and emissions, allowing the emission profiles to be
predicted using a diffusion-based emission model. Prototype reference materials were
then created using 0.025 cm thick PC films loaded with known quantities of
formaldehyde. The D and K values for the formaldehyde/PC system were determined to
be (1.9±0.3)xlO"13 m2/s and 230±40 respectively for the 0.025 cm thick PC according to
sorption/desorption test under 0 RH condition at 25 °C.
Although some deviation existed between chamber-measured and model-predicted
concentrations, overall the ASTM D-5116-10 small-scale chamber testing of
formaldehyde emissions reference materials suggests that formaldehyde emissions could
be predicted by a fundamental emissions model.
Tests showed that the aluminium foil packaging and dry-ice shipping and storage
methods could successfully retard the loss of formaldehyde from the polymer films,
although loss of formaldehyde was not completely prevented.
A series of shelf life tests showed that emission chamber measurements agreed well with
the model predicted emission profiles, except that the model tends to underestimate the
formaldehyde concentrations after the first 40 hours. This observation could be due to
slow depolymerisation of formaldehyde from the surface of the film or some
inconsistency in the model parameters. When reference materials are tested in small
chambers by different laboratories, the reference emission profiles predicted by the model
can be compared to the observed emission profiles to validate different emissions testing
methods, to evaluate the test performance of individual laboratories, and to help identify
the root causes of variability.
Although the effect was weak, RH test results indicated that the presence of water could
affect the formaldehyde emission rate. This could be due to water molecules plasticizing
32
-------
CONCLUSIONS
the PC film resulting in an increase in both K and D. Another possible reason might be
that moisture affected formaldehyde polymerization on the PC film surface.
While further refinement and testing is needed, particularly with regard to the chemistry
of formaldehyde, the results obtained in this study suggest that it is possible to create a
viable reference material for formaldehyde emissions testing using the proposed
approach.
33
-------
RECOMMENDATIONS
6 RECOMMENDATIONS
Based on the results of this project, there are four recommendations for further work:
First, additional work is needed to identify and quantify the behavior of formaldehyde in
PC films and in test chamber systems. Specifically the chemical behavior that affects
formaldehyde mass-transfer on and within the PC films and chamber surfaces is not
sufficiently understood. Because this behavior may be influenced by moisture and
temperature, the relationship between humidity, temperature, and formaldehyde mass-
transfer warrants further investigation. Although the idea that formaldehyde diffuses into
and from the PC film as well as adsorbs to the surface is consistent with experimental
results, the effect of formaldehyde reactivity and the conditions that facilitate
formaldehyde reactions need to be further investigated and quantified.
Second, additional work is needed to evaluate packaging materials, storage methods, and
other conditions that affect the shelf-life of the reference materials. The results suggest
that a small amount of formaldehyde is emitted from the PC films, albeit at a slow rate,
even while tightly wrapped in foil and stored at -12 °C. A packaging method and/or
sealing technique that could extend the shelf life and avoid the need for shipping in the
presence of dry-ice would be of real value.
Third, experience has shown that substantial testing and evaluation is needed to perfect
the development of such reference materials and build confidence in their practical
application. Although the initial results are promising, it is nevertheless necessary to
further test the procedures to develop and validate the current formaldehyde reference
material for small-scale chambers, including more substantial inter-laboratory studies.
Finally, the reference material should be scaled-up for use in large chambers with larger
or more numerous films subject to similar extensive testing and inter-laboratory studies.
34
-------
REFERENCES
7 REFERENCES
Arts, J.H.E., Muijser, H., Kuper, C.F. and Woutersen, R.A. (2008) Setting an indoor air
exposure limit for formaldehyde: Factors of concern, Regulatory Toxicology and
Pharmacology, 52, 189-194.
Asian, H., Songur, A., Tune, A.T., Ozen, O.A., Bas, O., Yagmurca, M., Turgut, M.,
Sarsilmaz, M. and Kaplan, S. (2006) Effects of formaldehyde exposure on granule
cell number and volume of dentate gyrus: A histopathological and stereological study,
Brain Research, 1122, 191-200.
ASTM (2010) ASTM D5116-10, Standard Guide for Small-scale Environmental Chamber
Determinations of Organic Emissions from Indoor Materials/Products, ASTM
International, West Conshohocken, PA.
ASTM (2008) ASTM D5157-97, Standard Guide for Statistical Evaluation of Indoor Air
Quality Models, ASTM International, West Conshohocken, PA.
ATSDR (2008) Toxicological profile for Formaldehyde. (Draft for Public Comment). Atlanta,
GA: U.S. Department of Health and Human Services, Public Health Service.
Bizzari, S.N. (2009) CEH Marketing Research Report: Formaldehyde, SRI International,
Menlo Park, CA.
Braswell, J.R., Spiner, D.R. and Hoffman, R.K. (1970) Adsorption of formaldehyde by
various surfaces during gaseous decontamination, Applied Microbiology, 20, 765-769.
Brown, N (1976) Polymerization of Formaldehyde, Journal of Macromolecular Science: Part
A - Chemistry: Pure and Applied Chemistry, 1, 209-230.
Cox, S.S., Liu, Z., Little, J.C., Howard-Reed, C., Nabinger, S.J. and Persily, A. (2010)
Diffusion-controlled reference material for VOC emissions testing: proof of concept,
Indoor Air, 20, 424-433.
Cox, S.S., Zhao, D. and Little, J.C. (2001) Measuring partition and diffusion coefficients for
volatile organic compounds in vinyl flooring, Atmospheric Environment, 35, 3823-
3830.
Crank, J. (1975) The Mathematics of Diffusion, Second Edition, Clarendon Press, Oxford,
England.
Deng, Q., Yang, X. and Zhang, J.S. (2009). Study on a new correlation between diffusion
coefficient and temperature in porous building materials, Atmospheric Environment,
43, 2080-2083.
Fang, L., Clausen, G. and Fanger, P.O. (1999) Impact of temperature and humidity on
chemical and sensory emissions from building materials, Indoor, 9, 193-201.
Hennebert, P. (1988) Solubility and diffusion coefficients of gaseous formaldehyde in
polymers, Biomaterials, 9, 162-167.
Howard-Reed, C., Little, J.C., Marand, E., Cox, S.S., Nabinger, S.J. and Persily, A. (2007)
'Improving the reliability of VOC emissions testing of building products' In:
Proceedings of ASHRAE IAQ 2007, Baltimore, MD, American Society of Heating,
Refrigerating and Air-Conditioning Engineers (ASHRAE)' conference.
Howard-Reed, C., Liu, Z., Benning, J., Cox, S.S., Samarov, D., Leber, D., Hodgson, A.T.,
Mason, S., Won, D. and Little, J.C. (2011) Diffusion-controlled reference material for
volatile organic compound emissions testing: pilot inter-laboratory study, Building
and Environment, 46, 1504-1511.
35
-------
REFERENCES
IARC (2006) IARC monographs on the evaluation of carcinogenic risks to humans. Vol 88.
Formaldehyde, 2-Butoxyethanol and 1-tert-Butoxypropan-2-ol, International Agency
for Research on Cancer, Lyon, France.
Kelly, T.K., Smith, D.L. and Satola, J. (1999) Emission rates of formaldehyde from materials
and consumer products found in California homes, Environmental Science and
Technology, 33, 81-88.
Kiburn, K.H. (1994) Neurobehavioral impairment and seizures from formaldehyde, Archives
of Environmental Health, 49, 37-44.
Kim, E., Little, J.C. and Chiu, N. (2004) Estimating exposure to chemical contaminants in
drinking water, Environmental Science and Technology, 38, 1799-1806.
Koss, G. and Tesseraux, I. (1999) Hydrocarbons, In: Marquardt, H., Schafer, S.G., McClellan,
R. and Welsch, F., eds, Toxicology, Academic Press, New York.
Little, J.C., Hodgson, A.T. and Gadgil, A.J. (1994) Modeling emissions of volatile organic
compounds from new carpets, Atmospheric Environment, 28, 227-234.
Meyer, B. and Boehme, C. (1997) Formaldehyde emission from solid wood, Forest Products
Journal, 47, 45-48.
NIOSH (1994) Formaldehyde by VIS 3500. In: NIOSH Manual of Analytical Methods
(NMAM), Fourth Edition, NIOSH, Washington, DC.
NTP (2011) Report on Carcinogens, Twelfth Edition, U.S. Department of Health and Human
Services, National Toxicology Program.
Parthasarathy S., Maddalena R.L., Russell M.L., and Apte M.G. (2011) Effect of temperature
and humidity on formaldehyde emissions in temporary housing units, Journal of the
Air & Waste Management Association (Air & Waste Management Association), 61,
689-695.
Paustenbach, D., Alarie, Y., Kulle, T., Schachter, N., Smith, R., Swenberg, J., Witschi, H.
and Horowitz, S.B. (1997) A recommended occupational exposure limit for
formaldehyde based on irritation, Journal of Toxicology and Environmental Health,
50, 217-263.
Risholm-Sundman, M., Larsen, A., Vestin, E. and Weibull, A. (2007) Formaldehyde
emission-comparison of different standard methods, Atmospheric Environment, 41,
3193-3202.
Rock, F., Barsan, N. and Weimar, U. (2010) System for dosing formaldehyde vapor at the
ppb level, Measurement Science and Technology, 21, 115201-115207.
Sakanashi, T.M., Rogers, J.M., Fu, S.S., Connelly, L.E. and Keen, C.L. (1996) Influence of
maternal folate status on the developmental toxicity of methanol in the CD-I mouse,
Teratology, 54, 198-206.
Salthammer, T., Mentese, S. and Marutzky, R. (2010) Formaldehyde in the indoor
environment, Chemical Reviews, 110, 2536-2572.
US EPA (1990) Compendium method IP-6A, Determination of Formaldehyde and Other
Aldehydes in Indoor Air Using a Solid Adsorbent Cartridge, US Environmental
Protection Agency, Research Triangle Park, NC.
Walker J.F. (1975) Formaldehyde, Third Edition, Robert E. Krieger Publishing Company,
Huntington, New York.
Weigl, M., Wimmer, R., Sykacek, E. and Steinwender, M. (2009) Wood-borne formaldehyde
varying with species, wood grade, and cambial age, Forest Products Journal, 59, 88-
92.
36
-------
REFERENCES
WHO (1989) Formaldehyde (Environmental Health Criteria, No. 89), World Health
Organization, Geneva, Switzerland.
Wolkoff, P. (1998) Impact of air velocity, temperature, humidity, and air on long-term
vocemissions from building products, Atmospheric Environment, 32,14-15.
Xiong, J., Yan, W. and Zhang Y. (201 la) Variable volume loading method: A convenient
and rapid method for measuring the initial emittable concentration and partition
coefficient of formaldehyde and other aldehydes in building materials, Environmental
Science and Technology, 45, 10111-10116.
Xiong, J., Yao, Y. and Zhang, Y. (201 Ib) C-history method: rapid measurement of the initial
emittable concentration, diffusion and partition coefficients for formaldehyde and
VOCs in building materials, Environmental Science and Technology, 45, 3584-3590.
Xu, J., Zhang, J.S., Liu, X., Gao, Z. (2012) Determination of partition and diffusion
coefficients of formaldehyde in selected building materials and impact of relative
humidity, Journal of the Air & Waste Management Association (Taylor & Francis
Ltd), 62, 671-679.
Yu, P.H., Wright, S., Fan, E.H., Lun, Z.R. and Gubisne-Harberele, D. (2003) Physiological
and pathological implications of semicarbazide-sensitive amino oxidase, Biochimica
et Biophysica Acta, 1647, 193-199.
Zhang, Y., Luo, X., Wang, X., Qian, K. and Zhao, R. (2007) Influence of temperature on
formaldehyde emission parameters of dry building materials, Atmospheric
Environment, 41, 3203-3216.
37
-------
APPENDIX A
APPENDIX A: MICROBALANCE SORPTION/DESORPTION TEST RESULTS
In addition to Figure 6-8, replicate sorption/desorption tests were also performed. The results and
analysis of all the sorption/desorption tests for PC and PMP are shown below. The legend in all
the figures is the same as Figure 6-8: blue dots are the microbalance measured weight of the film
during the sorption/desorption cycle; purple lines show the linear mass gain due to
polymerization; green dots show the mass gain of monomer formaldehyde due to diffusion; and
red lines are the model prediction based on Equation (1).
161.67 i 1 161.67
161.66
161.63
200000 400000 600000 800000
Time (s)
(a) sorption
100000 200000 300000 400000
Time (s)
(b) desorption
Figure Al. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.48 g/m , D:
1.6x10'13m2/s, K:210).
201.72
201.68
S01.64
201.6
201.56
201.59
200000 400000 600000
Time (s)
(a) sorption
100000 200000 300000 400000
Time (s)
(b) desorption
Figure A2. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 1 g/m3, D:
2.1xlO'13m2/s, K:270).
38
-------
APPENDIX A
161.67
161.66
161.67
400000
Time (s)
(a) sorption
161.63
800000
0 100000200000300000400000500000
Time (s)
(b) desorption
Figure A3. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.48 g/m , D:
1.5xlO'13m2/s, K: 190).
201.72
201.56
250000
Time (s)
(a) sorption
201.59
500000
250000
Time (s)
(b) desorption
500000
Figure A4. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.96 g/m3, D:
2.2xlO'13m2/s, K:280).
39
-------
APPENDIX A
146.34
146.45
146.43
146.41
146.39
146.37
200000 400000 600000
Time (s)
(a) sorption
200000 400000 600000
Time (s)
(b) desorption
Figure A5. Sorption/desorption data and analysis for a 0.025 cm thick PC (y;n: 0.86 g/m3, D:
151.02
150.82
2.2xlO'13m2/s, K:210).
151
jjfl50.98
^150.96
150.94
150.92
200000 400000
Time (s)
(a) sorption
600000
300000 600000
Time (s)
(b) desorption
Figure A6. Sorption/desorption data and analysis for a 0.051 cm thick PC (y;n: 0.86 g/m , D:
4.0xlO'13m2/s, K: 155).
40
-------
APPENDIX A
146.23 r
146.18
'146.13
146.08
146.03
146.24
146.08
200000 400000 600000 800000
Time (s)
(a) sorption
100000 200000 300000 400000
Time (s)
(b) desorption
Figure A7. Sorption/desorption data and analysis for a 0.051 cm thick PC (y;n: 0.97 g/m , D:
3.7x10'13m2/s, K: 180).
170.5
170.48
170.46
170.44
170.42
1000000
Time (s)
(a) sorption
2000000
170.5
170.49
.•*
^170.48
170.47
200000 400000
Time (s)
(b) desorption
600000
Figure A8. Sorption/desorption data and analysis for a 0.025 cm thick PMP (y;n: 1.73 g/m , D:
3.5xlO'14m2/s, K:40).
41
-------
APPENDIX B
APPENDIX B: MICROBALANCE SORPTION DATA FOR EMISSION TEST
Microbalance sorption data
The microbalance sorption data of loading process for Batch 1, Batch 2 and Batch 3 is
shown below. Blue dots are the microbalance measured weight of the film during the
sorption process; purple lines show the linear mass gain due to polymerization; green
dots show the mass gain of monomer formaldehyde due to diffusion; and red lines are the
model prediction based on Equation (1) using D = 1.9xlO"13 m2/s.
108.09
108.07-
108.05-
i 108.03
108.01
107.99
100000 200000 300000 400000
Time (s)
Figure Bl. Sorption data of loading process for VT 10 (Batch 1).
42
-------
APPENDIX B
156.62
156.54
100000 200000 300000 400000 500000
Time (s)
Figure B2. Sorption data of loading process for VT 11 (Batch 2).
CO
j§ 187.80
200000 400000
Time (s)
600000
Figure B3. Sorption data of loading process for VT 12 (Batch 3).
43
-------
APPENDIX B
Calculation of Co
The size of the PC films used in microbalance testing was 3.6 cm x 3.6 cm and the
thickness was 0.025 cm. Thus the volume of the film was 3.3><10"7 m3 and the surface
area was 2.6^10"3 m2 (double sides). Using data obtained with the microbalance (Figure
Bl, B2 and B3), the mass gain due to diffusion of these three loading processes was
0.056 - 0.068 mg, 0.052 - 0.056 mg and 0.052 - 0.060 mg; the mass gain due to
polymerization was 0.030 mg, 0.031 mg and 0.053 mg. Based on data above, the Co
resulting from these three loading processes was 190±27 g/m3, 160±8.9 g/m3 and 170±19
g/m3.
44
-------
APPENDIX C
APPENDIX C:
Table Cl
Tube ID
ID3622
ID3623
ID3624/ID3625
ID3628
ID3626
ID3629
ID3630/ID3631
ID3632
ID3633
ID3634
ID3635
ID3636/ID3637
ID3640
ID3642
ID3645
Table C2
Tube ID
ID3700
ID3703
ID3704/ID3705
ID3706
ID3707
ID3708
ID3710/ID3711
ID3715
ID3716
ID3717
ID3721
ID3722/ID3723
ID3728
ID3729
ID3733
SMALL-SCALE CHAMBER
Preliminary small-scale chamber
Description
SES-SCH-Fl-O.lhr
SES-SCH-Fl-0.2hr
SES-SCH-Fl-0.5hrA/B
SES-SCH-Fl-0.7hr
SES-SCH-Fl-lhr
SES-SCH-Fl-2hr
SES-SCH-Fl-4hrA/B
SES-SCH-Fl-6hr
SES-SCH-Fl-8hr
SES-SCH-Fl-lOhr
SES-SCH-Fl-24hr
SES-SCH-Fl-28hrA/B
SES-SCH-Fl-32hr
SES-SCH-Fl-48hr
SES-SCH-Fl-123hr
Preliminary small-scale chamber
TEST RAW
raw data for
Elapsed
time
(h)
0.08
0.21
0.50
0.74
1.03
2.00
4.01
6.01
8.06
10.00
24.00
28.01
32.02
48.52
122.91
raw data for
_ . . Elapsed time
Description /ux
SES-SCH-F2-0.1hr
SES-SCH-F2-0.2hr
SES-SCH-F2-0.5hrA/B
SES-SCH-F2-0.7hr
SES-SCH-F2-lhr
SES-SCH-F2-2hr
SES-SCH-F2-4hrA/B
SES-SCH-F2-6hr
SES-SCH-F2-8hr
SES-SCH-F2-10hr
SES-SCH-F2-24hr
SES-SCH-F2-28hrA/B
SES-SCH-F2-31hr
SES-SCH-F2-76hr
SES-SCH-F2-99hr
w
0.08
0.21
0.48
0.72
0.99
1.99
3.99
5.99
7.99
10.46
24.03
28.02
31.45
76.41
99.33
DATA
Batch 1 —
Sample
volume
(L)
6.83
3.38
3.44
3.45
7.04
6.92
7.41
7.58
46.71
41.56
41.89
41.77
43.54
85.07
180.35
Batch 1 —
Sample
volume
(L)
7.00
3.50
3.47
3.49
7.00
6.99
7.00
6.97
41.84
41.79
41.61
42.88
42.83
41.94
54.43
filmBlFl.
Formaldehyde
in air
(ug/m3)
209.03
329.41
459.84
528.70
515.79
480.89
309.77
257.69
197.49
169.30
70.51
54.38
56.66
15.41
1.75
filmB!F2.
Formaldehyde
in air
(ug/m3)
131.33
231.40
384.86
457.54
501.54
546.78
444.29
353.30
270.57
193.06
36.78
25.84
19.79
4.79
3.48
45
-------
APPENDIX C
Table C3. Preliminary small-scale chamber raw data for Batch 1 -
Tube ID
ID3738
ID3737
ID3760/ID3761
ID3762
ID3763
ID3764
ID3766/ID3767
ID3768
ID3772
ID3773
ID3790
ID3791/ID3792
ID3793
ID3794
Table
Tube ID
ID4115
ID4120
ID4123
ID4125/ID4126
ID4129
ID4134/ID4145
ID4138
ID4140
ID4144/ID4145
ID4148
ID4150
ID4152
ID4154/ID4155
ID4158
ID4161
ID4164/ID4165
ID4175
Description
SES-SCH-F3-0.1hr
SES-SCH-F3-0.2hr
SES-SCH-F3-0.5hrA/B
SES-SCH-F3-0.7hr
SES-SCH-F3-lhr
SES-SCH-F3-2hr
SES-SCH-F3-4hrA/B
SES-SCH-F3-6hr
SES-SCH-F3-8hr
SES-SCH-F3-10hr
SES-SCH-F3-24hr
SES-SCH-F2-28hrA/B
SES-SCH-F3-31hr
SES-SCH-F3-52hr
C4. Small-scale chamber raw
Description
SES-SCHl-FAl-WO-O.lOhr
SES-SCHl-FAl-WO-0.25hr
SES-SCHl-FAl-WO-0.50hr
SES-SCHl-FAl-WO-0.75hr A/B
SES-SCHl-FAl-WO-l.Ohr
SES-SCHl-FAl-WO-2.0hr A/B
SES-SCHl-FAl-WO-4.0hr
SES-SCHl-FAl-WO-6.0hr
SES-SCHl-FAl-WO-8.0hr A/B
SES-SCHl-FAl-WO-lOhr
SES-SCHl-FAl-WO-24hr
SES-SCHl-FAl-WO-28hr
SES-SCHl-FAl-WO-32hr A/B
SES-SCHl-FAl-WO-48hr
SES-SCHl-FAl-WO-108hr
SES-SCHl-FAl-WO-123hrA/B
SES-SCHl-FAl-WO-146hr
Elapsed
time
(h)
0.10
0.20
0.47
0.72
0.99
1.99
3.99
6.07
8.30
10.46
23.96
28.06
31.61
51.92
data for Batch 2
Elapsed
time
(h)
0.11
0.30
0.50
0.70
1.00
2.00
4.12
6.00
8.00
10.00
24.00
28.00
32.00
48.00
108.08
122.60
146.11
Sample
volume
(L)
4.26
3.50
3.49
3.50
6.99
6.97
6.96
13.93
67.28
41.73
41.99
41.21
41.77
45.17
— film
Sample
volume
(L)
3.20
3.50
3.54
3.53
3.52
7.01
6.94
7.23
7.04
7.05
21.01
20.82
20.83
42.15
63.67
145.92
64.30
-filmBlFS.
Formaldehyde
in air
(ug/m3)
151.25
202.13
331.68
395.28
433.28
482.72
419.71
336.04
252.78
179.42
38.53
26.13
19.70
7.06
B2FA1.
Formaldehyde
in air
(ug/m3)
527.19
696.70
803.92
825.94
929.58
886.83
740.58
564.39
411.42
307.23
60.42
40.32
32.58
13.96
5.97
4.47
4.93
46
-------
APPENDIX C
Table
Tube ID
ID4116
ID4122
ID4124
ID4125/ID4126
ID4130
ID4134/ID4145
ID4139
ID4141
ID4146/ID4147
ID4149
ID4151
ID4153
ID4154/ID4155
ID4159
ID4162
ID4166/ID4167
ID4176
Table
Tube ID
ID4201
ID4205
ID4207/ID4208
ID4211
ID4213
ID4215/ID4216
ID4219
ID4221
ID4223/ID4224
ID4227
ID4232
ID4234
ID4238
ID4241/ID4242
ID4242
ID4245
ID4250
ID4257/ID4258
C5. Small-scale chamber raw
Description
SES-SCH5-FA2-WO-0.10hr
SES-SCH5-FA2-WO-0.25hr
SES-SCH5-FA2-WO-0.50hr
SES-SCH5-FA2-WO-0.75hr A/B
SES-SCH5-FA2-WO-1.0hr
SES-SCH5-FA2-WO-2.0hr A/B
SES-SCH5-FA2-WO-4.0hr
SES-SCH5-FA2-WO-6.0hr
SES-SCH5-FA2-WO-8.0hr A/B
SES-SCH5-FA2-WO-10hr
SES-SCH5-FA2-WO-24hr
SES-SCH5-FA2-WO-28hr
SES-SCH5-FA2-WO-32hr A/B
SES-SCH5-FA2-WO-48hr
SES-SCH5-FA2-WO-108hr
SES-SCH5-FA2-WO-123hrA/B
SES-SCH5-FA2-WO-146hr
C6. Small-scale chamber raw
Description
SES-SChl-FA4-W2-0.10hr
SES-SChl-FA4-W2-0.25hr
SES-SChl-FA4-W2-0.50hrA/B
SES-SChl-FA4-W2-0.75hr
SES-SChl-FA4-W2-1.0hr
SES-SChl-FA4-W2-2.0hrA/B
SES-SChl-FA4-W2-4.0hr
SES-SChl-FA4-W2-6.0hr
SES-SChl-FA4-W2-8.0hrA/B
SES-SChl-FA4-W2-10hr
SES-SChl-FA4-W2-14hr
SES-SChl-FA4-W2-24hr
SES-SChl-FA4-W2-28hr
SES-SChl-FA4-W2-32hrA/B
SES-SChl-FA4-W2-32hrB
SES-SChl-FA4-W2-48hr
SES-SChl-FA4-W2-72hr
SES-SChl-FA4-W2-147hrA/B
data for Batch 2
Elapsed
time
(h)
0.11
0.30
0.50
0.70
1.00
2.00
4.12
6.00
8.00
10.00
24.00
28.00
32.00
48.00
108.08
122.60
146.11
data for Batch 2
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
8.00
10.00
14.37
24.00
28.00
32.00
32.00
48.00
72.22
147.39
— film B2FA2.
Sample
volume
(L)
3.20
3.51
3.52
3.46
3.53
6.96
6.92
7.21
7.05
7.05
21.21
20.76
21.19
41.66
63.58
145.94
64.38
Formaldehyde
in air
(ug/m3)
547.14
717.16
783.08
875.49
907.80
921.13
750.39
567.66
422.20
318.50
61.94
40.56
30.59
13.71
4.95
4.60
3.81
— film B2FA4.
Sample
volume
(L)
3.48
3.48
3.48
3.50
3.49
6.95
6.93
6.95
6.98
6.94
6.97
20.83
20.90
21.11
21.07
42.15
42.44
82.25
Formaldehyde
in air
(ug/m3)
296.52
467.97
610.29
677.94
752.40
857.25
718.47
534.76
399.88
300.20
168.65
59.83
42.89
31.63
32.46
13.11
7.79
3.75
47
-------
APPENDIX C
Table
Tube ID
ID4202
ID4206
ID4209/ID4210
ID4212
ID4214
ID4217/ID4218
ID4220
ID4222
ID4225/ID4226
ID4228
ID4233
ID4235
ID4239
ID4243/ID4244
ID4244
ID4246
ID4251
ID4259/ID4260
Table
Tube ID
ID4281
ID4283
ID4287/ID4288
ID4291
ID4293
ID4295/ID4296
ID4299
ID4301
ID4303/ID4304
ID4309
ID4312
ID4314
ID4319
ID4321/ID4322
ID4325
ID4330
ID4336
ID4340/ID4341
C7. Small-scale chamber raw
Description
SES-SCh5-FA5-W2-0.10hr
SES-SCh5-FA5-W2-0.25hr
data for Batch 2
Elapsed
time
(h)
0.10
0.30
SES-SCh5-FA5-W2-0.50hrA/B 0.50
SES-SCh5-FA5-W2-0.75hr
SES-SCh5-FA5-W2-1.0hr
SES-SCh5-FA5-W2-2.0hrA/B
SES-SCh5-FA5-W2-4.0hr
SES-SCh5-FA5-W2-6.0hr
SES-SCh5-FA5-W2-8.0hrA/B
SES-SCh5-FA5-W2-10hr
SES-SCh5-FA5-W2-14hr
SES-SCh5-FA5-W2-24hr
SES-SCh5-FA5-W2-28hr
SES-SCh5-FA5-W2-32hrA/B
SES-SCh5-FA5-W2-32hrB
SES-SCh5-FA5-W2-48hr
SES-SCh5-FA5-W2-72hr
0.70
1.00
2.00
4.00
6.00
8.00
10.00
14.37
24.00
28.00
32.00
32.00
48.00
72.22
SES-SCh5-FA5-W2-147hrA/B 147.39
C8. Small-scale chamber raw
Description
SES-SChl-FA6-W4-0.10hr
SES-SChl-FA6-W4-0.25hr
SES-SChl-FA6-W4-0.50hrA/B
SES-SChl-FA6-W4-0.75hr
SES-SChl-FA6-W4-1.0hr
SES-SChl-FA6-W4-2.0hrA/B
SES-SChl-FA6-W4-4.0hr
SES-SChl-FA6-W4-6.0hr
SES-SChl-FA6-W4-8.0hrA/B
SES-SChl-FA6-W4-10hr
SES-SChl-FA6-W4-14hr
SES-SCh 1 -FA6-W4 -24hr
SES-SChl-FA6-W4-28hr
SES-SChl-FA6-W4-32hrA/B
SES-SChl-FA6-W4-49hr
SES-SChl-FA6-W4-73hr
SES-SChl-FA6-W4-98hr
SES-SChl-FA6-W4-147hrA/B
data for Batch 2
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
8.00
9.50
14.23
24.00
28.00
32.00
48.50
72.57
98.10
147.47
— filmB2FA5.
Sample
volume
(L)
3.51
3.51
3.50
3.52
3.51
6.98
7.00
6.99
7.02
6.99
7.01
21.00
20.93
21.13
21.07
42.22
42.38
82.01
Formaldehyde
in air
(ug/m3)
334.56
513.53
636.25
682.99
815.26
887.86
721.73
549.05
410.56
309.66
175.30
61.35
45.17
33.40
33.68
14.10
7.90
4.04
— film B2FA6.
Sample
volume
(L)
3.48
3.48
3.49
3.48
3.49
6.96
6.94
6.93
6.94
6.99
6.94
20.88
20.94
20.95
41.85
44.64
80.35
74.43
Formaldehyde
in air
(ug/m3)
310.14
465.38
595.07
696.31
742.76
824.35
686.63
519.83
396.87
328.22
174.65
63.11
47.93
40.79
15.12
9.87
5.70
3.39
48
-------
APPENDIX C
Table
Tube ID
ID4282
ID4284
ID4289/ID4290
ID4292
ID4294
ID4297/ID4298
ID4300
ID4302
ID4305/ID4306
ID4310
ID4313
ID4315
ID4320
ID4323/ID4324
ID4326
ID4331
ID4337
ID4342/ID4343
Table
Tube ID
ID4370
ID4375
ID4377/ID4378
ID4381
ID4383
ID4385/ID4386
ID4390
ID4393
ID4395/ID4396
ID4399
ID4401
ID4403
ID4405
ID4407/ID4408
ID4411
ID4420
ID4414
ID4422/ID4423
C9. Small-scale chamber raw data for Batch 2
Description
SES-SCh5-FA7-W4-0.10hr
SES-SCh5-FA7-W4-0.25hr
SES-SCh5-FA7-W4-0.50hrA/B
SES-SCh5-FA7-W4-0.75hr
SES-SCh5-FA7-W4-1.0hr
SES-SCh5-FA7-W4-2.0hrA/B
SES-SCh5-FA7-W4-4.0hr
SES-SCh5-FA7-W4-6.0hr
SES-SCh5-FA7-W4-8.0hrA/B
SES-SCh5-FA7-W4-10hr
SES-SCh5-FA7-W4-14hr
SES-SCh5-FA7-W4-24hr
SES-SCh5-FA7-W4-28hr
SES-SCh5-FA7-W4-32hrA/B
SES-SCh5-FA7-W4-49hr
SES-SCh5-FA7-W4-73hr
SES-SCh5-FA7-W4-98hr
SES-SCh5-FA7-W4-147hrA/B
CIO. Small-scale chamber raw
Description
SES-SChl-FBl-W6-0.10hr
SES-SChl-FBl-W6-0.25hr
SES-SChl-FBl-W6-0.50hrA/B
SES-SChl-FBl-W6-0.75hr
SES-SChl-FBl-W6-1.0hr
SES-SChl-FBl-W6-2.0hrA/B
SES-SChl-FBl-W6-4.0hr
SES-SChl-FBl-W6-6.0hr
SES-SChl-FBl-W6-8.0hrA/B
SES-SChl-FBl-W6-10hr
SES-SChl-FBl-W6-14hr
SES-SChl-FBl-W6-24hr
SES-SChl-FBl-W6-28hr
SES-SChl-FBl-W6-32hrA/B
SES-SChl-FBl-W6-49hr
SES-SChl-FBl-W6-73hr
SES-SChl-FBl-W6-101hr
SES-SChl-FBl-W6-147hrA/B
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
8.00
9.50
14.23
24.00
28.00
32.00
48.50
72.57
98.10
147.47
data for Batch 2
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
7.83
10.12
14.65
24.22
28.10
31.97
48.72
72.82
100.63
147.10
— film B2FA7.
Sample
volume
(L)
3.48
3.49
3.47
3.49
3.50
6.95
6.95
6.94
6.93
7.01
6.97
20.88
20.99
21.00
42.04
44.49
80.36
74.40
— film
Sample
volume
(L)
3.51
3.53
3.52
3.52
3.52
7.03
7.03
7.03
7.02
7.01
7.00
23.90
25.01
23.12
41.74
44.08
62.60
102.17
Formaldehyde
in air
(ug/m3)
366.39
504.25
619.56
692.09
779.09
829.66
728.25
539.11
396.53
320.83
174.57
64.60
51.50
33.79
13.22
8.46
5.41
4.09
B2FB1.
Formaldehyde
in air
(ug/m3)
317.91
449.48
532.40
583.46
671.11
773.03
697.80
539.90
413.03
293.04
167.74
61.04
45.94
34.06
13.20
7.88
5.63
3.86
49
-------
APPENDIX C
Table
Tube ID
ID4371
ID4376
ID4379/ID4380
ID4382
ID4384
ID4387/ID4388
ID4391
ID4394
ID4397/ID4398
ID4400
ID4402
ID4404
ID4406
ID4409/ID4410
ID4412
ID4421
ID4419
ID4424/ID4425
Table
Tube ID
ID4445
ID4447
ID4449/ID4450
ID4455
ID4457
ID4461/ID462
ID4463
ID4465
ID4467/ID4468
ID4473
ID4475
ID4478
ID4480
ID4482/ID4483
ID4486
ID4490
ID4492
ID4498/ID4499
Cll. Small-scale chamber raw
Description
SES-SCh5-FB2-W6-0.10hr
SES-SCh5-FB2-W6-0.25hr
SES-SCh5-FB2-W6-0.50hrA/B
SES-SCh5-FB2-W6-0.75hr
SES-SCh5-FB2-W6-1.0hr
SES-SCh5-FB2-W6-2.0hrA/B
SES-SCh5-FB2-W6-4.0hr
SES-SCh5-FB2-W6-6.0hr
SES-SCh5-FB2-W6-8.0hrA/B
SES-SCh5-FB2-W6-10hr
SES-SCh5-FB2-W6-14hr
SES-SCh5-FB2-W6-24hr
SES-SCh5-FB2-W6-28hr
SES-SCh5-FB2-W6-32hrA/B
SES-SCh5-FB2-W6-49hr
SES-SCh5-FB2-W6-72hr
SES-SCh5-FB2-W6-101hr
SES-SCh5-FB2-W6-147hrA/B
C12. Small-scale chamber raw
Description
SES-SChl-FB3-W10-0.10hr
SES-SChl-FB3-W10-0.25hr
SES-SChl-FB3-W10-0.50hrA/B
SES-SChl-FB3-W10-0.75hr
SES-SChl-FB3-W10-1.0hr
SES-SChl-FB3-W10-2.0hrA/B
SES-SChl-FB3-W10-4.0hr
SES-SChl-FB3-W10-6.0hr
SES-SChl-FB3-W10-8.0hrA/B
SES-SChl-FB3-W10-10hr
SES-SChl-FB3-W10-15hr
SES-SChl-FB3-W10-24hr
SES-SChl-FB3-W10-28hr
SES-SChl-FB3-W10-32hrA/B
SES-SChl-FB3-W10-48hr
SES-SChl-FB3-W10-72hr
SES-SChl-FB3-W10-99hr
SES-SChl-FB3-W10-172hrA/B
data for Batch
Elapsed
time (h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
7.83
10.12
14.65
24.22
28.10
31.97
48.72
72.82
100.63
147.10
data for Batch
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
8.00
10.03
14.64
24.00
28.01
32.00
48.33
72.36
99.28
171.52
2 — film
B2FB2.
Sample Formaldehyde
volume (L) in air (ug/m3)
3.51
3.53
3.48
3.52
3.51
6.99
7.03
7.03
6.99
7.01
7.02
23.93
25.03
23.13
41.67
43.80
62.65
102.24
2 — film
Sample
volume
(L)
3.49
3.49
3.49
3.49
3.50
6.97
6.96
6.95
6.92
6.91
6.80
20.87
21.20
20.84
43.46
43.47
64.18
73.60
242.67
310.10
465.08
523.37
556.52
627.76
567.57
449.50
366.67
250.40
140.06
56.33
39.48
33.79
13.01
8.57
7.00
4.34
B2FB3.
Formaldehyde
in air
(ug/m3)
333.76
468.15
563.62
643.53
700.60
780.82
680.84
536.23
411.01
287.34
182.00
65.18
48.41
35.04
12.84
8.80
6.16
3.59
50
-------
APPENDIX C
Table
Tube ID
ID4446
ID4448
ID4453/ID4454
ID4456
ID4458
ID4460/ID4459
ID4464
ID4466
ID4469/ID4470
ID4474
ID4476
ID4479
ID4481
ID4484/ID4485
ID4487
ID4491
ID4493
ID4501/ID4500
Table
Tube ID
ID4117/ID4121
ID4160
ID4168
ID4184
ID4186
ID4195/ID4196
ID4240
ID4268
ID4318
ID4353/ID4354
ID4413
ID4431
ID4435
ID4440
ID4488/ID4489
CIS. Small-scale chamber raw data for Batch
Elapsed
Description time
(h)
SES-SCh5-FB4-W10-0.10hr 0.10
SES-SCh5-FB4-W10-0.25hr 0.30
SES-SCh5-FB4-W10-0.50hrA/B 0.50
SES-SCh5-FB4-W10-0.75hr 0.70
SES-SCh5-FB4-W10-1.0hr 1.00
SES-SCh5-FB4-W10-2.0hrA 2.00
SES-SCh5-FB4-W10-4.0hr 4.00
SES-SCh5-FB4-W10-6.0hr 6.00
SES-SCh5-FB4-W10-8.0hrA/B 8.00
SES-SCh5-FB4-W10-10hr 10.03
SES-SCh5-FB4-W10-15hr 14.64
SES-SCh5-FB4-W10-24hr 24.00
SES-SCh5-FB4-W10-28hr 28.01
SES-SCh5-FB4-W10-32hrA/B 32.00
SES-SCh5-FB4-W10-49hr 48.33
SES-SCh5-FB4-W10-72hr 72.36
SES-SCh5-FB4-W10-99hr 99.28
SES-SCh5-FB4-W10-172hrA/B 171.52
C14. Small-scale chamber raw data for Batch
Elapsed
Description time
(h)
SES-SCH6-PF-2.5hrA/B 2.50
SES-SCh6-PF-48hr 48.00
SES-SCh6-PF-122hr 121.77
SES-SCh6-PF-170hr 169.74
SES-SCh6-PF-218hr 218.28
SES-SCH6-PF-264hrA/B 292.05
SES-SCh6-PF-312hr 323.50
SES-SCh6-PF-516hr 516.03
SES-SCh6-PF-683hr 682.60
SES-SCH6-PF-855hrA/B 855.42
SES-SCh6-PF-1046hr 1045.90
SES-SCh6-PF-1212hr 1211.53
SES-SCh6-PF-1334hr 1334.04
SES-SCh6-PF-1553hr 1552.97
SES-SCH6-PF-1693hrA/B 1692.69
2 — film
Sample
volume
(L)
3.49
3.49
3.48
3.50
3.50
6.97
6.94
6.95
6.94
6.91
6.75
20.77
21.12
20.93
43.45
43.49
64.20
73.60
3 — film
Sample
volume
(L)
104.19
104.70
102.77
115.85
104.05
104.20
416.42
448.81
471.57
578.27
538.23
508.25
518.28
499.62
508.67
B2FB4.
Formaldehyde
in air
(ug/m3)
378.09
489.24
592.31
664.11
721.07
770.16
689.26
530.87
405.52
289.24
190.93
65.01
48.13
36.66
12.03
9.69
6.25
4.23
B3FA3.
Formaldehyde
in air
(ug/m3)
147.49
64.86
17.24
10.56
6.20
3.53
2.05
.87
.27
.27
.12
.11
.60
.36
.11
51
-------
APPENDIX C
Table
Tube ID
ID4614
ID4615
ID4616/ID4617
ID4618
ID4619
ID4620/ID4621
ID4624
ID4625
ID4627/ID4628
ID4629
ID4630
ID4632
ID4634/ID4635
ID4636
ID4637
ID4639
ID4642/ID4643
ID4646
Table
Tube ID
ID4545
ID4547
CIS. Small-scale chamber raw data for
Description
SES-SCHl-FA2-071612-0%RH-0.10hr
SES-SCHl-FA2-071612-0%RH-0.25hr
SES-SCHl-FA2-0%RH-0.50hrA/B
SES-SCHl-FA2-071612-0%RH-0.75hr
SES-SCHl-FA2-071612-0%RH-1.0hr
SES-SCHl-FA2-0%RH-2.0hrA/B
SES-SCHl-FA2-071612-0%RH-4.0hr
SES-SCHl-FA2-071612-0%RH-6.0hr
SES-SCHl-FA2-0%RH-8.0hrA/B
SES-SCHl-FA2-071612-0%RH-10hr
SES-SCHl-FA2-071612-0%RH-14hr
SES-SCHl-FA2-071612-0%RH-24hr
SES-SCHl-FA2-0%RH-28hrA/B
SES-SCHl-FA2-071612-0%RH-32hr
SES-SCHl-FA2-071612-0%RH-48hr
SES-SCHl-FA2-071612-0%RH-78hr
SES-SCHl-FA2-0%RH-122hr A/B
SES-SCHl-FA2-071612-0%RH-145hr
C16. Small-scale chamber raw data for
Description
SES-SCHl-FCl-071612-70%RH-0.10hr
SES-SCHl-FCl-071612-70%RH-0.25hr
Batch 3 — film B3FA2.
Flamed Sample
L/ldUaCU 1
.. f.^ volume
time ml
Ulll^ \11J ,r ^
0.10 3.47
0.30 3.48
0.50 3.48
0.70 3.47
1.00 3.47
2.00 6.95
4.00 6.93
6.01 7.33
8.00 6.95
10.00 6.94
14.47 6.84
24.37 41.76
28.01 21.12
32.00 20.76
48.47 45.78
77.83 71.56
121.50 62.57
145.28 63.39
Batch 3— film B3FC1.
Elapsed Sample
time volume
(h) (L)
0.10 3.46
0.30 3.48
ID4549/ID4550 SES-SCHl-FCl-071612-70%RH-0.50hrA/B 0.50 3.48
ID4453
ID4555
ID4557/ID4558
ID4561
ID4563
ID4565/ID4566
ID4569
ID4571
ID4574
ID4576/ID4577
ID4580
ID4582
ID4586
ID4590/ID4591
ID4593
SES-SCHl-FCl-071612-70%RH-0.75hr
SES-SCHl-FCl-071612-70%RH-1.0hr
SES-SCHl-FCl-071612-70%RH-2.0hrA/B
SES-SCHl-FCl-071612-70%RH-4.0hr
SES-SCHl-FCl-071612-70%RH-6.0hr
SES-SCHl-FCl-071612-70%RH-8.0hrA/B
SES-SCHl-FCl-071612-70%RH-10hr
SES-SCHl-FCl-071612-70%RH-14hr
SES-SCHl-FCl-071612-70%RH-24hr
SES-SCHl-FCl-071612-70%RH-28hrA/B
SES-SCHl-FCl-071612-70%RH-32hr
SES-SCHl-FCl-071612-70%RH-48hr
SES-SCHl-FCl-071612-70%RH-72hr
SES-SCHl-FCl-071612-70%RH-96hrA/B
SES-SCHl-FCl-071612-70%RH-145hr
0.70 3.47
1.00 3.47
2.00 6.93
4.00 6.93
6.00 6.91
8.00 6.99
10.00 6.93
14.17 6.92
24.00 20.76
28.00 20.80
32.00 20.78
48.02 41.60
71.83 41.52
96.00 62.40
144.99 63.20
Formaldehyde
in air
(ug/m3)
266.76
454.80
592.42
638.81
749.78
803.93
701.81
526.20
416.21
333.35
215.23
89.46
71.06
54.02
21.57
9.37
5.54
5.28
Formaldehyde
in air
(ug/m3)
479.00
620.44
675.37
683.92
698.49
654.22
502.65
389.25
334.05
289.51
231.78
143.21
120.73
102.69
51.94
19.45
9.66
5.51
52
-------
APPENDIX C
Table
Tube ID
ID4546
ID4548
C17. Small-scale chamber raw data for Batch
Description
SES-SCH5-FC2-071612-50%RH-0.10hr
SES-SCH5-FC2-071612-50%RH-0.25hr
ID4551/ID4552 SES-SCH5-FC2-071612-50%RH-0.50hrA/B
ID4554
ID4556
ID4559/ID4560
ID4562
ID4564
ID4567/ID4568
ID4570
ID4572
ID4575
ID4578/ID4579
ID4581
ID4583
ID4587
ID4588/ID4589
ID4594
Table
Tube ID
ID4597
ID4599
ID4600/ID4601
ID4603
ID4602
ID4604/ID4605
ID4606
ID4607
ID4608/ID4609
ID4610
ID4611
ID4613
ID4622/ID4623
ID4626
ID4633
ID4638
ID4640/ID4641
ID4644
SES-SCH5-FC2-071612-50%RH-0.75hr
SES-SCH5-FC2-071612-50%RH-1.0hr
SES-SCH5-FC2-071612-50%RH-2.0hrA/B
SES-SCH5-FC2-071612-50%RH-4.0hr
SES-SCH5-FC2-071612-50%RH-6.0hr
SES-SCH5-FC2-071612-50%RH-8.0hrA/B
SES-SCH5-FC2-071612-50%RH-10hr
SES-SCH5-FC2-071612-50%RH-14hr
SES-SCH5-FC2-071612-50%RH-24hr
SES-SCH5-FC2-071612-50%RH-28hrA/B
SES-SCH5-FC2-071612-50%RH-32hr
SES-SCH5-FC2-071612-50%RH-48hr
SES-SCH5-FC2-071612-50%RH-72hr
SES-SCH5-FC2-071612-50%RH-96hrA/B
SES-SCH5-FC2-071612-50%RH-145hr
CIS. Small-scale chamber raw data for Batch
Description
SES-SCH6-FAl-071612-50%RH-0.10hr
SES-SCH6-FAl-071612-50%RH-0.25hr
SES-SCH6-FAl-50%RH-0.50hrA/B
SES-SCH6-FAl-071612-50%RH-0.75hr
SES-SCH6-FAl-071612-50%RH-1.0hr
SES-SCH6-FAl-50%RH-2.0hrA/B
SES-SCH6-FAl-071612-50%RH-4.0hr
SES-SCH6-FAl-071612-50%RH-6.0hr
SES-SCH6-FAl-50%RH-8.0hrA/B
SES-SCH6-FAl-071612-50%RH-10hr
SES-SCH6-FAl-071612-50%RH-14hr
SES-SCH6-FAl-071612-50%RH-24hr
SES-SCH6-FAl-50%RH-28hrA/B
SES-SCH6-FAl-071612-50%RH-32hr
SES-SCH6-FAl-071612-50%RH-48hr
SES-SCH6-FAl-071612-50%RH-73hr
SES-SCH6-FAl-50%RH-102hrA/B
SES-SCH6-FAl-071612-50%RH-146hr
3 — film
Elapsed
time
(h)
0.10
0.30
0.50
0.70
1.00
2.00
4.00
6.00
8.00
10.00
14.17
24.00
28.00
32.00
48.02
71.83
96.25
145.03
3 — film
Elapsed
time (h)
0.10
0.30
0.50
0.70
1.00
2.00
4.03
6.03
8.00
10.00
14.37
24.07
28.00
32.00
47.92
72.52
101.88
145.55
B3FC2.
Sample
volume
(L)
3.45
3.47
3.47
3.47
3.47
6.93
6.92
6.91
6.95
6.92
6.90
20.76
20.80
20.71
41.44
41.57
62.45
64.82
B3FA1.
Sample
volume
(L)
3.51
3.50
3.49
3.49
3.66
6.95
8.34
7.02
6.95
6.96
6.80
22.39
20.86
20.85
20.85
45.92
71.71
62.30
Formaldehyde
in air
(ug/m3)
458.99
604.71
648.17
746.50
714.31
672.03
539.61
403.13
360.90
313.83
254.97
145.33
119.98
97.75
44.83
16.43
8.03
3.53
Formaldehyde
in air
(ug/m3)
324.08
443.45
542.31
630.45
629.62
633.46
489.47
410.94
334.09
289.22
221.01
134.69
110.84
94.18
49.23
14.73
5.95
4.67
53
-------
APPENDIX C
Elapsed
time
(h)
0
0.25
0.5
1
2
3
4
6
8
24
28
48
72
144
Table C19
Formaldehyde
in air
(ug/m3)
B3FB1
12.0
355.3
278.0
209.2
178.9
152.4
120.8
151.2
65.5
43.2
27.6
22.3
14.0
Small-scale chamber raw data for
Formaldehyde
in air
(ug/m3)
B3FB2
12.2
429.3
605.9
419.4
330.6
254.5
212.0
192.4
84.1
64.1
33.9
21.6
13.9
Formaldehyde
in air
(ug/m3)
B3FB3
14.0
422.0
312.3
222.7
210.3
194.3
151.7
139.0
64.4
56.2
26.7
20.0
Lab A.
Formaldehyde
in air
(ug/m3)
B3FB4
13.1
492.7
472.7
377.7
332.7
267.3
234.3
208.0
79.5
71.8
34.3
21.3
Formaldehyde
in air
(ug/m3)
B3FB5
12.7
424.0
548.3
497.3
370.0
315.3
275.7
242.0
97.8
83.9
38.3
22.7
54
-------
APPENDIX D
APPENDIX D: FORMALDEHYDE MASS BALANCE ANALYSIS
To evaluate the soundness of the overall approach to predicting and measuring
formaldehyde emissions from the PC films, a mass balance analysis was performed by
comparing the actual monomer formaldehyde to the model predicted monomer
formaldehyde emitted during the first 144 hours of each chamber test. Table Dl shows
the monomer formaldehyde loaded into each film as measured using the microbalance.
Total formaldehyde in the films as measured using the microbalance
Table Dl. Total formaldehyde in the film as measured using the microbalance.
Test ID
Formaldehyde mass
(Hg)
VT 10
VT 11
VT 12 (10 cm x 10 cm)
VT 12 (8.5 cm x 8.5 cm)
480
410
430
310
Total formaldehyde in the films according to experimental data
Table D2. Formaldehyde mass-balance calculation using chamber test data.
Film ID
Total formaldehyde mass in the
film
Emitted formaldehyde mass in
small chamber tests Recovery
B1F1
B1F2
B1F3
B2FA1
B2FA2
B2FA4
B2FA5
B2FA6
B2FA7
B2FB1
B2FB2
B2FB3
B2FB4
B3FA1
B3FA2
B3FB2
B3FB4
B3FB5
B3FC1
B3FC2
480
480
480
410
410
410
410
410
410
410
410
410
410
430
430
310
310
310
430
430
360
360
310
570
580
510
530
520
520
500
440
520
530
580
570
340
290
330
610
620
75%
75%
65%
140%
140%
120%
130%
130%
130%
120%
110%
130%
130%
130%
130%
110%
94%
110%
140%
140%
55
-------
APPENDIX E
APPENDIX E: MATLAB PROGRAMS OF MODELS
Sorption/desorption model program
close all;
clear all;
T=3600*24*20;
time step=300;
n=0;
L=0. 0254*0. 01;
D=2.1*10A-13;
N=T/time_step+l ;
ratio=zeros (N, 1) ;
for time=0 : time step:T
n=n+l
i=0;
s um= 0 ;
step_step=8/ ( (2*i+l) A2*3.141592654A2) *exp(-
1*D* (2*i+l) A2*3.141592654A2*time/LA2) ;
sum=sum+step step;
while (abs (step_step) /sum>0. 00001)
step_step=8/ ( (2*i+l ) A2*3 . 141592654 A2 ) *exp (-
2*3.141592654A2*time/LA2) ;
sum=sum+step step;
end
ratio (n, 1) =l-sum;
end
for i=l: 1: (T/time_step+l)
xaxis (i)=300* (i-1) ;
end
induceddata=load ( ' checkeddata . csv' ) ;
plot (induceddata ( : , 1 ) , induceddata ( : , 4 ) , 'r', xaxis, ratio, 'b')
Emission model program
clear all;
Q=0.001/60;
V=0.0209;
56
-------
APPENDIX E
L=0. 0254/100/2;
A=8. 5/100*8. 5/100*2;
D=1.9*10A (-13) ;
K=233;
Cinitial=l. 628E+08;
h=Q/A/D/K;
k=V/A/K;
x=L;
T=24*3600*6;
time_step=360 ;
time index=0;
for t=0 : time_step:T
time_index=time_index+l
i=l;
s um= 0 ;
root (i ) =qiuun (h,k,L,i) ;
term_for_sum(i) =exp (-D*root (i) *root (i) *t) * (h-
k*root (i) *root (i) ) *cos (root (i) *x) / (L* (h-k*root (i) *root (i) ) * (h-
k*root (i) *root (i) ) +root (i) *root (i) * (L+k) +h) /cos (root (i) *L) ;
sum=sum+term for sum(i) ;
while (abs ( term_f or_sum) /sum>0 . 0001)
root (i ) =qiuun (h,k,L,i) ;
term_for_sum(i) =exp (-D*root (i) *root (i) *t) * (h-
k*root (i) *root (i) ) *cos (root (i) *x) / (L* (h-k*root (i) *root (i) ) * (h-
k*root (i) *root (i) ) +root (i) *root (i) * (L+k) +h) /cos (root (i) *L) ;
sum=sum+term for sum(i) ;
end
y(time index) =2*Cinitial*sum/K;
clear sum;
clear term for sum;
clear root;
end
qiuun
function root=qiuun (h, k, L,m)
if m==l
LB=0.5*pi/ (2*L) ;
UB=l*pi/ (2*L) ;
else
57
-------
APPENDIX E
LB=(2*m-3)*pi/(2*L);
UB=(2*m-l)*pi/(2*L);
end
root=LB;
DX=UB-LB;
error=l;
while (abs(error)>0.000000001)&&(abs(DX)>O.OOOOOOOOOOOOOOi;
DX=DX/2;
QN=root+DX;
error=tan(QN*L)-(h/QN)+(k*QN);
if error<0
root=QN;
end
end
Uncertainty of emission model program
clear all;
L=0.0254/100/2;
A=8.5/100*8.5/100*2;
Q=0.001/60;
V=0.0209;
Cinitial_mean=l.701E+08;
D_mean=l.9*10A(-13) ;
K_mean=233;
Cinitial_sd=l.916068973E+07;
D_sd=0.3*10A(-13) ;
K_sd=40;
Time=3600*24*6;
tt=360;
for 1=1:1:10000
Cinitial(i)=0;
D(i)=0;
K(i)=0;
while ((Cinitial(i)<=0)|(D(i)<=0)|(K(i)<=0))
Cinitial(i)=normrnd(Cinitial mean, Cinitial sd);
D(i)=normrnd(D_mean, D_sd);
58
-------
APPENDIX E
K(i)=normrnd(K mean, K sd) ;
end
cci=Cinitial (i) ;
dd=D(i);
kk=K(i);
i
gas cone(i,:)=steve(dd,kk, cci, L, A, Q, V,Time, tt);
end
steve
function y=steve(dd,kk, cci, L, A, Q, V,T, time step)
h=Q/A/dd/kk;
k=V/A/kk;
x=L;
for t=0 : time_step:T
i=l;
s um= 0 ;
root (i ) =qiuun (h,k,L,i) ;
term_for_sum(i) =exp (-dd*root (i) *root (i) *t) * (h-
k*root (i) *root (i) ) *cos (root (i) *x) / (L* (h-k*root (i) *root (i) ) * (h-
k*root (i) *root (i) ) +root (i) *root (i) * (L+k) +h) /cos (root (i) *L) ;
sum=sum+term_for_sum(i) ;
while (abs(term for sum) /sum>0 . 001 )
root (i ) =qiuun (h,k,L,i) ;
term_for_sum(i) =exp (-dd*root (i) *root (i) *t) * (h-
k*root (i) *root (i) ) *cos (root (i) *x) / (L* (h-k*root (i) *root (i) ) * (h-
k*root (i) *root (i) ) +root (i) *root (i) * (L+k) +h) /cos (root (i) *L) ;
sum=sum+term for sum(i) ;
end
time index=t/time step+1;
y (time_index) =2*cci*sum/kk;
clear sum;
clear term_f or_sum;
clear root;
end
59
-------
APPENDIX F
APPENDIX F: VT QUALITY ASSURANCE
PROJECT PLAN
60
-------
Quality Assurance Project Plan
Developing a Reference Material for Formaldehyde Emissions Testing
(ver. 2)
Submitted to:
Dr. Xiaoyu Liu
U.S. Environmental Protection Agency (E305-03)
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Indoor Environment Management Branch
Research Triangle Park, NC27711, USA
Phone: (919) 541-2459
Fax: (919)541-2157
Email: liu.xiaoyu@epa.gov
Submitted by:
Dr. John Little
Department of Civil and Environmental Engineering
Virginia Tech
Blacksburg, VA 24061-0246
Phone: (540)231-8737
Fax (540)231-7916
Email: jcl@vt.edu
Date: August 30,2011
-------
TITLE AND APPROVAL PAGE
QAPP- Developing a Reference JVlaterial for Formaldehyde Emissions Testing .
Document Title
jQhn_C.__Little,.PhD, Department of Civil and Environmental Engineering, Virginia Tegh
Prepared By
405 Durham Hall. Virginia Tech. Blacksburg. VA 24061: (540) 231-8737
Address and Telephone Number
August 30, 2011
Date
Project Manager:
Signature
C . UrruE
Printed Name/Date
Project QA Officer:
Signature
Printed Name/Date
U . S . EP A Proj ect Manager Approval :
Signature
uiunJU LiU
Printed Name/Date
-------
DISTRIBUTION LIST
QAPP Recipients
Xiaoyu Liu
John C. Little
Julie Petruska
Steven S. Cox
Zhe Liu
Cynthia Howard-
Reed
Charles E. Frazier
Title
EPA Project
Manager
VT Principal
Investigator
VTProjectQA
Officer
VT Team Member
VT Team Member
NIST Team Member
VT Project
Consultant
Organization
U.S. EPA
Virginia Tech
Virginia Tech
Virginia Tech
Virginia Tech
NIST
Virginia Tech
Email Address
liu.xiaoyu@epa.gov
jcl@vt.edu
juniper@vt.edu
stcox2@vt.edu
liuzhe@vt.edu
cynthi a. reed@ni st.gov
cfrazier@vt.edu
-------
TABLE OF CONTENTS
1. PROJECT DESCRIPTION AND OBJECTIVES 1
1.1 Background 1
1.2 Project objectives 3
2. ORGANIZATION AND RESPONSIBILITIES 3
2.1 Personnel 3
2.2 Project schedule 4
3. SCIENTIFIC APPROACH 5
3.1 Analyte of interest and matrices under study 5
3.2 Analytical approach 5
3.2.1 Measure formaldehyde gas stream 5
3.2.2 Infuse formaldehyde and determine Co 6
3.2.3 Determine D and K 7
3.3 Modeling approach 8
3.4 Method performance metrics 9
3.4.1 Measure formaldehyde gas stream 9
3.4.2 Infuse formaldehyde and determine Co 10
3.4.3 Determine D and K 11
3.4.4 Ship and store the reference materials 11
3.4.5 Compare chamber tests and model predictions 12
4. SAMPLING PROCEDURES 12
4.1 Requirements for samples 12
4.2 Preparation of samples 13
4.3 Sample preservation 13
4.4 Sample numbering 14
5. MEASUREMENT PROCEDURES 14
6. METHOD PERFORMANCE METRICS 14
7. DATA ANALYSIS, INTERPRETATION, AND MANAGEMENT 15
7.1 Data reporting requirements 15
7.2 Data validation 15
7.3 Data summarization 15
7.4 Data storage 15
8. REPORTING 16
9. REFERENCES 16
-------
1. PROJECT DESCRIPTION AND OBJECTIVES
1.1 Background
Wood and composite wood products usually emit formaldehyde (Meyer and Boehme,
1997) with emission rates that vary greatly and even unpredictably (Weigl et al., 2009).
Formaldehyde is classified as a known carcinogen and various international guidelines
and recommendations for formaldehyde in indoor air have been established (Salthammer
et al., 2010) including new legislation recently passed by the US Congress and signed by
the President. To demonstrate compliance with these regulations, manufacturers and
independent laboratories conduct formaldehyde emissions testing in chambers with
environmental conditions similar to a real building. Unfortunately, there are substantial
uncertainties involved in chamber measurements, with published inter-laboratory studies
for emissions of volatile organic compounds (VOCs) showing coefficients of variation
between measured emission rates on the order of 50 % and as large as 300 % (Howard-
Reed et al., 2007). For formaldehyde alone, standard emission testing methods vary
substantially in different countries (Risholm-Sundman et al., 2007). Inter-laboratory
studies can be expensive and time-consuming, and may lead to inconclusive results,
especially since there is no way to identify which laboratory's results are correct.
Therefore a well characterized reference material for formaldehyde emissions testing is a
critical prerequisite for improving formaldehyde emissions measurement methods.
yin=o,Q
x=L
x=0
V
y(t)
A ^(x=L) i, yinterface
A Tl
: D
y(t), Q
Figure 1. VOC source in test chamber showing mechanisms and parameters controlling
the emission rate.
In collaboration with the National Institute of Standards and Technology (NIST), the PI
has been developing reference materials for emissions testing of VOCs, whose emissions
can be accurately predicted by a fundamental emission model. The project began with the
development of a reference material for toluene (Cox et al., 2010; Howard-Reed et al.,
2011), and is currently being extended to include n-butanol, a polar and more difficult
VOC to handle experimentally. The emission model on which the reference materials are
based is shown in Figure 1. The mechanisms of VOC emissions from the material include
(1) internal diffusion within the material, characterized by material-phase diffusion
coefficient, D (m2/s), and initial material-phase concentration, Co (g/m3); (2) partition at
the material/air interface, characterized by partition coefficient between the material and
air, K (dimensionless); and (3) convective mass transfer through the boundary layer near
the material surface, characterized by convective mass-transfer coefficient, hm (m/s). But
1
-------
for diffusion controlled materials, external convective mass-transfer resistance is
relatively small compared to that due to internal diffusion and therefore, the emission rate
is largely controlled by internal diffusion and partition at the material/air interface while
convective mass transfer through the boundary layer is negligible. To predict the
emission rate and resulting gas-phase concentration profile in a well-mixed chamber, we
simply need to measure the three parameters (Co, D and K) and apply the emission model
(Cox et al., 2002).
The development of the reference material for toluene has been described in detail by
Cox et al. (2010) and Howard Reed et al. (2011), and involves the following basic steps:
(1) polymethyl pentene (PMP) was selected as the polymer substrate; (2) thin-film
samples of the PMP were loaded with toluene using small loading vessels held at a
constant gas-phase toluene concentration (the toluene diffused into the polymer film until
the material-phase toluene concentration in the polymer reached equilibrium with the
gas-phase toluene concentration in the vessels); (3) the mass-transfer properties for
toluene/PMP (the material-phase diffusion coefficient, Z), the material/air partition
coefficient, K, and the material-phase toluene concentration of the loaded samples, Co)
were measured gravimetrically using a recording microbalance; (4) a small well-mixed
environmental chamber was used to measure gas-phase toluene emissions from the pre-
loaded PMP films; (5) a fundamental emissions model was used together with the
independently measured toluene/PMP parameters (Co, K and D), the dimensions of the
PMP film (thickness, L, and surface area, A) and the chamber operating configuration (air
flow rate, Q, and chamber volume, V) to predict the toluene concentration profile; and, (6)
the measured emissions profiles were compared to the predicted emissions profile.
The initial results for toluene and n-butanol are very promising and a similar procedure is
proposed for the development of a reference material for formaldehyde. This will involve
the following experimental steps: (1) create a constant gas-phase concentration using a
diffusion vial containing paraformaldehyde (which depolymerizes into formaldehyde
(Rock et al., 2010)) so that PMP films can be loaded with formaldehyde in a small
loading vessel, (2) measure the formaldehyde/PMP parameters (Co, K and D) using a
microbalance, (3) test the reference material by placing the pre-loaded PMP samples into
a small emissions chamber and comparing the measured formaldehyde concentration
profile with that predicted by the model. The success of this method depends on two key
criteria: (a) the formaldehyde needs to be sufficiently "soluble" in the PMP (a large
enough K) so that Co is high enough for the formaldehyde emissions profile to be at the
desired concentration; and (b) the diffusion of formaldehyde in the PMP needs to be ideal
or "Fickian" in nature.
Although monomeric formaldehyde should be somewhat soluble in the PMP polymer
matrix, more will dissolve into polar matrices such as nylon, polyacrylamide,
polycarbonate (PC) and polyvinyl chloride (PVC), as reported by Hennebert (1988).
While the -OH or -NH groups in nylon and polyacrylamide would react with the
formaldehyde monomers by forming a -QHbOH group, PC and PVC should not be
reactive. If PMP does not absorb sufficient formaldehyde, PC and PVC represent viable
alternatives, and the fact that it is not reactive, suggests that diffusion in PC and PVC
-------
may be ideal (Hennebert, 1988). We will focus on PMP and PC in the development of the
formaldehyde reference material. Because the presence of humidity creates the potential
for formaldehyde to polymerize, we will initially use dry conditions in developing and
deploying the reference material. We are optimistic that this will enable emissions to be
predicted using our simple modeling approach, as proven for toluene in PMP. Although
humidity does not appear to affect either the toluene/PMP or butanol/PMP systems, it
may be that the presence of humidity will cause formaldehyde to behave in a non-ideal
fashion, given that it tends to promote polymerization. We will therefore do a careful
check of the influence of humidity on the performance of the formaldehyde reference
materials (both PMP, which absorbs little water, and PC, which is expected to absorb a
fair amount of water). It may be that we will have to change to a non-ideal diffusion
mechanism to predict the emission rate of formaldehyde in the presence of water vapor.
Either way, the emissions process should still be highly reproducible, which is the
essential nature of a reference material.
1.2 Proj ect obj ectives
The purpose of this project is to develop a reference material for formaldehyde using
similar procedure to those we have developed for reference materials for toluene and n-
butanol. Our specific research objectives are to:
1) Develop a procedure to create a range of constant gas-phase concentrations of
formaldehyde using solid paraformaldehyde (which depolymerizes into
formaldehyde) in temperature controlled diffusion vials;
2) Investigate the potential for two polymer materials (PMP and PC) to be used as
reference materials for formaldehyde by measuring the diffusion coefficient (D)
and partition coefficient (K) in a microbalance over a range of gas-phase
concentrations;
3) Determine the impact of humidity on the mass transfer of formaldehyde in
PMP/PC using microbalance sorption/desorption tests;
4) Develop the test protocol and test the overall performance of the reference
materials. This will be performed in two steps: firstly, EPA and VT will establish
a test protocol for the reference material and then PMP/formaldehyde and
PC/formaldehyde will be tested by measuring formaldehyde emissions in small
chambers at EPA and comparing the observed concentration profiles with model
predicted values; and secondly, emission chamber tests will be conducted at both
EPA and NIST to further evaluate the test protocol and the performance of the
prototype reference materials.
2. ORGANIZATION AND RESPONSIBILITIES
2.1 Personnel
Dr. John Little will serve as the principal investigator (PI) at Virginia Tech (VT) and is
responsible for leading the work at VT, and submitting monthly reports documenting
progress and the final project report. Dr. Julie Petruska, the Environmental Laboratory
Supervisor for the Environmental Engineering Program at VT, will serve as the VT
project QA Officer and ensure that the project is implemented according to the QAPP and
that the data collected meet project objectives. Dr. Steven Cox, a Senior Research
Associate at VT, will be responsible for laboratory infrastructure and will direct and
-------
supervise the overall laboratory safety and the quality of the analytical and experimental
data. Mr. Zhe Liu, a Graduate Research Assistant, will be responsible for the day-to-day
activities including carrying out the experimental work, analyzing experimental data,
developing models and preparing the draft report summarizing the results. Dr. Little will
meet with Dr. Cox, Dr. Petruska and Mr. Liu weekly to review progress and discuss the
results. In addition, Dr. Charles Frazier, Professor of Wood Science and Forest Projects
and Director of the Virginia Tech Wood-Based Composites Center (a National Science
Foundation Industry/University Cooperative Research Center), will serve as a consultant
to this project, providing advice on polymer properties and formaldehyde chemistry, as
well as acting as liaison with the representatives of the Wood-Based Composites Industry.
In addition to the project team at VT, a team at EPA will work with the VT team for
object 4 to develop a test protocol for the reference material and test the
PMP/formaldehyde and PC/formaldehyde in small emission chambers. The EPA team
will be responsible for conducting emission tests, analyzing the emission testing data, and
evaluating the test protocol. A team at NIST will be then involved to test the prototype
reference materials and NIST team will report data to the EPA team for data analysis.
During the project period working on objective 4, a conference call will be held weekly
between the VT and EPA teams to discuss the test protocol, emissions testing plans, and
testing results. The following project organizational chart illustrates the group hierarchy.
EPA Proj ect Manager
XiaoyuLiu
I
VT Principal Investigator
John Little
VT Proj ect QA Officer
Julie Pestruska
VT Project Consultant
Charles Frazier
I
VT Team Member
Steven Cox
VT Team Member
Zhe Liu
EPA Team
NIST Team
2.2 Project schedule
The following chart shows the project schedule in terms of the specific project objectives
outlined in section 1.2. The project duration is from June 1, 2011 to February 15, 2012.
Tasks
1
2
3
4
Qtr 1 Qtr 2 Qtr 3
Dr. Little (PI) and the VT team plan to visit EPA and NIST during the project period. It is
also anticipated that representatives from EPA and NIST will visit VT during the project
-------
to further familiarize themselves with the procedures being used at VT. Dr. Little will be
responsible for submitting monthly reports documenting progress. Dr. Little will ensure
completing the development of the formaldehyde reference material and submitting the
preliminary results to the EPA Project Officer by January 15, 2012. A final project report
that includes a data quality review will be submitted to the Project Officer by February 15,
2012.
3. SCIENTIFIC APPROACH
3.1 Analyte of interest and matrices under study
To load either PMP or PC with formaldehyde, we will create a continuous gas stream
with constant formaldehyde concentrations using a dynamic formaldehyde generation
system (objective 1). The system consists of diffusion vials maintained at elevated
temperatures with purge flow and dilution flow regulated by mass-flow controllers
(accurate flow rate confirmed by bubble meters). Solid paraformaldehyde contained in
the diffusion vials depolymerizes into monomeric formaldehyde and enters the purge gas
(dry and clean air). The purge gas containing formaldehyde is then mixed with dilution
air. The concentration of formaldehyde in the gas stream is adjusted by controlling the
temperature of the diffusion vials and the flow rate of the dilution gas. The humidity of
the gas stream will be adjusted by changing the humidity of the dilution air flow (e.g. 0%
RH for/) and K determination in objective 2 and infusing formaldehyde into PMP/PC in
objective 4; 50% RH for objective 3). This gas stream will then pass through PMP and
PC samples in loading vessels to infuse formaldehyde into the PMP and PC films. It will
also enable us to conduct sorption/desorption tests using the microbalance system to
determine D and K.
PMP and PC films will be purchased without any additives. Their purity will be checked
by a dynamic microbalance. Before each test, dry and clean air will be passed through the
microbalance on which the samples are suspended for at least 24 hours to check for
volatile contaminants. All the samples will be obtained from a single roll/batch of
material from the manufacturers to ensure they are identical and uniform.
3.2 Analytical approach
3.2.1 Measure formaldehyde gas stream
The gas-phase formaldehyde concentration from the dynamic formaldehyde generation
system will be determined by three different approaches: gravimetric method, visible
absorption spectrometry, and electrochemical sensor. Their principles and procedures are
described below. Although the first two measurement methods are not continuous, the
gravimetric method is expected to provide a quantifiable release rate from the diffusion
vials while the spectrometry method can provide direct and accurate measurement of the
formaldehyde concentration in the gas stream. The results of these two methods will be
compared with each other and used to assess the performance of the dynamic
formaldehyde generation system. The continuous monitoring data employing an
electrochemical sensor will confirm the consistency of the concentration throughout the
test periods.
-------
Gravimetric method: The diffusion vials containing paraformaldehyde will be weighed
by a high-resolution electronic balance over appropriate time intervals to determine the
release rate of formaldehyde. The gas-phase formaldehyde concentration from the
dynamic generation system can thus be calculated by dividing the release rate by the total
flow rate of purge flow and dilution flow.
Visible absorption spectrometry: The formaldehyde concentration in the gas stream from
the dynamic generation system will be measured directly by visible absorption
spectrometry, following the NIOSH Analytical Method 3500. Briefly, an appropriate
volume of gas stream will be pumped through two impingers in series containing 20 mL
1% sodium bisulfite solution so that gas-phase formaldehyde will be completely absorbed
by the aqueous solution (the backup impinger will be used to check the collection
efficiency); then 4-mL aliquots of impinger solution will be transferred to a flask and 0.1
mL 1% chromotropic acid and 6 mL concentrated sulfuric acid will be added to the
sample solution; the sample solution will heated at 95°C for 15 min and then maintained
at room temperature for 2 hours so that the chromophore can be fully developed; and
finally the absorbance at 580 nm of the sample solution will be measured using a
spectrophotometer. Meanwhile, at least six calibration standards with different known
formaldehyde concentrations and a blank will be treated with the reagents and analyzed
by the spectrophotometer for absorbance at 580 nm. Therefore, a calibration curve
(absorbance versus formaldehyde concentration in the sample solution) can be
constructed and the formaldehyde concentration in the tested solution sample can be
obtained from the calibration curve. Finally, formaldehyde concentration in the gas
stream can be derived using appropriate aliquot factor and gas sample volume.
Electrochemical sensor: The gas stream is continuously monitored using commercially
available formaldehyde detection and measurement instrument based on electrochemical
sensors (Formaldehyde Meter Z-300XP, Environmental Sensors Co., FL). The measured
data of formaldehyde concentration in the gas stream will be recorded automatically.
3.2.2 Infuse formaldehyde and determine Co
To infuse formaldehyde into PMP and PC samples, formaldehyde gas stream from the
dynamic generation system will be passed into a stainless steel loading vessel, with
several PMP or PC samples secured on stainless steel screen fixtures (Figure 2). Gas-
phase formaldehyde in the gas stream will diffuse into the samples until sorption
equilibrium is reached between the material phase and the gas phase. To determine the
material-phase concentration of the samples, a microbalance is connected downstream
with effluent from the loading vessel passing across an additional PMP or PC sample on
the microbalance. Therefore, the material-phase concentration of formaldehyde at the end
of the loading process (Co) can be determined gravimetrically using the final mass
increase of the extra sample and then dividing by its volume. The loading vessels will be
maintained at 23 °C to eliminate variations in Co caused by temperature.
-------
Microbalance
Exhaust
Loading Vessel
— I
PMP
Sample
,r
4
Inlet
Figure 2. Infusing formaldehyde in loading vessels and measuring Co
Desorption Test:
Clean air
Sorption Test:
Air with CH7O
"1 ^-
>o
><
Ph
IP
Samples
1
Tare
Weight
Figure 3. Microbalance sorption/desorption tests
3.2.3 Determine D and K
To determine D and K, a well-developed microbalance method will be employed (Cox et
al., 200la). As shown in Figure 3, a test sample (a PMP film here) is continuously
measured gravimetrically by a microbalance. During the sorption test when the air stream
containing formaldehyde (from the dynamic formaldehyde generation system) is passed
across the clean sample, formaldehyde diffuses into the material and the gain in mass of
the sample is recorded by the microbalance, generating a sorption curve. Once the sample
has reached equilibrium with formaldehyde in the gas stream, then clean air is passed
through the sample for the desorption test and a desorption curve is generated by
measuring the mass loss from the sample over time. The sorption/desorption tests will be
performed under at least three different formaldehyde concentrations.
-------
1000
00
^r 800
o
A yo=l .6 g/m
D y0=l g/m3
O y =0.5 g/m3
600-
a
o
O
8
£ 400-
Model (3.6xlO"14m2/s)
yo=1.6g/m3
D y0=l g/m3
O y =0.5 g/m3
0.5
150
200
Gas Phase Concentration (g/m ) Time (h)
(a) (b)
Figure 4. Determining K and D using microbalance data (toluene), (a) linear regression of
material-phase concentration and gas-phase concentration in equilibrium to determine K
(b) diffusion model fitted to sorption/desorption data to determine D
As shown in Figure 4(a) for toluene, a linear correlation can be constructed for the gas-
phase formaldehyde concentration in the gas stream and the material-phase concentration
in equilibrium, and the slope of the linear regression line is K. D will be determined by
fitting a Fickian diffusion model to the sorption and desorption data. Under the
experimental conditions, the mass change caused by Fickian diffusion of formaldehyde
into the film is given by (Crank, 1975):
—L = i~X r~2'exP\——2 —r 0)
CO ^ J
where Mt (mg) is the total formaldehyde mass that has entered or left the film in time t (s),
Moo (nig) is the formaldehyde mass in the film when air-phase/material-phase partition
equilibrium is reached, and 2L (m) is the film thickness. Figure 4(b) shows the example
of fitting the Fickian diffusion model to the normalized sorption/desorption data of
toluene to obtain D.
All the sorption/desorption tests will be carried out at 23 °C to eliminate temperature
dependence ofD and K. The sorption/desorption tests will be carried out first at 0% RH
for determining D and K under dry condition (objective 2), and then at 50% RH to
evaluate the impacts of humidity on mass transfer (objective 3).
3.3 Modeling approach
The fundamental model describing VOC emissions from a homogeneous, diffusion-
controlled source is briefly reviewed. Figure 1 shows the mechanisms governing
emissions of VOCs from a material source in a test chamber. If we assume that the
external convective mass-transfer rate is fast, the chamber wall sink-effect is negligible,
and that the initial material-phase VOC concentration is uniform with depth (Co) then we
can develop a fundamental emission model to predict the chamber concentration. The
transient diffusion equation in the material slab is
-------
~.^ ~\1^l
rj( rj (
^-°W <2)
where C (mg/m3) is the material-phase concentration of a VOC, D (m2/s) is the material-
phase diffusion coefficient, t (s) is time, and x (m) is distance from the base of the slab.
The initial condition assumes a uniform material-phase concentration of the VOC in the
slab, Co (mg/m3). The first boundary condition assumes there is no flux from the base of
the slab. The second boundary condition is imposed via a mass balance on the VOC in
the chamber air, or
— •V = Q-ym-D — -Q-y (3)
^ •*-•• •/ in ^ •*-- •/ \ /
dt dx X=L
where yin (|ig/m3) and y (|ig/m3) are the concentrations of the VOC in the influent and
chamber air respectively, Q (m3/s) is the volumetric air flow rate, V (m3) is the well-
mixed chamber volume, A (m2) is the exposed surface area of the slab, and L (m) is the
thickness of the slab. A linear and instantaneously reversible equilibrium relationship is
assumed to exist between the slab surface and the chamber air, or
C
K = ^±- (4)
y
where K (dimensionless) is a material/air partition coefficient with units of mass per
volume/mass per volume. The instantaneously reversible assumption implies that
resistance to mass transfer between the material surface and the bulk chamber air is
negligible, which has been shown to be the case for the PMP/toluene system (Cox et al.,
2010). Assuming yin is zero and D and K are independent of concentration, an analytical
solution to these equations was given by Little et al. (1994):
(5)
[Lh-kq+qL + k+h}cosqnL
where
* = -2- (6)
ADK
k = — (7)
AK
and the qn are the roots of
qntan(qnL) = h-kq2n (8)
When material-phase concentration over time is given by equation (5), gas concentration
is the chamber can be simply obtained by equation (4). This model has proven to be
effective for predicting emissions of VOCs from vinyl flooring (Cox et al., 2002). The
three key model parameters (Co, K, and D) were measured completely independently of
the chamber experiments (Cox et al., 2001a; Cox et al., 2001b).
3.4 Method performance metrics
3.4.1 Measure formaldehyde gas stream
As described in section 3.2.1, a continuous gas stream with constant formaldehyde
concentration will be generated from a dynamic generation system. Three different
approaches, including gravimetric method, visible absorption spectrometry, and
-------
electrochemical sensor, will be used to determine the formaldehyde concentration and
evaluate the performance of the dynamic generation system.
The gravimetric method will be performed over appropriate time intervals to determine
the release rate of formaldehyde from each diffusion vial in the dynamic generation
system. A high-resolution electronic balance (±10 jig) will be used for the gravimetric
method. The electronic balance will be externally calibrated every day and maintained
according to its manual. Briefly, the calibration procedure includes setting the zero and
establishing the full capacity weight for the balance using certified calibration weights
provided by the manufacturer of the balance. For each generation period during which
gas stream with a constant formaldehyde concentration is generated continuously, the
diffusion vial will be weighed at least five times, e.g. one measurement at the beginning
and one at the end, and three during the period. Since the weight decrease rate is exactly
the formaldehyde release rate from the diffusion vial, the linearity between the measured
weight and time will be examined to evaluate whether the release rate is constant during
the period. The average release rate (slope of the linear regression line for weight versus
time) will be used to calculate the average formaldehyde concentration during the period.
The spectrometry method will be used to measure the formaldehyde concentration of the
gas stream directly. It will be carried out at least five times during each generation period,
e.g. one measurement at the beginning and one at the end, and three during the period.
The visible absorption spectrometry standard method (NIOSH Analytical Method 3500)
will be followed strictly and quality control procedures will be performed. For example,
dual impingers will be used in series to absorb gas-phase formaldehyde to ensure
efficient collection of formaldehyde; and at least six working standards and blanks will
be tested by the spectrophotometer every time to construct the calibration line. Since the
purge gas and dilution gas is clean air, there will be no oxidizable organic compounds
other than formaldehyde in the gas stream and therefore no interferences for the
spectrometry measurement.
The gas stream will also be continuously monitored using a commercially available
formaldehyde detection and measurement instrument based on electrochemical sensors
(Formaldehyde Meter Z-300XP, Environmental Sensors Co., FL). The instrument will be
maintained according to its manual and sent to the manufacturer for calibration using
standard calibration gas routinely (every few months).
3.4.2 Infuse formaldehyde and determine Co
As shown in Figure 2, PMP and PC samples will be loaded with formaldehyde in
stainless steel loading vessels and the mass change of a representative sample during the
loading process will be continuously monitored by a high-resolution microbalance (±0.1
jig). The mass of the representative sample will be measured and recorded every 5
minutes throughout the loading process. Although gas streams with different
formaldehyde concentration can be generated from the dynamic generation system, a
high concentration gas stream is preferred for loading to achieve greater mass gain of the
samples and higher material-phase concentration of formaldehyde (Co). Greater mass
gain can be more accurately measured by the microbalance (with high signal-to-noise
10
-------
ratio) and higher Co leads to higher emission rates of the reference materials, facilitating
the measurement of formaldehyde in the emission tests. For the same reasons and better
control on the loading process, loading will last until partition equilibrium has been
reached between the materials and loading gas stream, i.e., after the mass measured by
the microbalance becomes stable.
3.4.3 Determine D and K
As described in section 3.2.3, microbalance sorption/desorption tests will be carried out
to determine D and K for PMP and PC. The mass of samples will be measured and
recorded every 5 minutes during the tests. As for the toluene case shown in Figure 4,
sorption/desorption tests will be performed under at least three different concentration
levels and the sorption/desorption cycle will be repeated three times as replicates under
each concentration level. As shown in Figure 4(a), the linearity between gas-phase
formaldehyde concentration and corresponding material-phase concentration in
equilibrium will be examined statistically to test the assumption that the simple linear
sorption isotherm (equation 4) is applicable for PMP/formaldehyde and PC/formaldehyde
system and that K is constant within the tested concentration range. From the linear
regression, the uncertainty of K will also be evaluated. As Figure 4(b) shows, the Fickian
diffusion model will be fitted to the sorption/desorption data under different
concentration levels. The assumption that D is constant over the tested concentration
range will be examined statistically. Uncertainty ofD will also be estimated from the
model fitting.
3.4.4 Ship and store the reference materials
The validity and performance of the formaldehyde reference materials can only be tested
in very rigorous emission chamber tests. For this purpose, loaded PMP/PC samples
infused with known amounts of formaldehyde will be sent to EPA and later to NIST for
emission chamber tests. To prevent or minimize the potential loss of formaldehyde
during the shipping and handling procedures, three packaging approaches will be tried
and tested: sealed aluminum zip bags, a cryogenic package, and an equilibrium container.
Sealed aluminum zip bag: Each loaded PMP/PC sample will be removed rapidly from the
loading vessels and placed in a small sealed aluminum zip bag. Air will be evacuated
from the bag before sealing the bag to minimize the headspace inside. The package will
be delivered by express mail to EPA and NIST and then stored at room temperature prior
to the emission tests.
Cryogenic package: The aluminum zip bags containing PMP/PC samples will be put into
coolers with dry ice and then shipped to EPA and NIST. When received, samples in zip
bags will be stored in freezers at -20°C until the emission tests. Low temperature during
the shipping and storage period will reduce D and increase K of formaldehyde
significantly and this reduces the volatile loss of formaldehyde from the samples.
However, low temperature may also change the properties of the PMP/PC matrices and
promote polymerization of formaldehyde.
11
-------
Equilibrium container: Each PMP/PC sample will be placed in a sealed aluminum case
which is filled with air containing formaldehyde (from the dynamic generation system)
and kept at ambient temperature during the shipping. When received, the sample will be
retained in the original package and maintained for a period at 23 °C. Because the
material-phase concentration of the sample is always at equilibrium with the gas-phase as
in the loading vessel, it should remain in its original condition when taken out for the
emission tests. The equilibrium container also eliminates cryogenic conditions.
3.4.5 Compare chamber tests and model predictions
A detailed test protocol for reference material will be developed by EPA and VT. The
PMP/formaldehyde and PC/formaldehyde will be tested by EPA in small emission
chambers following the test protocol. After the test protocol is developed by EPA and VT,
the reference materials will also send to NIST for testing and comparison. At least three
batches of PMP and three batches of PC will be produced by VT team with six identical
samples in a single batch (three for NIST and three for EPA with different packaging
approaches). PMP/PC samples will be tested in small-scale chambers for emissions at
NIST and EPA. Emission tests will be carried out following standard test method for
formaldehyde emissions from wood products (ASTM D 6007-02, 2008) and the test
protocol developed by EPA and VT. Chamber air samples will be collected at 0.5 h, 1 h,
2 h, 4 h, 8h, 24h, 32 h, 48h, 54 h, and 72 h in each test, with a minimum of 3 duplicate air
samples. Other procedures and chamber operating configurations, such as chamber
volume, chamber airflow rate and mixing fan will not be specified to validate the
reference material under various chamber configurations. The chamber test results will be
analyzed by the EPA team to determine within-batch and between-batch variations, and
evaluate the test protocol.
Model prediction will be performed for each chamber test, with parameters obtained from
chamber tests (Q, V, and A) or microbalance data (Z), K and Co). Uncertainties associated
with each model parameters will be taken into account in the prediction using the Monte
Carlo Method (Cox et al., 2010). The chamber concentrations predicted by the model will
be statistically compared (for example, using paired t-test) to the measured concentrations
in corresponding chamber test to determine whether the model prediction matches the
experimental results. Furthermore, the comparison between model predictions and
chamber test results will help to diagnose problems and uncertainties in the development
of the formaldehyde reference material.
4. SAMPLING PROCEDURES
4.1 Requirements for samples
Samples in this project may involve various matrices in different processes, including:
gas samples of formaldehyde from the dynamic generation system (throughout the entire
project); PMP/PC samples for microbalance sorption/desorption tests (objectives 2 and 3);
and PMP/PC samples loaded with formaldehyde for chamber tests (objective 4).
Gas samples of formaldehyde from the dynamic generation system: The gas samples are
obtained from the continuous stream from the dynamic generation system and passed
through impingers for determining the gas-phase concentration of formaldehyde in the
12
-------
gas stream (see section 3.2.1, visible absorption spectrometry method for determining
formaldehyde concentration). As described in section 3.4.1, at least five samples will be
obtained during each generation period. The volume of each gas sample will be
calculated by gas stream flow rate and collection time. The concentration of
formaldehyde in the samples is expected to be relatively constant during a single
generation period but will vary significantly in different generation periods (from 100
ppm to 1000 ppm level).
PMP/PC samples for microbalance sorption/desorption tests: PMP/PC samples will be
obtained from a single roll/batch of PMP/PC purchased from manufacturers without any
additives. For consistency, samples will have a uniform size (for example, 3.6 cm x 3.6
cm x 0.0254 cm of PMP will be used). Before being tested, each sample will be put on
the microbalance and swept by clean air until its weight is stable (indicating no volatile
contaminants remaining in it). During sorption/desorption tests, the mass change of each
sample will be monitored by the microbalance.
PMP/PC samples loaded with formaldehyde for chamber tests: PMP/PC samples will be
obtained from the same single roll/batch as those for sorption/desorption tests. The size of
each sample can vary to accommodate specific requirements of the chamber tests. The
formaldehyde concentration in the samples is proportional to the gas-phase concentration
of formaldehyde in the gas stream for loading but a high concentration is preferred to
facilitate the accurate determination of Co and the subsequent emission chamber tests.
4.2 Preparation of samples
As detailed in section 3.1 and 4.1, gas samples of formaldehyde are obtained and
processed in the laboratory from the continuous gas stream and the dynamic generation
system. As in section 4.1, PC/PMP samples are obtained from a single roll/batch of
pristine materials purchased from the manufacturer and then prepared and processed in
the laboratory (infused with formaldehyde). The loaded PC/PMP samples need to be sent
to EPA and NIST for chamber tests according to the procedures described in section 3.4.4.
4.3 Sample preservation
As detailed in section 3.1 and 4.1, the continuous gas stream containing formaldehyde is
generated from the dynamic generation system. It will be instantly measured for
formaldehyde concentration (see section 3.2.1) or used (passed to loading vessels for
infusing formaldehyde into PMP/PC or to microbalance for sorption/desorption tests). No
preservation is required.
Loaded PMP/PC samples will be packed and sent to EPA and NIST once removed from
the loading vessels. The shipping and storage requirements are described in section 3.4.4.
In addition, samples will be tested in emission chambers as soon as possible after being
received by EPA and NIST to minimize shelf-life and potential volatile loss of
formaldehyde.
4.4 Sample numbering
13
-------
Gas samples of formaldehyde from the dynamic generation system will be labeled
according to specific generation information, sampling time, and sample volume.
PMP/PC samples for microbalance sorption/desorption tests will be numbered according
to specific testing information and time. Loaded PMP/PC samples will be labeled
according to the loading information (time and gas-phase concentration of formaldehyde),
positions in loading vessels, and shipping and storage procedures.
5. MEASUREMENT PROCEDURES
Measurements involved in this project mainly include: gas-phase concentration of
formaldehyde from the dynamic generation system (section 3.2.1); and weight of
PMP/PC samples by microbalance for determining D, K and Co (sections 3.2.3 and 3.3.3).
The sample preparation, calibration, measurement and quality control procedures for
visible absorption spectrometry will be carried out following the NIOSH Analytical
Method 3500. Other measurements require routine laboratory procedures and instruments
(for example, bubble meters for gas flow rate, electronic balance for weight of diffusion
vials, electrochemical sensors for direct formaldehyde measurement, and microbalance
for PMP/PC mass), which are generally calibrated and operated according to product
manuals.
6. METHOD PERFORMANCE METRICS
Method performance metrics have been provided in section 3.4. The two most important
steps involved in the reference material development are: determining whether constant
formaldehyde concentrations can be generated from the dynamic generation system
(section 3.4.1); and whether sorption/desorption of formaldehyde in PMP/PC is governed
by Fickian diffusion (section 3.4.3). These are the two key challenges that we have to
address in this project. Formaldehyde concentration generated from the dynamic
generation system will be measured by three methods over a period of time and the
concentration measurements by each method will be statistically tested to determine
whether the concentration varies over time. For example, the linearity of the weight of
diffusion vial will be checked by assessing the R2 value of the linear regression, i.e., if R2
is larger than 0.99, the release rate from the diffusion vial is regarded constant and the
formaldehyde concentration can be calculated from the slope of the linear regression line.
The direct measurement of formaldehyde concentration data will be compared to the
value calculated from the release rate of the diffusion vial to determine whether they are
statistically equal at significance level of 0.05 (t-test). If significant variation of
concentration occurs or significant difference is found between results by two methods,
the system will need modification and improvement until the requirement of constant
concentration is met. The second challenge will be evaluated by fitting the Fickian
diffusion model to the microbalance sorption/desorption data (as in Figure 4 for toluene).
According to ASTM Standard D5157-97, a correlation coefficient (R) of 0.9 or greater
generally indicates adequate model performance and it is therefore regarded that the
sorption/desorption data can be described by Fickian diffusion principles. If the
sorption/desorption data do not match the Fickian diffusion model statistically, implying
that mass transfer of formaldehyde in PMP/PC does not follow Fickian diffusion, we will
have to resort to other approaches or other polymer substrates.
14
-------
In validating the reference material using emission chamber tests, two criteria should be
met. Firstly, the same reference material samples should generate the same emission
profiles under identical chamber testing configurations (chamber concentration at a
certain time should be the same across all the tests). This repeatability will be tested by
comparing the measured chamber concentration of different runs using paired t-test.
Secondly, the model should be able to predict the emission test results reasonable well.
Paired t-test can be used to determine whether model predicted chamber concentration is
equal to measured value but a maximum percent difference of 10% between the model
predicted concentration and measured value is also acceptable.
7. DATA ANALYSIS, INTERPRETATION, AND MANAGEMENT
7.1 Data reporting requirements
The primary parameters to be determined in this study include diffusion coefficient (D),
the partition coefficient (K), and the initial material-phase concentration (Co) for
developed formaldehyde reference materials (with PMP and PC as substrate). The gas-
phase concentration of formaldehyde under which PMP/PC substrate samples are loaded
and microbalance sorption/desorption tests are performed will be also reported. SI units
will be used throughout this project.
Data reduction procedures mainly involve determination of/), K and Co. D and K will be
determined from microbalance sorption/desorption data and detailed methods and
equations have been provided in section 3.2.3. Co will be determined from the total mass
gain measured by microbalance, which is then divided by sample volume (section 3.2.2).
7.2 Data validation
The gas-phase concentration of formaldehyde under which PMP/PC substrate samples
are loaded and microbalance sorption/desorption tests are performed will be measured by
three methods to ensure accuracy and reliability (section 3.2.1). D, K and Co will be used
as model parameters to predict emission profiles of formaldehyde reference materials in
emission chamber tests (section 3.3). The model predictions will be compared to
emission chamber results to validate the values of/), K and Co.
7.3 Data summarization
During the determination of/), K and Co, associated uncertainties will be analyzed. Mean
and standard error of/), K and Co will be reported.
7.4 Data storage
A dedicated laboratory notebook will be maintained for all the experimental effort. All
the data will be transcribed to EXCEL spreadsheets each day that they are generated. The
EXCEL files will be backed-up to electronic media on a weekly basis and also sent to Dr.
Little for separate storage on a weekly basis.
8. REPORTING
Monthly reports documenting progress will be submitted during the projection period.
Detailed methodologies and preliminary results will be reported in a final report to EPA.
Quality assurance results will be included as an appendix to the final report. During the
15
-------
project period for objective 4, a conference call will be planned weekly to discuss
shipping reference materials issues and chamber tests with the EPA team.
In addition to the final report to EPA, the methodologies and results will be summarized
into one or two manuscripts for journal publication.
9. REFERENCES
ASTM (2008) Standard Test Method for Determining Formaldehyde Concentrations in
Air from Wood Products Using a Small-scale Chamber (ASTMD 6007-02,
reapproved2008). West Conshohocken, PA: ASTM International.
Cox, S.S., Hodgson, A.T. and Little, J.C. (2001b) Measuring concentrations of volatile
organic compounds in vinyl flooring. Journal of Air and Waste Management
Association. 51: 1195-1201.
Cox, S.S., Little, J.C. and Hodgson, A.T. (2002) Predicting the emission rate of volatile
organic compounds from vinyl flooring. Environmental Science and Technology,
36: 709-714.
Cox, S.S., Liu, Z., Little, J.C., Howard-Reed, C., Nabinger, S.J. and Persily, A. (2010)
Diffusion-controlled reference material for VOC emissions testing: proof of concept.
Indoor Air. 20:424-433.
Cox, S.S., Zhao, D. and Little, J.C. (2001a) Measuring partition and diffusion coefficients
for volatile organic compounds in vinyl flooring. Atmospheric Environment, 35:
3823-3830.
Crank, J. (1975) The Mathematics of Diffusion, 2nd Edition. Oxford University Press,
New York.
Hennebert, P. (1988) Solubility and diffusion coefficients of gaseous formaldehyde in
polymers. Biomaterials, 9: 162-167.
Howard-Reed, C., Little, J.C., Marand, E., Cox, S.S., Nabinger, S.J. and Persily, A. (2007)
"Improving the reliability of VOC emissions testing of building products" In:
Proceedings ofASHRAEIAQ 2007, Baltimore, American Society of Heating,
Refrigerating and Air-Conditioning Engineers (ASHRAE)' conference.
Howard-Reed, C., Liu, Z., Benning, J., Cox, S.S., Samarov, D., Leber, D., Hodgson, A.T.,
Mason, S., Won, D. and Little, J.C. (2011) Diffusion-controlled reference material
for volatile organic compound emissions testing: pilot inter-laboratory study.
Building and Environment, 46: 1504-1511.
Little, J.C., Hodgson, A.T. and Gadgil, A.J. (1994) Modeling emissions of volatile
organic compounds from new carpets. Atmospheric Environment, 28: 227-234.
Meyer, B. and Boehme, C. (1997) Formaldehyde emission from solid wood. Forest
Products Journal, 47: 45-48.
Risholm-Sundman, M., Larsen, A., Vestin, E. and Weibull, A. (2007) Formaldehyde
emission-comparison of different standard methods. Atmospheric Environment, 41:
3193-3202.
Rock, F., Barsan, N. and Weimar, U. (2010) System for dosing formaldehyde vapor at
the ppb level. Measurement Science and Technology, 21(11): 115201-115207.
Salthammer, T., Mentese, S. and Marutzky, R. (2010) Formaldehyde in the indoor
environment. Chemical Reviews, 110: 2536-2572.
16
-------
Weigl, M., Wimmer, R., Sykacek, E. and Steinwender, M. (2009) Wood-borne
formaldehyde varying with species, wood grade, and cambial age. Forest Products
Journal 59: 88-92.
17
-------
APPENDIX G
APPENDIX G: EPA QUALITY ASSURANCE
PROJECT PLAN
61
-------
A, :;x,r :,1 jV'i
FWl :' ' ' » ."''-•* . •-•
Imagine the result
&EPA
Green Building Research: Indoor
Source Emissions and Sink Effect
Study of Formaldehyde
Quality Assurance Project Plan
Category III / Measurement Project
Revision: A4-1.1
QAPP Addendum 4: July 2012
Procedure for Evaluation of Reference Material
for Formaldehyde Emissions Testing using
Small Environmental Chambers
-------
Green Building Research:
Indoor Source Emissions and
Sink Effect Study of
Formaldehyde
Quality Assurance Project Plan
Addendum 4-1.1
Category III / Measurement
/V i / •
f /M^^l L.i>
Xiaoyu Liu, PhD
U.S. EnvironmerMal Protection Agency
Work Assignment Manager
Date
Prepared for
Prepared for;
U.S. Environmental Protection Agency
Air Pollution Prevention and Control
Division (APPCD)
Research Triangle Park, HC 27711
7/12/2012
Nancy Roache
ARCADIS U.S., Inc
Work Assignment Leader
Date
Robert Wright
U.S. Environmental Protection Agency
Quality Assurance Representative
ionAc
Date
Laura Beach Nessley
ARCADIS U.S., Inc.
Quality Assurance Officer
8/15/2012
Date
Prepared by:
ARCAD1S U.S., Inc.
4915 Prospectus Drive
Suite F
Durham
North Carolina 27713
Tel 919 544 4535
Fax 919 544 5690
Our Rel:
RN990273.0015
Date:
July 2012
This document is intended only for the use
of the individual or entity for which it was
prepared and may contain information that
is privileged, confidential, and exempt from
disclosure under applicable law. Any
dissemination, distribution, or copying of
this document is strictly prohibited.
Project
Revision: A4-1.1
-------
Document Revisions
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: i
Version
Author
Date
Description
1.0
EPA
7/21/2008
Initial release
1.1
EPA
10/27/2008
Revision:
Editorial correction based on QA comments;
Order of tests (Table 3-3, 3-7);
Number of pieces of wallboard for sink tests (Section 3.1.4);
Sampling location for large chamber baby crib tests (Section
3.3.2);
Test schedule;
Addition:
Large chamber air velocity measurement (Appendix 1);
FLEC measurement for baby crib tests (Section 3.3.2);
High temperature chamber method to determine formaldehyde
content in solid materials (Section 3.1.5);
Dr. David Marr as a participant.
1.2
EPA
12/11/2009
Revision:
Editorial correction based on Ronald Rogers' comments
Section 3.3.1 deleted gypsum wallboard tests in the large
chamber
Section 3.3.3, changed baby furniture test condition to be
conducted under different air exchange rate; reduced number of
tests
A1-1.0
ARCADIS
8/31/2010
Addendum 1 Initial release: August 23, 2010
EPA comments: September 21, 2010, QTRAK No. 08028, QA
Category III "Document has EPA QA requirements that must be
addressed before the QAPP can be approved."
Response to comments: September 30, 2010
ARCADIS
11/16/2010
Resubmitted: November 16, 2010
EPA Comments: 11/30/2010, QLOG. A-00241/QTRAK 08028,
QA Category III." The revised quality assurance project plan
(QAPP)addendum is acceptable with minor revisions."
Response to comments: December 27, 2010
Resubmitted: December 30, 2010
A3-1.0
ARCADIS
7/29/2011
Initial release 1.0
Addendum 3 "Measurement of Henry's Law Constants of
Aqueous Formaldehyde Solutions using a Novel Headspace
Extraction Method"
A2-1.0
ARCADIS
06/20/2011
Initial release: June 20, 2011 To Libby Nessley and Xiaoyu Liu
Procedure for Air Resources Board (ARB) Interlaboratory
Comparison of Composite Wood Product Third Party Certifiers
Resubmitted: December 20, 2011
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: ii
ARCADIS
3/28/2012
Revision: Comments from QA-Liu, received July 15, 2011,
QLOG No. A-00241 / QTRAK 08028, QA Category III.
"acceptable with minor revisions".
Changes made including changing to ARCADIS format and
resubmitted to WAM 3/30/2012
ARCADIS
2/30/2012
Added Appendix D: "Test Plan for Repeat of Secondary Test
Method for iHWPW
A4-1.0
ARCADIS
4/4/2012
Initial release 1.0 April 4, 2012
Procedure for Evaluation of Reference Material for
Formaldehyde Emissions Testing using Small Environmental
Chambers
A4-1.1
ARCADIS
July 11, 2012
Response to EPA QA comments, QLOG No. A-00241 /
QTRAK 08028, QA Category III (May 3, 2012), with the
recommendation "acceptable with minor revisions" which
allows data collection to start in parallel with the revision of the
QAPP Addendum.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: iii
Distribution List
NRMRL
Xiaoyu Liu
Robert Wright
Robert Thompson
(PI/WA Manager)
(QA Officer)
(IEMB Chief)
Contractor Staff
Nancy Roache
Corey Mocka
Robert Pope
Russell Logan
Libby Nessley
(WA Leader)
(Chemist)
(Chemist)
(Technician)
(QA Officer)
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: iv
List of Acronyms
ACH Air changes per hour
Co Initial formaldehyde concentration
COC Chain of Custody
CWP Composite wood product
DAD Diode array detector
DAS Data acquisition system
DNPH 2,4-Dinitrophenylhydrazine
EPA U.S. Environmental Protection Agency
FSCWA Formaldehyde Standard for Composite Wood Products Act
HLPC High-performance liquid chromatography
HTC High temperature chamber
IARC International Agency for Research on Cancer
IEMB Indoor Environment Management Branch
u-CTE Markes Micro-Chamber/Thermal Extractor
NIST National Institute for Standards and Technology
NRMRL National Risk Management Research Laboratory
PI Principal investigator
ppm parts per million
QA Quality Assurance
QAPP Quality Assurance Project Plan
QC Quality Control
RH Relative Humidity
RTP Research Triangle Park
SOP Standard operating procedure
TCSA Toxic Substances Control Act
TPC Third-party certifier
VT Virginia Polytechnic Institute and State University
VOC Volatile organic compound
WA Work Assignment
WAL Work Assignment Leader
WAM Work Assignment Manager
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: v
Contents
Document Revisions i
Distribution List iii
Contents v
1. Project Description and Objectives 2
1.1 Background 2
1.2 Project Objectives 2
1.3 Facility Location and Description 3
1.3.1 Small Environmental Chamber 3
1.3.2 Markes Micro-Chamber/Thermal Extractor 4
1.3.3 EPA High Temperature Chamber 6
2. Project Organization 7
2.1 EPA Staff 7
2.2 In-house Contractor (ARCADIS) Staff 7
3. Experimental Approach 9
3.1 Test Materials 9
3.2 Test Procedures and Sampling Schedules 10
3.2.1 Task 1: Initial Small Chamber Testing 10
3.2.2 Task 2: Evaluation of Methods to Determine C0 11
3.2.2.1 Liquid Extraction Procedure 13
3.2.2.2 Markes u-CTE Procedure 15
3.2.2.3 EPA HTC Procedure 18
3.2.3 Task 3: Evaluation of Packaging Integrity and Shelf Life 19
3.2.3.1 Small Chamber Procedure 19
3.2.3.2 Co u-CTE Procedure 24
3.2.4 Task 4: Evaluation of Formaldehyde Emission Rates from the
Reference Material at Varying RH 26
3.2.5 Test and Sample Identification 26
3.3 DNPH Sampling Method 27
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: vi
4. QA/QC 28
4.1 Data Reporting 28
5. References 29
Tables
Table 2-1. Key Points of Contact 7
Table 3-1. Chamber Parameters for Initial Small Chamber Tests 10
Table 3-2. Proposed Film Tests for Determination of Experimental C0 17
Table 3-3. Proposed Schedule for Collection of DNPH Cartridge Samples for the
Film u-CTE Time Series Method at 60 °C 17
Table 3-4. Proposed Small Chamber Tests for Reference Material 20
Table 3-5. Proposed Sampling Schedule for Small Chamber Tests 21
Table 3-6. Sample Schedule for DNPH for the Package Evaluation Test 24
Table 4-1. Data Reporting Requirements 28
Figures
Figure 1-1. Small Chamber 4
Figure 1-2. Markes Micro-Chamber/Thermal Extractor (u-CTE) 5
Figure 1-3. Diagram of a Single Micro-Chamber 5
Figure 1-4. High Temperature Chamber 6
Figure 3-1. Results of Initial Small Chamber Tests 11
Figure 3-2. Liquid Extraction of Film 13
Figure 3-3. Film Frame with Film in Chamber 22
Figure 3-4. Reference Material Film after SCh Test 23
Figure 3-5. Film Slicer 25
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 1
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 2
1. Project Description and Objectives
The U.S. Environmental Protection Agency (EPA) has collaborated with Virginia
Polytechnic Institute and State University (Virginia Tech, VT) to develop and test the
emission characteristics of a polymer film being developed as a carrier for a
formaldehyde reference material. This reference material will be used for improving the
quantification of uncertainties associated with formaldehyde emission measurements
in environmental chambers for the purpose of validation of source emission
characteristics.
1.1 Background
Formaldehyde is produced on a large scale worldwide. One major use includes the
production of wood-binding adhesives and resins. One of the major sources of
exposure is inhalation of formaldehyde emitted from composite wood products (CWP)
containing urea-formaldehyde resins. The International Agency for Research on
Cancer (IARC) reclassified formaldehyde from "probably carcinogenic to humans" to
"carcinogenic to humans" in 2004, based on the increased risk of nasopharyngeal
cancer.
On July 7, 2010, President Obama signed the Formaldehyde Standard for Composite
Wood Products Act (FSCWA) Senate Bill 1660 into law.[1] This Act amends the Toxic
Substances Control Act (TSCA) as Title VI and requires EPA, by July 1, 2011, to
promulgate regulations that ensure compliance with the standards. For industry to
comply with the regulations, both internal and independent testing of formaldehyde
emissions from the manufactured products must be conducted. The primary and
secondary methods for evaluation of formaldehyde emissions involve the use of both
large and small environmental chambers. This reference material would then be used
to validate chamber testing methods including those used by third-party certifiers
(TPCs) to confirm compliance of manufacturers of CWPs to these new regulations.
1.2 Project Objectives
The overall objective of this project is to collaborate with Virginia Tech researchers to
test the emission characteristics of the polymer film that will be developed as a carrier
for a formaldehyde reference material. Virginia Tech has developed the process for
loading a known concentration of formaldehyde onto a polymer film. The final product
needs to be tested and evaluated using small environmental chambers to determine its
usefulness as a reference material for validating formaldehyde emissions in small
chamber effluents. This Addendum outlines the responsibilities of ARCADIS for
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 3
conducting tests in the EPA small chamber laboratory to assess the performance of
the reference material film during environmental chamber testing. Details of the
preparation of the films are presented in the approved QAPP by Dr. John Little,
Developing a Reference Material for Formaldehyde Emissions Testing (ver. 2) [2],
Department of Civil and Environmental Engineering, Virginia Tech, Blacksburg, VA. All
of the materials to be tested on this project will be prepared and supplied to EPA by
Virginia Tech. All questions about the films and how they are prepared should be
addressed in the VT QAPP.
The resources of the small chamber include the 53-L small environmental chambers,
the Markes Micro-Chamber/Thermal Extractor (u-CTE), the EPA high temperature
chamber (HTC), and a modified EPA Method 8315a for liquid extraction of
formaldehyde to evaluate the initial concentration (C0) on the film, the effectiveness of
the packaging (shelf-life), and actual small environmental chamber emission rates from
the supplied reference materials. An interlaboratory comparison with at least one other
laboratory may also be conducted once the procedure has been established.
This addendum is a continuation of the approved October 2008 Quality Assurance
Project Plan (QAPP) "Green Building Research: Indoor Source Emissions and Sink
Effect Study of Formaldehyde" located in the following directory:
L:\l_ab\NRML PublicMHCHO 2008\QA Documents\QAPP.
1.3 Facility Location and Description
The EPA Indoor Environment Management Branch (IEMB) small chamber laboratory is
located at the EPA Research Triangle Park (RTP), North Carolina campus in room
E378A. The proposed research will be conducted in this laboratory (E378A). The
sections below detail the chamber systems that are located in E378A, which will be
used for this project.
1.3.1 Small Environmental Chamber
The small environmental chamber system (Figure 1-1) consists of two large
temperature-controlled incubators, which house a total of eight 53-liter stainless steel
environmental chambers, a clean air system consisting of high pressure Volatile
Organic Compound (VOC) and oil-free compressed house air, an AADCO 737-11 Pure
Air generator, an OPTO 22 data acquisition system (DAS), and a Blue M temperature-
controlled water bath for controlled humidification. The support laboratory facilities are
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 4
located in rooms E383, E375A and E377A. These laboratories contain equipment such
as a large drying oven, micro-balance, and deionized water system. The Agilent 1200
High Performance Liquid Chromatography (HPLC) with a Diode Array & Multiple
Wavelength Detector (DAD) in room E383A will be used for the 2,4-
Dinitrophenylhydrazine (DNPH - Silica Gel Cartridge (Waters Sep-Pak®) - Short Body,
55-105um) extract analysis. The detailed description of the small chamber system is
documented in the Facility Manual for the Small Chamber Laboratory (August 2003)
(L:\l_ab\NRML_Public\APPCD Facility Manuals\Small Chamber).
Figure 1-1. Small Chamber
1.3.2 Markes Micro-Chamber/Thermal Extractor
The u-CTE system (Figures 1-2 and 1-3) consists of six micro-chambers that allow
surface or bulk emissions to be tested simultaneously from up to six samples at the
same temperature and flow rate. Each micro-chamber consists of an open-ended
cylinder (cup) constructed of Silicosteel® (silicone-coated stainless steel) measuring
30-mm deep with a diameter of 45 mm and a volume of 44 ml. The system has
temperature control that allows the tests to be conducted at ambient temperature or at
elevated temperatures up to 120 °C. The chamber's flow distribution system maintains
a constant flow of air through each sample chamber, independent of sorbent-tube
impedance and whether or not a sorbent tube is attached. The flow rate is controlled
by the source air pressure and the flow distribution device in the unit. For all of the
evaluation tests, the high flow-rate option (50 mL/min to 500 mL/min) will be selected.
According to the vendor, surface air velocities are roughly uniform across the surface
of the sample, and these surface air velocities range from approximately 0.5 cm/s at an
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 5
inlet gas flow rate of 50 mL/min to approximately 5 cm/s at an inlet gas flow of 350
mL/min..
Figure 1-2. Markes Micro-Chamber/Thermal Extractor(u-CTE)
Micro-
chamber
PUF sampling tube
O-ring specific to tube type
Detachable micro-
chamber sample top
-Heated air supply
Flow control
Figure 1-3. Diagram of a Single Micro-Chamber
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 6
1.3.3 EPA High Temperature Chamber
The high temperature chamber (HTC), also located in room E378A, is made of
electro-polished stainless steel with the size of 17.8 cm (depth) by 2.5 cm (height) by
25.4 cm (width) (Figure 1-4). The HTC is designed to have well-distributed airflow and
well-mixed exhaust across a test coupon and to provide heating up to 220 °C. The test
coupon rests on the bottom of the chamber. A front drawer is held in place by two cam-
activated clamps. The drawer is sealed to the chamber body by a Teflon-encased
Viton O-ring. The entire unit is encased in an insulated aluminum case. A 122-cm long
air heater is at the inlet to the chamber. The line heater heats the incoming air stream
independently. Zero-percent relative humidity (RH) air is generated by the small
chamber clean air generation system. A mass flow controller is used to regulate the air
flow through the chamber. Relative humidity is monitored at the exhaust and recorded
to the OPTO DAS. The temperature is set and maintained by the control panel of the
apparatus at 60 °C for the liquid and film tests. The temperature of the effluent is
monitored and recorded to the OPTO DAS. Figure 1-4 shows a picture of the HTC.
Figure 1-4. High Temperature Chamber
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 7
2. Project Organization
The organizational table for this project, presented in Table 2-1, tabulates the key
points of contact for this project along with email and phone contact information. The
roles and responsibilities of the project personnel are discussed in the following
paragraphs. Virginia Tech is responsible for production and delivery of the
formaldehyde reference material for these tests, therefore, the EPA PI for this project
will be responsible for communication with Virginia Tech. The contact will be Dr. John
Little (jcl@vt.edu), Department of Civil and Environmental Engineering, Virginia Tech,
Blacksburg, VA.
Table 2-1. Key Points of Contact
Affiliation and Project Role
EPA PI (This Task)
EPA QA Officer
ARCADIS QA Officer
ARCADIS (WAL)
ARCADIS Chemist
ARCADIS Chemist
ARCADIS Technician
Contact
Xiaoyu Liu
Bob Wright
Libby Nessley
Nancy Roache
Corey Mocka
Robert Pope
Russell Logan
Phone Number
541-2459
541-4502
328-5588
541-0365
541-2862
541-2013
541-3810
Email Address
Liu.Xiaovu@epa.gov
Wrig ht.Bob@epa.gov
Libby.Nessley@arcadis-us.com
Nancy.Roache@arcadis-us.com
Corey.Mocka@arcadis-us.com
Robert.Pope@arcadis-us.com
Russell.Logan@arcadis-us.com
2.1 EPA Staff
Principal Investigator (PI), Dr. Xiaoyu Liu: Dr. Liu will serve as the PI and has the
responsibility for developing the work plan, experimental design, data analysis,
reporting, project management and communication with Virginia Tech.
QA Representative, Mr. Bob Wright: Mr. Wright will be responsible for review and
approval of the QA project plan (QAPP) and other deliverables of this project and
provide assistance on all QA-related issues.
2.2 In-house Contractor (ARCADIS) Staff
ARCADIS Work Assignment Leader (WAL), Nancy Roache: Ms. Roache is responsible
for writing this addendum, setup and monitoring of all large and small chamber tests,
the initial review of HPLC data for formaldehyde from samples collected from the small
chamber and large chamber tests, preparing test materials, and collecting air samples
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 8
for both small chamber and large chamber tests. In addition, she is responsible for
communicating any delays in scheduling or changes in cost to the EPA as soon as
possible.
ARCADIS Chemist, Corey Mocka: Mr. Mocka is responsible for the operation and
calibration of the HPLC analytical system for formaldehyde analysis. He will perform
solvent extraction and derivatization of formaldehyde from DNPH cartridges, analyze
sample extractions, and report data. Mr. Mocka will report directly to the WAL, Nancy
Roache.
ARCADIS Chemist, Robert H Pope: Mr. Pope is responsible for setting up the small
chamber system to monitor the formaldehyde emissions from formaldehyde sources,
preparing test materials, and collecting air samples from the small chamber tests. He
will prepare test notebooks for all small chamber tests. Mr. Pope will report directly to
the WAL, Nancy Roache.
ARCADIS Technician, Russell Logan: Mr. Logan is responsible for staff support. Mr.
Logan will report directly to the WAL, Nancy Roache.
ARCADIS Quality Assurance Officer (QAO), Laura Nessley: Ms. Nessley will review
the QAPP and be responsible for ensuring that the contractor staff adheres to the
procedures described therein.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 9
3. Experimental Approach
The experimental design of this project consists of the following five tasks to be
performed byARCADIS:
1. Initial scouting small chamber testing of reference material - December 2011
2. Evaluation of three unique methods for the experimental determination of the initial
HCHO concentration (C0) on the film to validate the film loading concentration
determined by VT.
3. Additional small chamber testing with emphasis on the evaluation of the integrity of
the packaging and determination of shelf life of the reference material
4. Evaluation of formaldehyde emission rates from the reference material at varying
values of RH.
5. Conduct an interlaboratory comparison with at least one other laboratory (possibly
MIST).
The initial small chamber testing was performed in December with reference material
supplied by Virginia Tech. These small chamber tests were performed at 23 °C, 1 air
exchange rate (ACH). One test was conducted at 50% RH, and two tests were
conducted at approximately 0% RH. The results of these test initiated the need to
conduct follow-on testing of C0, shelf life, and variations in RH.
3.1 Test Materials
The reference materials to be tested on this project have been developed and
prepared by Virginia Tech. All of the materials to be tested on this project will be
prepared and supplied to the EPA PI by Virginia Tech.
According to the EPA PI, the date and time the film is removed from the loading vessel
is considered as time zero. Each film that is supplied to the EPA PI will be numbered
with the loading date and time as well as the position in the loading vessel. Films from
each loading batch are considered equal; however, the films from different loading
batches may vary slightly. The initial films for the December tests were packaged for
shipping wrapped individually in aluminum foil and placed in individual plastic zip bags.
The bags were placed into a cooler with dry ice and overnight shipped to EPA RTP.
This protocol will be followed for these tests.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 10
Upon receipt of the films from VT, the packaged films will be removed from the cooler
(remaining in their original packaging), photographed then placed in the freezer located
in E383A. The chain of custody (COC) form received with the samples will be dated,
initialed and placed with the samples. The freezer that will be used for this project is on
the EPA Metrology Laboratory monitoring program and is equipped with a HOBO®
thermocouple that records the temperature every hour. The data are downloaded by
the Metrology Laboratory every three months. A weekly spot check of the temperature
will be made while the films are being stored and the data will be recorded in the
laboratory notebook # 2287.
3.2 Test Procedures and Sampling Schedules
3.2.1 Task 1: Initial Small Chamber Testing
Three small chamber tests were conducted with reference material supplied by Virginia
Tech in December 2011. A set of three polycarbonate films measuring 10 cm x 10 cm
x 0.0254 cm was received from Virginia Tech on December 6, 2011. The predicted
monomer formaldehyde initial concentration on the surface (C0) was estimated to be
188 g/m3as reported by VT. Table 3-1 details the test parameters of each film tested.
Table 3-1. Chamber Parameters for Initial Small Chamber Tests
Test ID
SES-SCh-F1
SES-SCh-F2
SES-SCh-F3
Start date
End date
12/7/2011
12/12/2011
12/15/2011
12/19/2011
12/21/2011
12/23/2011
Age,
days3
2
8
14
ACH
1
1
1
Temp,
°C
23
23
23
RH, %
50
<11.3b
<11.3b
The age of the film is the number of days from removal from the loading vessel.
This reading for RH is below the lowest calibration point of 11.3% RH for the probe
The resulting data are presented in Figure 3-1. The chamber data show a measurable
difference between the predicted value and the measured concentrations from the
chamber data. There is also a noticeable difference in the emission rate between 50%
RH and 0% RH. These are issues that are being investigated with this study.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 11
a
o
o
U
OJ
•a
OJ
•a
Model prediction
Test 1:2 days old,
50% RH
Test 2: 8 days old,
0%RH
Test 3: 14 days old,
0%RH
Time (h)
Figure 3-1. Results of Initial Small Chamber Tests
3.2.2 Task 2: Evaluation of Methods to Determine Experimental Co
Virginia Tech will send six films (3.5 cm x 3.5 cm x 0.0254 cm) to EPA. The predicted
monomer formaldehyde initial concentration on the surface, C0 is estimated to be 190
g/m3; the polymerized formaldehyde concentration, CP on the film is estimated to be
0.034 g/m2. The total loading mass is 73 ug. These films will be divided in half and will
be used to evaluate methods for the experimental determination of C0 on the film. It is
assumed that at room temperature, formaldehyde depolymerization will not occur. At
25°C, the measured monomer formaldehyde on the film surface using selected
methods will be C0. At higher temperature, e.g. 40C and 60C, the measured
formaldehyde on the film will be the total loading of formaldehyde mass. Three
methods will be evaluated: liquid extractions (EPA Method 8315A
http://www.epa.gov/osw/hazard/testmethods/sw846/pdfs/8315a.pdf), the Markes u-
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 12
CTE at three temperatures (25 °C, 40 °C, and 60 °C), and the EPA HTC at 60 °C. One
of the methods will then be selected to determine the experimental C0 and the total
loading mass of formaldehyde on the reference materials that will be used for source
emission tests in the future. The selection criteria will be based on the recovery
efficiency of the monomer formaldehyde concentration as compared the initial C0and
the procedure simplicity.
Tests will be conducted using a formaldehyde standard solution with a concentration
similar to the expected concentration of the film to determine the recovery efficiency of
each method. The expected mass of the monomer formaldehyde in the film will be
calculated using Equation 1 below:
Cm0 = Lx |/i/x T x C0 x 1000000 {1}
Where:
Cm0 = Experimentally determined mass of the monomer formaldehyde on the film (ug)
L = Length of the film subsample (m)
W= Width of the film subsample (m)
T= Thickness of the film subsample (m)
C0 = Initial dosed concentration of monomer formaldehyde on the parent film, provided
by Virginia Tech (g/m3)
The expected mass of the polymerized formaldehyde in the film will be calculated
using equation 2 below.
CmP= L x W x CP x 1000000 {2}
Where:
CmP= Experimentally determined mass of the polymerized formaldehyde on the film
L = Length of the film subsample (m)
W= Width of the film subsample (m)
CP = concentration of polymerized formaldehyde on the parent film, provided by
Virginia Tech (g/m2)
The total mass of formaldehyde on the film will be the sum of the mass of the monomer
formaldehyde on the film (Cm0) + the mass of the polymerized formaldehyde on the film
(CmP).
Percent recovery of C0 will be calculated using Equation 3:
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 13
% Recovery =
* 100
{3}
Where:
C, = Liquid concentration of the integrated DNPH sample (ug/mL)
V= Final volume of the DNPH extract (mL)
C0 = Initial concentration (provided by VT) of the film subsample (ug)
N = Number of DNPH samples
The formaldehyde standard solution tests will be followed by testing with the reference
material. The following sections will give a description of each method and the
proposed test procedure.
3.2.2.1 Uquid Extraction Procedure
The liquid extractions of the films will follow a modified version of EPA Method 8315A.
Initially, a HCHO standard solution with a concentration of 3.2 mg/mL will be tested to
evaluate the proposed method. For the C0 evaluations the films will be removed from
the freezer and placed immediately into the extraction vessel with no temperature
equilibration period. The physical extraction of the HCHO from the film will be
accomplished by submerging a subsample of the film (1.75 cm x 3.5 cm) into a 60-mL
amber bottle (Figure 3-2) containing 60 mL of Honeywell Burdick & Jackson High
Purity Water for HPLC, Cat.# 365-4.
Figure 3-2. Liquid Extraction of Film
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 14
The bottle will be shaken vigorously using a VWR Model 3500 standard shaker in a
temperature-controlled incubator maintained to 23 °C initially for a period of 3.5 hours.
The resulting solution will be derivatized (yielding the hydrazone) using the modified
version of EPA Method 8315A, then analyzed by HPLC. If the results of this initial
extraction period show a lower concentration than expected, then the procedure will be
repeated extending the extraction period until 100% recovery is achieved, or the
recovery results are the same after two extended extraction periods. Procedure for the
modified EPA Method 8315A is listed below.
1. Extract the film by submerging a subsample of the film (1.75 cmx3.5 cm) in the 60
ml bottle containing 60 ml of water
2. Shake vigorously shake using a VWR Model 3500 standard shaker in a
temperature-controlled incubator maintained at 23 °C initially for a period of 3.5
hours. (This time will be extended if the recovery is less than 100 %)
3. Immediately after extraction, transfer 10 ml of the solution from the jar to a 125-mL
Erlenmeyer flask
4. Add 4 ml of citrate buffer (pH 5.0) and 3 ml of DNPH solution (3.00 mg/mL)
5. Cover and return to the orbital shaker inside the incubator at a temperature of 23
°C for exactly 1 hour
6. Add 20 ml of dichloromethane to the flask containing the HCHO solution and
separate the solvent portion containing the derivatized HCHO from the water using
a 250-mL separatory funnel; repeat three (3) times.
7. Dry the solution using sodium sulfate.
8. Transfer the samples to a 100-mL blow-down tube; wash the flask four (4) times
with 10-mL portions of methylene chloride and add the washes to the blow-down
tube.
9. Insert the samples into the RapidVap and blow the solutions down to
approximately 1.5 ml. The RapidVap settings are 12 psi nitrogen gas, 50 to 90%
speed (start at 50%, after 10 minutes increase to 90%), at 40 °C. The entire blow-
down process takes 35 minutes.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 15
10. Pipette the solution from the blow-down tube into a 10-mL volumetric flask; wash
the blow-down tube seven (7) times with approximately 1 ml of acetonitrile and
add the washes to the flask
11. Bring the solution to volume with acetonitrile and aspirate the solution 20 times
with a pipette
12. Transfer approximately 1.5 ml to a 2-mL amber vial
13. Analyze the sample on the Agilent 1200 HPLC/DAD (MOP 826
L:\l_ab\NRML_Public\APPCD MOPs and Facility Manuals\APPCD MOPS and
Facility manuals\800-Small Chamber\Appendix B - Operating Procedures)
3.2.2.2 Markes jj-C TE Procedure
The ease of use and versatility of airflow rate, temperature, and sample size makes
the Markes u-CTE procedure ideal for determining C0 using elevated temperature air
extraction. Each of the six chambers on the u-CTE will be evaluated individually with
the HCHO standard solution at 60 °C before testing the reference material. Operation
and detailed setup procedures are outlined in SOP 6962, Operation of the Markes
Micro-Chamber Thermal Extractor (L:\l_ab\NRML_Public\APPCD MOPs and Facility
Manuals\APPCD MOPS and Facility manuals\6900-PCP MOPs). The specific
procedures for the HCHO standard solution test are listed below:
1. Prepare a HCHO solution with a concentration of 75 ug/mL and a HCHO standard
solution with a concentration of 3.21 mg/mL.
2. Set the u-CTE to the desired temperature.
3. Set up u-CTE for high flow and adjust pressure to achieve the highest possible
steady air flow rate of approximately 300 to 450 mL/min (approximately 48 psi)
4. Attach a test DNPH (Silica Gel Cartridge (Waters Sep-Pak®) - Short Body, 55-
105um) to chamber 1 and measure the air flow through the DNPH cartridge using
a Gilibrator; record measurement in notebook.
5. Repeat step 4 for chambers 2 through 6.
6. Open each chamber lid.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 16
7. Attach a new DNPH silica gel cartridge to the lid of each of the chambers that will
be used for the test
8. Close each lid and record time
9. Collect empty chamber background sample for 60 min, remove, record time and
extract for analysis by HPLC
10. Attach a new DNPH cartridge to each of the six chambers' lids
11. Spike 1 ml of the 75-ug/mL HCHO solution directly into each chamber 2 through 6
12. Immediately close the lid and record time
13. Spike 1 ml of the 3.21-mg/mL HCHO solution directly into chamber 1
14. Immediately close the lid and record time
15. Collect an integrated sample for 150 min on chamber 2, remove and immediately
attach a second DNPH to collect for an additional 60 min; remove, record time,
and extract for analysis by HPLC
16. Collect integrated samples on chambers 3 through 6 for 200 min; remove, record
time, and extract for analysis by HPLC
17. Collect a time series of DNPH samples from chamber 1 following the sample
schedule in Table 3-3
18. After sampling is complete, clean u-CTE according to SOP 6962
The basic sampling scheme for collecting samples in the u-CTE during the HCHO
standard solution test is to collect DNPH according to the procedure detailed in MOP
812 (L:\Lab\NRML_Public\APPCD MOPs and Facility Manuals\APPCD MOPS and
Facility manuals\800-Small Chamber\Appendix B - Operating Procedures) The
sampling schedule should be a 30-second sample every 10 minutes for the first 30
minutes, then a 60-second sample every 10 minutes for the next 2 hours changing to a
10-minute sample every hour for the next 3 hours. The test duration will be 6.5 hours
with 20 samples and 1 field blank.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 17
After results from the HCHO standard solution tests have been reviewed, a series of
film tests will be defined. Table 3-2 details the proposed film tests for determination of
experimental C0. The setup and operation of the u-CTE will be the same as for the
HCHO standard solution tests (steps 2 through 10 above). The procedure for the film
test continues by placing the subsample of the film in the desired chamber and closing
the lid - record the start time in the laboratory notebook. For the integrated samples,
the chamber effluent will be collected onto a DNPH cartridge (MOP 812) fora minimum
of 5 hours. After removing the DNPH, a second DNPH cartridge is connected to the lid
and the chamber effluent is collected onto the cartridge for an additional 16 to 20
hours. The procedure for the time-series tests will follow the sampling schedule
outlined in Table 3-3. All DNPH cartridges will be extracted with acetonitrile using the
method outlined in MOP 812 and analyzed by HPLC (MOP 826).
Table 3-2. Proposed Film Tests for Determination of Experimental Co
Type of test
Time series
Integrated
Integrated
Integrated
u-CTE
temp., °C
60
60
40
25
Airflow
rate,
mL/min
350
350
350
350
Size of film
1. 75 cm x 3.5
cm
1. 75 cm x 3.5
cm
1. 75 cm x 3.5
cm
1. 75 cm x 3.5
cm
u-CTE #
1
1 and 2
1 and 2
1 and 2
Duplicate
yes
yes
yes
Table 3-3. Proposed Schedule for Collection of DNPH Cartridge Samples for the Film u-
CTE Time Series Method at 60 °C
Elapsed Time (Mrs)
-1 Empty Chamber Background
0.042 (2.5 min)
0.2(12min)
0.4 (24 min)
0.5(30 min)
0.7 (42 min)
0.9 (54 min)
1
Sample Duration, min
150
5
5
5
5
5
5
5
Sample Volume (L)
50
1.8
1.8
1.8
1.8
1.8
1.8
1.8
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 18
1.5
2
3
Total number of samples
Field blanks (daily)
Total number of samples per test
10
10
30
3.6
3.6
11
11
1
12
3.2.2.3 EPA HTC Procedure
The EPA HTC has been used successfully in previous research to collect emissions
from solid material at an elevated temperature. The proposed tests for this chamber
will be limited to one HCHO standard solution at 60 °C and one reference material test.
An SOP has not been written for the operation of this chamber. The specific
procedures for the HCHO standard solution test are listed below:
1. Prepare a HCHO solution with a concentration of 75 ug/mL
2. Attach clean air supply to pre-heat tube and set air flow rate to approximately 150
to 200 mL/min
3. Set the HTC to the desired temperature
4. Open the chamber door and place a new small Al dish on the floor of the chamber
5. Close and seal the door
6. Let system flush at set temperature for at least 1 hour before collecting a
background sample
7. Attach a test DNPH cartridge to the chamber effluent and measure the air flow
through the DNPH cartridge using a Gilibrator; record measurement in notebook
8. Attach a new DNPH cartridge to chamber exhaust point; record time
9. Collect empty chamber background sample for 200 minutes, remove the DNPH
cartridge, record time, and extract for analysis by HPLC
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 19
10. Open the door and quickly spike 1 ml of the 75-ug/mL HCHO solution into the Al
dish
11. Seal the door quickly; record the time
12. Collect an integrated sample for 200 min , remove the DNPH cartridge, and
immediately attach a second DNPH to collect for an additional 60 min, remove the
DNPH cartridge, record time, and extract for analysis by HPLC
13. After sampling is complete, clean the HTC with deionized water; elevate heat and
flush with clean dry air.
The procedure for the reference material film will follow steps 2 through 9 removing the
Al dish and replacing it with a film holder. The film will be placed in the chamber on the
film holder, and the door will be sealed immediately. The DNPH sample will be
collected for a minimum of five hours, and a second DNPH cartridge will be attached to
the exhaust and collected for an additional 16 to 18 hours. All DNPH cartridges will be
extracted with acetonitrile and analyzed by HPLC.
3.2.3 Task 3: Evaluation of Packaging Integrity and Shelf Life
3.2.3.1 Small Chamber Procedure
Task 3 will use the small environmental chamber systems set up in the six-chamber
incubator. The small chambers will be operated in accordance with approved MOPs
801, 802, 803, 804, and 806, located in L:\Lab\NRML_Public\APPCD MOPs and
Facility Manuals\APPCD MOPS and Facility manuals\800-Small Chamber. Three
chambers will be set up in the six-chamber So-Low Incubator with the following
parameters: 0% RH, 1 ACH, and 25 °C. Eleven (11) 10-cm x 10-cm films and five (5)
3.5-cm x 3.5-cm films will be requested from Virginia Tech. Table 3-4 outlines the
proposed tests for this task. The duration of each test will be a minimum of 96 hrs.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 20
Table 3-4. Proposed Small Chamber Tests for Reference Material
Time
WeekO
Week 2
Week 4
Week 6
Week 10
Test ID
SES-SCH#-F4&5-W(0)-
SES-SCH#-F6-W(0)-Packaging-
SES- MCH#-C0-F4&5-W(0)-
SES-SCH#-F7&8-W(2)-
SES- MCH#-C0-F7&8-W(2)-
SES-SCH#-F9&10-W(4)-
SES- MCH#- C0-F9&10-W(4)-
SES-SCH#-F11&12-W(6)-
SES- MCH#-C0-F11&12-W(6)-
SES-SCH#-F13&14-W(10)-
SES- MCH#-C0-F13&14-W910)-
Proposed small chamber tests and C0 determination
(2) SCh tests with 10-cm x 10-cm films - Duration 96 hours
(1) SCh test with 10-cm x 10-cm wrapped and sealed film - Duration 2 weeks
(2) C0 test with 3.5-cm x 3.5-cm film divided in half- u-CTE
(2) SCh tests with 10-cm x 10-cm films - Duration 96 hours
(2) C0 test with 3.5-cm x 3.5-cm film divided in half- u-CTE
(2) SCh tests with 10-cm x 10-cm films - Duration 96 hours
(2) C0 test with 3.5-cm x 3.5-cm film divided in half- u-CTE
(2) SCh tests with 10-cm x 10-cm films - Duration 96 hours
(2) C0 test with 3.5-cm x 3.5-cm film divided in half- u-CTE
(2) SCh tests with 10-cm x 10-cm films - Duration 96 hours
(2) Co test with 3.5-cm x 3.5-cm film divided in half- u-CTE
3.2.3.1.1 Shelf Life Tests
Table 3-5 outlines the proposed sampling schedule for the small chamber tests
evaluating the shelf life of the films. These tests will be conducted with duplicated
identical environmental parameters. Upon completion of the small chamber emissions
test the 10 cm x 10 cm film will be removed. A photograph will betaken on a grid
surface, and two small sections will be removed from the film (Figure 3-4). The picture
will be available to determine the surface area of each section using Auto CAD if
needed. The smaller sections will be placed in separate u-CTE chambers at a
temperature of 60 °C and a flow of 350 mL/min. An integrated DNPH sample will be
collected for a minimum of 5 hours to determine the residual formaldehyde on or in the
film.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 21
Table 3-5. Proposed Sampling Schedule for Small Chamber Tests
Elapsed Time (Mrs)
-1 - Background
0.1 (6 min)
0.3(18 min)
0.5 (30 min)
0.7 (42 min)
1
2
4
6
8
10
24
28
32
48
72
96
Total number of primary
samples
Duplicates
Field blanks
Field controls (when
needed)
Total number of samples
per test
Primary Sample
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Duplicate Sample
X
X
X
X
X
X
Sample Volume (L)a
35
3.5
3.5
3.5
3.5
3.5
7
7
7
7
7
21
21
21
42
60
60
16
6
3
3
28
3 Samples will be collected at 350 mL/min
The specific procedures for the small chamber tests are listed below:
1. Clean three chambers as detailed in MOP 801 (L:\l_ab\NRML_Public\APPCD
MOPs and Facility Manuals\APPCD MOPS and Facility manuals\800-Small
Chamber\Appendix B - Operating Procedures).
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 22
2. Set up chamber to specified parameters: fan blowing upward and placed to the
rear of the chamber, 0% RH, 1 ACH, and 25 °C.
3. Flush for at least 24 hours - collect single DNPH sampling cartridge background for
2 hours at 350 mL/min.
4. Analyze sample, if results show elevated formaldehyde area response above 4.5,
re-clean chamber.
5. Repeat steps 2-4.
6. Clean the film holder using the same process detailed in MOP 801 at least one day
prior to the beginning of the test. Place the empty holder in each of the small
chambers (Figure 3.3).
Figure 3-3. Film Frame with Film in Chamber
7. Reseal chamber and flush under set parameters for a minimum of 16 hours-
collect duplicate DNPH sampling cartridge background for two hours at 350
mL/min.
8. After the background samples are removed, open the chamber and remove the
film holder.
9. Remove the test film from freezer. Let the film equilibrate to room temperature
approximately 5 min before placing it in the holder. Make sure that the films are
from the same batch and insert the film into the film holder for each chamber
(Figure 3.3).
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 23
10. Place the holder with the film in the chamber (Figure 3.3).
11. Reseal chamber and begin sampling as outlined in Table 3.5.
12. After test is complete remove the film from the chamber.
13. Photograph on a grid surface (Figure 3-4) for determination of surface area using
Auto CAD (SOP 6016) if needed.
Figure 3-4. Reference Material Film after SCh Test
14. Cut two sections from the film to be placed in the u-CTE chambers at a
temperature of 60 °C and a flow of 350 mL/min.
15. Collect an integrated DNPH sample for a minimum of 5 hours to determine the
residual formaldehyde on or in the film.
16. Wrap remaining film in three layers of aluminum foil, place in a plastic zip bag and
place in the freezer in E383.
17. The chamber will be completely disassembled, cleaned (MOP 801) and readied for
the next test starting with step 1.
3.2.3.1.2 Packaging Test
One small chamber test, SES-SCH#-F6-W(0) will be conducted to investigate the
effectiveness of the packaging. This test will continue for least a two weeks and may
be extended depending on the data that is collected. The sampling schedule can be
adjusted to incur no weekend sampling for this test.
1. Prepare small chamber as described in steps 1-8 above.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 24
2. As soon as films are received, remove one film from the shipping cooler,
photograph and remove any extra packaging such as plastic bags.
3. Place the aluminum foil wrapped film on to the film stand (Figure 3-3) and place in
the chamber.
4. Seal the chamber and begin sampling according to the sample schedule in Table
3-6.
Table 3-6. Sample Schedule for DNPH for the Package Evaluation Test
Elapsed Time (Mrs)
-1 - Background
Day 1
Day3
Day5
Day 8
Day 10
Day 12
Day 16
Total number of primary
samples
Duplicates
Field blanks
Field controls (when
needed)
Total number of samples
per test
Primary Sample
X
X
X
X
X
X
X
X
Duplicate Sample
X
X
X
Sample Volume (L)a
35
100
100
100
100
100
100
100
8
2
2
3
15
' Samples will be collected at 350 mL/min
3.2.3.2 C0 jJ-CTE Procedure
On the same day that the film tests start, a 3.5 cm x 3.5 cm film from the same batch
will be extracted thermally using the u-CTE to determine C0. The following section
details the procedure for determining C0.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 25
1. For the C0, clean two micro-chambers and set test parameters to 60 °C and the
pressure to approximately 48 psi (300 -450 mL/min)
2. Flush the system with clean dry air from the small chamber clean air system for at
least 24 hours, then collect backgrounds from each chamber for 5 hours
3. Extract and analyze samples. If the results show elevated formaldehyde area
response above 4.5, re-clean chamber.
4. Repeat steps 6-8
5. Once the background passes, continue to flush with the system air until time for
the test to begin
6. On the day the small chamber test starts, remove one of the 3.5 cm x 3.5 cm films
from the freezer. Make sure that it is from the same batch as the films being used
with the small chamber tests.
7. Immediately place the film in the film slicer (Figure 3-4) and divide film in half
Figure 3-5. Film Slicer
8. Put a DNPH cartridge on the lid of each of two chambers that have been prepared
for testing
9. Place each half of the divided film into one of the u-CTE chambers, seal the lid and
record the time
10. Collect the chamber effluent onto a DNPH cartridge for 5 hours
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 26
11. Remove the DNPH cartridge and replace with a second DNPH cartridge to collect
overnight
12. Remove DNPH cartridge, extract and analyze samples
13. Clean chambers
3.2.4 Task 4: Evaluation of Formaldehyde Emission Rates from the Reference Material at
Varying RH
The performance of this task will depend on the available funding and resources. For
this task one chamber test at 0% RH, one test at 70% RH and a set of duplicate
chamber tests at 50% RH are proposed. Details of this task are TBD. The sampling
schedule for these tests will follow the schedule outlined in Table 3.5.
3.2.5 Task 5: Interlaboratory Comparison
Once the details of packing, shipping and chamber testing have been establish, EPA
has proposed a small interlaboratory comparison for the process with at least one other
laboratory. The details of this task are currently under discussion.
3.2.6 Test and Sample Identification
In order to differentiate individual tests, each test will have a unique identification. The
nomenclature of the conditioning phase of the test will be as follows:
[SES]-[Chamber#]-[Film Test #]-[Week #]-[Type of Test]-[Elapsed Time]-[Replicate]
where:
SES = Standard Emissions Source
Chamber = (MCH = uCTE (#1-6), SCH = small chamber (#1-8), HTC = high
temperature chamber, LE = liquid extraction)
Film Test # = F1 - TBD
Week Number = W(#)
Type of Test = [C0 Eval, Packaging, Shelf life (WO -W10), Interlaboratory comparison
(1C)]
Type or Elapsed Time (BKG = chamber background, Elapsed Time = #hr),
Replicate ID = A, B.
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 27
A typical sample name would be:
SES- SCH1-F10-WO-Bkg (-24hr) A
3.3 DNPH Sampling Method
All air samples will be collected on DNPH-treated silica gel for analysis of
formaldehyde by Agilent 1200 HPLC. This analysis is a standardized method used
extensively in the source characterization laboratory[3].
The commercially available cartridges (Waters Sep-Pak DNPH Silica Gel Cartridge,
Waters Associates, Milford, PA) contain 350 mg of a 55- to 105-um chromatographic-
grade silica gel coated with DNPH. Samples will be collected on the cartridges by
drawing air from the sampling line located within 0.3 m of the source using mass flow
controllers and vacuum pumps at a sampling rate of 100 to 300 mL/min. The sampling
rate flow rate will be set with the mass flow controller and then measured with a
Gilibrator. After collection, DNPH cartridge samples will be capped, placed back in their
original air-tight re-sealable relabeled pouch and stored in the freezer located in E378A
before solvent extraction. DNPH cartridges will be extracted with 5 ml of acetonitrile
(HPLC grade) before analysis and must be extracted within 2 weeks of collection.
Sample information will be recorded on labels affixed to the pouch in which samples
are stored, in the Excel notebook, and in the laboratory logbook. Details of the
sampling and analysis procedure are given in MOPs 812, 826 and 827 of the small
chamber MOPs (L:\l_ab\NRML_Public\APPCD MOPs and Facility Manuals\APPCD
MOPS and Facility manuals\800-Small Chamber).
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 28
4. QA/QC
Work on this project will be performed following the general guidelines for QA/QC
described in the Facility Manual for the Small Chamber Laboratory (August 2003) and
the Facility Manual for the Large Indoor Air Quality Environmental Test Chamber
(2006). Section 6 (Quality Assurance/Quality Control) of this approved QAPP details
the Data Quality Indicator goals that are to be followed for this study. Any deviations
will be noted in the report.
4.1 Data Reporting
Data from all tests will be reported in electronic files. Reported data will include the
concentration of target carbonyls at each sampling time, operating parameters of tests,
and results for analyses of QC samples. Written and verbal communications with
Virginia Tech will be prepared by the WAM/PI, Xiaoyu Liu, for this project. Table 4-1
details the data reporting requirements associated with these tests.
Table 4-1. Data Reporting Requirements
Measurement Parameters
Carbonyl Concentrations
Inorganic Gases Concentrations
Temperature
Pressure
Relative Humidity
Air Exchange Rate
Air Flow
Mass
Unit
ug/m3
ppb
°C
Pa
%
h-1
mL/min
g
-------
Project No.: RN990272.0015
QAPP: GBR Sink Formaldehyde
Reference Material Evaluation Addendum 4.1.1
Date: July 11,2012
Page: 29
5. References
[1] S. 1660 (111th): Formaldehyde Standard for Composite Wood Products Act
(FSCWA) http://www.aovtrack.us/conciress/bill.xpd?bill=s111-1660 , Site last
accessed May 16, 2012.
[2] Little , John, Quality Assurance Project Plan, Developing a Reference Material for
Formaldehyde Emissions Testing (ver. 2); Department of Civil and Environmental
Engineering, Virginia Tech, Blacksburg, VA.
[3] EPA method IP-6A, Determination of Formaldehyde and Other Aldehydes in
Indoor Air Using a Solid Adsorbent Cartridge, U. S. Environmental Protection
Agency
-------
SOP No. 880
Revision 1
June 29, 2012
Pagel of 23
U.S. EPA Office of Research and Development
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Standard Operating Procedure 880
Test Procedure for Formaldehyde Reference Material Testing Using Small Chamber
Authored by:
"fl
« f
6/29/2012
Nancy Roache
ARCADIS Task Manager
Date
Approved by:
/±
.\t*ATvkA
Xiaoyu Liu 7
EPA Princip^
nvestigator
Date
Libby Nessley
ARCADIS QA Officer
6/29/2012
Date
Robert Wright
APPCD QA Manager
Date
Keywords: reference material, small chamber, film, formaldehyde
-------
SOP No. 880
Revision 1
June 29,2012
Page 2 of 23
Distribution List
APPCD
Xiaoyu Liu
Bob Wright
ARCADIS
Nancy Roache
Corey Mocka
Robert Pope
Libby Nessley
Laboratory Copy: maintained by Xiaoyu Liu
Electronic Filing: Scientific Data Share L:\NRMRL_PCB_Mitigation\QA Documents\SOPs
Revision Record
Revision
0
1
Date
June 21, 2012
June 27, 2012
June 28, 2012
Responsible Person
Nancy Roache
EPAQA
Nancy Roache
Description of Change
Initial release
QA comments for - QLOG No. A-1621 1 / QTRAK 1 1037,
QA Category IV - "acceptable with minor revisions"
Revised to reflect comments from QA office
-------
SOP No. 880
Revision 1
June 29,2012
Page 3 of 23
Table of Contents
1. TITLE
2. PURPOSE AND SCOPE
3. SUMMARY OF METHOD
4. DEFINITIONS
5. HEALTH AND SAFETY
6. INTERFERENCES
7. RELATED DOCUMENTS
8. EQUIPMENT AND SUPPLIES
9. PROCEDURE
9.1 Chamber Cleaning
9.2 Chamber Reassembly
9.3 Chamber Preparation for Testing
9.4 Reference Materials
9.5 DNPH Sample Extraction and Analysis
10. Reporting Data
4
4
4
4
5
5
5
6
7
7
10
12
13
Tables
Table 1. Small Environmental Chamber System Components
Table 2. Sampling System Components
Table 3. Experimental Conditions for Small Chamber Reference Material Tests
Table 4. Sampling Schedule for Small Chamber Tests
Table 5. Data Reporting Requirements
6
7
9
12
13
Figures
Figure 1. Film Holder with Film in Chamber
Appendices
APPENDIX A: Chain of Custody Form
APPENDIX B: See Attached Excel File
10
-------
SOP No. 880
Revision 1
June 29,2012
Page 4 of 23
1. TITLE
Test Procedure for Formaldehyde Reference Material Small Chamber Testing
2. PURPOSE AND SCOPE
This SOP provides a written, repeatable procedure for the set up, sampling, analysis and data
reduction of the small chamber testing for the Formaldehyde Reference Material.
The scope of this project is to collaborate with Virginia Polytechnic Institute and State University
(Virginia Tech or VT) researchers to test the emission characteristics of the polymer film that will be
developed as a carrier for a formaldehyde reference material whose emissions can be predicted by
a fundamental emission model. Small environmental chambers will be used to determine the
emission rate of formaldehyde from a polycarbonate film that has been dosed with formaldehyde
gas.
3. SUMMARY OF METHOD
Tests to evaluate the emission rate of formaldehyde from a polycarbonate film will be conducted in
small environmental chambers operated at specified environmental conditions. The source for the
chamber tests will be a product that has been developed by Virginia Tech as a proposed reference
material for evaluating the uncertainties of small chamber testing for formaldehyde. The reference
material will be sealed in the chamber and a series of timed air samples collected from the chamber
effluent onto 2,4-dinitrophenylhydrazine (DNPH) cartridges. DNPH cartridges will be extracted and
analyzed by high performance liquid chromatography (HPLC) and the data reported to the EPA
Principal Investigator (PI).
4. DEFINITIONS
• Small environmental chamber - 53-liter electropolished stainless steel chambers that meet the
specifications in ASTM Standard Guide D5116-10 — Standard Guide for Small-Scale
Environmental Chamber Determinations of Organic Emissions from Indoor Materials/Products
(ASTM, 2010)
• DNPH Cartridge - silica gel cartridge coated with 2,4-dinitrophenylhydrazine (DNPH)
• Sampling Pump System- consists of vacuum pump connected by a valve system to 4 mass
flow controllers (MFC) whose output is controlled by a mass controller box (CB). The four pump
lines are designated as pump 1 (P1), pump 2 (P2), pump 3 (P3), and pump 4 (P4).
-------
SOP No. 880
Revision 1
June 29,2012
Page 5 of 23
• Practical Quantification Limit (PQL) (MOP 826) is defined as the lowest standard on the
calibration curve if the following conditions are met:
o PQL peak is identifiable with a precision of 15% and accuracy of 15%
o PQL response is at least 5 times that of the IDL.
• Dl Water - Deionized water.
• Reference Material - A polymer film designed as a carrier for formaldehyde.
• Film Holder -150 mm x 150 mm aluminum holder with wire cross hairs to hold film and feet to
stand upright (provided by VT)
5. HEALTH AND SAFETY
Good laboratory practice will be followed including all safety procedures outlined in the "Chemical
Hygiene Plan" revised in February 2012.
6. INTERFERENCES
The presence of moisture creates the potential for formaldehyde to polymerize, therefore it is
recommended that the films should be handled and stored in environments with relative humidities
lower than 20%.
7. RELATED DOCUMENTS
• MOP 802 - Operation of Small Emissions Chambers during Testing (under revision 2012)
• MOP 803 - Operation of the Opto Display Software Data Acquisition System (DAS) in the Small
Chamber Laboratory (under revision 2012)
• MOP 806 - Operation of the Clean Air System for the Small Chamber Laboratory (under
revision 2012)
• MOP 808 - Determination of Small Chamber Formaldehyde Emission using DNPH-Coated
Silica Gel Cartridges
• MOP 811 - Collecting Air Samples Using Sorbent Tubes
• MOP 812 - Collection and Extraction of Air Samples on DNPH-Silica Gel Cartridges
• MOP 826 - High Pressure Liquid Chromatography (HPLC) Analytical Procedures
-------
SOP No. 880
Revision 1
June 29,2012
Page 6 of 23
• MOP 871 - Glassware Cleaning Procedures for Small Chamber Lab
8. EQUIPMENT AND SUPPLIES
Below is a list of equipment associated with the small chamber laboratory that is used for this
procedure.
• Small environmental chamber system (Table 1)
Table 1. Small Environmental Chamber System Components
Component
Manufacturer
Model #
Manufacturer's Location
Clean Air System
Pressure Regulator
Compressed Air Dryer
Carbon Trap
Moisture Trap
Pure Air Generator
Pressure Regulator
Mass Flow Controllers (8)
Constant Temperature Bath
1000ml_ Round Bottom Flask (4)
Midget Impinger Bubbler (4)
Wilkerson
Hankison
Supelco
Supelco
Aadco
Norgren
Teledyne
BlueM
Prism Glass
Prism Glass
B18-03-FKOO
SSRD1 0-300
24565
23992
737-11 A
B736-2AK-API-RMG
HFC-E-202
MR 3240C-1
PRG-5795-03
PRG-5030-23
Richland, Ml
Ocala, FL
Bellafonte, PA
Bellafonte, PA
Cleves, OH
Littleton, CO
Hampton, VA
Blue Island, IL
Raleigh, NC
Raleigh, NC
Chamber System
Incubator
Inlet RH Probes (4)
Internal RH Probes (4)
Thermocouples (4)
So-Low
Vaisala
Vaisala
Pyromation
C-SCN4-52-8
HMT333
HMT335
E-Type
Cincinnati, OH
Helsinki, Finland
Helsinki, Finland
Fort Wayne, IN
DAS System
Opto Control System
Electrical Control Box
Opto Operation Computer
Opto 22
Carotek, Inc
Dell
B3000
AT-607983
Optiplex 745
Temecula, CA
Mathews, NC
Round Rock, Tx
-------
SOP No. 880
Revision 1
June 29,2012
Page 7 of 23
Sampling Vacuum System (Table 2)
Table 2. Sampling System Components
Component
Vacuum Pump
Mass Flow Controllers (4)
Mass Flow Control Box
Gilibrator
Flexible PFA (Perfluoroalkoxy) Tubing
Flexible Silicone Tubing
3-Port Glass Sampling Manifold
Manufacturer
Welch
Coastal Instruments
Porter Instrument Co.
Sensidyne
Fisher Scientific
Fisher Scientific
Prism Research Glass
Model #
2565B-50
FC-260
CM4
800286
—
—
—
Manufacturer's Location
Skokie, II
Burgaw, NC
Hatfield, PA
Clearwater, FL
—
—
Raleigh, NC
• DNPH-Silica Gel Cartridge (Waters Sep-Pak®) - Short Body, 55-105um
http://www.waters.com/waters/partDetail.htm?partNumber=WAT037500&locale=en_US
• Extraction Supplies
o Acetonitrile, HPLC Grade (Fisher Scientific)
o 5 %" Glass Pasteur Pipettes (Fisher Scientific)
o 5 ml Glass Syringe (Fisher Scientific)
o 5 ml Class A volumetric flasks (Fisher Scientific)
• Film Holder- provided by Virginia Tech
9. PROCEDURE
The following procedure is specifically for the small chamber tests associated with the
Formaldehyde Reference Material. Chamber preparation, setup, and empty chamber background
checks should be completed before the arrival of the test films.
9.1 Chamber Cleaning
1. Wash chambers, faceplate, and chamber O-ring thoroughly in the sink with detergent (Liquinox
or Sparkleen) and hot water.
2. Follow with three rinses of warm water and then three rinses of Dl water. Dry chambers and
faceplate with Kimwipes (Fisher Scientific).
-------
SOP No. 880
Revision 1
June 29,2012
Page 8 of 23
3. Diffusers and sampling manifolds are cleaned in a similar manner and then placed in the oven
at 125 °C. Fans are cleaned with isopropanol wipes then baked at 125 °C for 10 minutes then
wiped again with an isopropanol wipe and dried with a Kimwipe. Nitrile gloves should be worn
while cleaning the chambers.
9.2 Chamber Reassembly
1. Reassemble the chamber with the small-hole diffuseron the inlet air side of the chamber cover
and the large-hole diffuser on the outlet side of the cover.
2. If a mixing fan is desired for the chamber test, install a 40mm brushless DC cooling fan by
snaking two wires through the inlet diffuser holes on the inside of the chamber and connecting
them to the fan wires. Test the fan before sealing to ensure proper wiring.
3. The fan is suspended in the back center of the chamber by attaching thin steel wire diagonally
from the end of each diffuser creating an "X" between the inlet and outlet diffuser manifolds.
The fan is attached at the center of the "X" with about 11/4" of small steel wire.
4. The fan is operated at the 12 volt setting on an AC/DC converter that rotates the fan at 6500
rpm ±10% and a nominal airspeed at the surface of 7.7CFM ± 10%.
9.3 Chamber Preparation for Testing
Experimental conditions for the reference material testing are shown in Table 3. The chambers
should be set to the test parameters and operated with the film holder in place at least 16 hours
prior to collection of a chamber background sample.
-------
SOP No. 880
Revision 1
June 29,2012
Page 9 of 23
Table 3. Experimental Conditions for Small Chamber Reference Material Tests
Parameter
Temperature (in chamber) a, °C
Relative humidity(inlet air) b, %RH
Airflow0, m3/h
Air exchange rate, h-1
Test materials
Loading factor d, m"1
Substrate surface area e, m2
Test period f, days
Value
25 ±0.5
50 ±5
or otherwise determined
0.053 ± 5%
1 ± 0.05
VT Reference material
0.397 ± 0.02
0.021 ± 0.001
or otherwise determined
4 or longer
As measured by a Pyromation E-Type thermocouple (calibrated annually by EPA Metrology Laboratory).
b As measured by a Viasala model HMT333 Humidity Transmitter with HUMICAP180 humidity sensor
(calibrated annually by EPA Metrology Laboratory).
Air input to the 53 liter chamber as measured by a Gilibrator.
d Loading factor (m"1) = surface area of the substrate (m2)/volume of the chamber, (m3).
e Information provided by Virginia Tech or measured with NIST calibrated calipers.
f Testing period is a nominal value that can be extended or stopped at anytime during the test.
The specific procedures for the small chamber tests are listed below:
1. At least 4 days before film arrival, clean chambers as detailed above.
2. Set up chambers to specified parameters: mixing fan blowing up and placed to the rear of the
chamber.
3. Flush for at least 16 hours - collect single DNPH background for 2 hours at 350 mL/min to
validate cleanliness of chamber.
4. Analyze the samples, if the results show an elevated formaldehyde area response above the
PQL response, repeat steps 1-4.
5. At least one day before the film test is to be conducted, clean the film holder using the same
procedure as used for the chamber parts and place a holder without the film in each of the
small chambers
6. Reseal chamber and flush under test parameters for a minimum of 16 hours- collect duplicate
DNPH backgrounds for 2 hours at 350 mL/min.
-------
SOP No. 880
Revision 1
June 29,2012
Page 10 of 23
7. Analyze the samples, if the results show an elevated formaldehyde area response above the
PQL response then re-clean holder and repeat steps 5-7.
8. Remove and analyze the background samples. The chambers are ready for testing.
9.4 Reference Materials
The reference materials to be tested on this project have been developed and prepared by Virginia
Tech. The samples will be overnight shipped from VT.
The date and time the film is removed from the loading vessel is considered as time zero. Each film
that is supplied will be numbered with the loading date and time as well as the position in the
loading vessel. A chain of custody will be delivered with each sample batch (Appendix A). The gas-
phase formaldehyde concentration and the polymerized formaldehyde concentration will also be
included on the COC. Films from each loading batch are considered equal; however, the films from
different loading batches may vary slightly. The films will be removed from the loading vessel and
immediately packaged for shipping by Virginia Tech then overnight shipped to Testing Laboratory.
Upon receipt of the films from VT, the packaged films will be removed from the cooler documented
as to their ID, temperature upon receipt (if provided), and size. The chain of custody (COC) form
received with the samples will be dated, initialed and placed with the samples.
Figure 1. Film Holder with Film in Chamber
1. Remove the test film from its packaging. Make sure that the films have been individually
wrapped (no piggy-back films). If multiple films are in the package select another package.
10
-------
SOP No. 880
Revision 1
June 29,2012
Page 11 of 23
2. Open the chamber and remove the film holder
3. Place the film in the holder (Figure 1) and place the holder back in the chamber.
4. Reseal chamber, record time (time zero for the start of the test) and begin sampling as outlined
in Table 4.
5. After the test is complete remove the film from the chamber and wrap with 3 layers of aluminum
foil, label and place in freezer until results are reviewed.
Table 4 outlines the proposed sampling schedule for the small chamber tests evaluating the shelf
life of the films. Single samples will be collected at 350 mL/min from one port of the 3-port glass
manifold. Duplicate samples will be collected at 350 ml/min from 2-ports of the 3-port glass
manifold (MOP812). The elapsed time is calculated using the time zero start time of the test and
the midpoint of the sampling duration. The sampling duration and vacuum flow rate through the
sample media determine the volume. The volume of sample air to be collected is determined by
the calibration range of the instrument being used to analyze the extracted samples. A large
enough volume must be collected to at least provide a target analyte concentration above PQL. An
optimum sample volume would give the recovery of the target analyte concentration in the mid-level
of the calibration. With DNPH extracts if the target analyte is above the highest calibration standard
concentration the solution can be diluted to the acceptable range.
11
-------
SOP No. 880
Revision 1
June 29,2012
Page 12 of 23
Table 4. Sampling Schedule for Small Chamber Tests
Elapsed Time (Hrsf
-1 - Background
0.1 (6min)
0.3(18 min)
0.5(30 min)
0.7 (42 min)
1
2
4
6
8
10
24
28
32
48
72
96
Total number of primary
samples
Duplicates
Field blanks
Field controls (when
needed)
Total number of samples
per test
Primary Sample
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Duplicate Sample
X
X
X
X
X
X
16
6
3
3
28
Sampling time in hours from start of chamber test to midpoint of sample duration
9.5 DNPH Sample Extraction and Analysis
All DNPH sample cartridges will be extracted and analyzed using established methods: MOP 812 -
Collection and Extraction of Air Samples on DNPH-Silica Gel Cartridges and MOP 826 - High
Pressure Liquid Chromatography (HPLC) Analytical Procedures. Any deviation from these methods
will be reported in the final report.
12
-------
SOP No. 880
Revision 1
June 29,2012
Page 13 of 23
10. Reporting Data
Data from all tests will be reported in electronic files. An Excel notebook (Appendix B) will be sent to
the tester for data reporting. Reported data will include the concentration of target carbonyls at each
sampling time, operating parameters of tests, and results for analyses of QC samples. Written and
verbal communications will be with Dr. Xiaoyu Liu (EPA PI). Table 5 details the data reporting
requirements associated with these tests
Table 5. Data Reporting Requirements
Measurement Parameters
Formaldehyde concentrations
Volume
Temperature
Relative Humidity
Air Exchange Rate
Air Flow
Elapsed time
Test duration
Unit
ug/m3 and ppb
liters
°C
%
h'1
mL/min
hour
hours
13
-------
SOP No. 880
Revision 1
June 29,2012
Page 14 of 23
APPENDIX A
Chain of Custody Form
14
-------
SOP No. 880
Revision 0
June 29,2012
Page 15 of 23
CHAIN OF CUSTODY FORM
Page_1 of
coc#
Sent from:
Contacts Company Name
A
B
City, State, Zip
c
5 reject NanefLocstiort (City, Statel
D
Item
#
1
Temp
Mann
Film ID
xxxxxx-xx-xxxxx
elephme:
E
F
G
FILM SIZE
Project #:
H
Film removed from loading vessel
Date
3/27/2012
erature upon receipt ° C =
acturer ;Model#
Time
10:00
Batch #
1
Position #
6
Monomer HCHO
cone
g/m3
192.00
Polymerized
HCHO cone
g/m2
0.0034
U
o
X
o
o
X
in
CO
x
o
LO
CO
Measured by: Devi
Relinquished By
Signature-
™
3 ate/Time:
Received By
5 linted Name
signature-
,,.
JatefTime:
REMARKS
ce ;
Relinquished By
Printed M em e:
Signs! ijies
""
Oa^Time
Received By
Printed Nsme
Signature:
F!ml
M"T"
15
-------
SOP No. 880
Revision 0
June 29,2012
Page 16 of 23
APPENDIX B
Electronic Notebook Data File
(See attached excel file)
16
-------
SOP No. 880
Revision 0
June 29,2012
Page 17 of 23
'•cunmfcation ic- fa)u
Test R&TMxl:
FI ea-M-r -natbe
Test <=*Mtidrtic.n.s:
Te-s.t Kfcvt«-i :*• Conditions :
.ee the trK^rr*>e^r conditions b-sto-w •fo'T eiacBi -ctnaar
OH^ m b« r 1 D
'=.-;-- -_-.;-
-,--: =--^ - — =
-r:^,f^i=^r ..^..^-^ L
— -::-- - ;. ,
_ = - l-^r - ~ -. : . ==l l-i
c: -us n- r>e r = - ~ ^
r;.- —a or t>s- r T t- 1— r -= - .r= r .. - =? ;^:
Far-
ir»iarr^b^-r ID
et= — T ~ = ,'O- = T-:zser :IT i-s=--z: 3^T = - — ^— - -
Tesi ~ - •::: T -—re
Te^-t Liuratic'-;. -Hr
Ofrvarnbeir Vo i_^i-^ l_
•=- :~ — '" -n . — ---. -— ~ = --. • .
n - u - _-_- _ ; ..— :*-*
:Z - - ^ -:..•' r T =• —
C ' -3 — - :;-^r ' T=- - -- --^ - = : _ - s C:
Far
^^'™^^-c^=:ssri*^
ber
Average
Average
STO'
STTJ
¥-irtSD
•=*4*CSI>
Te -j T F" I j - 1 F3 r-c- pxi :s--ec3
5i
L
^C-
^C-
2B
Ties* F» Lan P r-o|K35.<^
-------
SOP No. 880
Revision 0
June 29,2012
Page 18 of 23
Narrative
Description of Test procedure and problems encountered
18
-------
SOP No. 880
Revision 0
June 29,2012
Page 19 of 23
Chamber Preparation
Input required information in blue cells
Parameter
Temperature (*O)
Relative Humidity (%)
( li.miE.i i- Airflim Kate i in ' h i
Chamber Mivni-j L .111
Chamber Volume (in )
Cliamber Conditioned at Setpoints It with
1 "ham her Background Concentration of Foi
Frame (h)
mnldehyde (u,g/m [
Setpnint \"alue
25
50
O.O53
\'cs no
Input value
at least 16 h
at least 16 h
<2
Aetual \_ajue
l_ncertamtv
Measurement Instrument
19
-------
SOP No. 880
Revision 0
June 29,2012
Page 20 of 23
Sample Preparation
input required information in blue cells
Film II) =
Vpproiimailf Action
Duration Comments
1 . Remove film shipping package
2. Record date and. sample number (he ^ure to *i;ive spreadsheet wiih sample number in name)
3. Wail 5 minutes lo allow Him package to reach room temperature
4. Unwrap Sample (wear nylon or similar gjoves).
5. 1 'sing Uveeswrs. place reference male rial in sampk' holder (refei" to photo he low). If possible, do
llm procedure wrtli sample holder inside chamber.
(>. Seal chamber - This is lime B.TO tor the Icsl
7. Sample loading complete
Picture of Poly film in Film Frame in Chamber
Photo of Him Frame with film in EPA small chamber.
If possible, please place holder In chamber where airflow isthesameonetther side of ttie sample.
For example, in this picture tKe holder is parallel to the main airflow streamlines of the mixing fan, which Is blowing upward.
-------
ir Sampling hiformaiion
SOP No. 880
Revision 0
June 29,2012
Page 21 of 23
to finer sampling Jala following schedule above.
•iwnpk ". . >i
SOWOIIIII \k*l
^,nnr- ••
':!' il
!>!'. ?
;i]\ .
[IIS I!
•••r>IV (1
:.r, :
!,U 1!
IH\ li
•!>]'. !!
"IH1. ii
IH\ II
DQ ii
"I«V i!
•-IHV il
I)J', !i
v H\ IF
III !:
s H ;;
- IV
•- I1. 11
]'. D
"Dn1 -in
vlHV 41!
[>]'.' :.'
i>n ,f
"H1. , '
r*!1. i '
; i i •• • •
il', :.:
-E>]\ t>!
•-tMV d'
im »j
•l|\ ,:;
.•IH\ 11'
•;.np. • '
-F>]\ ft!
•t>]V 11'
••iil^i;'
= iMV 0-
Ml', ii'
*aiv »!
(H\ I)'
,if, r
•.IH\ ii'
..U. :,'
-IH\ ir
;>i\ :i'
AlBMfr \KTXlfc-
Swl*l*
VDlumt
i
EUM
Q.«l
.. n
1.1 fi'j
M [II !
OflCt
i, ,i
n.0',1
ii in.>
ii',.i
U.HO
uiltt
u fKi
l.i (HI
n.JHt
CLflO
mm
Ci.DO
11(111
II III'
U
-------
SOP No. 880
Revision 0
June 29,2012
Page 22 of 23
LC Calibration
Input required information in blue cells
Calibration Information
Formaldehyde Cone.
(ug/ml,)
RF
RF
( :ili lira 1 hill \ ;llufS
Awraie
Std. Drviiition
%KSI)
UT (mill)
RF
22
-------
SOP No. 880
Revision 0
June 29, 2012
Page 23 of 23
Concentrations of Formaldehyde Measured during Test
input required information in blue tells
Reporting Format:
nnn
nrvi
In calibration
Estimated value above quantification limit (bold italics)
Estimated value below quantification limit (bold suites-through)
Estimated value below detection limit (bold underlined)
Sample ID
Description
Khipscd Time
(hi
Sumple Volume
(D
Formaldehyde
Concentration (ue/m3)
Formaldehyde
C onrcntntlion (ppb)
Formaldehyde Emission (ppb)
0.40 0.60 040
Elapsed Time (h)
------- |