HI
C3
UNITED STATES ENVIRONMENTAL
PROTECTION AGENCY
Drywall Sampling Analysis
Background
Consumers from more than 10 States and the District of
Columbia have reported concerns related to drywall
imported from China that is in their houses. The
Consumer Product Safety Commission (CPSC) is the lead
federal agency for this issue. The U.S. Environmental
Protection Agency (EPA) is working with CPSC and the
Centers for Disease Control and Prevention-Agency for
Toxic Substances and Disease Registry (CDC-ATSDR), in
coordination with State and local authorities, to investigate
this matter.
To gather more information about Chinese drywall, CDC-
ATSDR requested that EPA conduct an elemental analysis
of Chinese drywall and compare it with drywall
manufactured in the United States.
Analysis of Drywall Samples
With CDC-ATSDR's concurrence, two wallboard samples
from Florida houses known to have been manufactured in
China were selected by the Florida Department of Health
(FDOH) for analysis. Additionally, four samples of U.S.-
manufactured drywall were purchased by EPA from local
stores in Edison, New Jersey and included in the analysis.
Prior to analysis, the thin layer of paint was scraped off of
the two Chinese drywall samples for metals analysis. The
paper was then separated from the solid (gypsum) material
of all six drywall samples and placed into separate glass
jars. The paper portions of the samples were analyzed for
metals, semi volatile organic compounds (SVOCs) and
formaldehyde. The gypsum samples were analyzed for
metals, SVOCs, volatile organic compounds (VOCs),
formaldehyde, sulfide, water soluble chlorides, total
organic carbon (TOC), pH and loss on ignition (LOI).
The results of this analysis will inform additional testing
by CPSC to help determine the compounds that may be
affecting residents and their houses.
Results
The results of the analysis are noted below. It is important
to note that the analysis included a very small sample size,
and the results of this testing may not be representative of
all drywall products. The analysis was conducted to
identify the elemental material contained in the drywall
samples and is not itself intended to establish a definitive
link between the drywall and the conditions being
observed in houses.
• Sulfur was detected at 83 parts per millions (ppm)
and 119 ppm in the Chinese drywall samples. Sulfur
was not detected in the four US-manufactured
drywall samples.
• Strontium was detected at 2,570 ppm and 2,670 ppm
in the Chinese drywall samples. Strontium was
detected in the US-manufactured drywall at 244 ppm
to 1,130 ppm. Total acid soluble sulfides were not
detected in any samples.
• Iron concentrations of 1,390 ppm and 1,630 ppm
were detected in the Chinese drywall samples and in
the range of 841 ppm to 3,210 ppm for the US-
manufactured drywall samples. Additional drywall
samples will be tested to determine whether the iron
is present as oxide, sulfide or sulfate.
EPA's analysis showed the presence of two organic
compounds in the Chinese drywall that are associated with
acrylic paints: propanoic acid, 2-methyl-, 2,2-dimethyl-l-
(2-hydroxy-l-methylethyl) propyl ester at estimated
concentrations of 58 ppm and 92 ppm, and propanoic acid,
2-methyl-, 3-hydroxy-2,4,4-trimethylpentyl ester at
estimated concentrations of 50 ppm and 84 ppm. These
compounds were not detected in the US-manufactured
drywall.
EPA will continue to work with its federal and state
partners to respond to this issue. EPA also is working with
a multi-agency and state technical group to develop an
indoor sampling protocol for use by CPSC and states to
conduct indoor air testing in houses suspected of
containing Chinese drywall. The group's goal is to
complete the protocol by June 30, 2009. EPA expects that
results from the indoor sampling will be evaluated by
CDC-ATSDR for possible health implications.
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
ENVIRONMENTAL. RESPONSE TEAM
Edison, New Jersey 08837
May 7, 2009
Ms. Lynn Wilder
Environmental Health Scientist
Agency for Toxic Substances and Disease Registiy
Department of Homeland Security
4770 Bufoid Highway, NE
Mailstop F-57
Atlanta, GA 30341-3717
Subject: Drywall Sample Analysis
Deai Ms,, Wilder,
The Agency fbi Toxic Substances and Disease Registry (ATSDR) contacted the
Enviionmental Response Team (ERT) of the USEPA Office of Superfund Remediation
and Technology Innovation (OSRTI) for analytical assistance with the Chinese-
manufactured diywall used in Florida, On March 5,2009, a teleconfeience was held
with ERT, ATSDR and the Floiida Department of Health (FDOH), The FDOH
piovided background information, including the work that had been previously
performed by contractors from Lennar and Knauf (a German company that
manufactures diywall in China) ATSDR requested that ERT conduct an independent
elemental analysis of'the Chinese diywall and compare it with the diywall
manufactured in the U S, With ATSDR's concurrence, six wallboard samples were
selected for analysis. Two diywall samples known to have been manufactured in China
were extracted by FDOH from affected homes in Florida, Four samples of U.S.-
manufactured diywall were purchased from local stores in Edison, New Jersey,
Diywall Sample Analysis
ATSDR requested that the ERT analytical laboratory provide support to analyze
drywall samples from China suspected of emitting rotten egg odors and causing copper
corrosion (e..g.., power switches, appliances) throughout the houses with complaints. The
corrosion of copper containing items may lead to releases of chlorofluoiocarbons
(CFCs) and natural gases, depending on their construction materials. Individuals
complaining about the diywall in their1 homes have also reported health issues such as
problems with asthma, respiratory irritation, breathing difficulties, coughing, insomnia,
eye irritation and headaches. At this time, FDOH has been unable to determine if these
issues are directly linked to the suspect drywall, To date, a relatively low number1 of
E
-------
samples have been analyzed, and the emission levels detected from samples tested in
the laboiatoiy are far lower than those typically associated with such symptoms.
Two Chinese painted dry wall samples extracted from Florida homes by FDOH
were shipped to Edison for' analysis by USEPA/ERT ERI purchased four US-
mamrfactured drywall samples from local stores for comparison.. First, the thin layer of
paint was scraped off'of two Chinese drywall samples for metals analysis, The top and
bottom layers of paper were separated from the solid (gypsum) material of all six
diywall samples and placed into separate glass jars,. The paper portions of the samples
were analyzed for' metals, semi volatile organic compounds (SVOCs) and
formaldehyde., The gypsum samples were analyzed for metals, SVOCs, volatile organic
compounds (VOCs), formaldehyde, sulfide, water soluble chlorides, total organic
carbon (TOC), pH and loss on ignition (LOI), Also, an optical microscopic examination
was conducted to determine the presence of'fly ash.
The diywall sample manufacturers and product names are as follows: US
Gypsum/Hamilton (US); PROROC/Certainteed (US); National Gypsum/Gold Bond
(US); GP/Tough Rock (US); Knauf/33928-20055 (China); and MIC/33966-12077
(China).. The ERT/REAC analytical methods were modified to analyze these samples,
as standard methods were not available in the area of sample digestion/preparation
procedures,
Analytical Methods
Semi Volatile Organic Compounds: The gypsum and paper portions of the
diywall samples were analyzed using ERT/REAC SOP #1805. A specific weight of
sample in giams is extracted with a 1:1 methylene chloride/acetone mix in a Soxtherm
extractor. The extract is concentrated, spiked with an internal standard mixture and
subsequently analyzed by gas chromatography/mass spectrometry (GC/MS), Target
analytes are identified by comparing the measured mass spectra and retention times
with those obtained from calibration standards acquired under the same operating
conditions used for the samples, Quantitation of each identified target analyte is
calculated based on the internal standard method.. The method was modified to
determine the presence of any non-target compounds via a library search for the
purpose of tentative identification,. The NIST/EPA/NIH Mass Spectral Library
containing more than 100,000 spectra was used. The elemental sulfur was analyzed
using the sample extracts by GC/MS using an ERT/REAC modified method,
Volatile Organic Compounds: The two Chinese and one US-manufactured
drywall gypsum samples were analyzed using ERT/REAC SOP #1807 A known
amount of gypsum is weighed into a 40-milliliter (mL) TeflonD-lined septum vial, 5
niL of'commercially available water suitable for' VOC analysis is added, and the sealed
vial is placed in the auto sampler, An additional 5-mL portion of V DC-free water
containing surrogate/internal standards is added by the autosampler In order to purge
the compounds out of the dry wall, the samples were heated for five minutes at 75 °C,
These samples were then purged with helium for 20 minutes at the same temperature,
-------
desorbed (trapped) onto the top for four minutes and injected into the GC and detected
using a 5975 MSB, The method was modified to deteimine the presence of any non-
target compounds via a library search for the purpose of tentative identification, The
NIST/EPA/NTH Mass Spectral Library containing more than 100,000 spectia was used.,
Metals; The gypsum samples were first screened using a NIT ON x-ray
fluorescence detector (XRF) to deteimine the presence of any metals. The XRJ will
help to ascertain whether additional metals that are not included in the Target Analyte
List (TAL) routinely analyzed by the laboratory need to be added. The gypsum, paper
and paint samples were analyzed for TAL metals using ERT/REAC SOP #1811,
Determination ofMetak by Inductively Coupled Plasma (ICP) Methods, and SOP
#1832, Determination of Mercury by Cold Vapor Atomic Absorption (CVAA), Based on
the XRF screening, strontium and sulfur were added to the list of analytes,
Formaldehyde, Sulfide, Total Organic Carbon: Analyses for these
compounds were contracted to outside laboratories.. Formaldehyde was analyzed by
high pressure liquid cinematography (HPLC), ultraviolet detection (UV) in accordance
with modified NIOSH Method 2016, For acid soluble sulfides, the gypsum samples
were distilled using EPA SW-846 Method 9030B, which separates the sulfides from the
matrix by adding sulfuric acid to the sample and heating to 70°C. The sulfide was
quantified using an iodometric method,, TOC was determined using a caibonaceous
analyzer in accordance with EPA Region II SOP #C-88..
Water Soluble Chlorides: A specific weight of'sample was mixed with a
known volume of water prior' to analysis.. Samples were analyzed using a five-point
calibration curve by a modified fenicyanide spectrophotometric technique, as outlined
in the Standard Methods for the Examination of Water and Wastewatei, Method 4500-
Cl-E,
Loss on Ignition and pH: Loss on ignition data were obtained by weighing a
known amount of sample into a crucible and igniting at 750°C using the modified
Standard Methods for the Examination of Water and Wastewatei, Method 2540G, A 5
percent weight by volume of a gypsum sample in water was prepared and mixed using a
magnetic stiner. The pH of the resulting aqueous solution was measured
electrometrically using a calibrated pH meter,
Alkalinity and Sulfate: Alkalinity was performed in accordance with the
Standard Methods fbi the Examination of Watei and Wastewater, Method 2320B, that
uses an acid titrant to measure the buffering capacity or' ability to react with acids to a
specific pH.. Sulfates were determined using EPA Region II SOP #C-19.
Optical Microscopic'Examination: The optical microscopic examination was
performed at the ERT-Las Vegas laboratory using an Olympus optical microscope.
-------
Discussion of the Results:
The significant differences between the Chinese diywall and the US-
manufactured diywall analysis aie as follows:
ERT analysis shows the presence of sulfui at 83 ppm and 119 ppm in the
Chinese diywall samples and sulfui' not detected in foui US-manufactured diywall
samples., The metal analysis shows the presence of strontium at 2,570 ppm and 2,670
ppm in the Chinese diywall samples, wheieas strontium was detected in the US-
manufactured diywall at 244 ppm to 1,130 ppm., The total acid soluble sultldes weie not
detected in any of the diywalls Further investigation is ciitical to deteimine the
presence of strontium as strontium sulfate or strontium sulfide using
x-iay diffraction
Iron concentrations of 1,390 ppm and 1,630 ppm were detected in the Chinese
diywall samples and in the range of 841 ppm to 3,210 ppm foi the US diywall samples,
The highest concentration of'iron detected in the National Gypsum/Gold Bond diywall
was twice as high as the amount found in the Chinese diywall An investigation will be
done using additional diywall samples to deteimine whether the iron is present as oxide,
sulfide 01 sulfate,
No evidence of'fly ash in the Chinese drywall samples was noted based on the
optical microscopic examination.
The ERT/REAC SVOC analysis results show the presence of'two organic
compounds in the Chinese dtywall, as tentatively identified by the mass spectrometiy
library search for the Chinese diywall. The FDOH has requested that ERT further
investigate these compounds.. The two compounds were propanoic acid, 2-methyl-, 2,2-
dimethyl-l-(2-hydioxy-l-methylethyl) propyl ester (CAS # 74367-33-2) at estimated
concentrations of'58 and 92 ppm, and propanoic acid, 2-methyl-, 3-hydroxy-2,4,4-
tiimethylpentyl ester (CAS # 74367-34-3) at estimated concentrations of 50 and 84
ppm, These compounds were not detected in the US-manufactured diywall, ERT
analyzed two samples foi VOCs by GC/MS. The analyses confirm the presence of'the
above two compounds in the Chinese drywall, as tentatively identified by the mass
spectrometiy library search, ERT is in the process of'obtaining standards of'propanoic
acid, 2-methyl-,.2,2-dimethyl-l-(2-hydroxy-l-methylethyl) propyl ester (CAS # 74.367-
33-2) and piopanoic acid, 2-methyl-, 3-hydioxy-2,4,4-trimethylpentyl ester (CAS #
74367-34-3) to confirm the findings,. The literature search reveals that these
compounds are found in acrylic paints as reported in the following website:
http://www2.mst.dk/comiTion/Udgi vramme/Fiame.asp?http://www2.rnst.dk/udgi
v/publications/2008/978-87-7052-763-7/html/kap02 eng.htm
The summary of analytical results of'the six drywall (gypsum, paper, and paint)
samples is presented in Summary Table 1, The semi-quantitative XRF data for gypsum
4
-------
analysis are presented in Table 2, The tentatively identified compounds detected by the
GC/MS library search for the SVOC analysis are presented in Table 3 for the gypsum
and paper portions of the drywall samples,.
Work in Progress
The additional diywall samples to be received from CPSC will be analyzed
semi-quantitatively for1 calcium sulfate, strontium sulfide, strontium sulfate, pyiites and
iron oxide by x-ray diffraction. The drywall samples from the United States and China
will also be analyzed for VOCs, SVOCs, metals including strontium, sulfide, sulfite,
formaldehyde, TOC and LOT, An optical microscopic examination for fly ash will also
be conducted,. Based on these analyses and the chamber study, ERT will conduct indoor
aii monitoring in Florida and Louisiana in three test houses for predetermined
parameters. A QAPP is under preparation for the Technical Workgroup to review based
on the available infbimation to date, and will be modified based on any new
information.
If there are any questions, please call me at 732-321-6761
Sincerely
Enclosures
cc: David Krause, FDOH .
Barnes Johnson, OSRTI
Arnold Layne, OSRTI/TIFSD
Jeff Heimerman, OSRTI/TIFSD
Dave Wright, ERT
Harry Compton, ERT
-------
e
a
4>
3
«
•r
[Sample No.
<
K
5
c
£
c
o
m
•o
o
3ypsum/G
c
S
I
c
1
1
€£>
to
y
to
M
CO
CM
55
W
C
3
m
X
a.
Sample ID
M
US
0) (
(S
.= -
I f
O
CO T
-j (.
Method
=
•*? m
c-i ^_
.8
^
t *
3 J
CO
3 g
|)H of 5% sluriy
C3
"£
E
I
e
c
I
S'
01
E
ai
a
p
=?
as
§ P
O JJ
£
^ CM <
*i|
0 °? t
S3:
§ 5 S
; 8 g
l§f
5 eo a
REAC SOP
i r i
111
s £ ^
11
D
g a
^
\*
S ?
3 O
: ^
Chromium
^
V
I
$
o
V
cc
o
3
q
m
REAC SOP
a>
0
rfS
O
s
S
V
cc
R
s
as
REAC SOP
OJ
^
^
1
O
ac
o
a
0
1
(XI
D
O
i
£
a
"8
^
3
5
S
«^
V
?
CO
i
ej
33
REAC SOP
9
"8
i
3
S
C
c5
s
a
I
1
1
£>
D
REAC SOP
U
•g
^
I
c-
i
ifi
pv,
co
CD
s
CO
REAC SOP
03
1
JS
| Manganese
c
V
§
o
K
O
S
o
JO
d
§
(D
REAC SOP
a
H
1
1
1
10
a •
31
II
*3 ;
CO CE
/> w
o C
yj i.
® a
(S S
Q 1
p i
ga
(_. y
I|
§1
1 S
is
i?
? D
I REAC SOP
0
1
S*
ur
V
1
V
1
V
CD
1
V
B
1 REAC SOP
at
TJ
o
•**•
i
o
S70/57
S
5
i REAC SOP
3
O
'C
1
ra
(Stfontium IDrvv»
i
2
Z
SNJ
CM
i
®
i REAG SOP
rr
O
,-,
Strontium (Pain
1
1
f
o
1
V
1
SM2320S
?n'
i
O
S
1
ca
1
i
v
f
S&A232Q8
j2
=
3'
3s
15
-i£
s
f
*
t
•*
7
S
I
o
V
e*
qj
V
o
V
5*
|
•^
J3
5
T!
««
3
C3
i
5680QO
§
(0
r\
d
I
CO
0
0
5
S
Q
t/3
a
CB
I
3
40
5
O3
S3
I
S
CM
S
O
CO
J=
o
S
o
isWM^plioiqS
V
s
V
s
*>
ffl
V
»
5
o
1
DC
"O
TJ
O
•"
a
014/0.67 '
0.54/ND
1
1
•**
^
Q
_,
IO
O
led MOSH 2
T3
2
a.
ft
>
£r-
Fom^aldehvde (C
.
1
to
s
O
R
o
[TOTAL_ORgANl
<
I!
i
-------
a:
x
E
£
5
«
a.
>i
a
Q
H—
O
a
0)
3
a
CM
J)
Xi
fO
to
1
Q
D
I
CS
CO
3
1
s
2
o
t—
o
co
o
O5
—.
-r
o
^
0
*-
*
^
is
to
Z>
3
|
x
*>
«
D
T-
o
O
c
0
^J
-f-
o
0
T-
+
0
CN
3
^
O
uf/33928-20055
g
•N
•"»
CN
CO
o
03
^
7
*
o
\
CN
CO
_)
oc/Certamyeed
3
-*
to
1
o
—.
^"
o
c\
o
a
+
X
r\
»
_ (
T!
O
J
13
5
^
03
2.
O
5
5
3
*
tf<
CM
^
CO
p
~i
>
t_>
^~
o
T-
+
^
N
OJ
CO
— >
J
3
a
O
.3
•£>
M
a
m
(A
i p-
11
0 ^
£ X
3 e
t s
Q. Z
W £
a. *
a S
X (o
O (U
£ Q.
s= |
,ac SO
S *
S. 8
-5 *
0 C
3 £
trontium, Sulfur
line appears as
not be quar»tifie<
E
-------
a
=3
g
l
s
3
a
e
jf 4
i 10 CO
81
g
•s
is
s
ill
------- |