EPA/600/R-20/286 | August 2020
www.epa.gov/research
United States
Environmental Protection
Agency
&EPA
Characterization of Drywall
Products for Assessing
Impacts Associated with
End-of-Life Management
Office of Research and Development
Homeland Security Research Program
-------
EPA/600/R-20/286
September 2020
Characterization of Dry wall Products for
Assessing Impacts Associated with End-of-
Life Management
by
Xiaolan Huang Ph.D.
Pegasus Technical Services Inc.
Cincinnati, Ohio
EP-C-1 5-01 0
Thabet Tolaymat Ph.D.
Homeland Security and Materials Management Division
Center for Environmental Solutions and Emergency Response
Cincinnati, Ohio, 45268
-------
Notice/Disclaimer Statement
The preparation of this report was supported by funding by the Office of Research and Development of the U.S.
Environmental Protection Agency (EPA) under contract numbers EP-C-15-010 with Pegasus Technical Services
Inc. This report has been subject to both internal and external Agency review and has been approved for
publication as an EPA document.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency
strives to formulate and implement actions leading to a compatible balance between human activities
and the ability of natural systems to support and nurture life. To meet this mandate, EPA's research
program is providing data and technical support for solving environmental problems today and building
a science knowledge base necessary to manage our ecological resources wisely, understand how
pollutants affect our health, and prevent or reduce environmental risks in the future.
The Center for Environmental Solutions and Emergency Response (CESER) within the Office of
Research and Development (ORD) conducts applied, stakeholder-driven research and provides
responsive technical support to help solve the Nation's environmental challenges. The Center's research
focuses on innovative approaches to address environmental challenges associated with the built
environment. We develop technologies and decision-support tools to help safeguard public water
systems and groundwater, guide sustainable materials management, remediate sites from traditional
contamination sources and emerging environmental stressors, and address potential threats from
terrorism and natural disasters. CESER collaborates with both public and private sector partners to foster
technologies that improve the effectiveness and reduce the cost of compliance while anticipating
emerging problems. We provide technical support to EPA regions and programs, states, tribal nations,
and federal partners, and serve as the interagency liaison for EPA in homeland security research and
technology. The Center is a leader in providing scientific solutions to protect human health and the
environment.
Gregory Sayles, Director
Center for Environmental Solutions and Emergency Response
iv
-------
Table of Contents
Notice/Disclaimer Statement iii
Table of Figures vi
Table of Tables viii
Acronyms and Abbreviations ix
Executive Summary xi
1. Introduction 1
2. Materials and Methods 3
2.1. Sample Collection and Preparation 3
2.2. Drywall Characterization 3
2.2.1. Moi sture content 3
2.2.2. Total carbon and sulfur content by combustion 4
2.2.3. Organic compounds in drywall 4
2.2.4. X-ray diffraction (XRD) Analysis 6
2.2.5. Total metals by EPA Method 3051A and acidic extraction 6
2.2.6. Cumulative water-soluble SO4 and metals 8
2.3. Drywall Leaching Behavior 8
2.3.1. Leaching processing kinetics 8
2.3.2. Liquid-solid partitioning as a function of liquid-to-solid ratio in solid
materials - USEPA Method 1316 8
2.3.3. Mass transfer rates -EPA Method 1315 9
2.4. Statistical Analysis 11
2.5. Quality Metrics 11
3. Physical and Chemical Properties 14
3.1. Moi sture C ontent 14
3.2. Total Sulfur and Carbon Content 17
3.3. Trace Organic Compounds 21
3.4. Crystalline Mineral Phases 22
3.5. Total Acid Extractable Sulfur and Metals 23
3.6. Water-Extractable Sulfur and Metals 41
4. Leaching Behavior of Drywall 48
4.1. Kinetics of Leaching Processing 48
4.2. Liquid-Solid Partitioning Tests 57
4.3. Monolithic Leaching Tests (MLs) 75
5. Conclusions 85
5.1. Drywall Characteristics 85
5.2. Leaching Behavior of Drywall 86
6. References 88
v
-------
Table of Figures
Figure 3-1: WLOI of gypsum of drywall board 15
Figure 3-2: Distribution of Formaldehyde content in drywall board 22
Figure 3-3: Typical XRD pattern of gypsum of drywall board 23
Figure 3-4: Distribution of the macro-elements in the gypsum from drywall. The box-and-whisker plots
show the following: the minimum value, the 25th quartile, the median, the 75th quartile, the
maximum value 25
Figure 3-5: Distribution of the micro-elements in the gypsum from drywall. The box-and-whisker plots
show the following: the minimum value, the 25th quartile, the median, the 75th quartile, the
maximum value 26
Figure 3-6: Positive correlations among elements in the gypsum from drywall 28
Figure 3-7: Effect of dilution temperature on the relative metal content of gypsum from drywall 28
Figure 3-8: Effect of extraction time on the calcium, sulfur and strontium content in gypsum from
drywall 31
Figure 3-9: Effect of extraction time on the other components in gypsum from drywall 32
Figure 3-10: Calcium and sulfur content in the gypsum from drywall by three methods 33
Figure 3-11: Al, Si, Fe, Ti, Mg, K and Na (mg kg"1) content in the gypsum from drywall by three
methods 35
Figure 3-12: Sr, Ba, Mn, Cu, Zn, Ni, Se and P (mg kg"1) content in the gypsum from drywall by three
methods 38
Figure 3-13: Water-extractable sulfate, calcium, and CaSC>4 in the gypsum from drywall 41
Figure 3-14: The molar ratio of Ca/S in the water extraction in the gypsum from drywall 42
Figure 3-15: Effect of gypsum sources on the concentration of Ca and SO/dn the water extraction 43
Figure 3-16: pH of the extraction in the gypsum from drywall 43
Figure 3-17: Water-extractable elements (other than Ca) in the gypsum from drywall 44
Figure 3-18: Water-extractable Sr in the gypsum from drywall 45
Figure 3-19: Water-extractable Ba in the gypsum from drywall 45
Figure 3-20: Water-extractable Mg in the gypsum from drywall 46
Figure 3-21: Water-extractable P in the gypsum from drywall 47
Figure 3-22: Water-extractable Si in the gypsum from drywall 47
Figure 4-1: Kinetics of SO4 and Ca of different drywalls 48
Figure 4-2: Kinetics of saturation index (SI) of anhydrite, gypsum and CaCC>3 H2O 49
Figure 4-3: Kinetics of conductivity of different drywall 50
Figure 4-4: Kinetics of pH of different drywalls 50
Figure 4-5: Kinetics of Sr concentration and saturation index (SI) of celesite (SrSC>4) in the solution... 51
Figure 4-6: Kinetics of Si concentration and saturation index (SI) of SiC>2 (am, gel) in the solution 52
Figure 4-7: Kinetics of Ba concentration and SI of barite (BaS04) and witherite (BaCCb) in the solution
53
Figure 4-8: Kinetics of P concentration and calcium phosphate double function plot of the solubility for
different drywalls 55
Figure 4-9: Equilibrium time and composition changes in drywall (group I, average of A, B, G and I). 56
Figure 4-10: Equilibrium time and composition changes in drywall (group II, Drywall-L) 57
Figure 4-11: Effect of L/S ratio on the pH of drywall leachate 58
Figure 4-12: Effect of L/S ratio on the concentrations of different constituents in the Drywall-A
leachates 60
Figure 4-13: Effect of L/S ratio on the concentrations of Sr, Mg, B, and DOC in five drywall leachates62
Figure 4-14: Effect of L/S ratio on the concentrations of Si in five drywall leachates (the linear
correlation for Drywall L was from the ratio 40 to 400) 63
vi
-------
Figure 4-15: Effect of L/S ratio on the concentrations of P in five drywall leachates 64
Figure 4-16: Effect of L/S ratio on the concentrations of Ba in five drywall leachates 66
Figure 4-17: Effect of L/S ratio on the concentrations of Fe in five drywall leachates 67
Figure 4-18: Effect of L/S ratio on the release amount of Ca, S, Ba, Sr, and Se from Drywall- A 72
Figure 4-19: Effect of L/S ratio on the release amount of P from different drywall leachates 72
Figure 4-20: Effect of L/S ratio on the release amount of Si from different drywall leachates 72
Figure 4-21: Effect of L/S ratio on the release amount of S from different drywall leachates 73
Figure 4-22: Effect of L/S ratio on the amount of Ca released from different drywall leachates 73
Figure 4-23: Effect of L/S ratio on the amount of B released from different drywall leachates 74
Figure 4-24: Effect of L/S ratio on the amount of Sr released from different drywall leachates 74
Figure 4-25: Effect of L/S ratio on the amount of Ba released from different drywall leachates 74
Figure 4-26: Effect of L/S ratio on the amount of Se released from different drywall leachates 75
Figure 4-27: Kinetics of pH and EC in M1315 leaching process 76
Figure 4-28: Cumulative loss vs total leaching time for the different components in drywall 79
Figure 4-29: Flux vs mean interval time for the different components in drywall 82
Figure 4-30: Dynamics of De of different compositions in drywall A 84
vii
-------
Table of Tables
Table 2-1: Drywall sample identification list 3
Table 2-2: Target organic compounds and detection limits (Dg kg"1) 4
Table 2-3: Elemental recovery (%) by nitric acid extraction at 90 oC 7
Table 2-4: Summary of Experimental QA/QC Checks 12
Table 3-1: The average of WLOI content of gypsum of drywall (%) 15
Table 3-2: The average of MC of gypsum for drywall after drying procedures (%) 15
Table 3-3: The MC of drywall board at 105°C (%) 17
Table 3-4: The sulfur content (%) of gypsum of drywall (air-dried weight basis) 19
Table 3-5: The sulfur content (%) of gypsum of drywall (air-dried weight basis) 19
Table 3-6: The carbon content (%) of gypsum of drywall (oven-dried weight basis with no follow-up air
drying) 19
Table 3-7: The carbon content (%) of gypsum of drywall (oven-dried weight basis followed by air
drying) 20
Table 3-8: The carbon content (%) of drywall (105 °C dry-weight basis) 20
Table 3-9: The sulfur content (%) of drywall (105 °C dry-weight basis) 20
Table 3-10: The content of PAHs in drywall (Dg kg"1) 21
Table 3-11: The Semiquantitative analysis (%) of mineral phases of gypsum from drywall 23
Table 3-12: The sulfur and metal content of gypsum from drywall by USEPA M3051A 24
Table 3-13: Effect of dilution temperature on the elemental composition of gypsum from Drywall-L .. 29
Table 3-14: Elemental composition of gypsum from drywall (mg/kg) 29
Table 3-15: Sulfur content of gypsum in drywall (%) 31
Table 3-16: Calcium content of gypsum in drywall (%) 32
Table 3-17: Strontium content of gypsum in drywall (%) 33
Table 3-18: Calcium and sulfate content of gypsum in drywall (%) by three methods 34
Table 3-19: Al, Si, Fe, Ti, Mg, K and Na content of gypsum in drywall (%) by three methods 36
Table 3-20: Sr, Ba, Mn, Se, P, Cu, Mn, Zn and Ni content of gypsum in drywall by three methods 37
Table 3-21: Elemental recovery (%) by nitric acid extraction at 90 °C * 39
Table 3-22: The sulfur and metal content of the 10-drywall paper samples by "Nitric acid extraction"
and EPA M3051A 39
Table 3-23: The sulfur and metal content of the 10 drywall boards by EPA M3051A 40
Table 3-24: The sulfur and metal content of the 10-drywall boards by "Nitric-acid extraction" 40
Table 3-25: The water cumulative extractable sulfate and metal content of gypsum from drywall 44
Table 4-1: Saturation index of minerals related to calcium in leachates 59
Table 4-2: Saturation index of minerals related to Sr in leachates 61
Table 4-3: Saturation index of minerals related to Si in the leachate 63
Table 4-5: Saturation index of minerals related to Ba in leachate 66
Table 4-6: Saturation index of minerals related to iron in leachate 68
Table 4-7: Summary of the linear dependence logarithmic concentration on the logarithmic L/S ratio in
drywall leachates 69
Table 4-8: Pore water concentration of drywall (mg L"1)* 71
Table 4-9: Cumulative release composition (%) in M1315 leaching processing 76
Table 4-10: Slopes of the linear equation of the logarithmic cumulative released compositions and
logarithmic total leaching time for drywall 77
Table 4-11: Slope and r2 of the equation between logarithmic flux and logarithmic mean leaching time82
Table 4-12: Weighted arithmetic mean De of drywall board (cm2 s"1) 83
Table 4-13: Leachability index (LX) of drywall board 84
viii
-------
Acronyms and Abbreviations
AES Atomic Emission Spectroscopy
ANS American Nuclear Society
ASTM American Society for Testing and Materials, now ASTM International
CDD Construction and Demolition Debris
CESER Center for Environmental Solutions and Emergency Response
De Diffusion Coefficient
DI Deionized (Water)
DOC Dissolved Organic Carbon
DCP Dicalcium phosphate
EC Electrical conductivity
ECD Electron Capture Detector
EPA U.S. Environmental Protection Agency
FGD Flue-gas desulfurization
GC Gas Chromatography
GFAA Graphite Furnace Atomic Absorption
H Hour
HPLC High-performance liquid chromatography
IC Ion Chromatography
ICDD International Centre for Diffraction Data
ICP Inductively Coupled Plasma
IR Infrared
L Liter
LOQ Limit of Quantification
LSP Liquid-solid partitioning
L/S Liquid-to-solid
LX Leachability Index
MC Moisture Content
ML Monolithic Leaching
MDL Method Detection Limit
MRL Method Reporting Limit
MS Mass Spectrometry
NIST National Institute of Standards and Technology
ORD Office of Research and Development
OCP Octocalcium phosphate
PAHs Polynuclear Aromatic Hydrocarbons
PDF Powder diffraction file
PCBs Polychlorinated Biphenyls
QA/QC Quality Assurance/ Quality Control
RIR Reference Intensity Ratio
RL Reporting limit
RPM Rotation per minute
RPD Relative percent difference
RSD Relative Standard deviation
SI Systeme International; International System of Units
SIM Selected ion Monitoring
SVOC Semivolatile Organic Compound
ix
-------
s/s
Solidification and stabilization
TCP
Tricalcium phosphate
TBT
Tributyltin
U.S.
United States
VOC
Volatile Organic Compound
WLOI
Weight Loss on Ignition
-------
Executive Summary
Drywall, also known as plasterboard, wallboard, sheetrock, and gypsum board, is a panel made
of gypsum plaster pressed between two thick sheets of paper. It is one of the major types of construction
and demolition debris in modern society. An estimated 3.9x 109 m2 drywall is used per year in North
America, representing approximately 50% of the world's use. In recent years, the generation of drywall
waste has been on the rise in North America, and most of this waste results from both construction and
demolition projects. The disposal of drywall waste in landfills is common, but this can potentially lead
to environmental concerns such as hydrogen sulfide gas generation and chemical leaching.
Given the potential concerns associated with drywall disposal in landfills, a scientific study is
needed to better understand the chemical composition of drywall products and how chemical
constituents may leach from these products upon contact with water. To produce such data, ten drywall
samples from the U.S. marketplace (produced by two different manufacturers) were collected,
representing a variety of different drywall products (e.g., regular, fire-resistant, and mold-resistant).
Drywall product characteristics examined in this study included mineral analysis, moisture content, total
sulfur, metal composition, water-soluble sulfur and metal concentrations, trace organic chemical
analysis, and two different leaching tests (M1315, and M1316). The results also provide methodological
insights related to the evaluation of solid wastes and similar materials with high calcium sulfate content.
The main results of this study were as follows:
1. Moisture content (MC) of the drywall was related to the temperature used for its determination.
The average MC of gypsum from drywall measured at 45, 105, 230, 400, and 550 °C was 0.45,
15.4, 20.4, 20.7, and 21.6%, respectively. The average MC of the drywall samples tested at 105
°C was 15.4%). The MC results at 150 °C were unstable because calcium sulfate exists at three
levels of hydration. Samples that were air-dried (not oven-dried at high temperatures) were
employed in this work, since the mineral phases can change during the high-temperature MC
analysis.
2. Drywall, including the gypsum core of the drywall board, contains a small amount of organic
carbon. The average total carbon and sulfur content of the gypsum samples using a combustion
methodology were 0.87 and 17.6%, respectively (air-dry weight basis). Formaldehyde was
detected at a concentration ranging from 500 to 8,500 jug kg"1, with a median and an average of
1,800 and 3,700 |ig kg"1, respectively. Tributyltin (TBT) was also detected in some samples,
especially in the mold control drywall. Polynuclear aromatic hydrocarbons (PAHs) were also
detected in some samples, attributed to the paper fraction of the drywall product.
3. The dominant mineral in the drywall products was gypsum, accompanied by small amounts of
hemihydrate and anhydrite. Calcium and magnesium carbonate and silica were also detected.
4. The total acid-extractable sulfur and metal concentrations of the gypsum core of the drywall
samples were investigated using different methods. The sulfur and calcium content average in
the gypsum samples was 13.7, 17.7, 18.3%, and 18.6, 24.5, 24.0%, using USEPA M3051A, 0.25
M HC1 extraction (24 hours (h)) and 10% HNO3 at 90 °C for 16h methods, respectively. The
average strontium content was 140, 175, and 189 mg kg"1, respectively. Re-precipitation is a
common occurrence after microwave digestion of materials with high amounts of calcium sulfate
minerals, and the re-precipitation was confirmed using digestion experiments followed by
dilution at different temperatures. The results suggest that analysts should be cautious of
measuring elemental concentrations of gypsum materials using USEPA M3051A; use of this
method might significantly underestimate the content of sulfur, calcium, strontium, and other
compositions. A new acid extraction procedure (10% HNO3) at water sub-boiling temperatures
(90 °C) for 16h) was developed in this work and is recommended for future work. The results of
extractable sulfur in this new procedure were similar to the total sulfur concentration measured
xi
-------
using the combustion technique and significantly higher than that measured using USEPA
M3051A. The results for calcium and strontium using the new procedure were also significantly
higher than the calcium and strontium measured using USEPA M3051A.
5. Water-extractable sulfate and inorganic element concentrations were studied by repeating a
water extraction procedure four times. Very high cumulative water-extractable sulfate and
calcium were observed in the gypsum samples tested. The other detectable components of the
water extracts were Sr, Ba, Mg, Fe, P, and Si. The average cumulative water-extractable SO4,
Ca, and Sr concentrations from the gypsum samples were 54.4±1.5%, 22.1±0.5%, and 193±211
mg kg"1, respectively. Based on the total sulfur content by combustion, 98.2% of the water-
extractable sulfur was in the form of sulfate (SO4). The average water-extractable calcium and
strontium content in the gypsum samples was 90%, and 95%, respectively, when the cumulative
water extraction concentrations were compared to those measured using the new acid extraction
procedure.
6. Kinetics leaching experiments were conducted for periods up to 2 months using five of the
drywall products at a fixed liquid-to-solid ratio (L/S = 20) and room temperature. The chemical
concentrations, pH, and conductivity in the leachates were measured and based on the chemical
measurements, and MINTEQ modeling, the kinetics of saturation index (SI) of controlling
minerals were assessed. Chemical equilibrium is a dynamic process, and there is no universal
time at which chemical equilibrium is reached for all constituents in the leachate. In many cases,
a constituent concentration (e.g., calcium) is not controlled by a single mineral phase, and the
changes in leachate concentration over time are related to changing mineral phases; however, for
most samples, equilibration time of one week was found appropriate.
7. Liquid-solid partitioning of inorganic constituents from the drywall samples was examined on
five drywalls using a modified EPA Method 1316 with ten different L/S ratios (from 2.5 to 400).
The linear dependence of logarithmic constituent concentration as a function of the logarithmic
L/S ratio was observed and found to be dependent on the saturation index (SI) of the minerals
controlling constituent equilibrium. When controlling mineral for a leached constituent was at an
unsaturated status (SI<0), the linear dependence was found valid for all the studied samples.
These relationships were further used to estimate the constituent concentrations in the pore
water. The estimated average of pore water concentrations of Sr, B, Ba, Zn, Cu, Mn, Ni, Co, Mo,
Cd, and Se were 10, 11.5, 0.29, 5.6, 6.3, 4.1, 0.2, 0.2, 0.23, 0.02, and0.6mgL_1, respectively.
The linear dependence relationship was noted when the constituent was released from a single
mineral phase.
8. Monolithic leaching tests were conducted using USEPA Method 1315 for five different drywall
products. A linear relationship was observed between logarithmic cumulative released
constituent concentrations, and logarithmic total leaching time with gypsum was the controlling
solid phase. The slopes of the linear equation indicated that the leaching process was controlled
by dissolution and not diffusion. The dominant species (sulfate, Ca), as well as Sr, leached
following a dissolution mechanism. A surface wash-off pattern, a delayed-release pattern, and a
depletion pattern were also observed for the other minor elemental constituents depending on the
mineral source and composition of the gypsum. The diffusivity of the leached constituents, as
well as the leachability index, were further calculated. The average weighted arithmetic means
diffusion coefficient (De) of S, Ca, Sr, Zn, Mn, Mg, P, Ba, Si, Fe and dissolved organic carbon
(DOC) from the samples was 60.2, 55.4, 41.9, 18.1, 6.5, 5.9, 0.45, 0.38, 0.30, 0.21, and 0.02 xlO"
8 cm2 s"1, respectively. The more highly leachable constituents were Zn, Sr, SO4, and Ca; the
moderately leachable constituents were Mg and Mn; and the relatively slow release constituents
were P, Fe, Si, Ba, and DOC. The leachability index of most constituents was between 8 and 9.
xii
-------
1. Introduction
Gypsum drywall - also commonly referred to as plasterboard, wallboard, sheetrock, or gypsum
board - is a panel comprised of the mineral gypsum (CaSC>4 2H2O) pressed between two sheets of
paper. Drywall is a construction product used for interior walls and ceilings in buildings and serves as a
more rapidly installed alternative to the traditional lath and plaster construction method. Approximately
4 billion m2 of drywall has been estimated to be used annually in North America, representing
approximately 50% of the overall amount used in the world (Founie, 2006). The home-building-and-
remodeling markets in North America over recent decades have increased demand for building
materials, and the gypsum drywall was one of the biggest beneficiaries of increased construction activity
as "an average new American home contains more than 7.31 metric tons of gypsum" (Olson, 2001).
When discarded, this material represents one of the larger components of construction and demolition
debris (CDD) in modern society (USEPA, 1998a; Townsend et al., 2004; USEPA, 2006; Somasundaram
et al., 2014; USEPA, 2014a; Jimenez Rivero et al., 2016).
Drywall is manufactured by first calcining source gypsum to remove part of the water (resulting
in CaSO4'0.5H2O) and then rehydrating the gypsum to produce a slurry. The slurry is then spread onto a
moving continuous sheet of paper, which is sandwiched between another layer of paper. After initial
drying, the drywall becomes hard and ready to cut into panels for a final drying process. The drywall is
then trimmed to the dimensions required, bundled, and sent to the market (USEPA, 2015). A variety of
gypsum drywall products are manufactured and sold in the U.S. and Canada, including regular
whiteboard, and products modified to provide greater fire-resistance, mold-resistance, and soundproofing
ability. Examples of drywall currently available on the market include:
• Regular whiteboard comes in thicknesses ranging from H-inch to 3/4-inch thickness.
• Fire-resistant ("Type X") drywall comes in different thicknesses and includes additives such as
glass fibers to provide necessary properties for improving fire resistance. In some cases, perlite,
vermiculite, and boric acid are added to additionally improve fire resistance.
• Green board is a drywall product that contains oil-based additives to provide moisture resistance
for applications such as bathrooms; the product is so-named because of the green-colored paper.
• Blue board includes a skim coat of plaster finish to provide additional water and mold resistance.
• Sound dampening drywall incorporates additional materials to limit sound transmission.
• Paperless drywall products do not include the paper coating and backing and are said to pose
fewer mildew problems if exposed to water.
• Specialty drywall products as lined with lead (to be used in the walls around radiological
equipment) and foil (to serve as a vapor barrier) are also manufactured and marketed.
The primary component of gypsum drywall is the gypsum, a mineral also known as calcium
sulfate dihydrate (CaSC>4 2H2O). Each gypsum molecule is composed of one molecule of calcium
sulfate (CaSC>4) and two molecules of water (FhO). By weight, the compound is 21% water, but by
volume, it is nearly 50% water. The source of gypsum used in drywall manufactured includes naturally
occurring gypsum deposits (geologically deposited from lakes and seawater). Recycled gypsum is also
widely used in drywall manufacture (Pedreno-Roj as et al., 2019; USEPA, 2015). The introduction of the
Clean Air Interstate Rule by the U.S. Environmental Protection Agency (EPA) in March 2005 required
power plants to "cut sulfur dioxide emissions, which necessitated that coal-fired power plants install
scrubbers (industrial pollution control devices) to remove sulfur dioxide present in the output waste gas.
l
-------
Scrubbers use the technique of flue-gas desulfurization (FGD), which produces synthetic gypsum as a
by-product. Though issues such as mercury release during calcining of FGD gypsum have been raised
(Heebink and Hassett, 2005), use of FGD gypsum as the gypsum source for drywall manufacture is very
common. In addition, post-consumer recycled gypsum (recovered from construction debris) is used as a
gypsum source (CDRA, 2015; USEPA, 2015).
The other main component of the drywall board is paper, which is often produced using recycled
paper products. Small amounts of other additives may be included in the manufacturing process,
including starch, paper pulp, unexpanded vermiculite, and those additional components required for
specialty products (see product descriptions above). Starch is added to help the paper adhere to the
gypsum core, and paper pulp is added to increase the core's tensile strength (resistance to lengthwise
pressure).
Gypsum drywall enters the waste stream from product manufacture, during building
construction, and as a result of building demolition. A variety of potential environmental concerns such
as generation and emission of hydrogen sulfide and leachate with elevated levels of minerals have been
raised regarding the management of the end-of-life of drywall (Venta, 1997; Townsend et al., 2004;
USEPA, 2006, 2014a, 2015; Jimenez Rivero et al., 2016), and with the many different types of drywall
products historically and currently in use, it is useful to better understand the trace constituents of
drywall products. The environmental impacts of drywall disposal in landfill are dependent on its
chemical and leaching characteristics. To address this need, ten drywall samples from the US market,
representing two different manufacturers and a variety of drywall products, were randomly selected for
evaluation in this study. A variety of chemical properties were measured on each of the drywall
products, including mineral analysis, moisture content (MC), total sulfur and metal concentration, water-
soluble sulfur and metals concentrations, selected organic constituents, and constituent leachability
using two different US. EPA Leaching Environmental Assessment Framework (M1315 and M1316).
The results add to the existing database on gypsum drywall properties that can be used to guide
sustainable materials management decision making. Lessons from several issues encountered during
drywall analysis will be of value to future researchers examining similar material streams.
2
-------
2. Materials and Methods
2.1. Sample Collection and Preparation
Ten dry wall samples were collected from home improvement and construction product retail
stores and represented two different product manufacturers (X and U). The samples collected and
analyzed are designated by Sample ID as A, B, C, D, E, F, G, H, I, and L (see Table 2-1).
Table 2-1: Drywall sample identification list
Sample
ID
Manufacturer
Type
Thickness
Date of
Manufacture
A
X
Lift Lite "TE"
0.5 inch
7/29/2014
B
X
Regular
3/8 inch
7/15/2014
C
X
Mold Defense
0.5 inch
7/24/2014
D
X
Firecheck type X
5/8 inch
7/22/2014
E
U
UL-Regular
0.5 inch
6/2/2015
F
U
UL-MoldTough
0.5 inch
5/28/2015
G
U
UL-Firecheck X
5/8 inch
11/7/2014
H
U
UL-regular
0.5 inch
8/11/2015
I
U
Regular
3/8 inch
7/20/2015
L
U
Mold Tough
0.5 inch
8/12/2015
Upon receipt at the laboratory, gypsum drywall sheets were logged and stored in a storage unit.
The paper was removed from the drywall for each sheet, and the paper and gypsum were stored as two
separate components (paper and gypsum). The gypsum component was further cut into squares of
approximately 6 mm and run through a rock crusher. The processed gypsum was then ground to a fine
powder and sieved using a USA Standard Testing Sieve system using an ASTM International (ASTM)
#10 sieve (2 mm) to yield the final sized particles. The paper component was cut into 2 to 3 mm squares.
The weight of each drywall component (paper and gypsum) was measured to determine their percentage
weight.
2.2. Drywall Characterization
2.2.1. Moisture content
The MC of the gypsum drywall samples was measured after the sample size was reduced to less
than 2 mm. The samples were placed in an oven, and MC was determined by ASTM C471M-16
(ASTM-International, 2016) and Method D-2216-10 (ASTM-Internationl, 2010). In Method C417M-16,
the weight loss on ignition (WLOI) at 45 °C was designated as the "free water" and the WLOI at 230 °C
as "combined water." This manner of describing MC is supported by the infrared spectroscopy and
thermal gravimetric analysis results from gypsum, as reported by Reidy et al. (2014). Carbon dioxide
(CO2) has been observed to be released from drywall gypsum at higher temperatures; the WLOI
between 230 °C and 550 °C reportedly relates to the decomposition of the organic matter, while the
WLOI between 550 °C and 1000 °C relates to the decomposition of carbonate (e.g., calcite or dolomite)
(Heiri et al., 2001; Wang et al., 2011; Galan et al., 2013). The MC of the paper samples was measured at
105 C based on Method D-2216-10 (ASTM-Internationl, 2010).
3
-------
2.2.2. Total carbon and sulfur content by combustion
The carbon and sulfur content of the gypsum samples was determined by infrared absorption on
a Leco CS230 Carbon/Sulfur Analyze (LECO Corporation, St. Joseph, Michigan, USA) (LECO, 2005).
Samples (pre-heated at different temperatures) were added to a crucible along with an accelerator and
then placed in an induction furnace combustion chamber. After closing the furnace, oxygen was purged
into the combustion chamber, and the induction furnace was started. The oxygen-rich environment,
combined with the sample inductive properties and the accelerator, resulted in sample combustion.
Carbon dioxide (with some CO) formed as did SO2. The gases were swept into the carrier stream, and
SO2 was measured in the first infrared cell (IR cell). Any CO present in the carrier stream was converted
to CO2 in the catalytic heater assembly, and the CO2 was measured in the second IR cell. The
concentrations of CO2 and SO2 were determined through a reduction in the level of energy at the
detector. Sample processing (the pre-heated temperature and time) and MC played a role in the results of
the total sulfur content.
2.2.3. Organic compounds in drywall
Concentrations of volatile organic compounds (VOCs) in the gypsum samples were tested using
EPA Method 8260B using gas chromatography (GC)/mass spectrometry (MS) (USEPA, 1996b). The
samples were purged with inert gas, and the effluent gas passed through a sorbent trap where the volatile
organics were trapped. After purging, the sorbent trap was rapidly heated and back-flushed onto the
head of a GC column. The GC column was temperature-programmed to separate the volatile
compounds, which were subsequently detected and identified using MS. The target VOCs included 48
compounds (e.g., acetone, bromodichloromethane, chloroethane, cyclohexane, dichlorodifluoromethane,
ethylbenzene, 2-hexanone, methylene chloride, styrene, trichloroethene, trichlorofluoromethane, and
total xylenes). The full list of these compounds and their detection limits and limits of quantification are
provided in Table 2-2
Semivolatile organic compound (SVOC) concentrations, including polynuclear aromatic
hydrocarbons (PAHs), in the gypsum samples, were determined using USEPA Method 8270C with
GC/MS (USEPA, 2014b). Briefly, SVOCs were extracted using acetone and methylene chloride in a
microwave (100 °C, 10 min, USEPA Method 3546 (USEPA, 2007d)). The target SVOCs included 65
compounds (e.g., acetophenone, anthracene, benzo(a)anthracene, benzo(b)fluoranthene,
benzo(g,h,i)perylene, benzo(k)fluoranthene, benzo(b)fluoranthene, di-//-butylphthalate, caprolactam,
chrysene, bis(2-ethylhexyl)phthalate, fluoranthene, indeno(l,2,3-cd)pyrene, phenanthrene, phenol, and
pyrene). The full compound list and their detection and quantification limits are provided in Table 2-2.
Table 2-2: Target organic compounds and detection limits (fig kg"1)
Compounds
MDL1
LOQ2
Compounds
MDL
LOQ
VOCs
SVOCs
Acetone
690
2.000
Dibenzofuran
50
99
Benzene
49
490
3,3 '-Dichlorobenzidine
300
990
Bromodichloromethane
99
490
2.4-Dichlorophenol
50
99
Bromoform
99
490
Diethyl phthalate
200
500
Bromomethane
200
490
2.4-Dimethylphenol
50
99
2-Butanone
400
990
Dimethyl phthalate
200
500
Carbon Disulfide
99
490
4.6-Dinitro-2-methylphenol
500
1.500
Carbon Tetrachloride
99
490
2,4-Dinitrophenol
890
3.000
Chlorobenzene
99
490
2.4-Dinitrotoluene
200
500
Chloroethane
200
490
2,6-Dinitrotoluene
50
99
Chloroform
99
490
Fluorene
10
51
Chloromethane
200
490
Hexachlorobenzene
10
51
4
-------
Compounds
MDL1
LOQ2
Compounds
MDL
LOQ
Cyclohexane
99
490
Hexachlorobutadiene
50
99
l,2-Dibromo-3-chloropropane
200
490
Hexachlorocyclopentadiene
500
1.500
Dibromochloromethane
99
490
Hexachloroethane
99
500
1,2-Dibromoethane
99
490
Isophorone
50
99
1,2-Dichlorobenzene
99
490
2-Methylnaphthalene
10
51
1,3 -Dichlorobenzene
99
490
2-Methylphenol
50
99
1,4-Dichlorobenzene
99
490
4-Methylphenol
50
99
Dichlorodifluoromethane
200
490
Naphthalene
10
51
1,1 -Dichloroethane
99
490
2-Nitroaniline
50
99
1,2-Dichloroethane
99
490
3-Nitroaniline
200
500
1,1 -Dichloroethene
99
490
4-Nitroaniline
200
500
cis-1,2-Dichloroethene
99
490
2-Nitrophenol
50
99
trans-1,2-Dichloroethene
99
490
4-Nitrophenol
500
1.500
1,2-Dichloropropane
99
490
N-Nitroso-di-«-propylamine
50
99
cis-1,3 -Dichloropropene
99
490
N-Nitrosodiphenylamine
50
99
trans-1,3-Dichloropropene
99
490
Di-«-octyl phthalate
200
500
Ethylbenzene
99
490
Pentachlorophenol
99
510
Freon 113
200
990
2,4,5-Trichlorophenol
50
99
2-Hexanone
300
990
2,4,6-Trichlorophenol
50
99
Isopropylbenzene
99
490
Acetophenone
50
99
Methyl Acetate
200
490
Anthracene
10
51
Methyl Tertiary Butyl Ether
49
490
Benzo(a)anthracene
10
51
4-Methyl-2-pentanone
300
990
Benzo(a)pyrene
10
51
Methylcyclohexane
99
490
Benzo(b)fluoranthene
10
51
Methylene Chloride
200
490
Benzo(g,h,i)perylene
10
51
Styrene
99
490
Benzo(k)fluoranthene
10
51
1,1,2,2-Tetrachloroethane
99
490
Di-«-butyl phthalate
200
500
T etrachloroethene
99
490
Caprolactam
99
500
Toluene
99
490
Chrysene
10
51
1,2,4-Trichlorobenzene
99
490
bis(2-Ethylhexyl) phthalate
200
510
1,1,1 -Trichloroethane
99
490
Fluoranthene
10
51
1,1,2-Trichloroethane
99
490
Indeno( 1,2,3 -cd)pyrene
10
51
Trichloroethene
99
490
Nitrobenzene
50
99
T richlorofluoromethane
200
490
Phenanthrene
10
51
Vinyl Chloride
99
490
Phenol
50
99
Xylenes (Total)
99
490
Pyrene
10
51
SVOCs
I'CHs
Acenaphthene
lu
51
PCB-1016
3.5
17
Acenaphthylene
10
51
PCB-1221
4.5
17
Atrazine
99
500
PCB-1232
7.9
17
Benzaldehyde
200
500
PCB-1242
3.3
17
l,l'-Biphenyl
50
99
PCB-1248
3.3
17
4-Bromophenyl-phenylether
50
99
PCB-1254
3.3
17
Butylbenzyl phthalate
200
500
PCB-1260
4.8
17
Carbazole
50
99
4-Chloro-3-methylphenol
50
99
4-Chloroaniline
99
200
bis (2 -Chloroethoxy)methane
50
99
Formaldehyde
500
1500
bis (2 -Chloroethyl)ether
50
99
2-Chloronaphthalene
20
98
2-Chlorophenol
50
99
4-Chlorophenyl-phenylether
50
99
5
-------
Compounds
MDL1
LOQ2
Compounds
MDL
LOQ
2,2'-oxybis( 1 -Chloropropane)
50
99
Tributyltin (TBT)
1.5
3.0
Dibenz(a,h)anthracene
10
51
MDL - method detection limit; 2LOQ - limit of quantification
Polychlorinated biphenyl (PCB) concentrations in the gypsum samples were analyzed using EPA
Method 8082 with the GC/electron capture detector (ECD) (USEPA, 2007b). PCBs were extracted using
1:1 acetone and methylene chloride in a microwave (100 °C, 10 mins, EPA Method 3546 (USEPA,
2007d)). Formaldehyde represents a target organic compound related to air quality and building
materials; it can be present in the drywall as a constituent of the glues used in drywall manufacture.
Formaldehyde concentrations in the gypsum samples were analyzed using EPA Method 8315A via high-
performance liquid chromatography (HPLC) (USEPA, 1996a). The sample was extracted using 0.1 M
ammonium acetate (pH 4.91- 4.95) for 18 hours (h) at room temperature with a 20:1 liquid-to-solid
(L/S) ratio. The concentration of tributyltin (TBT), a biocide used in some anti-fouling paints and
reportedly used in the manufacturer of some drywall products, was measured by first extracting with a
tropolone and hexane mixture, followed by analysis by GC/MS with selected ion monitoring (SIM)
(Krone et al., 1989).
Based on initial test results for the organic chemical constituents only the drywall paper was
further tested as the concentration of these constituents in gypsum was negligible.
2.2.4. X-ray diffraction (XRD) Analysis
Crystalline mineral phases in the gypsum samples were investigated from 5 to 110° 29 on a
Philips X'Pert Pro Diffractometer (Philips, Almelo, The Netherlands) using cobalt Ka radiation. The
powder diffraction file (PDF) patterns database from the International Centre for Diffraction Data
(ICDD) was employed for the search, match, and identification steps. A subset of reference patterns was
built for all drywall samples. The semi-quantitative phase analysis was performed by the X'Pert
HighScore Plus software (PANalytical B V Alemo, The Netherlands) using the CHUNG Normalized
Reference Intensity Ratio Method (RIR) (Chung, 1974). The relative intensity of each phase was given
by a scale factor determined by a least-squares fit through all matching reference pattern lines in X'Pert
HighScore. The concentration X of phase a was calculated using:
'(hklW
I re I
(hid),,
I(hM)jRIRjl(hkl).
where RIRoc = Reference Intensity Ratio (based on the relative net peak height ratio of the strongest line
(Fel = 100%) of the phase and of the strongest line of corundum, measured with copper Ka radiation in a
mixture of equal weight percentages), and I(hki)K = Intensity of reflection of hkl in-phase a (hkl are the
reflection indices). The normalization used in this method assumed that the sum of all identified phases
was 100% and that no unidentified crystalline phases or amorphous phases were present in the sample.
Only under these conditions can meaningful semiquantitative results be obtained.
2.2.5. Total metals by EPA Method 3051A and acidic extraction
The drywall samples (<2 mm), including both the paper and gypsum components, were acid-
digested using USEPA Method 3051A (USEPA, 2007c). A maximum of 0.25 g of representative solid
sample (air-dried) was digested in 10 mL of trace metal grade concentrated nitric acid (HNO3) using a
microwave heating processing with a two-stage program. Initially, the digestion vessel was heated to
175 ± 5 °C within 10 minutes, and then the vessel was maintained at 175 ± 5 °C for an additional 5
minutes. After digestion, samples were cooled to room temperature, and then the digested solution was
transferred to a centrifuge tube and brought to a volume of 50 mL with deionized (DI) water. This
6
-------
solution was centrifuged for 5 minutes at 5000 revolutions per minute (rpm) and filtered using a 0.45-
|j,m membrane filter to obtain a clear solution for elemental analysis; dilution was performed as
necessary. Elemental composition (Ag, Al, B, Ba, Ca, Cd, Co, Cr, Cu, Fe, K, Na, Mg, Mn, Mo, Ni, P, S,
Sb, Se, Si, Sr, Ti, V, and Zn) was measured by inductively coupled plasma atomic emission
spectroscopy (ICP-AES, Thermo Spectra-Tech, Shelton, CT) (EPA Method 6010C (USEPA, 2007a)).
Lead and As were further analyzed using Graphite Furnace Atomic Absorption (GFAA) by EPA
Method 7010 (USEPA, 1998b). The method detection limits (MDL) for elements in the aqueous phase
and the limits of quantification (LOQ) for elements in the solid phase for ICP-AES and GFAA are
presented in Table 2-3. Standard reference material 1633C (Coal Fly Ash), blank, spikes, and dry wall
sample spikes were also digested in every batch for quality control.
Table 2-3: Elemental recovery (%) by nitric acid extraction at 90 oC
Name
MDL
(mg L"1)
LOQ
(mg kg x)
Sample A-G (%)
Sample L-G (%)
Blank
(%)
Avg
(%)
Level I
Level II
Level-1
Level-II
Al
0.25
50
105
105
126
120
105
112
As
0.007
1.5
98
99
101
99
95
98
B
0.044
8.9
99
100
102
100
95
100
Ba
0.005
0.98
93
93
96
95
98
95
Cd
0.001
0.11
95
96
97
96
104
98
Co
0.001
0.27
93
94
95
94
104
96
Cr
0.038
7.63
92
92
94
93
99
94
Cu
0.005
1.07
102
104
103
102
106
103
Fe
0.030
6.00
101
100
101
98
106
101
K
0.568
114
119
119
115
117
98
114
Mg
0.170
34.1
107
117
97
97
108
105
Mn
0.007
1.4
98
98
98
97
105
99
Mo
0.004
0.76
99
99
101
100
104
101
Na
0.097
19.5
107
107
111
110
99
107
Ni
0.002
0.46
95
95
97
96
105
98
P
0.006
1.15
103
105
109
108
103
105
Se
0.015
3.1
101
102
104
103
98
102
Sr
0.005
0.98
99
97
113
106
107
104
Ti
0.020
4.0
99
100
104
103
102
101
V
0.026
5.3
98
98
99
98
105
100
Zn
0.015
2.9
97
97
100
99
96
98
Acid-extractable SO4 (0.25 M HC1) and metals concentrations were further determined based on
Sun and Barlaz (2015). Equilibrium time was 1, 4, 24, 72, 168, 336, and 772 hours at room temperature
and S/L ratio of 200. Calcium and sulfate solubility was reported to change depending on acid
concentration and environmental temperature (Hulett, 1902; Marshall and Jones, 1966; Freyer and
Voigt, 2003; Li and Demopoulos, 2006; Wang et al., 2013). In addition, a new approach to assess sulfate
and metal total amounts in the dry wall samples was tested by using 10% HNO3 in a water bath or oven
at 90 °C for 16 h. Acid concentration and temperature control calcium sulfate solubility and also can
affect the solubility of other elements (Wollmann and Voigt, 2008; Zeng and Wang, 2011; Wang et al.,
2015). A liquid-to-solid (L/S) ratio of 250 was used for all extractions, and samples were filtered (0.45
|j,m) and analyzed using ICP-AES and ion chromatography (IC). In a manner similar to M3051A, blank
spikes and drywall sample spikes were also included in the analysis.
7
-------
2.2.6. Cumulative water-soluble SO4 and metals
Water-extractable sulfate and metals cumulative concentrations were tested using the modified
method described by Sun and Barlaz (2015). The L/S ratio was 200:1 with an end-over-end rotation at
30±2 rpm, for 60 minutes at ambient temperature with the repeated replacement of DI water. All
solutions were filtered (0.45-|j,m) after extraction, and the residue and used membrane were returned to
the container for the next extraction. The same procedure was repeated four times until the conductivity
(EC) of the extraction was less than 50 |j,S/cm.
The content of SCUin each extraction after filtering was determined by IC. The metals in each
extraction after filtering were acidified (HNO3) and then analyzed by ICP-AES. The SO42" water-
extractable (Ldiw) results, as mmol kg"1, were calculated from the individual extract concentrations. The
concentration of SO42" (Ci) in each extraction was measured as a concentration in mmol L"1, and the
cumulative mass of SO42" (Ldiw-scm) leached from each sampling event was calculated as:
Ldiw-so4 = £Vl x Ci/MR
where Vl= volume of the extraction fluid and Mr= dry mass of test material in the extractor (in kg).
The total sulfur (St) content measured through combustion using the LECO CS230
Carbon/Sulfur Analyzer (LECO Corporation, St. Joseph, Michigan, USA) was converted to units of
mmol kg"1, and the percentage of water-extractable SO42" (Sdiw, %) was determined as:
w= ^04x100
The cumulative water-extractable metals, especially Ca, Mg, and Sr, were calculated similarly to sulfate.
2.3. Drywall Leaching Behavior
The leaching behavior of drywall was investigated using several different approaches.
Experiments were performed to examine: 1) drywall leaching kinetics to determine an appropriate
equilibrium time for drywall, 2) leaching as a function of liquid-to-solid ratio (USEPA method M1316)
(USEPA, 2014c), and 3) leached constituent mass transfer rate using the semidynamic tank leaching
procedure (USEPA Method 1315) (USEPA, 2014c).
2.3.1. Leaching processing kinetics
Metal and sulfur release kinetics studies were conducted to determine equilibrium time and
conditions. A crushed paper and gypsum mixture at the appropriate weight percentages were used, and
DI water was used as an extraction solvent at L/S = 20. Extractions times were: 1 h, 4 h, 24 h, 120 h, 168
h (1 week), two weeks, four weeks, six weeks, and 8 weeks. These experiments were conducted at room
temperature (20 °C) using 25 g of the mixture in 500 mL DI water in triplicate. Analytical aliquots of the
extracts were filtered (0.45 |j,m), collected, and preserved as described in the methods to be performed.
Target metals (e.g., Ca, As, Pb, Zn, Se, and Sr) and sulfate concentrations were determined by ICP-AES
and IC (for anions), respectively. The leachate pH and electrical conductivity (EC) were also measured.
VMINTEQ 3.1 was used for chemical speciation, based on the MINTEQA2 (version 4.0) database
(Gustafsson, 2016). The ionic strength was given in the model based on the empirical relationship
between ionic strength and electrical conductivity (Griffin and Jurinak, 1973).
2.3.2. Liquid-solid partitioning as a function of liquid-to-solid ratio in solid
materials - USEPA Method 1316
Liquid-solid partitioning (LSP) of the inorganic constituents from five drywall samples (a
mixture of crushed paper and gypsum at appropriate weight percentages) as a function of L/S ratio were
determined at a neutral pH. Conditions used approached chemical equilibrium (USEPA Method 1316).
This method consists of 10 parallel extractions of samples over a range of L/S ratios from 80 to 200 mL
8
-------
eluate g"1 material (2.5, 5.0, 7.5, 10, 20, 40, 75, 100, 200 and 400), each with three replications. Also, a
blank was carried out through the entire procedure. The samples (both mixed crushed paper and
gypsum) were tumbled in an end-over-end fashion for one week, as determined in the previous kinetics
studies. After tumbling, the liquid and solid phases were separated via a filter (0.45 |j,m). Extract pH and
electrical conductivity (EC) were measured, and extracts were collected and preserved to meet the
requirements of the determinative methods to be performed (anions by IC and metals by ICP,
respectively). VMINTEQ 3.1 was also employed for the chemical species. The pore volume of drywall
was determined by the following monolithic leaching test (Section 2.3.3).
2.3.3. Mass transfer rates -EPA Method 1315
The long-term leachability of five different drywall products was evaluated using the
semidynamic tank leaching procedure (EPA Method 1315). These tests were conducted simultaneously
on the monolithic specimens with a fixed surface area (215±5 cm2). The liquid/surface ratio (9 m3 m"2)
was maintained constant for each leachate renewal. The polyethylene tanks were closed to prevent air
penetration and water evaporation during the leaching time. The solution was renewed after 2 h, 24 h, 48
h, 7 d, 14 d, 28 d, 42 d, 49 d, and 63 d. The sample was freely drained, and the weight was recorded to
monitor the amount of eluent absorbed into the solid matrix at the end of each leaching interval. The
leachate at each period was collected and filtered by membranes of different sizes (5, 0.45, and 0.05 |j,m)
and preserved based on the determinative methods to be performed. The eluate pH and specific
conductance were also measured for each time interval. The anion concentration was measured by IC,
and metals were determined by ICP.
The interval mass release of each sample was calculated for different leaching intervals as
follows:
Ci X Vi
Mt¦ = :—
tl A
where Mt = mass release during the current leaching interval, i (mg m"2), Ci = concentration of
composition I in the eluate for interval (mg L"1), V, = eluate volume in the interval i (i), and A =
specimen external geometric surface exposed to the eluent (m2).
The flux of the composition in an interval was further plotted as a function of the generalized
mean of the cumulative leaching time (Vt). The flux across the exposed surface of the sample was
calculated as follows:
Mi
Pi =
ti ti-i
where F; = flux for interval i, (mg m~V), M, = mass released during the current leaching interval i (mg
m"2), ti = cumulative time at the end of the current leaching interval i (s), and t i-i = cumulative time at
the end of the previous leaching interval i-1 (s).
The time used to plot each interval mass was the generalized mean of the square root of the
cumulative leaching time using the cumulative time at the end of the ith interval, ti, and the cumulative
time at the end of the previous interval, ti-i.
l i
+ f 2
n-l
By applying this method, the cumulative fraction of constituents leached from the drywall samples
versus time was determined. The cumulative released composition i (or loss composition, PCUm-i) was
evaluated by the percentage of the total amount of the composition or the leachable amount in the
drywall. In this work, the total amount of the composition was assumed to be that measured through
9
-------
nitric acid extraction" (10% HNO3, 90 °C, 16 h).
P,
cum—i
Total or leachable amount (composition i)
The constituents were assumed to be released primarily through a diffusion-controlled process. Usually,
a mathematical diffusion model based on Fick's second law is used to evaluate the leaching rate with
respect to time (Crank, 1975; De Groot, 1993). The American Nuclear Society (ANS) has standardized a
Fick's law-based mathematical diffusion model CANS. 1986") as follows:
where De = effective diffusion coefficient (cm2/s), an = contaminant loss (mg) during the particular
leaching period with index n, Ao = initial amount of contaminant present in the specimen (mg), V =
volume of the specimen (cm3), S = surface area of the specimen (cm2), Atn = duration of the leaching
period in seconds, and T = "mean time" for the leaching interval n in seconds, which can be calculated
as described above.
De values from the above equation shown are termed "effective" because diffusion occurs in the
liquid filling the interstitial spaces of a porous body. Therefore, the actual liquid path is longer than the
one assumed by the model. The exact solution of the diffusion equation depends on the initial and
boundary conditions. The document "Measurements of the Leachability of Solidified Low-Level
Radioactive Wastes" by the American Nuclear Society (ANS 16.1) also suggested using the leachability
index (LX) to estimate and compare the behavior of different compositions or different wastes (ANS,
1986). The LX is calculated using the diffusion coefficient from the equation shown above and is the
average of the negative logarithm of the effective diffusivity terms (expressed in cm2/s). Therefore, the
leachability index is defined as follows:
where n is the number of the particular leaching period, and m is the total number of individual leaching
periods. The relative mobility of different contaminants can be evaluated by this index, which varies
from 5 (De = 10~5 cm2/s, very mobile) to 15 (Ds = 10~15 cm2/s, immobile) (Dermatas et al., 2004; Moon
and Dermatas, 2007).
These equations are valid only when a leachable constituent is leached by diffusion from a
uniform regularly shaped solid, and when the leachable constituent is less than 20% of the total amount
(ANS, 1986). In these cases, the cumulative mass release can be described by one-dimensional semi-
infinite geometry if the composition of concern is not depleted over the time of interest (Crank, 1975;
De Groot, 1993). Depletion was assumed to occur when more than 20% of the total leachable content
had been released. In Under these conditions, the effective diffusivity can be calculated from a shape-
specific solution of the mass transport equations (Anders, 1978; Godbee et al., 1980; Kosson et al.,
2002). However, the equations are still not available for the parallelepipeds used in this study.
Therefore, the values of De or LX in this work were calculated until the total loss composition was more
than 20% of the total. The total leaching time for the calculations was not necessarily as long as the real
The effective diffusion coefficients were calculated using this model as follows:
m
n = i
10
-------
total experiment time (63 d). The calculated cumulative time was only a partial experimental time. For
most constituents in the drywall samples (e.g., Ca and S), the time to reach the leaching equilibrium was
approximately one to two weeks. However, extreme situations were observed (24 h (e.g., Mg in one
drywall [Drywall B])); and 63 d (e.g., dissolved organic carbon (DOC) in all samples)). The weighted
arithmetic mean of the De estimates was employed for different drywall or different constituents; the
weighting was based on its calculated cumulative time in the model. LX has been used as a performance
criterion for the solidification and stabilization (S/S) of wastes in several studies (Canada, 1991;
Dermatas et al., 2004; Moon and Dermatas, 2007). If the LX value is greater than 9, then the S/S wastes
can be used in "controlled utilization", provided that the information on the S/S wastes is acceptable for
a specific utilization such as quarry rehabilitation, lagoon closure, or road-base material. If the S/S
wastes have the LX value higher than 8, they can be disposed of in sanitary landfills. If the S/S wastes
have the LX value lower than 8, they are not considered appropriate for disposal.
Based on the diffusion theory model developed by de Groot and van der Sloot (1992), the
cumulative maximum release of the component (Bt in mg m 2) is expressed as:
where Ds = effective diffusion coefficient in cm2 s 1 for component x, t = contact time in s, Umax =
maximum leachable quantity expressed in mg kg-1, and d = bulk density of the product in kg m 3. The
three mechanisms potentially controlling composition release (i.e., wash-off, diffusion, and dissolution)
can be distinguished by evaluating the slope of the curve in the equation above. Slope values close to 0.5
indicate that the constituent release is slow and controlled by diffusion. Slope values close to 1 indicate
that dissolution is the controlling mechanism, whereas the slope values close to 0 would suggest that the
constituent release is controlled by wash-off, occurring when a soluble layer exists on the surface of the
material.
2.4. Statistical Analysis
Statistical analysis and graphical representation of the data were performed using Microsoft
Excel 2013, JMP 9.0, and SigmaPlot 11.0. The statistical analysis technique was chosen based on the
properties of parameters. The mean, standard error, minimum and maximum values were used to
summarize the content of elements and mineral phases. Box plots with mean diamonds were employed
to graphically depict groups of numerical data through their summaries (minimum, lower quartile,
median, mean, upper quartile, and maximum). Care must be taken when evaluating the data presented
herein due to the exploratory nature of the experiment, the numerous comparisons being made, and the
methods being followed, especially when using different methods.
2.5. Quality Metrics
Accuracy checks, precision, calibration of instrumentation, and determination of detection limits
were used to ensure quality control and the confidence level of the obtained results. Precise,
documented, and valid data are needed for the ultimate decisions to be made. To ensure the quality of
the data, all instruments were regularly calibrated. Quality assurance (QA)/quality control (QC) checks,
as presented in Table 2-4, were conducted to ensure the precision and accuracy of the data.
11
-------
Table 2-4: Summary of Experimental QA/QC Checks
Test
Frequency
Measurement
Experimental
QC
Acceptance
Criteria
Corrective Action
Acid digestion
Once
Total metals
Method blank
Less than 3 times
MDL
Vessel Cleaned
Ref. Std. NIST5
1633C or
2865C
Recovery 70-130
%
Procedure repeated
Triplicates
% RSD6 < 10%
Procedure repeated
Total Hg
Once
Total Hg
Method blank
Less than 3 times
MDL
Vessel Cleaned
Ref. Std. NIST
2865C
Recovery 85 - 115
%
Procedure repeated
triplicates
% RSD < 10%
Procedure repeated
Total S and C
Once
Total S
and C
Method blank
Less than 3 times
MDL
Vessel Cleaned
Ref. Std. NIST
2865C for S
Recovery 90 - 110
% for S;
Procedure repeated
Triplicates
% RSD < 10% for
S
Procedure repeated
Water-
extractable
sulfur and
metals -
Method of
Musson et al.,
2008
Once
S042+, other
anions, and
cations K+,
Na+, Ca2+, and
Mg2+)
Triplicates
% RSD < 10%
Procedure repeated
Method blank
Less than 3 times
MDL
The problem was
investigated,
analysis repeated if
necessary
HC1
Extractable
sulfur and
metals
Once
S042+, other
anions, and
cations (K+,
Na+, Ca2+, and
Mg2+)
Triplicates
% RSD < 10%
Procedure repeated
Method blank
Less than 3 times
MDL1
The problem was
investigated,
analysis repeated if
necessary
hno3
Extractable
sulfur and
metals
Once
S and metals
Triplicates
% RSD2 < 10%
Repeat procedure
Method blank
Less than 3 times
MDL
The problem was
investigated,
analysis repeated if
necessary
Formaldehyde
Once
Formaldehyde
Method blank
-------
Test
Frequency
Measurement
Experimental
QC
Acceptance
Criteria
Corrective Action
Laboratory
Control Sample
70-130% Recovery
analysis repeated if
necessary
a matrix spike
70-130% Recovery
SVOCs
Once
SVOCs
Method blank
20% of
the samples), the
problem was
investigated and the
whole experiment
was repeated when
possible
Method blank
Less than 3 times
MDL
Investigate problem,
repeat if necessary
Method 1315
9 time
points
pH, EC, Total
alkalinity,
Dissolved
metals,
Inorganic
anions, and
DOC
Triplicates
% RSD < 15%
Occasional data
outside acceptance
limits were flagged.
In case of frequent
violation (>20% of
the samples), the
problem was
investigated, and the
whole experiment
was repeated if
possible
Method blank
Less than 3 times
MDL
Investigate problem,
repeat if necessary
MDL - Method detection limit; 2RSD - Relative standard deviation; 3RL - Reporting limit; 4RPD - relative percent
difference;5 NIST - National Institute of Standards & Technology;6 RSD -relative standard deviation
13
-------
3. Physical and Chemical Properties
3.1. Moisture Content
MC measurements of the drywall samples were made at several different temperatures. Calcium sulfate
exists predominantly at three levels of hydration in nature: dihydrate (gypsum, CaSC>4 2H2O),
hemihydrate (CaSC>4 0.5(H2O), a-hemihydrate and P-hemihydrate) and anhydrous state (anhydrite,
CaSC>4). When heated, gypsum converts to a partially dehydrated mineral called calcium sulfate
hemihydrate, calcined gypsum, or plaster of Paris. This material has the formula CaSC>4 (wThO), where
0.5 CaS04 -H20 +1 -H20 T
The endothermic property of this reaction is relevant to the performance of drywall, as it confers
fire resistance to the drywall. When calcined gypsum is mixed with water at ambient temperatures, it
quickly returns to the preferred dihydrate form, while physically "setting" to form a rigid and relatively
strong gypsum crystal lattice.
1 1
CaS04 —H20 +1 — H20 -> CaS042H20
This reaction is exothermic and is responsible for the ease with which gypsum can be cast into
various shapes, including sheets (for drywall), sticks (for blackboard chalk), and molds.
Upon heating to 180 °C, the nearly water-free form, called y-anhydrite (CaSC>4 wThO where n =
0 to 0.05), is formed. The y-anhydrite reacts slowly with water to return to the dihydrate state, a property
exploited in some commercial desiccants. At temperatures above 250 °C, the completely anhydrous
form called P-anhydrite or "natural" anhydrite is formed. Natural anhydrite does not react with water,
even over geological timescales, unless very finely ground.
The results of weight loss on ignition (WLOI) measurements of the gypsum samples at different
temperatures are presented in Figure 3-1 and Table 3-1. These results demonstrate that MC
measurement results at different temperatures depend on the testing procedures used. After reaching 230
°C, gypsum mass was relatively stable, supporting the concept of "combined water" in the ASTM
C471M-16 (ASTM-International, 2016). The WLOI measured between 230 to 550 °C can be related to
inherent organic matter decomposition in the drywall gypsum core (Heiri et al., 2001; Wang et al.,
2011). The MC measured using the "freeze-drying processing" (0.76%) was significantly higher than
the MC measured at a temperature of 45 °C (0.35%), but much lower than the MC measured at 105 or
230 °C (13.6 and 19.7%) (Table 3-1).
14
-------
o
h-l
-80 20 120 220 320
Temperature (°C)
A-G
—•—B-G
C-G
—•—D-G
E-G
F-G
H-G
—G-G
I-G
—•—L-G
420
520
Figure 3-1: WLOI of gypsum of drywall board
Table 3-1: The average of WLOI content of gypsum of drywall (%)
Temp (°C)
Mean
Stdev
Min
Max
-80
0.76
0.25
0.45
1.29
45
0.35
0.07
0.25
0.45
105
13.64
1.31
10.9
15.4
230
19.43
0.62
18.1
20.3
400
19.74
0.74
18.1
20.6
550
20.82
0.71
19.3
21.9
The moisture content at 105 °C presented great variation, even after 100 h (four times the drying time
for most soil and rocks as minerals), as mineral phases are still transforming (Table 3-2). However, there
was almost no difference in MC among samples dried at 230 °C, as all water should be lost by this point.
In addition, MC in gypsum and paper from the same drywall board were quite different, ranging from
12.3-18.3% and 11.2-13.8% (Table 3-3).
Table 3-2: The average of MC of gy
)sum for drywall after drying procedures (%)
Average Stdev
CV
Free water content <7, 45 °C
A-G
0.26
0.02
8.43
B-G
0.25
0.02
7.34
C-G
0.32
0.03
9.35
D-G
0.38
0.01
1.95
E-G
0.32
0.03
9.98
15
-------
Average
Stdev
cv
F-G
0.41
0.04
8.62
G-G
0.40
0.03
6.94
H-G
0.34
0.03
9.07
I-G
0.45
0.03
5.72
L-G
0.38
0.02
5.66
Average of 10
0.35
0.07
18.7
MC (ci) 105 °C
A-G
17.36
0.14
0.82
B-G
18.26
0.50
2.71
C-G
16.46
0.85
5.18
D-G
16.30
2.22
13.6
E-G
14.43
2.79
19.4
F-G
15.97
1.52
9.52
G-G
15.79
0.96
6.09
H-G
17.13
0.41
2.37
I-G
12.34
2.63
21.3
L-G
14.36
2.49
17.4
Average of 10
15.84
1.73
10.9
The combined water (a\ 230 °C
A-G
24.61
0.11
0.46
B-G
25.54
0.12
0.46
C-G
23.91
0.50
2.08
D-G
24.93
0.20
0.82
E-G
24.22
0.97
4.01
F-G
23.88
0.31
1.30
G-G
22.06
0.13
0.57
H-G
24.74
0.15
0.62
I-G
23.63
0.06
0.25
L-G
23.71
0.07
0.29
Average of 10
24.12
0.95
3.93
MC (ci) 400 °C
A-G
24.60
0.04
0.16
B-G
25.60
0.27
1.05
C-G
26.02
0.11
0.41
D-G
25.04
0.05
0.20
E-G
25.44
0.02
0.10
F-G
24.56
0.10
0.42
G-G
22.16
0.10
0.45
H-G
25.11
0.11
0.45
I-G
23.34
0.08
0.36
L-G
24.24
0.08
0.31
Average of 10
24.61
1.15
4.66
16
-------
Table 3-3: The MC of drywall board at 105°C (%)
Paper
Gypsum
Drywall
board
Average
Stdev
Average
Stdev
A
11.97
0.13
17.36
0.14
16.72
B
12.83
1.74
18.26
0.50
17.44
C
13.26
1.47
16.46
0.85
16.09
D
13.80
2.22
16.30
2.22
16.05
E
11.71
1.13
14.43
2.79
14.04
F
11.92
1.21
15.97
1.52
15.37
G
11.60
0.98
15.79
0.96
15.32
H
13.62
1.22
17.13
0.41
16.51
I
11.68
1.10
12.34
2.63
12.21
L
11.24
1.59
14.36
2.49
13.83
Average of 10
12.36
0.93
15.84
1.73
15.36
Dry-weight composition comparison for other chemical constituents is difficult as the temperature used
for MC analysis in drywall or other wastes with large amounts of calcium sulfate usually is not
described due to change in composition and volatilization lost (Musson et al., 2008; Reidy et al., 2014;
Sun and Barlaz, 2015).
3.2. Total Sulfur and Carbon Content
As expected, gypsum total sulfur content increased as drying temperature increased due to
weight loss (Table 3-4). The total sulfur content was relatively constant at different temperatures in all
samples (e.g., 20 °C, 400 °C and 550 °C); however, differences among different drywall products were
observed (p<0.01) (Tables 3-4 and 3-5). In addition, the sulfur content in oven-dried gypsum samples
was greater compared to air-dried, indicating rapid rehydration upon air contact. Thus, it is better to
compare gypsum chemical compositions in air-dried samples as MC will be unstable in oven drying
(105°C). Therefore, all results in this report were expressed on an air-dried basis (referred to as air-dry
weight basis), unless otherwise noted.The total sulfur content of pure control gypsum (CaSC>4 2H2O),
hemihydrate (CaS04 (H20)o5,a-hemihydrate, and P-hemihydrate) and anhydrite (CaS04) was 18.6,
22.0, and 23.5 %, respectively.
Conversely, total carbon measurements in gypsum samples decreased as a result of organic
matter heating decomposition (Table 3-6). Total carbon content at 20 °C was much higher (0.87%) than
at 400 °C (0-34-0.43%) or 550 °C (0.11-0.14%), respectively (p<0.01) (Tables 3-6 and 3-7). The small
amounts of organic carbon in the gypsum samples (0.3-1.9%) likely originated from additives in the
manufacturing process.
Based on the weight percentage of paper and gypsum in drywall products, the average total
sulfur and carbon content of 10 drywall samples was 16.7±0.77 and 4.47±0.95%, respectively (Tables 3-
8 and 3-9).
17
-------
Table 3-4: The sulfur content (%) of gypsum of drywall (air-dried weight basis)
20 °C
105 °C
400 °C
550 °C
A-G
17.77
21.88
22.40
22.37
B-G
18.23
22.61
22.67
23.02
C-G
17.93
21.33
22.22
22.74
D-G
18.12
21.97
22.87
23.06
E-G
17.18
20.85
21.80
22.47
F-G
17.29
21.19
21.64
21.40
G-G
16.45
19.86
20.49
21.19
H-G
17.35
21.60
22.33
22.69
I-G
17.88
21.13
22.13
22.60
L-G
17.44
20.63
21.64
22.55
Average of 10
17.56
21.31
22.02
22.41
Table 3-5: The sulfur content (%) of gypsum of drywall (air-dried weight basis)
20 °C
105 °C
400 °C
550 °C
Average
Stdev
CV (%)
A-G
17.77
18.64
17.98
17.85
18.06
0.40
2.22
B-G
18.23
19.12
18.05
18.25
18.41
0.48
2.63
C-G
17.93
18.32
17.63
17.93
17.95
0.28
1.58
D-G
18.12
18.89
18.29
18.32
18.40
0.34
1.86
E-G
17.18
18.22
17.38
17.67
17.60
0.46
2.63
F-G
17.29
18.27
17.37
16.96
17.47
0.56
3.21
G-G
16.45
17.15
16.77
17.09
16.86
0.34
1.99
H-G
17.35
18.44
17.85
17.72
17.84
0.46
2.58
I-G
17.88
18.81
17.94
17.89
18.13
0.46
2.51
L-G
17.44
18.04
17.42
17.73
17.65
0.29
1.66
Average
17.56
18.39
17.67
17.74
17.84
0.58
3.25
Table 3-6: The carbon content (%) of gypsum of drywall (oven-dried weight basis
with no follow-up air drying)
20 °C
105 °C
400 °C
550 °C
A-G
0.54
0.59
0.24
0.09
B-G
0.30
0.45
0.00
0.02
C-G
1.85
2.18
0.24
0.12
D-G
0.42
0.63
0.09
-0.01
E-G
1.07
1.20
0.58
0.26
F-G
1.06
1.33
0.59
0.36
G-G
0.79
0.99
0.56
0.31
H-G
1.00
1.19
0.55
0.07
I-G
0.67
0.73
0.83
0.07
L-G
1.02
1.11
0.66
0.08
Average of
0.87
1.04
0.43
0.14
19
-------
Table 3-7: The carbon content (%) of gypsum of drywall (oven-dried weight basis
followed by air drying)
20 °C
105 °C
400 °C
550 °C
A-G
0.54
0.51
0.19
0.07
B-G
0.30
0.38
0.00
0.01
C-G
1.85
1.87
0.19
0.09
D-G
0.42
0.54
0.07
0.00
E-G
1.07
1.05
0.46
0.21
F-G
1.06
1.14
0.47
0.29
G-G
0.79
0.85
0.46
0.25
H-G
1.00
1.02
0.44
0.06
I-G
0.67
0.65
0.67
0.06
L-G
1.02
0.97
0.53
0.06
Average of 10
0.87
0.90
0.35
0.11
Table 3-8: The carbon content (%) of drywall (105 °C dry-weight basis)
Gypsum part
Paper
part
Drywall
joard
Average
Stdev
Average
Stdev
Average
Stdev
A
0.59
0.03
28.08
1.41
3.62
0.06
B
0.45
0.03
25.55
1.42
4.03
0.00
C
2.18
0.06
22.35
1.31
4.16
0.19
D
0.63
0.06
20.85
1.90
2.51
0.23
E
1.20
0.01
28.60
2.06
4.93
0.20
F
1.33
0.03
30.58
1.55
5.35
0.24
G
0.99
0.11
31.83
0.71
4.22
0.30
H
1.19
0.05
27.46
2.06
5.57
0.24
I
0.73
0.05
24.35
1.22
5.21
0.18
L
1.11
0.07
25.94
1.45
5.11
0.39
Average of 10
1.04
0.43
26.56
3.46
4.47
0.95
Table 3-9: The sulfur content (%) of drywall (105 °C dry-weight basis)
Gypsum part
Paper part
Drywall board
Average
Stdev
Average
Stdev
Average
Stdev
A
18.64
0.09
5.22
0.83
17.12
0.18
B
19.12
0.42
6.69
0.91
17.32
0.49
C
18.32
0.28
8.39
0.56
17.21
0.31
D
18.89
0.38
8.80
1.00
17.91
0.44
E
18.22
0.32
5.30
0.87
16.40
0.40
F
18.27
0.54
3.89
0.44
16.21
0.52
G
17.15
0.46
3.12
0.20
15.63
0.43
H
18.44
0.33
5.40
0.50
16.20
0.36
I
18.81
0.57
6.42
0.77
16.42
0.61
L
18.04
0.39
6.53
0.72
16.13
0.45
Average of 10
18.39
0.55
5.97
1.79
16.66
0.70
20
-------
3.3. Trace Organic Compounds
In all the samples, most SVOC compounds and all VOC and PCB concentrations were below the
method detection limits (data not shown). In addition, most PAH concentrations were between the MDL
and RL and ranged from 14-622 |ig kg"1 (Table 3-10). Detected PAHs included carcinogenic compounds
such as benzo(a)anthracene, benzo(b)fluoranthene, benzo(g,h,i)perylene, benzo(k)fluoranthene,
benzo(b)fluoranthene, di-//-butyl phthalate, caprolactam, chrysene, bis(2-ethylhexyl)phthalate,
fluoranthene, indeno(l,2,3-cd)pyrene, phenanthrene, phenol, and pyrene (Table 3-10). Drywall sample E
presented the highest concentration among those tested. This sample was re-tested, and results showed
that PAH concentrations in the paper were almost ten times the PAH concentration of the composite
sample, indicating that the PAH source was not from gypsum but from the paper and/or chemical
additives or glues used for drywall manufacturing.
Moreover, formaldehyde concentrations in most of the drywall samples ranged from 500 up to
8,500 |Lxg kg"1, with a median and average value of 1,800 and 3,700 |Lxg kg"1, respectively (Figure 3-2).
Tributyltin (TBT) concentrations in two of the three mold-resistant samples exhibited TBT
concentrations above detection (19 and 59 |Lxg kg"1). However, no difference in TBT concentration in the
paper and gypsum fractions was observed in these drywall samples. In addition, these results might
differ from the results for old drywall samples as concentrations of formaldehyde and other organics
decrease with time.
Table 3-10: The content of PAHs in drywall (fig kg"1)
PAH compounds
Drywall
MDL
LOQ
A
B
C
D
E
F
G
H
I
L
Median
Acetophenone
ND1
ND
ND
ND
ND
91
ND
ND
ND
ND
91
50
99
Di-«-butyl phthalate
ND
ND
ND
ND
280
ND
ND
320
ND
ND
300
200
500
Caprolactam
ND
130
ND
ND
170
ND
ND
ND
ND
ND
150
99
500
bis(2-Ethylhexyl)phthalate
410
220
ND
ND
580
350
330
650
460
380
395
200
510
Indeno( 1,2,3 -cd)pyrene
ND
ND
ND
ND
21
ND
ND
ND
ND
ND
21
10
51
Nitrobenzene
ND
ND
ND
ND
ND
61
ND
ND
ND
ND
61
50
99
Phenol
ND
90
63
100
93
120
650
53
ND
ND
93
50
99
Anthracene
ND
ND
ND
ND
19
ND
ND
ND
ND
ND
19
10
51
Benzo(a)anthracene
19
24
47
18
69
17
18
ND
ND
ND
19
10
51
Benzo(a)pyrene
19
19
10
31
67
21
24
11
13
ND
19
10
51
Benzo(b)fluoranthene
22
26
12
41
56
26
23
16
21
15
23
10
51
Benzo(g,h,i)perylene
10
14
ND
14
42
15
15
11
ND
ND
14
10
51
Benzo(k)fluoranthene
16
15
ND
21
38
17
16
11
ND
ND
16
10
51
Chrysene
29
28
22
23
96
23
32
12
14
ND
23
10
51
Fluoranthene
23
23
ND
11
95
ND
33
18
10
ND
23
10
51
Phenanthrene
24
26
ND
12
86
30
34
24
14
11
24
10
51
Pyrene
32
25
ND
22
180
33
39
13
12
ND
29
10
51
Carcinogenic PAHs
194
200
91
193
748
182
234
116
84
26
188
100
510
Total PAHs
604
640
154
293
1892
804
1214
1139
544
406
622
759
2368
1ND - Not detected
21
-------
1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
Formaldehyde content ( ug/kg)
Figure 3-2: Distribution of Formaldehyde content in drywall board
3.4. Crystalline Mineral Phases
Gypsum component mineral phases of the drywall sample were characterized using XRD. A
typical gypsum XRD pattern is presented in Figure 3-3. The details of the mineral phases of each
sample, including the peak and pattern lists, are presented in Appendix A.
CHL Standard R
un - Baumann Co tube B-G
A-. 1
11111111 n|T IT i i
30
llll -l.ihijLll
1111 nri 11111 in ifi 11111 in 11111111
40 50 60 70
-
mr
iiTjTi fi irf
120
rTTTTTJ
80
100
TTTTTTJTTT
110
"I I"
130 140
Position [°2Theta] (Cobalt (Co))
22
-------
Figure 3-3: Typical XRD pattern of gypsum of drywall board
As expected, the dominant mineral in the samples was gypsum (reference code 01-074-1433, and
00-033-0311), with a small amount of hemihydrate (reference code 01-083-0438) and anhydrite
(reference code 01-072-0503) also observed. In addition, Ca and Mg carbonate was detected (reference
code 01-072-1652 and 01-086-2236), as was silica (reference code 01-085-0457).
As shown in Table 3-11, the drywall gypsum component (i.e., drywall without paper) was made
of 90% gypsum in all samples. The XRD technique is semiquantitative; a built-in assumption is that all
phases add to 100%. Based on these semiquantitative results, the average total sulfur content in these
samples was 17.9%, which is similar to the total sulfur content detected by the LECO CS230
Carbon/Sulfur Analyzer.
Table 3-11: The Semiquantitative analysis (%) of mineral phases of gypsum from
drywall
Compound
Calcium Sulfate
Silicon
Carbonate
Chemical
CaS04(H20)2
CaS04
CaS04( H20 )0 5
Si02
CaC
(Mg.i29 Ca.87i)
Ref. Code
01-074-
00-033-
01-072-
01-083-0438
01-085-
01-072-
01-086-2336
1433
0311
0503
0457
1652
A-G
52
40
2
2
3
1
1
B-G
53
40
1
2
3
1
1
C-G
51
39
3
2
4
1
1
D-G
53
40
2
2
2
1
E-G
53
38
1
3
2
1
1
F-G
51
39
1
4
2
1
1
G-G
49
38
1
6
3
1
2
H-G
52
40
1
3
2
1
1
I-G
50
40
1
4
2
1
2
L-G
51
39
1
5
2
1
2
Average of
52
39
1
3
3
1
1
10
3.5. Total Acid Extractable Sulfur and Metals
Acid-extractable sulfur content was lower (13.6%) compared to combustion technique sulfur
(18.3%), LECO CS230 Carbon/Sulfur Analyzer) (Tables 3-9 and 3-12). The 30% difference might
indicate issues during acid digestion using USEPA 3051 A. Besides, total acid-extractable calcium
average concentration was 18.6%, and the molar ratio of Ca to S was 1.09 (Table 3-12).
Acid-extractable macro- and micro-element results are presented in Figures 3-4 and 3-5. As
expected, positive correlations (both Person and Spearman) were observed between elements (p<0.01)
(e.g., Ca vs S and Na; A1 vs K; Fe vs K, Mn, Ni, Cu, Zn, and Ba; Mg vs Mn, Na, P, and Zn; Na vs Se,
Zn, and Hg) (Figure 3-6).
23
-------
Table 3-12: The sulfur and metal content of gypsum from drywall by USEPA
M3051A
Mean
Std Dev
Lower 95%
Upper 95%
Median
Minimum
Maximum
Ca (%)
18.56
2.82
17.51
19.62
18.7
12.59
23.75
S(%)
13.62
2
12.87
14.36
13.68
9.59
17.37
Mg (mg/kg)
2500
2120
1710
3290
2280
160
7790
K (mg/kg)
340
575
129
560
120
110
2200
Na (mg/kg)
170
63
142
190
150
86
280
Fe (mg/kg)
1030
1230
570
1490
570
250
4780
A1 (mg/kg)
550
1000
180
920
240
100
3710
Si (mg/kg)
280
198
200
350
200
120
800
Sr (mg/kg)
140
148
85
196
65
43
472
Ba (mg/kg)
19
30
8.2
30
7.3
4.8
115
Cu (mg/kg)
2.4
3.2
1.22
3.6
1.0
0.97
12.6
Mn (mg/kg)
10
14
5.1
16
5.2
2
55
P (mg/kg)
130
90
92
159
150
9.1
241
Ni (mg/kg)
3.5
7.9
0.5
6.4
0.8
0.39
28.5
Se (mg/kg)
4.1
1.2
3.7
4.6
3.7
3
5.9
Zn (mg/kg)
11
3.4
9.49
12
10.7
4.4
18
As ((ig/kg)
125
126
78
170
84
80
530
Pb ((-ig/kg)
560
270
460
660
440
120
1370
Hg (f-ig/kg)
165
85
130
200
150
43
360
24
-------
Sample
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
Sample
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
7000
5000 ¦
4000 -
3000 -
2000 -
3
1000 -
t
500 ¦
400 -
300 ¦
100-
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
sample
Sample
900-
800-
700-
600-
% 500 -
E,
^ 400 -
300-
200
100
:S
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
Sample
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
Sample
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
A-G B-G C-G D-G E-G F-G G-G H-G l-G L-G
Figure 3-4: Distribution of the macro-elements in the gypsum from drywall. The
box-and-whisker plots show the following: the minimum value, the 25th quartile, the
25
-------
median, the 75th quartile, the maximum value.
Figure 3-5: Distribution of the micro-elements in the gypsum from drywall. The
box-and-whisker plots show the following: the minimum value, the 25th quartile, the
26
-------
median, the 75th quartile, the maximum value.
All the calcium, sulfate, and other metals from the gypsum samples were assumed to be
dissolved in the heating process by microwave. The dilution of samples with DI water before analysis,
however, required the temperature to be lowered to room temperature. Upon dilution, the acid
concentration in the final digested solution was further decreased (< 2 M), and this is suspected to have
resulted in some dissolved calcium and sulfate (and other metals) re-precipitating, in agreement with the
transformation and solubility of gypsum (CaSC>4 2FhO) and anhydrite (CaSC>4) at changing pH and
temperature (Freyer and Voigt, 2003; Li and Demopoulos, 2005; Shukla et al., 2008; Wang et al., 2013).
Because of re-precipitation, solid residues were common in the final diluted digestates (EPA
3051 A), and thus, centrifugation or filtration was required. In these cases, the extracted analyte
concentrations may not reflect the true total concentrations in the sample (USEPA, 2007c); the
concentrations of sulfate, calcium, and other constituents assessed through the use of M3051A may be
underestimated. This underestimation was further demonstrated by additional testing on one sample
(gypsum from Drywall L) diluted at four different temperatures (4, 20, 37, and 80 °C) (five replications
at each temperature). The concentrations of calcium, sulfate, strontium, silicate, and titanium in this
sample were positively related to the temperature for dilution. The concentrations of iron, aluminum,
phosphorus, and magnesium were relatively stable with temperature (Figure 3-7, Table 3-13). These
observations are supported by other research on the solubility of calcium sulfate in complex systems
(Wollmann and Voigt, 2008; Zeng and Wang, 2011; Wang et al., 2015)
27
-------
Figure 3-6: Positive correlations among elements in the gypsum from drywall
O
00
-------
Table 3-13: Effect of dilution temperature on the elemental composition of gypsum
from Drywall-L
4 °C
20 °C
37 °C
80 °C
Ca (%)
15.20
±
1.64
16.52
±
1.09
19.99
±
3.21
20.98
±
0.38
S(%)
10.73
±
1.22
11.81
±
0.76
14.10
±
2.29
14.75
±
0.36
Mg (mg/kg)
3598
±
59
3540
±
43
3502
±
28
3494
±
17
K (mg/kg)
117
±
0.95
118
±
0.79
117
±
1.42
117
±
1.47
Na (mg/kg)
938
±
87.0
1047
±
23.5
893
±
74.3
975
±
72.0
Fe (mg/kg)
479
±
11.0
474
±
12.1
477
±
3.92
476
±
11.4
Al (mg/kg)
108
±
15.3
104
±
7.01
103
±
10.7
104
±
6.74
Si (mg/kg)
421
±
8.15
518
±
43.7
620
±
11.1
717
±
33.9
Ti (mg/kg)
18.9
±
2.25
19.0
±
1.06
22.0
±
2.42
22.1
±
1.12
Sr mg/kg)
47.1
±
2.77
50.1
±
2.11
58.2
±
6.43
60.6
±
0.8
Ba (mg/kg)
5.02
±
0.63
4.90
±
0.34
5.57
±
0.43
5.47
±
0.20
Cu (mg/kg)
1.97
±
0.02
1.98
±
0.01
1.97
±
0.02
1.96
±
0.02
Mn (mg/kg)
7.72
±
0.48
7.30
±
0.19
7.56
±
0.19
8.05
±
0.45
P (mg/kg)
179
±
1.66
182
±
2.29
182
±
1.53
182
±
2.51
Ni (mg/kg)
1.18
±
0.01
1.19
±
0.01
1.18
±
0.01
1.18
±
0.01
Se (mg/kg)
6.50
±
0.22
6.27
±
0.17
6.70
±
0.44
6.39
±
0.32
Zn (mg/kg)
9.07
±
0.07
9.44
±
0.75
9.06
±
0.11
9.23
±
0.35
In a previous study, calcium content average in drywall samples (n = 20) was approximately
10% (9.5±0.5%), or less than half of the theoretical calcium content in the pure gypsum (CaSC>4 2H2O,
23.3%) (Reidy et al., 2014). However, a study from The Division of Hazard Analysis, U.S. Consumer
Product Safety Commission found an average content of calcium of 24.9±7.9% (Garland and Greene,
2009).
Gypsum inorganic trace constituent concentrations compared to previously reported studies are
summarized in Table 3-14. The concentration differences may be related to gypsum sources or
extraction methods used. Regardless, the evidence suggested that using the EPA 3051A digestion
method may be questionable when the method is used for drywall or other materials containing high
amounts of calcium sulfate, and thus, a new approach is necessary.
Table 3-14: Elemental composition of gypsum from drywall (mg/kg)
Element
Study
Mean
Std Dev
Media
25%
75%
Range
Max
Min
Mg
2009*
5404
5720
4800
989
7270
18015
18200
185
2014**
1885
1180
1715
1087
3016
3759
3880
121
This work
2503
2199
2780
618
3246
7392
7552
161
All data
5249
5476
3121
1115
8070
19879
20000
121
Al
2009
874
792
726
234
1330
2541
2720
179
2014
2783
1806
2540
1527
3845
5722
6190
468
This work
550
1038
226
191
287
3386
3499
113
All data
1759
1609
1330
281
3255
6077
6190
113
Fe
2009
1413
861
1350
663
1860
2926
3270
344
2014
1574
959
1080
984
2340
2775
3590
815
This work
1033
1276
574
463
959
4341
4596
255
All data
1470
951
1100
710
1910
4341
4596
255
Ba
2009
45.7
60.7
17.5
7.46
80.3
227
229
2.50
29
-------
Element
Study
Mean
Std Dev
Media
25%
75%
Range
Max
Min
2014
18.9
16.5
13.2
9.71
27.2
51.0
53.9
2.88
This work
19.3
30.6
7.2
5.66
20.0
99.0
105
5.58
All data
43.8
51.3
17.5
7.54
79.0
226.5
229.0
2.50
Sr
2009
1598
1545
776
303
2890
4170
4310
140
2014
389
115
338
322
464
389
661
272
This work
140
153
67.2
52.5
191
424
467
42.8
All data
1054
1164
467
229
1740
4267
4310
42.8
Mn
2009
47.9
35.0
46.0
9.25
78.4
97.1
101
3.92
2014
24.8
21.6
18.4
11.3
31.5
63.7
68.7
4.99
This work
10.4
14.5
5.11
4.20
9.30
48.9
51.0
2.05
All data
40.1
34.1
24.9
8.09
72.0
105.0
107.0
2.05
Pb
2009
2.57
3.84
1.42
1.35
2.02
15.1
16.4
1.29
2014
1.79
0.96
1.37
1.22
2.15
2.93
4.03
1.10
This work
0.56
0.25
0.47
0.35
0.85
0.60
0.91
0.31
All data
1.93
2.48
1.37
1.06
1.96
16.1
16.4
0.31
K
2009
403
319
340
252
586
1279
1320
41
This work
349
592
147
114
250
1914
2027
112
All data
382
438
264
116
384
1985
2027
41
Na
2009
247
165
162
114
371
445
553
108
This work
166
65
150
111
232
180
271
91
All data
214
138
162
114
264
462
553
91
Cu
2009
2.43
1.64
1.97
1.12
3.22
6.19
6.86
0.67
This work
2.47
3.19
1.43
1.00
2.02
10.4
11.4
0.99
All data
2.44
2.32
1.62
1.10
2.75
10.8
11.4
0.67
Zn
2009
3.42
2.18
2.86
1.77
4.43
7.77
8.52
0.75
This work
10.8
3.44
10.9
8.41
12.4
11.9
17.7
5.81
All data
6.36
4.55
5.31
2.64
10.6
17.0
17.7
0.75
Ni
2009
1.90
1.37
1.33
0.96
2.19
4.81
5.46
0.65
This work
3.47
8.22
0.83
0.69
1.2
26.4
26.8
0.40
All data
2.53
5.20
1.30
0.83
1.85
26.4
26.8
0.40
Hg
2009
0.22
0.30
0.12
0.05
0.26
1.20
1.24
0.04
This work
0.17
0.09
0.16
0.11
0.21
0.31
0.35
0.04
All data
0.20
0.24
0.15
0.08
0.21
1.20
1.24
0.04
Se
2009
4.06
3.44
2.20
2.06
4.11
10.18
12.20
2.02
This work
4.15
1.16
4.09
3.00
5.17
2.78
5.75
2.97
All data
4.09
2.72
3.40
2.15
5.02
10.2
12.2
2.02
Cr
2009
2.84
4.17
1.87
1.20
2.68
17.02
17.70
0.68
2014
18.91
9.43
20.15
7.23
28.18
23.86
29.40
5.54
All data
8.02
9.13
3.07
2.05
11.7
28.7
29.4
0.68
As
2009
2.60
0.91
2.28
2.25
2.37
3.55
5.70
2.15
V
2009
2.80
2.45
2.19
1.98
2.89
10.56
11.20
0.64
Co
2009
1.10
1.38
0.54
0.46
1.06
4.97
5.40
0.43
Cd
2014
0.26
0.03
0.25
0.23
0.28
0.10
0.31
0.22
Cs
2014
0.33
0.17
0.29
0.23
0.37
0.56
0.73
0.17
Ga
2014
0.33
0.19
0.28
0.19
0.40
0.61
0.77
0.16
Rb
2014
3.51
3.22
2.68
1.58
4.11
10.6
11.2
0.57
P
This work
126
93
150
29
208
225
235
9.8
Si
This work
278
201
209
152
310
617
744
127
* Garland and Greene, 2009; " Reidy et al., 2014
30
-------
Other acidic extraction methods (e.g., 0.25 M HC1) and different S:L ratio (200:1) were also used
for sulfur content extraction in high calcium sulfate wastes at room temperature (Sun and Barlaz, 2015).
This method indicated that extractable sulfur was released quickly, but no increase was observed after
24 h up to 1 month (Figure 3-8). The sulfur concentration measured using this approach was close to the
results obtained by the combustion methodology (p>0.05) (Table 3-15) but greater than that measured
using USEPA 3051A (p<0.01).
270.0
180.0 §
90.0 —
0.0
Extraction time (h)
Figure 3-8: Effect of extraction time on the calcium, sulfur and strontium content in
gypsum from drywall
The concentrations of other inorganic constituents were also monitored at different extraction
times (4 h to 1 month). Similarly, calcium concentrations were stable after 24 h, and they were
significantly higher compared to the EPA 3051A method (p<0.01) (Table 3-16). Strontium release was
stable after 96 h (Figure 3-9 and Table 3-17). However, most of the elements (e.g., Cu, Ni, Fe, Mn, Zn,
and Mg) needed a much longer time to reach the maximum release (Figure 3-10). Concentrations
measured at these longer extraction times (e.g., 2 weeks) were significantly higher than those measured
at the shorter times (4 h and 24 h) (p<0.01).
Table 3-15: Sulfur content of gypsum in drywall (%)
0.25 MHC1 (200:1)
USEPA M3051A
Total S by combustion
Average
Stdev
CV (%)
Average
Stdev
CV (%)
Average
Stdev
CV (%)
A
17.86
0.13
0.73
13.36
0.88
6.58
17.77
0.57
3.19
B
17.76
0.21
1.16
13.69
1.58
11.57
18.23
0.11
0.60
C
17.04
0.96
5.61
11.32
1.61
14.19
17.93
0.34
1.87
D
17.41
0.05
0.28
12.84
1.40
10.87
18.12
0.25
1.37
E
16.48
0.55
3.33
16.30
0.99
6.09
17.18
0.32
1.88
F
17.51
0.74
4.24
13.84
2.66
19.23
17.29
0.39
2.23
G
16.86
0.10
0.61
15.36
0.23
1.53
16.45
0.13
0.79
H
18.89
0.07
0.37
15.50
1.13
7.26
17.35
0.14
0.81
I
18.60
0.42
2.25
12.40
1.57
12.62
17.88
0.24
1.36
L
18.04
0.14
0.77
11.62
0.46
3.99
17.44
0.28
1.58
Average
17.65
0.75
4.25
13.62
1.67
12.30
17.56
0.53
3.04
Relative %
100
78
100%
Average of 10 drywall
25.0 -
~ 20.0 -
g 15.0 --
10.0 -
0.5
32
128
512
31
-------
Table 3-16: Calcium content of gypsum in drywall (%)
0.25 MHC1 (200:1)
EPA M3051A
Relative
Average
Stdev
Average
Stdev
%
A
23.90
0.22
17.84
1.20
133.9
B
23.99
0.08
18.03
2.03
133.1
C
23.82
1.50
15.01
2.22
158.7
D
24.97
0.12
17.00
1.82
146.9
E
24.64
0.30
22.38
1.23
110.1
F
24.43
1.05
19.22
3.47
127.1
G
23.89
0.19
21.29
0.55
112.2
H
24.84
0.14
21.37
1.61
116.2
I
25.39
0.29
17.43
1.94
145.7
L
25.15
0.34
16.16
0.52
155.6
Average
24.50
0.58
18.57
2.43
134.0
225.0
200.0
175.0
150.0
125.0
100.0
75.0
50.0
25.0
0.0
Zn
i- i
*3-
u
*3"
u
125.0
100.0
75.0
50.0
25.0
0.0
Mn
n
*3"
u
Fe
125.0 T
100.0
75.0
5o.o
25.o
0.0 :
I1
<-M ld
o ™
125.0
100.0
75.0
50.0
25.0
0.0 =
III
~ j. fM 0) (!)
tn U —
*3-
U
Mg
125.0 j
loo.o
75.0
5o.o
25.0
0.0 =
pn
•n)-
u
•n)-
u
Figure 3-9: Effect of extraction time on the other components in gypsum from
drywall
32
-------
Table 3-17: Strontium content of gypsum in drywall (%)
0.25 MHC1 (200:1)
USEPA M3051A
Relative
Average
Stdev
Average
Stdev
%
A
608.6
7.86
467.0
5.99
130.3
B
91.2
0.46
72.5
8.48
125.8
C
547.5
32.74
381.7
43.25
143.4
D
167.9
1.76
128.0
12.50
131.2
E
58.9
0.52
61.9
4.21
95.2
F
102.7
4.60
88.9
12.82
115.5
G
63.2
0.28
54.0
1.14
117.1
H
63.5
0.51
57.0
2.92
111.5
I
60.9
0.36
48.3
4.44
126.2
L
59.4
0.59
42.8
0.20
138.9
Average
182.4
211.7
140.2
153.1
123.5
The results for the other constituents extracted by 0.25 M HC1 for the ten gypsum samples were
compared to the results measured by extracting according to M3051A. In general, the concentrations
from the 0.25 M HC1 extraction of Cu and Ni were significantly higher (p<0.05), while the
concentration of Ba was significantly lower (p<0.05). However, the concentrations of most elements
were related to extraction time (Figure 3-10). All these results further suggest low efficiency of the EPA
3051A method for solid wastes containing large concentrations of gypsum (e.g., drywall). Gypsum
solubility decreases, and re-precipitation occurs at room temperature in an acidic medium (Van
Driessche et al., 2019). Moreover, calcium, sulfur, and strontium results after HC1 extraction (0.25 M,
L/S=200, room temperature) were stable after 24 h and higher compared to the 3051A method (p<0.01),
indicating that HC1 extraction is an easy and reliable alternative approach. However, results for other
metals (e.g., Cu, Ni, and Zn by HC1) were extremely affected by extraction time. Some longer
extractions (e.g., two weeks) exhibited significantly higher concentrations than those conducted over a
shorter time (4 h and 24 h).
Figure 3-10: Calcium and sulfur content in the gypsum from drywall by three
methods
Since temperature plays a key role in controlling the solubility of calcium sulfate and phase
transformation in an acidic medium, 10% HNO3 extraction at 90 °C (sub-boiling, nitric acid extraction,
16 h) was further employed for the gypsum composition of drywall. There were no differences between
calcium and sulfur concentration when comparing results using nitric acid extraction and HC1 extraction
33
-------
(24 h) (p>0.05). However, concentrations were significantly higher than those measured using from the
EPA 3051A method (p<0.01) (Table 3-18, Figure 3-11).
Table 3-18: Calcium and sulfate content of gypsum in drywall (%) by three methods
Calcium
Sulfur
3051A
HC1 extraction
HNO3 extraction
3051A
HC1 extraction
HNO3
Mean
18.57
24.51
24.03
13.62
17.51
18.34
Std Dev
2.43
0.64
0.50
1.68
1.30
0.48
Lower 95%
16.83
24.05
23.67
12.42
16.58
17.99
Upper 95%
20.31
24.97
24.38
14.82
18.44
18.68
Minimum
15.01
23.48
23.03
11.32
14.46
17.37
Maximum
22.38
25.17
24.78
16.30
18.90
19.05
34
-------
9000
8000 -
7000
6000
5000
4000 -
3000 -
2000 -
1000
Figure 3-11: Al, Si, Fe, Ti, Mg, K and Na (mg kg"1) content in the gypsum from
drywall by three methods
The concentrations of macro- and microelements in the ten gypsum samples using the nitric acid
extraction at 90 °C are presented in Tables 3-19 and 3-20 and Figure 3-12. Si, Fe, and Al exhibited the
highest concentrations. In addition, Sr, Ba, Mn, and Se presented concentrations higher than those
35
-------
measured using the other extraction methods (in most cases). However, Cu, Zn, and Ni concentrations
were lower compared to the concentrations measured using HC1 and long extraction time (Figure 3-10).
Therefore, the extraction method using 10% nitric acid at a sub-boiling temperature presented
satisfactory results and recoveries, and it was considered the best approach for measuring inorganic
element concentrations in the gypsum samples. Matrix spike recovery for most elements was between
95-105%, except Al, K, and Na (Table 3-19). The composition of drywall paper was also investigated by
the nitric acid extraction method (10% HNO3, 90 °C, 16 h) and EPA 3051A (Table 3-21). Some
differences were also noted between these two methods. The concentrations of Mg, Al, Si, K, Na, Sr,
and Ba using the nitric acid extraction technique were higher, while the content for Ca, S, and Ni was
lower. While the paper was purposefully separated from the drywall products during the initial sample
preparation, the paper samples always had some attached gypsum (thus leading to elevated Ca and S).
Table 3-19: Al, Si, Fe, Ti, Mg, K and Na content of gypsum in drywall (%) by three
methods
Element
Method
Mean
Std
Dev
Lower
95%
Upper
95%
Min
Max
Aluminum
mg/kg
3051A
550
1040
0
1290
113
3500
HC1-24 h
288
517
0
658
53.2
1750
HNO3, 90 °C
753
1060
0
1510
278
3770
Silicon
mg/kg
3051A
278
201
134
421
127
744
HC1-24 h
406
794
0
970
43.2
2650
HNO3, 90 °C
1690
1690
486
2900
616
5900
Iron
mg/kg
3051A
1030
1280
119
1950
255
4600
HC1-24 h
509
438
196
823
204
1710
HNO3, 90 °C
1010
1220
141
1880
248
4410
Titanium
mg/kg
3051A
42.6
104
0
117
5.6
338
HC1-24 h
15.9
32.4
0
39.1
3.18
108
HNO3, 90 °C
51.6
120
0
138
8.86
394
Magnesium
mg/kg
3051A
2500
2200
930
4080
161
7550
HC1-24 h
2270
1700
1050
3490
144
5380
HNO3, 90 °C
2690
2320
1040
4350
179
8010
Potassium
mg/kg
3051A
349
592
0
773
112.2
2026
HC1-24 h
168
235
0.35
336
56.6
827
HNO3, 90 °C
485
645
23.5
946
145.3
2311
Sodium
mg/kg
3051A
166
64.8
119
212
90.8
271
HC1-24 h
115
64.6
68.7
161
42.6
246
HNO3, 90 °C
155
59.2
113
198
76
240
36
-------
Table 3-20: Sr, Ba, Mn, Se, P, Cu, Mn, Zn and Ni content of gypsum in drywall by
three methods
Element
Method
Mean
Std
Dev
Lower
95%
Upper
95%
Min
Max
Sr, mg/kg
3051A
140
153
30.7
250
42.8
467
HC1-24 h
175
212
23.3
327
58.7
607
HN03, 90 °C
189
217.4
33.4
345
62.1
621
Ba, mg/kg
3051A
19.3
30.6
0
41.1
5.58
105
HC1-24 h
8.65
7.54
3.26
14.2
2.76
27.6
HN03, 90 °C
22.9
29.4
1.93
44.3
7.62
104
Mn, mg/kg
3051A
10.4
14.5
0
20.7
2.05
50.9
HC1-24 h
8.12
8.66
1.92
14.3
1.9
31.5
HN03, 90 °oC
11.5
17.4
0
23.9
2.18
60.4
3051A
4.15
1.16
3.32
4.97
2.97
5.75
Se, mg/kg
HC1-24 h
2.98
1.13
2.17
3.79
1.74
4.77
HN03, 90 °C
4.21
1.37
3.23
5.18
2.40
5.86
3051A
126
93.3
58.9
193
9.83
235
P, mg/kg
HC1-24 h
128
96.8
59.0
198
8.09
260
HN03, 90 °C
126
92.7
59.4
192
10.8
236
Cu, mg/kg
3051A
2.47
3.19
0.18
4.75
0.99
11.4
HC1-24 h
2.47
1.33
1.51
3.42
1.71
6.12
HN03, 90 °C
2.51
2.64
0.62
4.39
1.34
9.93
Zn, mg/kg
3051A
10.8
3.44
8.30
13.2
5.81
17.7
HC1-24 h
7.30
2.48
5.53
9.07
4.48
11.9
HN03, 90 °C
8.53
3.06
6.34
10.7
5.84
16.2
Ni, mg/kg
3051A
3.47
8.22
0
9.34
0.4
26.8
HC1-24 h
2.59
4.11
0
5.53
0.94
14.2
HN03, 90 °C
3.52
7.66
0
9.00
0.69
25.3
37
-------
Figure 3-12: Sr, Ba, Mn, Cu, Zn, Ni, Se and P (mg kg"1) content in the gypsum from
drywall by three methods
38
-------
Table 3-21: Elemental recovery (%) by nitric acid extraction at 90 °C *
Element
Standard
composition
mg/L
Sample A-G
Sample L-G
Blank
Average
Level I
Level II
Level I
Level II
A1
1000
105
105
126
120
105
112
As
500
98
99
101
99
95
98
B
100
99
100
102
100
95
100
Ba
500
93
93
96
95
98
95
Cd
200
95
96
97
96
104
98
Co
1000
93
94
95
94
104
96
Cr
500
92
92
94
93
99
94
Cu
500
102
104
103
102
106
103
Fe
1000
101
100
101
98
106
101
K
1000
119
119
115
117
98
114
Mg
1000
107
117
97
97
108
105
Mn
200
98
98
98
97
105
99
Mo
200
99
99
101
100
104
101
Na
1000
107
107
111
110
99
107
Ni
500
95
95
97
96
105
98
P
200
103
105
109
108
103
105
Pb
500
90
92
93
93
99
94
Se
500
101
102
104
103
98
102
Sr
200
99
97
113
106
107
104
Ti
200
99
100
104
103
102
101
V
500
98
98
99
98
105
100
Zn
200
97
97
100
99
96
98
* Level I: added 2 mL standard solution in 50 mL extractions; Level II: added 3 mL standard solution in 50 mL
extractions
Based on the percentage of paper and gypsum in the drywall board, the composition of the drywall
board is presented in Tables 3-22 to 3-24.
Table 3-22: The sulfur and metal content of the 10-drywall paper samples by "Nitric
acid extraction" and EPA M3051A
Nitric acid extraction
M3051A
Mean
STDev
Median
Min
Max
Mean
STDev
Median
Min
Max
S(%)
6.44
1.28
6.9
4.33
8.34
12.51
1.43
12.5
10.33
14.73
Ca (%)
9.02
1.57
9.44
6.15
11.22
17.19
2.06
16.78
14.51
20.4
Mg
1180
428
1180
683
1860
2350
1990
2580
253
6950
K (mg/kg)
588
588
242
202
1750
373
524
153
115
1830
Na
1440
276
1420
1020
1820
315
94
312
170
459
Fe
424
113
382
263
638
970
1140
558
265
4160
A1
1010
228
976
811
1580
605
939
325
201
3270
Si (mg/kg)
1190
592
881
711
2360
279
195
211
136
732
Sr
91.1
87.7
46.2
37
263
133
146
63.1
40.5
439
Ba
16.5
4.4
15.1
11.3
24.5
18.9
27.4
8.0
6.3
95.3
Cu
6.28
1.01
6.03
46.4
8.05
3.54
2.92
2.44
2.02
11.6
Mn
27
4.78
26.6
20.2
36.7
13.1
12.9
8.15
5.56
48.8
Ni
1.34
0.84
1.07
0.82
3.68
3.25
7.38
0.89
0.46
24.2
Zn(mg/kg)
83.1
134
22.7
16.5
411
23.0
21.6
14.2
7.81
76.5
39
-------
Nitric acid extraction
M3051A
Se
2.75
0.66
2.45
2.17
4.00
4.10
1.06
4.03
3.04
5.56
P (mg/kg)
302
267
215
45.8
692
156
118
175
16
323
Table 3-23: The sulfur and metal content of the 10 drywall boards by EPA M3051A
Mean
Std Dev
Lower 95%
Upper 95%
Median
Minimum
Maximum
Ca (%)
17.19
2.06
15.72
18.67
16.78
14.51
20.40
S(%)
12.51
1.43
11.48
13.54
12.50
10.33
14.73
Mg (mg/kg)
2350
1990
928
3780
2580
253
6950
K (mg/kg)
373
524
0
748
153
115
1830
Na (mg/kg)
315
94
248
382
312
170
459
Fe (mg/kg)
970
1140
148
1790
558
265
4160
A1 (mg/kg)
605
939
0
1280
325
201
3270
Si (mg/kg)
279
195
140
418
211
136
732
Sr (mg/kg)
133
146
29
237
63.1
40.5
439
Ba (mg/kg)
18.9
27.4
0
38.6
8.0
6.3
95.3
Cu (mg/kg)
3.54
2.92
1.45
5.63
2.44
2.02
11.6
Mn (mg/kg)
13.1
12.9
3.88
22.4
8.15
5.56
48.8
P (mg/kg)
156
118
71.8
240
175
16
323
Ni (mg/kg)
3.25
7.38
0
8.53
0.89
0.46
24.2
Se (mg/kg)
4.10
1.06
3.34
4.86
4.03
3.04
5.56
Zn (mg/kg)
23.0
21.6
7.53
38.5
14.2
7.81
76.5
Table 3-24: The sulfur and metal content of the 10-drywall boards by "Nitric-acid
extraction"
Mean
Std Dev
Lower
Upper
Median
Minimum
Maximum
Ca (%)
21.6
1.05
20.8
22.3
21.8
19.4
23
S(%)
16.4
0.91
15.7
17.1
16.6
14.7
17.9
Mg
2373
1923
998
3749
2582
252
6709
K (mg/kg)
513
602
82
602
291
155
2191
Na (mg/kg)
353
117
270
437
365
167
522
Fe (mg/kg)
933
1074
164
1701
540
251
3924
A1 (mg/kg)
828
1036
87
1568
520
356
3768
Si (mg/kg)
1639
1552
528
2749
815
630
5459
Sr (mg/kg)
179
202
34.6
324
79.75
59
578
Ba (mg/kg)
22.9
27.7
3.1
27.7
12.1
8.2
99
Cu (mg/kg)
3.05
2.35
1.37
4.73
2.35
2
9.7
Mn
13.39
14.52
3
23.78
8.2
5.2
53.8
P (mg/kg)
152
115
69
234
171.8
16.1
305
Ni (mg/kg)
3.27
6.9
0
8.21
1.15
0.7
22.9
Se (mg/kg)
3.98
1.21
3.11
4.85
4.15
2.4
5.4
Zn(mg/kg)
19.6
19.8
5.42
33.8
10.35
6.9
65.6
Ti (mg/kg)
47.5
110
0
126
11.65
8.4
361.2
V (mg/kg)
4.75
2.76
2.77
6.73
3.85
3.7
12.6
B (mg/kg)
12.5
9.5
5.75
19.3
9.9
6.1
38
40
-------
Mean
Std Dev
Lower
Upper
Median
Minimum
Maximum
Co (mg/kg)
0.79
1.1
0.003
1.58
0.4
0.3
3.8
3.6. Water-Extractable Sulfur and Metals
As expected, sulfate in the drywall products gypsum component was highly water-soluble and
present as sulfate (SO4)"2 (Figure 3-13). The average water-extractable sulfate content by IC
(cumulative, four times, Ldiw) in the gypsum samples was 54.4±1.5%, which indicated that almost all
sulfur (Sdiw, 98.2±1.7%) is in the water-extractable (SO4)"2 form (Figure 3-13). The cumulative water-
extractable S (as SO4) was significantly higher than the determination of the total S concentration
measured using the M3051A digestion procedure (p<0.001).
Water-extractable SO.
(%)
list
3rd
12nd
14th
Water-extractable Ca (%)
¦ 1st H2nd
25 ¦ 3rd ¦ 4th
Water-extractable
CaS04 (%)
1st
3rd
12nd
14th
60
80
20
15
10
< ca u 0 u- (b
-L \ 4 in the gypsum from
drywall
Like sulfate, water-extractable calcium was also rapidly released from the gypsum samples
(Figure 3-13). The average cumulative water-extractable Ca concentration was 22.1±0.5 %, which was
higher compared to the EPA 3051A method (p<0.01). Compared to the 0.25 M HC1 extractable Ca
41
-------
(24.5±0.64%) and the nitric acid-extractable Ca (24.0±0.50%), at least 90% of the Ca was dissolved by
water (> Ca diw, 90%).
The average content of the cumulative water-extractable CaSC>4 from the gypsum samples was
76.1±1.9% (Figure 3-13). When accounting for the combined water content (19.4 ±0.6 %) based on
WLOI determination at 230 °C, the average content of gypsum (CaSC>4 2FhO) in the water extraction
was 95.6±2.5%. The overall molar ratio of Ca to SO4 in the water extraction was close to 1 (0.98 ±0.02).
The ratio at the initial extractions (1st and 2nd) for all gypsum from drywall was close to 1.0 (<1.0),
whereas one in the last extraction (4th) was much higher (»1.0) (Figure 3-14), indicating that the initial
extractions had high concentrations of SO4 and Ca (saturated from gypsum), whereas the last extraction
had low concentrations of SO4 and Ca (some other calcium mineral, likely calcium carbonate (calcite)
controlled the equilibrium in the last extraction) (Figure 3-15), as supported by the high pH in the last
extraction (Figure 3-16), as well as the chemical species analysis by VMINTEQ.
Molar ratio of Ca to S04 in water extraction "1st u2nd 1 3rd «4th
. 1
, 1
¦ 1
1
1
J
1 J _ J
1 1 J
1 1 ¦
J
1J
1
A-G
B-G
C-G
D-G
E-G
F-G
G-G
H-G
l-G
L-G
CaS04
2H20
¦ 1st
0.93
1.02
0.96
0.99
0.94
0.99
0.98
0.98
0.97
0.98
1.03
¦ 2nd
0.95
0.95
0.94
0.97
0.93
0.96
0.98
0.96
0.97
0.99
0.96
3rd
1.13
1.07
1.11
1.05
1.14
1.06
1.38
1.07
1.14
1.10
1.17
¦ 4th
4.33
2.17
5.81
2.37
4.10
1.64
11.52
3.75
5.25
1.34
14.29
Figure 3-14: The molar ratio of Ca/S in the water extraction in the gypsum from
drywall
42
-------
500
50
O
0.5
1st
2nd 3rd
4th
¦ Gypsum from drywall board
¦ CaS04 2H20
500
50
0.5
1st
2nd 3rd
4th
Gypsum from drywall board
• CaS04 2H20
Figure 3-15: Effect of gypsum sources on the concentration of Ca and SO4 in the
water extraction
Figure 3-16: pH of the extraction in the gypsum from drywall
43
-------
The water-extractable concentrations of other elements (in addition to Ca and S) in the gypsum
samples were less than 0.2% of the total. The primary elements measured in these water extractions
were Si, P, Mg, and Sr (Figure 3-17, Table 3-25). Gypsum from drywall samples F and L exhibited
higher concentrations of water-soluble silicate and iron. Gypsum from drywall samples E, G, H, and I
exhibited relatively high concentrations of phosphate and sodium, while drywall samples A, B, C, and D
exhibited a larger concentration of strontium.
2500 -r
c
(D
E
(D
QJ
_Q
¦
¦ Si
¦ P
Na
2000 ¦
¦ Mg
¦ Fe
¦ K
¦
Sr
¦ Ba
Ti
1500 --
Ctf)
X — 1000
(D
i_
(D
500 --
CaS04 A-G B-G C-G D-G E-G G-G H-G l-G L-G F-G
Figure 3-17: Water-extractable elements (other than Ca) in the gypsum from
drywall
Table 3-25: The water cumulative extractable sulfate and metal content of gypsum
from drywall
Mean
Stdev
Lower 95%
Upper 95%
Minimum
Maximum
S04 %
54.4
2.0
53.6
55.1
50.5
59.7
Ca %
22.1
0.6
21.9
22.3
21.1
23.5
Ba, mg/kg
4.0
5.2
2.0
5.9
BDL*
16.5
Sr, mg/kg
193
211
115
272
65
631
Mg, mg/kg
139
196
65
212
BDL
546
Fe, mg/kg
20
35
7
33
BDL
149
P, mg/kg
103
89
70
137
BDL
228
Si, mg/kg
310
612
82
539
BDL
1675
BDL, below detection limit
Among the microelements, strontium had the largest water-extractable concentration (up to 1.8
mg/L) and with a cumulative amount of 193±211 mg/kg (predominantly from the first two extractions).
The cumulative water-extractable strontium was close to the value for the 0.25 M HC1 (24 h) or 10%
HNO3 (90 °C) extractions, but higher compared to M3051A (/K0.01) (Figure 3-18), possibly indicating
that most strontium in the drywall was in the form of strontium sulfate (SrSC>4).
44
-------
Water-extractable Sr (mg/kg)
Sr mg/kg
¦ 1st ¦ 2nd
¦ 3rd B4th
I
¦
1
1 1 1 1 1 1 1
_&» G> fo Q> Q> 0> 0> w0> vp> p> ,Q>
p v v 0 v e>' v v' v
800 T
600 --
400 --
200 --
Cumulative water
extraction
10% HN03 oven
extraction
I0.25M HCI extraction i
24h)
A-G B-G C-G D-G E-G G-G H-G l-G L-G F-G
Figure 3-18: Water-extractable Sr in the gypsum from drywall
Conversely, Ba was detected in only half of the studied samples in the water extractions (>0.005
mg L"1). The cumulative water-extractable Ba concentration was much lower than the value from the
three acidic extractions, probably due to barite (BaSO/O formation (Figure 3-19) (Alhajri et al., 2020).
The average of the cumulative water-extractable Ba concentrations of the ten gypsum samples was
3.3±4.3 mg kg"1, whereas the amounts measured using 10% HNO3, the 0.25 M HCI extraction, and the
M3051A digestion, were 23±29, 8.7±7.5, and 19±31 mg"1, respectively.
15 T
Water-extractable Ba (mg/kg)
¦ 1st
¦ 2nd
¦ 3rd
¦ 4th
1
1
5J
-------
extraction (e.g., G-G, and F-G), while other samples were only detectable at the third or fourth
extraction (e.g., H-G, I-G). This observation might be related to their mineral phases and pH in the water
extraction. In addition, Mg(OH)2 and MgCCb can be formed at pH above 7 (da Silva et al., 2018).
Figure 3-20: Water-extractable Mg in the gypsum from drywall
Water-extractable P had very low concentrations (<10 mg kg"1) in almost half of the samples,
whereas the other half of the samples were higher (170±42 mg kg"1) (Figure 3-21). The cumulative
water extraction concentrations from these six samples were very close, even higher than their
corresponding acidic extractions (although either HC1 or HNO3 extraction was not designed for total P).
Brushite (CaFtP04-2FL>0) has a higher water solubility than P and can coexist with gypsum in nature.
Like water-extractable P, water-soluble Si exhibited very low concentrations, being undetected
in most of the samples (7) (<5 mg kg"1). However, two samples (F and G) had a considerable amount
(>1000 mg kg"1) of water-soluble Si, though the Si took a long time to be released (Figure 3-22). This
behavior was similar to the behavior observed in the HC1 extractions. The amount of Si in the extraction
after two weeks was significantly higher than the amount of Si in 24 h (/K0.01), All three acidic
extraction methods (HNO3, HC1, or M3051A) were not designed for determination of total Si (only
partial Si was extracted by these acids). The behavior of water-extractable Si was also related to the
silicon minerals in the drywall.
46
-------
Water-extractable P (mg/kg)
250 j
200 :-
150 ;-
100 j-
50 :-
0
list ¦ 2nd
3rd B4th
_S« e> £> £> , (s fe v0> vp> <2> y O <5 V Cs * v v <<'
C?
300
200 --
100 ¦¦
Cumulative water extraction
10% HN03 oven extraction
I0.25M HCI extraction ( 24h)
I0.25M HCI extraction (2w)
M 3051A digestion
P mg/kg
A-G B-G C-G D-G E-G G-G H-G l-G L-G F-G
Figure 3-21: Water-extractable P in the gypsum from drywall
Water-extractable Si (mg/kg)
2000 T
1500 --
1000 --
500 --
list
12nd
13rd
14th
15000 -r
12000 --
9000 --
6000 --
3000 --
o
aCCQUQLLJlix-1"-^
Cumulative water extraction
10% HN03 oven extraction
10.25M HCI extraction ( 24h) ITI§/Kg
I0.25M HCI extraction (2w)
M 3051A digestion
¦ - ,
I ¦-Jj.iL
A-G B-G C-G D-G E-G G-G H-G l-G L-G F-G
Figure 3-22: Water-extractable Si in the gypsum from drywall
47
-------
4. Leaching Behavior of Drywall
4.1. Kinetics of Leaching Processing
As described earlier, the dominant component (by mass) of gypsum drywall is calcium sulfate,
which can exist at three levels of hydration in the drywall board: dihydrate (gypsum, CaSC>4 2H2O),
hemihydrate (CaS04(H20)o.5, a-hemihydrate and (3-hemihydrate) and the anhydrous state (anhydrite,
CaSC>4). In the kinetic leaching experiments, sulfate and calcium solution concentrations from each of
the five samples tested did not change significantly (p<0.05) under the experimental conditions (L/S
=20), although some variations were noted during the 2-month testing period (Figure 4-1). Based on the
chemical species model VMINTEQ (Gustafsson, 2016), all solutions were near saturation states for
gypsum (SI-gypsum approximately 0) and undersaturation for the anhydrite (Si-anhydrite <0), and both
saturation indexes (Sis) of gypsum and anhydrite did not change significantly from 1 h up to 2 months
(Figure 4-2), except for sample L.
Figure 4-1: Kinetics of SO4 and Ca of different drywalls
48
-------
E
3
to
Q.
>
<4—
o
on
O
u
ro
U
M—
o
on
0.04
0.16
Time ( d)
0.64 2.56
10.24
40.96
¦a
_c
c
<
4—
o
JTi
0
-0.2
-0.4
-0.6
-0.8
1 . 1 1 , 1 , n-
-1 . 1 1 1 . 1 1
T_ A Drywall-A ~ Drywall-B
1 Drywall-G ¦ Drywall-I
m w ~^ / la
0.04
-X— Drywall-L
0.16
0.15
0.64 2.56
Time (d)
10.24
40.96
-0.05 ¦¦
Drywall-A
Drywall-B —5K— Drywall-G M Drywall-I —X— Drywall-L
Figure 4-2: Kinetics of saturation index (SI) of anhydrite, gypsum and CaCCb'EhO
All five samples displayed similar patterns for EC (Figure 4-3), but the patterns of pH behavior
did vary among samples (Figure 4.-4) and can be differentiated between two groups. For Group I
(samples A, B, G, and I), solution pH was initially approximately neutral (pH 7) followed by an increase
over time. For Group II (sample L), solution pH was initially higher (up to 9) and then decreased to
more neutral conditions (pH 7) after one month. The different patterns may have been related to the
mineral composition of the individual drywall products (Table 3.4-1), further supported by the SI of
CaCC>3 H2O. The SI of CaCC>3 H2O in sample L changed from initially being negative, then becoming
positive, followed by a return to being negative (Figure 4-2). The SI in the other four samples was
always negative. The concentration of calcium in solution was not only related to the calcium sulfate
mineral phases present, but also to other calcium minerals (e.g., CaCCbFhO) in the system.
49
-------
Time (day)
Figure 4-3: Kinetics of conductivity of different drywall
Figure 4-4: Kinetics of pH of different drywalls
50
-------
Unlike the relatively consistent calcium and sulfate concentrations observed in the leachates
from the kinetic experiments, the concentrations of most minor ions changed with time. For example,
measured Sr concentrations increased with time, as did the saturation index for celesite (SrSC>4), which
was always below zero (under saturation), and increased with time up to 1 month (Figure 4-5).
Time (d)
Time (d)
0.04 0.16 0.64 2.56 10.24 40.96
Figure 4-5: Kinetics of Sr concentration and saturation index (SI) of celesite (SrS04)
in the solution
The same trend was also observed for Si. For the Group I samples, the concentration of Si
increased initially and then reached equilibrium (Figure 4-6); the SI of SiC>2 (am, gel) was always
negative. For the Group II sample (Dry wall L), the SI was initially negative and then became positive
after two weeks. The time needed for dissolved constituents to reach equilibrium for the samples tested
related to the specific mineral composition of the different samples. There was no single time required
for all samples to reach equilibrium in the kinetic leaching experiment.
Element concentrations in the leachate solutions may not be controlled by a single mineral but
rather may be influenced by several related minerals in the system (Gorski et al., 2017). For calcium in
the drywall-water system, the minerals involved included gypsum, anhydrite, and different calcium
carbonates. For Ba, the kinetics that dictate Ba concentrations relate to both barite (BaSO/O and witherite
(BaCCb) (Figure 4-7). In most cases, barite was over-saturated (SI >0) in solution, whereas the witherite
(SI<0) was unsaturated. As a result, the concentrations of Ba in the leachate were increased.
51
-------
204.8
12.8
J. 0.8
0.05
0.04 0.08 0.16 0.32 0.64 1.28 2.56
Time (d)
5.12
10.24 20.48 40.96
Figure 4-6: Kinetics of Si concentration and saturation index (SI) of SiCh (am, gel)
in the solution
52
-------
0.0040 . 1 1 1 1 1 1 . 1 . 1 1 1 1 1—
0.04 0.16 0.64 2.56 10.24 40.96
Time (d)
0.04 0.16 0.64 2.56 10.24 40.96
Time (d)
Figure 4-7: Kinetics of Ba concentration and SI of barite (BaS04) and witherite
(BaCCb) in the solution
For phosphorus in the drywall-water system, the minerals primarily involved are the different
calcium phosphates, including brushite (dicalcium phosphate, DCP), tricalcium phosphate (TCP), and
octocalcium phosphate (OCP). The calcium-phosphate double-function plot indicated that the control
mineral was related to the particular sample (Figure 4-8). TCP and OCP controlled the phosphate release
for samples A and B with lower concentrations of phosphate in solution. OCP and DCP controlled
phosphate release in samples G and I, while DCP controlled for Drywall L. The controlling mineral
shifted during the kinetics experiment (e.g., from the initial OCP to the final DCP for samples G and I).
53
-------
The dissolution of OCD from samples G and I may contribute to high phosphate concentration (Figure
4-8).
In general, the results demonstrate that both calcium and sulfate in the leachates from the
samples tested can reach equilibrium in a relatively short time (less 24 h), whereas the other constituents
(e.g., Si, Sr, Ba or P) in the leachates require a much longer time (Figure 4-8). This observation also
relates to the minerals found in the samples that control their equilibrium concentrations. The ambiguity
in the definition and measurement of equilibration times has been acknowledged as a major problem in
past kinetic studies (Boulding, 1996; Sparks, 2013).
Under the assumption that the concentration at 7 d represents 100%, the kinetic patterns of all
constituents are presented in Figures 4-9 and 4-10. Some constituents in Figure 4-9 and 4-10 A (e.g., Ca,
and S) may reach (or approach) relative equilibrium within a week, but others that are shown in Figure
4-9 and 4-10 B (e.g., Sr, P, Se) still change noticeably with additional time. EPA has suggested that the
equilibration time should be the minimum amount of time needed to establish a rate of change of the
solute concentration in solution equal to or less than 5% per 24-h interval (Rey et al., 1992). Therefore, a
one-week equilibrium time was employed in this work for further study
54
-------
DCP
OCP
TCP
HAP
1h
4h
24h
120h
a week
2 week
4 week
6 week
8 week
Drywall A
Drywall G
Drywall B
Drywall L
Drywall I
txo
E
c
o
c
(D
U
c
o
(J
(D
X
Q_
i
o
Q_
CM
X
O)
o
X
Q.
i
o
CL
CM
X
O)
o
10
12
log Ca + 2pH
14
12 14 16 18
log Ca + 2pH
Figure 4-8: Kinetics of P concentration and calcium phosphate double function plot
55
-------
of the solubility for different drywalls
0.04 0.16 0.64 2.56 10.24 40.96
Time (d)
Figure 4-9: Equilibrium time and composition changes in drywall (group I, average
of A, B, G and I)
56
-------
150 T
¦a
r*>
(D
3
(D
>
(D
CC
100
50 ¦¦
¦a
r*>
(D
>
Group II (Drywall L): A
¦Ca
¦ ¦a* ST'
, <*"
0 v i—r—P"'
0.04
0.16
0.64
2.56 ... 10.24
Time (d)
Group II (Drywall L): B
TD 50
cc
0.04
0.16
0.64
2.56 10.24
Time (d)
-i .—h
40.96
40.96
Figure 4-10: Equilibrium time and composition changes in drywall (group II,
Drywall-L)
4.2. Liquid-Solid Partitioning Tests
The purpose of this leaching experiment was to assess the concentrations of the constituents
leached from the samples at steady-state conditions. The plotting of pH according to the L/S ratio as well
as the plotting of constituent concentrations and/or release amount provides useful information on the
available quantities and solubility of different constituents. Figure 4-11 displays the pH levels of the five
samples tested. In general, pH decreased as the L/S ratio increased; also, the pH of Drywall L was
significantly higher than the pH values of the other four (Group I).
57
-------
0 I I | I I I I I I | I I I I Lj-J III I | I I I I
1 4 16 64 256
Liquid-to-solid ratio
Figure 4-11: Effect of L/S ratio on the pH of drywall leachate
Figure 4-12 shows constituent concentrations in leachates from sample A as a function of L/S
ratio. As expected, the leached concentration of most constituents decreased dramatically when the L/S
ratio decreased, except for the Ca and S concentration (slight decrease). The most dramatically changed
constituents can be described by linear dependence (logarithmic concentration vs the logarithmic L/S
ratio) (Figure 4-12). Gypsum is the key mineral to control both calcium and sulfate concentration in the
leachates at different L/S ratios. Gypsum at the studied conditions (L/S from 2.5 to 200) for all drywalls
(except drywall L) was confirmed to be over-saturated or close to saturation (SIgyPsum < 0 or ~ 0).
Moreover, calcium carbonates were also over-saturated (SIcaco3 >0) in many cases, which indicated that
calcium carbonate was the controlling mineral phase for the calcium concentration under high pH
leaching conditions (e.g., Drywall L) (Table 4-1).
58
-------
Table 4-1: Saturation index of minerals related to calcium in leachates
Drywall
Anhydrite
Gypsum
CaC03xH20(s)
Aragonite
Calcite
I.S 2.5
A
-0.232
0.017
-2.209
-1.017
-0.873
B
-0.227
0.021
-2.219
-1.026
-0.883
G
-0.207
0.041
-1.66
-0.467
-0.324
I
-0.216
0.033
-1.318
-0.125
0.019
L
-0.599
-0.35
1.941
3.133
3.277
L/S =5
A
-0.220
0.029
-2.192
-1.000
-0.856
B
-0.221
0.028
-2.251
-1.059
-0.915
G
-0.211
0.038
-1.880
-0.688
-0.544
I
-0.222
0.027
-1.467
-0.274
-0.131
L
-0.760
-0.511
2.017
3.209
3.353
I.S 20
A
-0.248
0.001
-2.364
-1.172
-1.028
B
-0.228
0.021
-2.37
-1.177
-1.034
G
-0.242
0.007
-1.93
-0.737
-0.594
I
-0.229
0.02
-1.829
-0.636
-0.493
L
-0.751
-0.502
2.006
3.198
3.342
I.S loo
A
-U.22U
0.029
-3.457
-2.121
-2.121
B
-0.236
0.012
-3.464
-2.128
-2.128
G
-0.231
0.018
-2.783
-1.447
-1.447
I
-0.222
0.027
-2.451
-1.115
-1.115
L
-0.674
-0.425
1.911
3.247
3.247
I.S 2<)(i
A
-0.218
0.031
-3.88
-2.688
-2.544
B
-0.23
0.019
-3.895
-2.703
-2.559
G
-0.244
0.005
-2.789
-1.597
-1.453
I
-0.216
0.033
-2.412
-1.219
-1.076
L
-0.58
-0.331
1.783
2.975
3.119
I.S 40(i
A
-U.321
-0.072
-2.786
-1.594
-1.45
B
-0.312
-0.063
-2.8
-1.608
-1.464
G
-0.355
-0.106
-2.751
-1.558
-1.415
I
-0.342
-0.093
-2.477
-1.285
-1.142
L
-0.533
-0.284
1.627
2.819
2.963
59
-------
L/S ratio L/S ratio
0.40 T
t 7 *
£ 6*
£
5
4
o
ro
4->
c
(D
O
c
o
u
3 :¦
2
i
0
1000 g-
tLO 100 : "
E
E
¦i 10 J-
c
QJ
U
c
o
u
1 :¦
0.1
¦Ca
¦K
Zn
P
Ba
—I 1—I—I I I I I I I 1—I—I I I I I I
10 L/S ratio 100
•S X DOC —!^Mg
¦Na —A—Sr
10 . /c .. 100
L/S ratio
¦B
¦Ba
Mn X Zn
Fe —Si
Figure 4-12: Effect of L/S ratio on the concentrations of different constituents in the
Drywall-A leachates
60
-------
These linearly dependent patterns were not limited to a specific drywall sample and can be
observed in all the samples. For example, the results for Sr from all five samples are presented in Figure
4-13. The linear correlations were valid (/K0.01), Assuming the drywall water-holding capacity as the
pore volume (L/S) (which might represent the real situation for drywall in a landfill), concentrations of
Sr as high as 5 to 17 mg L"1 (with an average of 10 mg L"1) are predicted. In these cases, the mineral
celesite (SrSC>4), which controls the Sr concentration in the leachate, was at unsaturation status, and the
corresponding SIceiesite in all leachates was negative (Table 4.-2). The saturation index of another related
mineral, strontianite (SrCCb), is also presented.
Table 4-2: Saturation index of minerals related to Sr in leachates
L/S ratio
Mineral
Drywall
A
B
G
I
L
2.5
Celestite
-0.165
-0.728
-0.699
-0.762
-0.942
Strontianite
-2.276
-2.853
-1.328
-1.997
1.464
5
Celestite
-0.254
-0.832
-0.961
-0.892
-1.209
Strontianite
-2.359
-2.996
-1.776
-2.271
1.433
10
Celestite
-0.344
-0.945
-0.973
-1.048
-1.444
Strontianite
-2.552
-3.19
-2.026
-2.634
1.264
20
Celestite
-0.34
-1.137
-1.262
-1.268
-1.639
Strontianite
-2.589
-3.413
-2.414
-3.002
0.984
40
Celestite
-0.492
-1.454
-1.474
-1.252
-1.629
Strontianite
-3.024
-3.999
-3.038
-3.128
0.995
100
Celestite
-0.628
-1.68
-1.643
-1.668
-1.955
Strontianite
-3.998
-5.042
-13.254
-4.031
0.496
200
Celestite
-0.757
-1.757
-1.842
-12.152
-1.917
Strontianite
-3.998
-5.042
-13.254
-4.031
0.496
400
Celestite
-0.952
-1.825
-1.979
-12.18
-2.01
Strontianite
-3.55
-4.446
-13.118
-14.449
0.016
The same patterns were observed for Mg and B in all the leached samples. The concentration of
Mg and B in the leachate increased significantly as the L/S ratio decreased, also exhibited linear
dependence (Figure 4-13). Based on these linear equations, the concentrations of B in pore water would
range from 5.5 to 18 mg L"1 with an average of 11 mg L"1. The concentration of Mg in pore water would
be as high as 2,000 mg L"1 with an average of 580 mg L"1 This pattern was also observed for the total
dissolved organic matter (DOC) in the leachates (Figure 4-13), with average concentrations of DOC in
pore water expected to be approximately 3600 mg L"1 (2500 to 4900 mg L"1).
61
-------
CuO
£
£
O
u
CuO
o
1.50
1.00
0.50
o.oo
-0.50
-l.oo
-1.50
~
i—i—i—I—i—i—i—i—I—i—i—i—i—I—i—i—i—i—I—i—i—i—i I i
0.00 0.50 1.00 1.50 2.00 2.50
Log (L/S ratio)
A A
~ B
G
¦ I
3.00 *L
y = -0.3304x +1.1268
R2 = 0.9928
y = -0.6341x +0.7381
R2 = 0.9795
y = -0.5248x +0.5059
R2 = 0.9959
y = -0.6473X + 0.6604
R2 = 0.9928
y = -0.6506X + 0.605
R2 = 0.9922
CuO
£
2.80
2.00 j-
1.20
C
o
(J
W)
o 0.40 :-
u
-o.4o =-
g> -1.20
tLO
E
c
o
(J
CO
Ctf)
o
0.000
3.00
^ 2.50 j-
2.00 j-
1.50 j-
1.00 j-
o.5o j-
0.00 =-
tLO
E
u
c
o
u
u
O
W
o
"
-+-
-+-
0.00 0.50 1.00 1.50 2.00 2.50
Log (L/S ratio)
0.500 1.000 1.500 2.000 2.500
Log (L/S)
I;
0.00
1.00 Log(L/S) 2.00
XG
3.00
X L
3.000
y = -0.8495X + 1.7856
A A R2 = 0.9985
y = -0.9669X + 2.0227
~B R2 = 0.9992
y = -0.9889X + 2.3248
R2 = 0.9993
y = -1.0446X + 1.7952
R2 = 0.9975
y = -0.9048X + 2.5742
R2 = 0.9986
y = -0.9526x +0.5935
R2 = 0.9995
y = -0.9224x +0.9545
R2 = 0.9995
y = -1.0366x +0.7815
R2 = 0.9972
y = -0.9172x +0.7394
R2 = 0.9964
y = -1.1057x +0.2809
R2 = 0.984
L
y = -0.9571x +3.3017
R2 = 0.9939
y = -1.0155x +3.3068
R2 = 0.9979
y = -0.975x + 3.1169
R2 = 0.9864
I y = -1.0017x +3.1843
R2 = 0.9972
3.00
y = -0.8573x + 2.9946
R2 = 0.9933
Figure 4-13: Effect of L/S ratio on the concentrations of Sr, Mg, B, and DOC in five
drywall leachates
The behavior of Si in the leachate was sample-specific (Figure 4-14 and Table 4-3). The linearly
dependent logarithmic Si concentration on the logarithmic L/S ratio patterns was only valid for the
Group I samples (A, B, G, and I), not the Group II sample (Drywall L with the low L/S ratio). As
described earlier, the SI of SiC>2 (am, ppt) was always negative for Group I, whereas the value became
positive for the Drywall L (Group II) at the low L/S ratio (L/S < 10).
62
-------
25 r
2 i"
131-
1 i"
0.5 I-
01-
-0.5 i-
-i I-
-1.5
-2 —r-
0.00
* XXX
X-
X.
S
»;S3
•X..
-x..
¦•X
J A y = -1.0146x +1.3088
R2 = 0.9915
~ B y = -0.7624x + 1.1478
R2 = 0.998
""•'i;--'-'.'* * G y = -0.9783x + 1.0915
R = °-9975
—I 1 I I 1 1 1 I I 1 1 1 I I 1 1 1 I I 1 1 1 I I 1 1 1—I
0.50 1.00 1.50 2.00 2.50 3.00
Log (L/S)
¦ | y = -0.9256x + 0.9811
R2 = 0.996
* L y = -0.9515x +3.1081
R2 = 0.9956
Figure 4-14: Effect of L/S ratio on the concentrations of Si in five drywall leachates
(the linear correlation for Drywall L was from the ratio 40 to 400)
Table 4-3: Saturation index of minerals related to Si in the leachate
L/S ratio
Mineral
Drywall
A
B
G
I
L
2.5
SiCh (am,gel)
-0.882
-0.866
-1.961
-1.096
0.186
SiCh (am,ppt)
-0.852
-0.836
-1.931
-1.066
0.216
Cristobalite
-0.242
-0.226
-1.321
-0.456
0.826
Quartz
0.408
0.424
-0.671
0.194
1.476
5
SiCh (am,gel)
-0.864
-1.108
-1.363
-1.414
0.094
SiCh (am,ppt)
-0.834
-1.078
-1.333
-1.384
0.124
Cristobalite
-0.224
-0.468
-0.723
-0.774
0.734
Quartz
0.426
0.182
-0.073
-0.124
1.384
10
Si02 (am,gel)
-1.379
-1.368
-1.625
-1.676
-0.016
Si02 (am,ppt)
-1.349
-1.338
-1.595
-1.646
0.014
Cristobalite
-0.739
-0.728
-0.985
-1.036
0.624
Quartz
-0.089
-0.078
-0.335
-0.386
1.274
20
Si02 (am,gel)
-1.731
-1.571
-1.89
-1.983
-0.175
Si02 (am,ppt)
-1.701
-1.541
-1.86
-1.953
-0.145
Cristobalite
-1.091
-0.931
-1.25
-1.343
0.465
Quartz
-0.441
-0.281
-0.6
-0.693
1.115
40
Si02 (am,gel)
-2.141
-1.84
-2.246
-2.209
-0.422
Si02 (am,ppt)
-2.111
-1.81
-2.216
-2.179
-0.392
Cristobalite
-1.501
-1.2
-1.606
-1.569
0.218
Quartz
-0.851
-0.55
-0.956
-0.919
0.868
100
Si02 (am,gel)
-2.43
-2.111
-13.287
-2.475
-0.738
Si02 (am,ppt)
-2.4
-2.081
-13.257
-2.445
-0.708
Cristobalite
-1.79
-1.471
-12.647
-1.835
-0.098
Quartz
-1.14
-0.821
-11.997
-1.185
0.552
200
Si02 (am,gel)
-2.43
-2.304
-13.287
-13.288
-0.992
Si02 (am,ppt)
-2.4
-2.274
-13.257
-13.258
-0.962
Cristobalite
-1.79
-1.664
-12.647
-12.648
-0.352
Quartz
-1.14
-1.014
-11.997
-11.998
0.298
400
Si02 (am,gel)
-13.288
-2.369
-13.288
-13.288
-1.277
Si02 (am,ppt)
-13.258
-2.339
-13.258
-13.258
-1.247
Cristobalite
-12.648
-1.729
-12.648
-12.648
-0.637
Quartz
-11.998
-1.079
-11.998
-11.998
0.013
63
-------
The relationship of phosphate was also sample-dependent with the phosphate minerals present
(Table 4-1 and Figure 4-15). Phosphate concentrations in the leachates from sample G were significantly
higher than all the others and might be controlled by the Ca3(PC>4)2 (beta); the low level of phosphates in
the other leachates might be controlled by Ca3(PC>4)2 (am) or other Ca-P minerals. The oversaturated
Ca3(P04)2 (beta) was confirmed in sample G, whereas the Ca3(PC>4)2 (am) values were at undersaturated
status for all samples (Table 4-4).
3 o ¦¦
CU)
E
? -1 +
o
c
o
U -2 4-
Q_ '
W
3 -3 4-
* „
*
~... *
-+-
-+-
0.00 0.50 1.00 1.50 2.00
Log (L/S ratio)
2.50
3.00
A A
~ B
G
* L
y = -1.1092x +0.0321
R2 = 0.988
y = -0.9589x-0.9228
R2 = 0.9994
y = -0.2629x +0.0627
R2 = 0.8163
y = -0.4741X + 0.0647
R2 = 0.9983
y = -0.4988x- 0.1101
R2 = 0.992
Figure 4-15: Effect of L/S ratio on the concentrations of P in five drywall leachates
L/S
Mineral
Drywall-A
Drywall-B
Drywall-G
Drywall-I
Drywall-L
2.5
Ca3(P04)2 (ami)
-3.058
-4.775
-1.466
-1.400
-0.312
Ca3(P04)2 (am2)
-0.308
-2.025
1.284
1.350
2.438
Ca3(P04)2 (beta)
0.362
-1.355
1.954
2.020
3.108
Ca4H(po4)3 3H20(s)
-0.997
-3.568
1.116
1.044
1.047
CaHP04(s)
-1.112
-1.966
-0.591
-0.729
-1.814
CaHP04:2H20(s)
-1.394
-2.247
-0.873
-1.011
-2.095
Hydroxyapatite
7.604
5.024
10.267
10.537
13.798
5
Ca3(P04)2 (ami)
-3.254
-5.388
-2.119
-1.852
-0.805
Ca3(P04)2 (am2)
-0.504
-2.638
0.631
0.898
1.945
Ca3(P04)2 (beta)
0.166
-1.968
1.301
1.568
2.615
Ca4H(P04)3 3H20(s)
-1.300
-4.472
0.247
0.440
0.270
CaHP04(s)
-1.219
-2.257
-0.807
-0.881
-2.099
CaHP04:2H20(s)
-1.500
-2.538
-1.089
-1.162
-2.380
Hydroxyapatite
7.318
4.087
9.177
9.783
13.096
10
Ca3(P04)2 (ami)
-4.276
-6.032
-2.349
-2.285
-1.086
Ca3(P04)2 (am2)
-1.526
-3.282
0.401
0.465
1.664
Ca3(P04)2 (beta)
-0.856
-2.612
1.071
1.135
2.334
Ca4H(P04)3 3H20(s)
-2.770
-5.395
-0.129
-0.110
-0.164
CaHP04(s)
-1.668
-2.536
-0.953
-0.998
-2.251
CaHP04:2H20(s)
-1.949
-2.817
-1.234
-1.279
-2.533
Hydroxyapatite
5.723
3.078
8.863
9.037
12.686
64
-------
L/S
Mineral
Drywall-A
Drywall-B
Drywall-G
Drywall-I
Drywall-L
20
Ca3(P04)2 (ami)
-4.992
-25.300
-2.535
-2.778
-1.348
Ca3(P04)2 (am2)
-2.242
-22.550
0.215
-0.028
1.402
Ca3(P04)2 (beta)
-1.572
-21.880
0.885
0.642
2.072
Ca4H(po4)3 3H20(s)
-3.820
-34.279
-0.353
-0.768
-0.540
CaHP04(s)
-2.002
-12.153
-0.991
-1.163
-2.365
CaHP04:2H20(s)
-2.283
-12.434
-1.272
-1.444
-2.646
Hydroxyapatite
4.626
-25.839
8.528
8.214
12.276
40
Ca3(P04)2 (ami)
-6.252
-25.644
-2.942
-2.909
-1.649
Ca3(P04)2 (am2)
-3.502
-22.894
-0.192
-0.159
1.101
Ca3(P04)2 (beta)
-2.832
-22.224
0.478
0.511
1.771
Ca4H(P04)3 3H20(s)
-5.569
-34.660
-0.799
-0.904
-0.971
CaHP04(s)
-2.490
-12.189
-1.030
-1.168
-2.495
CaHP04:2H20(s)
-2.771
-12.470
-1.312
-1.449
-2.776
Hydroxyapatite
2.594
-26.491
7.755
7.957
11.804
100
Ca3(P04)2 (ami)
-26.778
-26.801
-3.512
-25.428
-22.737
Ca3(P04)2 (am2)
-24.028
-24.051
-0.762
-22.678
-19.987
Ca3(P04)2 (beta)
-23.358
-23.381
-0.092
-22.008
-19.317
Ca4H(P04)3 3H20(s)
-35.953
-35.984
-1.392
-34.431
-32.575
CaHP04(s)
-12.348
-12.356
-1.053
-12.176
-13.012
CaHP04 2H20(s)
-12.629
-12.637
-1.334
-12.457
-13.293
Hydroxyapatite
-28.600
-28.639
6.636
-26.072
-19.854
200
Ca3(P04)2 (ami)
-27.451
-27.494
-3.981
-25.359
-22.641
Ca3(P04)2 (am2)
-24.701
-24.744
-1.231
-22.609
-19.891
Ca3(P04)2 (beta)
-24.031
-24.074
-0.561
-21.939
-19.221
Ca4H(P04)3 3H20(s)
-36.751
-36.808
-2.092
-34.347
-32.368
CaHP04(s)
-12.473
-12.487
-1.284
-12.161
-12.900
CaHP04 2H20(s)
-12.754
-12.768
-1.565
-12.443
-13.181
Hydroxyapatite
-29.821
-29.893
5.929
-25.949
-19.775
400
Ca3(P04)2 (ami)
-25.848
-25.888
-4.126
-25.484
-22.466
Ca3(P04)2 (am2)
-23.098
-23.138
-1.376
-22.734
-19.716
Ca3(P04)2 (beta)
-22.428
-22.468
-0.706
-22.064
-19.046
Ca4H(P04)3 3H20(s)
-34.893
-34.947
-2.328
-34.502
-32.027
CaHP04(s)
-12.219
-12.232
-1.376
-12.191
-12.734
CaHP04 2H20(s)
-12.500
-12.513
-1.657
-12.472
-13.015
Hydroxyapatite
-26.869
-26.937
5.731
-26.170
-19.590
A poor linear relationship for Ba in the leachates was observed in some samples (Figure 4-16).
The saturation index of barite (BaSCU) in these leachates was positive, especially at the low liquid-to-
solid ratio, though the witherite (BaCCb) was negative in many cases (Table 4-5), suggesting that the
linear relationships were related to the minerals that control dissolution. If a mineral were oversaturated,
65
-------
the linear equations would not be valid for this composition. Witherite might be the controller of Ba
samples L and I, whereas barite may control the concentration of Ba in samples A, B, and G.
o.oo
ao
£
£
O
ao
o
-2.50
¦ 1 ¦ ¦ ¦ ¦
*
*
X
<
r- X
•••ft.
••ii,..,.
~
i ¦
X
T
i
".V»
' i
0.00 0.50 1.00 1.50 2.00
Log (L/S ratio)
2.50
A A
~ B
G
3.00 *L
y = -0.1958x-0.9493
R2 = 0.9127
y = -0.290x-0.8752
R2 = 0.9546
y = -0.2332x-0.314
R2 = 0.8898
y = -0.5979x-0.7721
R2 = 0.9944
y = -0.5295x- 1.0261
R2 = 0.9843
Figure 4-16: Effect of L/S ratio on the concentrations of Ba in five drywall leachates
The same patterns were observed for Fe in the leachates (Figure 4-17). The oversaturated status
of iron oxides (e.g., ferrihydrite, goethite, magnetite, and hematite) was confirmed by the chemical
species model (Table 4-5 and 4-6) for all leachates at the low ratio (L/S <40), exhibiting a poor linear
dependence, as expected.
Table 4-4: Saturation index of minerals related to Ba in leachate
L/S ratio
Drywall A
Drywall B
Drywall G
Drywall I
Drywall L
Barite
Witherite
Barite
Witherite
Barite
Witherite
Barite
Witherite
Barite
Witherite
2.5
0.986
-5.185
1.044
-5.141
1.66
-3.986
1.043
-4.253
0.726
-0.928
5
0.088
-6.078
0.989
-5.235
-0.161
-6.024
0.936
-4.503
0.492
-0.926
10
0.909
-5.36
0.997
-5.308
-0.174
-5.952
0.946
-4.7
0.27
-1.082
20
0.913
-5.396
0.901
-5.435
-0.184
-6.065
0.94
-4.853
0.006
-1.431
40
0.878
-5.714
0.743
-5.862
-0.167
-6.388
0.958
-4.978
0.017
-1.418
100
0.852
-6.579
0.579
-6.843
1.268
-5.478
0.111
-6.312
-0.127
-1.736
200
0.731
-7.124
0.364
-7.495
0.078
-6.661
-0.044
-6.433
-0.225
-2.056
400
0.551
-6.108
0.135
-6.546
-0.089
-6.679
-0.171
-6.5
-8.836
-10.87
66
-------
1.00
0.50
1
0.00
CU)
E
c
-0.50
-------
Table 4-5: Saturation index of minerals related to iron in leachate
L/S ratio
Drywall
Ferrihydrite
Ferrihydrite (aged)
Goethite
Hematite
Maghemite
2.5
Drywall A
5.536
6.046
8.246
18.892
11.088
Drywall B
5.268
5.778
7.977
18.356
10.552
Drywall G
5.175
5.685
7.885
18.171
10.367
Drywall I
5.69
6.2
8.4
19.2
11.396
Drywall L
4.233
4.743
6.943
16.286
8.482
5
Drywall A
5.384
5.894
8.094
18.588
10.784
Drywall B
4.762
5.272
7.472
17.344
9.54
Drywall G
4.726
5.236
7.436
17.272
9.468
Drywall I
5.151
5.661
7.861
18.122
10.318
Drywall L
3.723
4.233
6.433
15.266
7.462
10
Drywall A
5.125
5.635
7.834
18.069
10.265
Drywall B
4.441
4.951
7.15
16.702
8.898
Drywall G
4.424
4.934
7.133
16.667
8.863
Drywall I
4.763
5.273
7.473
17.346
9.542
Drywall L
3.751
4.261
6.46
15.321
7.517
20
Drywall A
4.629
5.139
7.339
17.078
9.274
Drywall B
4.171
4.681
6.881
16.162
8.358
Drywall G
4.144
4.654
6.853
16.107
8.303
Drywall I
4.458
4.968
7.168
16.736
8.932
Drywall L
3.971
4.481
6.681
15.763
7.959
40
Drywall A
4.07
4.58
6.779
15.959
8.155
Drywall B
3.729
4.239
6.439
15.278
7.474
Drywall G
3.686
4.196
6.396
15.192
7.388
Drywall I
4.206
4.716
6.915
16.231
8.427
Drywall L
3.6
4.11
6.31
15.02
7.216
100
Drywall A
3.394
3.904
6.103
14.607
6.803
Drywall B
2.994
3.504
5.703
13.807
6.003
Drywall G
-6.571
-6.061
-3.861
-5.322
-13.126
Drywall I
-6.382
-5.872
-3.672
-4.944
-12.748
Drywall L
3.796
4.306
6.505
15.411
7.607
200
Drywall A
2.962
3.472
5.672
13.744
5.94
Drywall B
-7.119
-6.609
-4.41
-6.419
-14.223
Drywall G
-6.571
-6.061
-3.861
-5.322
-13.126
Drywall I
-6.392
-5.882
-3.683
-4.964
-12.768
Drywall L
4.126
4.636
6.835
16.071
8.267
400
Drywall A
-6.543
-6.033
-3.834
-5.267
-13.071
Drywall B
-6.546
-6.036
-3.837
-5.273
-13.077
Drywall G
-6.515
-6.005
-3.806
-5.211
-13.015
Drywall I
-6.376
-5.866
-3.667
-4.933
-12.737
Drywall L
3.926
4.436
6.636
15.672
7.868
Linear patterns have been observed and reported by other researchers (Tiruta-Barna et al., 2004;
Van Praagh and Persson, 2008). Table 4-7 provides a summary of all equations that described
constituent concentration as a function of the L/S ratio. The dominant mineral in drywall is gypsum
-------
(>95%); hence, the leachate composition was relatively simple compared to the other solid wastes,
especially when compared to actual landfill leachates.
Table 4-6: Summary of the linear dependence logarithmic concentration on the
logarithmic L/S ratio in drywall leachates
Concentration (mg/L)
Drywall
Equations
R2
Sr
A
log (Sr)=-0.3304 log (L/S)+1.1268
0.9978
B
log (Sr)=-0.6341 log (L/S)+0.7381
0.9795
G
log (Sr)=-0.5248 log (L/S)+0.5059
0.9959
I
log (Sr)=-0.6473 log (L/S)+0.6604
0.9928
L
log (Sr)=-0.6506 log (L/S)+0.605
0.9922
B
A
log (B)=-0.9526 log (L/S)+0.5935
0.9995
B
log (B)=-0.9224 log (L/S)+0.9545
0.9995
G
log (B)=-1.0379 log (L/S)+0.8339
0.9718
I
log (B)=-0.91721og (L/S)+0.7394
0.9964
L
log (B)=-0.90541og (L/S)+0.1561
0.9465
Mn
A
log (Mn)=-0.8943 log (L/S)+0.3308
0.9943
B
log (Mn)=-0.8295 log (L/S)+0.1524
0.998
G
log (Mn)=-0.8995 log (L/S)+0.4808
0.9988
I
log (Mn)=-0.939 log (L/S)+0.5049
0.9993
L
log (Mn)=-1.2917 log (L/S)-0.7258
0.9901
Zn
A
log (Zn)=-0.8785 log (L/S)-0.1453
0.9887
B
log (Zn)=-0.7228 log (L/S)-0.2072
0.9559
G
log (Zn)=-0.9948 log (L/S)+0.0885
0.9961
I
log (Zn)=-0.8795 log (L/S)+0.1392
0.9861
L
log (Zn)=-0.8743 log (L/S)+0.6153
0.9918
Ni
A
log (Ni)=-1.0207 log (L/S)-l.095
0.9993
B
log (Ni)=-0.9957 log (L/S)-1.2524
0.9947
G
log (Ni)=-0.9856 log (L/S)-0.5726
0.9997
I
log (Ni)=-1.04751og (L/S)-0.6979
0.999
L
log (Ni)=-0.8414 log (L/S)-1.152
0.9962
Co
A
log (Co)=-0.9623 log (L/S)-1.6158
0.9841
B
log (Co)=-0.9506 log (L/S)-1.3921
0.9927
G
log (Co)=-1.1052 log (L/S)-0.69
0.9918
I
log (Co)=-1.001 log (L/S)-1.2372
0.9988
L
Below detection limit
Cd
A
log (Cd)=-0.869 log (L/S)-1.8821
0.9977
B
log (Cd)=-0.7958 log (L/S)-2.5053
0.9882
G
log (Cd)=-0.9772 log (L/S)-2.1308
0.9876
I
Below detection limit
L
log (Cd)=-0.8281og (L/S)-2.3004
0.9955
Se
A
log (Se)=-0.5362 log (L/S)-0.9123
0.9977
B
log (Se)=-0.5651 log (L/S)-1.1754
0.9795
G
log (Se)=-0.73641og (L/S)-0.5576
0.9979
I
log (Se)=-0.7748 log (L/S)-0.3016
0.998
L
log (Se)=-0.7448 log (L/S)-0.413
0.9976
Mo
A
log (Mo)=-1.0087 log (L/S)-0.8572
0.9944
B
log (Mo)=-0.8507 log (L/S)-1.4736
0.9795
G
log (Mo)=-0.9035 log (L/S)-1.2703
0.9979
I
log (Mo)=-1.0139 log (L/S)-0.9726
0.998
69
-------
Concentration (mg/L)
Drywall
Equations
R2
L
log (Mo)=-1.0115 log (L/S)-1.0808
0.9998
DOC
A
log (DOC)=-0.9571 log (L/S)+3.3017
0.9939
B
log (DOC)=-1.01551og (L/S)+3.3068
0.9979
G
log (DOC)=-0.97451og (L/S)+3.1164
0.9904
I
log (DOC)=-0.9643 log (L/S)+3.1459
0.996
L
log (DOC)=-0.865 log (L/S)+3.002
0.9967
Mg
A
log (Mg)=-0.8495 log (L/S)+1.7856
0.9985
B
log (Mg)=-0.9669 log (L/S)+2.0227
0.9992
G
log (Mg)=-0.9889 log (L/S)+2.3248
0.9993
I
log (Mg)=-1.0446 log (L/S)+1.7952
0.9975
L
log (Mg)=-0.9048 log (L/S)+2.5742
0.9986
K
A
log (K)=-0.9978 log (L/S)+2.5148
0.999
B
log (K)=-1.0494 log (L/S)+2.6601
0.9962
G
log (K)=-0.5221 log (L/S)+1.5792
0.9876
I
log (K)=-0.9829 log (L/S)+1.5168
0.998
L
log (K)=-0.98211og (L/S)+1.519
0.9967
P
A
log (P)=-1.1092 log (L/S)+0.0321
0.988
B
log (P)=-0.9589 log (L/S)-0.9228
0.9994
G
log (P)=-0.2833 log (L/S)+0.0459
0.9067 (invalid)
I
log (P)=-0.4741 log (L/S)+0.0647
0.9983
L
log (P)=-0.4988 log (L/S)+0.1101
0.992
Si
A
log (Si)=-1.0146 log (L/S)+1.3088
0.999
B
log (Si)=-0.73196 log (L/S)+1.0971
0.9878
G
log (Si)=-0.9783 log (L/S)+1.0915
0.9975
I
log (Si)=-0.8894 log (L/S)+0.95
0.9928
L
log (Si)=-0.73326 log (L/S)+2.6642
0.9574 (invalid)
Ba
A
log (Ba)=-0.1958 log (L/S)-0.9493
0.9127 (invalid)
B
log (Ba)=-0.41551og (L/S)-0.7541
0.9384 (invalid)
G
log (Ba)=-0.2332 log (L/S)-0.3143
0.8898 (invalid)
I
log (Ba)=-0.60171og (L/S)-0.7721
0.9952
L
log (Ba)=-0.5295 log (L/S)-1.0261
0.9843
Fe
A
log (Fe)=-0.9186 log (L/S)+0.9871
0.9482 (invalid)
B
log (Fe)=-0.8008 log (L/S)+0.3974
0.9392 (invalid)
G
log (Fe)=-0.5467 log (L/S)-0.1125
0.7513 (invalid)
I
log (Fe)=-0.5181 log (L/S)+0.1155
0.8485 (invalid)
L
log (Fe)=-0.1585 log (L/S)-0.9264
0.3528 (invalid)
Logarithmic concentration linear dependence of the logarithmic L/S ratio relationship can be
applied if the corresponding constituent minerals exist in an unsaturation state (SI<0), at least for
drywall. This assessment is based on the chemical species in the leachate. Based on these assumptions,
the composition of the pore water for each drywall sample tested as might be expected in a landfill may
further be estimated based on the assumed L/S. In this case, the L/S ratio has been estimated from the
water absorption capacity determined from the monolithic leaching test (ML). Using this approach, the
average L/S was 0.44 (0.16 to 0.55), and this value was assumed to represent drywall pore water
concentrations in a landfill environment; the concentrations of leachate are presented in Table 4-8 from
these equations. As expected, significant correlations (Shapiro-Wilk Normality Test, Significance Level
= 0.05) between the total amount and the concentration at pore volume for most constituents were
observed.
70
-------
Table 4-7: Pore water concentration of drywall (mg L"1)*
Average
Min
Max
Sr
10.0
5.2
16.6
B
11.5
7.3
17.7
Si
375
15.2
1800
Mn
4.1
2.0
6.9
Cu
6.3
0.30
12.2
Zn
5.6
0.956
20.5
Ni
0.21
0.066
0.38
Co
0.20
0.045
0.56
Cd
0.017
0.005
0.023
Se
0.62
0.094
1.51
Mo
0.23
0.056
0.53
P
1.71
0.21
3.21
Mg
580
110
2000
DOC
3600
2500
4900
*
The pore water concentration was estimated by the corresponding linear equations, and the pore volume (L/S) was determined by the ML
test.
As expected, the amount released by drywall composition increased with an L/S ratio decrease,
as the constituents can be described by the linear dependence of composition logarithmic amount on the
logarithmic L/S ratio (e.g., Ca, S, Sr, Ba, Sr, and Se in sample A) (Figure 4-18). These compositions
usually come from one single dominant source of minerals (Table 4-1, 4-2, and 4-5) and do not relate to
saturation status.
If the leached constituents originate from different minerals, the assumption of linear
dependence may not be valid. For example, the linear dependence between the logarithmic released P
amount and logarithmic L/S ratio for samples I and L was determined valid, whereas no clear patterns
were observed for samples A and B (Figure 4-19). The linear dependence between the logarithmic
released Si amount and logarithmic L/S ratio was also observed only for sample B, not the other samples
(Figure 4-20).
71
-------
5
4
3 !-
2 I-
i i-
o :¦
-i
-2
ao
o
0
+'+ +
X
X
.
+ +
X X
~ ::: X X X
—I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1—
0.5 1 1.5 2
Log (L/S ratio)
X
X
—i—i—l—i—i—i—r
2.5 3
Drywall A
Release amount (mg/kg)
• Ca
~ S
X Ba
+ Sr
A Se
y = 0.9607X + 2.8807
R2 = 0.9991
y = 0.9887x +2.6985
R2 = 0.9989
y = 0.8042X - 0.9493
R2 = 0.9944
y = 0.6484x +1.1489
R2 = 0.9962
y = 0.4638x-0.9123
R2 = 0.9969
Figure 4-18: Effect of L/S ratio on the release amount of Ca, S, Ba, Sr, and Se from
Drywall- A
ao
£
T3
CD
(O
CD
CuO
O
1.5
Log (L/S ratio)
i A
~ B
XG
* L
Released P (mg/kg)
y = 0.0321x-0.0731
R2 = 0.1811
y = 0.0411x-0.9228
R2 = 0.7648
y = 0.7506X + 0.0307
R2 = 0.9723
y = 0.5259x +0.0647
R2 = 0.9986
y = 0.5012x-0.1101
R2 = 0.9921
Figure 4-19: Effect of L/S ratio on the release amount of P from different drywall
leachates
4.0
Released Si ( mg/kg)
w
w
E
3.0
~ A
y = -0.0758x +1.4171
R2 = 0.1529
X
~ B
y = 0.2376x +1.1478
00
¦a
2.0
A
R2 = 0.9795
(D
^
ro
(D
-------
leachates
The corresponding mineral information can be found in Table 4-4 (P) and Table 4-3 (Si),
respectively. The plots for the release amount of S, Ca, B, Ba, Sr, and Se from the different drywall samples
are presented in Figures 4-20 to 4-26, and no general patterns can be applied directly.
Despite the differences noted above, in general, the minerals in drywall are quite simple and
similar (Table 3-11). There were almost no differences among all samples relative to calcium and sulfate
release (Figure 4-21 and 4-22). These observations suggest that the patterns of the constituent
concentrations and mineral phases in a drywall source are more complex than the concentrations in
leachate, and the linear dependence of the logarithmic amount on the logarithmic L/S ratio needs further
confirmation for each circumstance (each composition with its corresponding minerals). In addition, the
compositions in leachate usually are not controlled by a single mineral.
Ctf)
E
¦a
(D
CU
(D
tLO
o
i"*'®""
.•W"
t—i—i I i—i—i—i I i—i—i—i I i—i—i—i I i—i—i—i I i—i—i—r
0.5 1 1.5 2
Log (L/S ratio)
2.5
Released S (mg/kg)
J A
y = 0.9506x +2.7782
R2 = 0.9988
~ B
y = 0.9887x +2.6985
R2 = 0.9989
XG
y = 0.9739x +2.7403
R2 = 0.9996
¦ 1
y = 0.9717x +2.7572
R2 = 0.9986
* L
y = 0.9418x +2.8289
R2 = 0.9981
Figure 4-21: Effect of L/S ratio on the release amount of S from different drywall
leachates
— 5 --
tLO
E
¦a
(D
(D
tLO
o
4 --
3 ¦¦
2 --
1 ¦¦
*
...I*"
—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r-
0.5
1 1.5
Log (L/S ratio)
2.5
Released Ca (mg/kg)
A A
~ B
XG
¦ I
51 L
y = 0.9615x +2.8772
R2 = 0.9988
y = 0.9607x +2.8807
R2 = 0.9991
y = 0.9639x + 2.8754
R2 = 0.9995
y = 0.9633x +2.8701
R2 = 0.9988
y = 0.9553x +2.848
R2 = 0.9994
Figure 4-22: Effect of L/S ratio on the amount of Ca released from different drywall
leachates
73
-------
euo
£
T3
-------
0.5
Q)
on
"S -0.5 ¦¦
tn
ro
_0J
2.5 units) during the
leaching process were observed for sample L, which may relate to its mineral composition (e.g., a small
amount of carbonates).
Released Se (mg/kg)
0.5
1 1.5
Log (L/S ratio)
2.5
A A
~ B
G
> L
= 0.4638x-0.9123
R2 = 0.9969
= 0.5827x- 1.295
R2 = 0.9655
= 0.2636x-0.5576
R2 = 0.9837
= 0.2252x-0.3016
R2 = 0.9766
= 0.2745x-0.4295
R2 = 0.9766
75
-------
0.01 0.10 1.00 10.00 100.00
Time ( day)
Figure 4-27: Kinetics of pH and EC in M1315 leaching process
The results for cumulative released constituent masses (or loss composition, PCum-Mi. %) are
summarized in Table 4-9, based in the cumulative constituent mass measured in the ML tests and the
masses of the total constituents presented earlier. Different size membranes (0.05, 0.45, and 5 |j,m) were
used for the filtration; however, no filter size impact was noted in elemental concentration and the data
represent the average of the three filter sizes (p>0.05). In addition, as several elements were below
detection limits, only 11 constituents (Ca, S, Sr, Ba, Mg, Si, Fe, Mn, Zn, P, and DOC) in five samples
were compared. As expected, the amount released (PCUm-L) was related to the mineral composition and
the specific sample source. There was almost no difference for the dominant constituents in drywall, Ca,
and SO4, but significant differences were observed for the minor constituents (e.g., Sr and Ba (Table 4-
9)).
Table 4-8: Cumulative release composition (%) in M1315 leaching processing
Average
STDEV
Min
Max
DOC
5.4
1.4
3.9
7.0
Fe
7.7
8.9
0.6
22.9
Si
15.5
19.4
0.9
49.0
P
18.8
8.5
5.8
27.5
Mg
19.1
17.8
4.4
48.0
Ba
24.8
14.9
10.8
50.3
Mn
39.7
15.4
13.1
52.0
76
-------
Average
STDEV
Min
Max
s
43.6
4.0
37.8
47.9
Ca
43.0
3.8
37.3
46.6
Sr
50.3
8.2
40.0
61.5
Zn
63.4
12.9
46.0
72.9
The cumulative released masses of constituents from the drywall samples were plotted as a
function of leaching time (Figure 4-28). A general linear dependence between the logarithmic cumulative
released compositions (PCUm-L mg m"2) and logarithmic total leaching time (tcum) was observed, although
there were some differences among the different samples (Table 4-10). For the dominant constituents S
and Ca, as well as the minor constituents Sr, Zn, and DOC, the variation of slopes of the best-fit equations
were below 20%. For the other minor constituents (Fe, Si, P, Mg, Ba, DOC, and Mn), the variation of
slopes was much higher, up to 50%. The variation of intercepts in the equations were much higher
compared to the changes in slopes, and there was no link between the intercept and its corresponding total
constituent amount in the sample.
Table 4-9: Slopes of the linear equation of the logarithmic cumulative released
compositions and logarithmic total leaching time for drywall
Sample
Slope
r2
S
Drywall A
0.724
0.997
Drywall B
0.757
0.992
Drywall G
0.672
0.993
Drywall I
0.710
0.998
Drywall L
0.697
0.999
Overall
0.712
0.982
Ca
Drywall A
0.734
0.997
Drywall B
0.767
0.994
Drywall G
0.675
0.996
Drywall I
0.716
0.998
Drywall L
0.712
0.998
Overall
0.721
0.982
Sr
Drywall A
0.736
0.997
Drywall B
0.721
0.985
Drywall G
0.631
0.994
Drywall I
0.647
0.998
Drywall L
0.653
0.997
Overall
0.678
0.702
Zn
Drywall A
0.351
0.983
Drywall B
0.356
0.991
Drywall G
0.378
0.956
Drywall I
0.358
0.990
Drywall L
0.292
0.981
Overall
0.347
0.568
Ba
Drywall A
0.648
0.996
Drywall B
0.606
0.992
Drywall G
0.730
0.995
Drywall I-lst period ( 0-28 d)
0.372
0.985
Drywall I-2nd period (28 to 63 d)
0.676
0.992
Drywall L-lst period ( 0-28 d)
0.314
0.976
Drywall L-2nd period (28 to 63 d)
0.711
0.980
Overall
0.554
0.619
77
-------
Sample
Slope
r2
Fe
Drywall A
0.348
0.967
Drywall B
0.241
0.973
Drywall G
0.401
0.962
Drywall I-lst period (0-7d)
0.383
0.964
Drywall I-2nd period (28 d-63 d)
1.482
0.970
Drywall L-lst period (0-28 d)
0.327
0.987
Drywall L-2nd period (28 d-63 d)
1.965
0.965
Si
Drywall A
0.359
0.989
Drywall B
0.450
0.980
Drywall G
0.371
0.992
Drywall I
0.338
0.994
Drywall L-lst period (0-42 d)
0.882
0.992
Drywall L-2nd period (42 d-63 d)
0.059
0.977
P
Drywall A-lst period (0-42 d)
0.227
0.984
Drywall A-2nd period (42 d-63 d)
0.644
0.964
Drywall B
0.227
0.992
Drywall G
0.625
0.986
Drywall I
0.677
0.999
Drywall L-lst period (0-28 d)
0.594
0.994
Drywall L-2nd period (28 d-63 d)
1.259
0.990
DOC
Drywall A
0.436
0.982
Drywall B-2nd period (0-7 d)
0.488
0.994
Drywall B-2nd period (7 d-63 d)
0.223
0.994
Drywall G
0.309
0.992
Drywall I
0.483
0.985
Drywall L-lst period (0-28 d)
0.262
0.993
Drywall L-2nd period (28 d-63 d)
0.911
0.973
Mg
Drywall A-lst period (0-2 d)
0.615
1.000
Drywall A-2nd period (2 d-63 d)
0.282
0.994
Drywall B~lst period (0-2 d)
0.627
0.996
Drywall B-2nd period (2 d-63 d)
0.109
0.983
Drywall G-lst period (0-2 d)
0.845
0.998
Drywall G-2nd period (2 d-63 d)
0.244
0.976
Drywall I-lst period (0-2 d)
0.563
1.000
Drywall I-2nd period (2 d-63 d)
0.377
0.958
Drywall L-lst period (0-2 d)
0.934
0.998
Drywall L-2nd period (2 d-63 d)
0.546
0.997
Mn
Drywall A-lst period (0-2 d)
0.566
1.000
Drywall A-2nd period (2 d-63 d)
0.230
0.979
Drywall B~lst period (0-2 d)
0.528
1.000
Drywall B-2nd period (2 d-63 d)
0.208
0.987
Drywall G-lst period (0-2 d)
0.659
0.999
Drywall G-2nd period (2 d-63 d)
0.445
0.988
Drywall I-lst period (0-28 d)
0.534
0.996
Drywall I-2nd period (28 d-63 d)
0.202
0.994
Drywall L-lst period (0-28 d)
0.337
0.958
Drywall L-2nd period (28 d-63 d)
1.224
0.937
78
-------
00
E
> —
j® e
E
3
u
oo
o
5.50 t-
5.oo
4.50
4.oo
3.50
3.00 :-
8
T 1 1 1 1 1
3.50 5.50
7.50
Log (leaching time, s)
DO
E
E
3
u
6 3"
5.5 i-
5
4.5
4 i-
3.5
3 :-
Ca
3.5
X
H
0
1—1—
4.5
—I 1 I I 1—
5.5
1 Drywall A
~ Drywall B
X Drywall G
~ Drywall I
Drywall L
I i ¦ i i I
6.5
Log (leaching time, s)
7.5
ao
E
E
=5
u
Sr
*
-+-
i i i i
3.5 5.5 7.5
Log (leaching time, s)
2 T
1.5 --
1 --
0.5 --
Zn
X
X xx
X
i
3.5
5.5
—I
7.5
3
£
b0
2.5
E
CD
2
tn
o
cu
1.5
>
_ro
1
E
X
u
0.5
¦U
bO
I
0
Log (leaching time, s)
Fe
3.5
5.5
7.5
Log (leaching time, s)
2 r
e 1.5
l --
0.5 --
0 --
-0.5
Ba
Sm
i
—i—i—i—r
—\
7.5
3.5 5.5
Log (leaching time, s)
3.5 5.5
Log (leaching time, s)
3.5 5.5 7.5
Log (leaching time, s)
2.5 T
3.5 5.5 7.5
Log (leaching time, s)
4.5 -r
o
o
Log (leaching time, s)
Figure 4-28: Cumulative loss vs total leaching time for the different components in
drywall
79
-------
Typically, leaching is assumed to be diffusion-controlled if the slope of a release equation is
0.5±0.15 (e.g., Zn, and Ba in most drywall). Dissolution or wash-out of mobile species was observed for
the dominant constituents in drywall, including S, Ca, and even Sr in all drywall, as the slope of these
equations was larger than 0.65. Meanwhile, the behavior of some constituents in drywall might be
controlled by multiple mechanisms, as the slopes in the linear equations were changed significantly
during the leaching process. A surface wash-off pattern, with initial rapid leaching followed by a lower
leaching rate, was observed in some cases (e.g., Mg in all samples and Si in sample L). A delayed
release pattern was also observed for Fe in samples I and L, P in samples A and L, and Mn in sample L.
In this pattern, a low initial release rate was observed, followed by a higher leaching rate toward the end.
A depletion pattern was not common in drywall and occurred only when the composition had been
leached out during the initial period (e.g., Si in sample L and Mn in samples A, B, and I).
The interval flux of different constituents was also plotted as a function of mean leaching time
(Figure 4-29). Although a linear dependence between the logarithmic flux (mg m"2s"') and logarithmic
mean leaching time (T) was suggested, the corresponding correlation coefficient (r ) was significantly
lower than the correlation coefficient of the cumulative released constituents as a function of total
leaching time (Table 4-11), especially for the constituents involving two different leaching processes
(e.g., Fe, Si in sample L). As stated earlier, several mechanisms may control the leaching behavior of
drywall.
Based on the diffusion model, the diffusivity and leachability index of the constituents were
calculated (ANS, 1986; Kosson et al., 2002); the results were presented as Table 4-12 and 4-13,
respectively. As expected, the highly leachable constituents were Zn, Sr, SO4, and Ca, the moderately
leachable constituents were Mg and Mn, and the relatively slow-release constituents were P, Fe, Si, Ba,
and DOC. However, the mobility of the constituents, especially the minor constituents, was closely
related to the minerals in the samples. The variations of De or LX were large (e.g., Mg, Fe, and Si)
among the different samples. De is a dynamic coefficient in the leaching process, and a large variation
was observed in the process of leaching for a single sample. The results of De for different constituents
in sample A are presented in Figure 4-30.
80
-------
£
ao
£.
X
3
l.OOE+OO n-
l.OOE-Ol : -
1.00E-02
l.OOE+OO t-
X X
• 8
ttJ
CuO
£_
X
3
1.00E-02
1.00E-03
1.00E-04
1.00E-05 ±
1.00E-06
0 01 Mean Interval (days) 100
Sr
~
A A
O.Ol 0.1,
1.00E-02
1.00E-03
1.00E-04
1.00E-05
1.00E-06
X
fi
T 1 I I Mill I 1 I I I Mil
0.01 0.1 1
Mean intervaf^fclays)100
Zn
x
X
X
~ ~ X ^
0
1 I I lllll I
10
I I I I Mil
100
Mean Interval (days)
5.00E-03
5.00E-04 =-
5.00E-05
X *
Si
X X
X
5.00E-06 —i—i i 11ml—i—i i 11ml
0.01 0.1 1
i i 11 ml—
10
I I I lllll
100
1.00E-03
1.00E-04 t
" 1.00E-05
£
CuO
E 1.00E-06 ; -
x
r 1.00E-07
X
&
Mean Interval (days)
X X X
~
I a Snfgj
~ ~
~~~
I I 11 lllll I I 11 lllll
0.01 0.1 1
I lllll I I 11 lllll
10 100
Mean Interval (days)
1.00E-02
x Ca
S
X X
Sx g
Drywall A
~ Drywall B
X Drywall G
~ Drywall I
> Drywall L
rami—
Tml—
0 01 °r3lean Intervarf&ays)100
1.00E-04 3-
Ba
X X v
1.00E-05 i-
1.00E-06 : -
1.00E-07
4 $ XX*X
S ~
X
i i i nil 1—i i i i nil
0.01 0.1 ivieah interva^days) 100
1.00E-02
Fe
7- 1.00E-03
to
E 1.00E-04
tLO
— 1.00E-05
X
3
^ 1.00E-06
—I 1 I I lllll
~
D %
I *||
—i 1 i i Mini
0.01
1.00E-02 ;
1.00E-03 |-
1.00E-04 ¦ -
1.00E-05
Mill
Mean'Werval (days)100
Mg
* *
~ ~ $ X
i i i iiiiiI i i 11mil -mnl
0.01 0.1 1 10 100
Mean Interval (days)
Mn
i
*»«
t—i i mill i—i i 11 ml
0.01 0.1 1
nil i—i i 11 nil
10 100
Mean Interval (days)
81
-------
5.00E-01 : -
5.00E-02
5.00E-03 : ¦
5.00E-04
0.01
0.1 1 10
Mean Interval (days)
100
Figure 4-29: Flux vs mean interval time for the different components in drywall
Table 4-10: Slope and r2 of the equation between logarithmic flux and logarithmic
mean leaching time
Sample
Slope
r2
S
Drywall A
-0.263
0.867
Drywall B
-0.253
0.795
Drywall G
-0.325
0.895
Drywall I
-0.264
0.888
Drywall L
-0.269
0.916
Ca
Drywall A
-0.253
0.868
Drywall B
-0.243
0.793
Drywall G
-0.318
0.912
Drywall I
-0.254
0.880
Drywall L
-0.249
0.892
Sr
Drywall A
-0.266
0.913
Drywall B
-0.300
0.827
Drywall G
-0.371
0.924
Drywall I
-0.327
0.923
Drywall L
-0.298
0.902
Zn
Drywall A
-0.613
0.929
Drywall B
-0.613
0.927
Drywall G
-0.697
0.900
Drywall I
-0.594
0.952
Drywall L
-0.700
0.909
Ba
Drywall A
-0.383
0.882
Drywall B
-0.472
0.922
Drywall G
-0.300
0.922
Drywall I
-0.551
0.922
Drywall L
-0.590
0.895
Fe
Drywall A
-0.669
0.887
Drywall B
-0.787
0.965
Drywall G
-0.577
0.884
82
-------
Sample
Slope
r2
Drywall I
-0.149
0.185
Drywall L
-0.373
0.451
Si
Drywall A
-0.606
0.928
Drywall B
-0.490
0.944
Drywall G
-0.618
0.967
Drywall I
-0.657
0.966
Drywall L
-0.433
0.509
P
Drywall A
-0.699
0.907
Drywall B
-0.796
0.952
Drywall G
-0.418
0.910
Drywall I
-0.326
0.936
Drywall L
-0.239
0.646
DOC
Drywall A
-0.687
0.946
Drywall B
-0.745
0.957
Drywall G
-0.775
0.973
Drywall I
-0.565
0.917
Drywall L
-0.537
0.831
Drywall A
-0.630
0.950
Drywall B
-0.850
0.918
Drywall G
-0.647
0.884
Drywall I
-0.537
0.926
Drywall L
-0.360
0.867
Mn
Drywall A
-0.709
0.930
Drywall B
-0.725
0.943
Drywall G
-0.504
0.875
Drywall I
-0.628
0.916
Drywall L
-0.529
0.743
Table 4-11: Weighted arithmetic mean De of drywall board (cm2 s"1)
Average
STDEV
Min
Max
S
6.02E-07
3.79E-07
1.98E-07
1.13E-06
Ca
5.54E-07
3.31E-07
1.66E-07
1.01E-06
Sr
4.19E-07
4.58E-07
9.26E-09
1.13E-06
Zn
1.81E-07
1.69E-07
8.35E-09
4.10E-07
Mn
6.49E-08
7.58E-08
1.39E-09
1.86E-07
Mg
5.89E-08
1.29E-07
8.44E-11
2.89E-07
P
4.47E-09
4.17E-09
3.05E-10
9.82E-09
Ba
3.84E-09
2.45E-09
1.45E-09
7.76E-09
Si
3.00E-09
5.84E-09
5.92E-12
1.34E-08
Fe
2.07E-09
4.02E-09
3.73E-12
9.23E-09
DOC
2.09E-10
1.48E-10
6.57E-11
4.49E-10
83
-------
Table 4-12: Leachability index (LX) of drywall board
Average
STDEV
Min
Max
Zn
7.0
0.78
6.2
8.1
Mn
7.5
0.92
6.5
OO
OO
Sr
8.1
0.15
7.9
8.2
S
8.3
0.23
7.9
8.5
Ca
8.3
0.22
8.0
8.6
Ba
8.5
0.41
7.9
9.1
P
OO
00
0.51
8.2
9.6
Mg
8.7
1.45
6.3
9.9
Si
9.5
1.09
8.1
11.1
DOC
9.8
0.43
9.2
10.2
Fe
10.3
0.85
9.4
11.6
1.00E-06
1.00E-07
0.06
Drywall A
i
0.6 6
Time (d)
60
Ca
S
Ba
P
Si
Mg
Fe
DOC
Figure 4-30: Dynamics of De of different compositions in drywall A
The weighted arithmetic means De was determined, which is weighted based on the time of
leaching; the arithmetic means De was often used in other studies. The weighted arithmetic means De
was surmised to be a more reasonable estimate for leachability than the arithmetic mean. LXs of most
constituents in the drywall were between 8 and 9. This finding supports the current management of
drywall landfill as Subtitle D landfill (Canada, 1991; Dermatas et al., 2004; Moon and Dermatas, 2007)
The results of the monolithic leaching tests for five different drywall products indicated that the
leaching processes occurring during the M1315 laboratory conditions were not controlled by diffusion
only, and the dominant constituents, S and Ca (along with some minor constituents, such as Sr), leached
out following a dissolution model. While the other leaching models (surface wash-off pattern, delayed-
release pattern, and depletion pattern) were also observed for some constituents among the samples,
most constituents displayed linear dependence between the logarithmic cumulative released
compositions and logarithmic total leaching time. Based on the theory of diffusion, the diffusivity and
leachability index of the constituents were calculated. The highly leachable constituents were Zn, Sr,
SO4, and Ca; the moderately leachable constituents were Mg and Mn; and the relatively slow release
constituents were P, Fe, Si, Ba, and DOC. The leachability index of most compositions was between 8
and 9.
84
-------
5. Conclusions
The findings presented in this report provide information regarding the mineral and chemical
characteristics of gypsum drywall products and add to the database of existing literature on the subject. A
total of 10 drywall samples were collected, and all or a subset of these samples were characterized using
multiple analytical procedures, including their leaching behavior by EPA Methods 1315 and 1316. The
major findings of this research are highlighted below.
5.1. Drywall Characteristics
Ten drywall products were evaluated for mineral analysis, moisture content, total sulfur, and
metal composition, water-soluble sulfur and metals, organic component analysis, and two different
leaching tests (M1315, and M1316). Overall, the primary constituents and minerals in drywall are quite
uniform, but the composition of minor constituents exhibited a large degree of variability. This
variability was attributed to differences in gypsum feedstock, conditions at the processing facilities, and
sample processing and analysis.
• MC of the drywall was related to the temperature used for its determination. The average MC of
gypsum from drywall measured at 45, 105, 230, 400 and 550 °C was 0.76, 0.35, 13.6, 19.4, 19.7
and 20.8%, respectively. The average MC of the drywall samples tested at 105 °C was 15.36%.
The MC results at 150 °C were unstable because calcium sulfate exists at three levels of
hydration at that temperature. Samples that were air-dried and not dried at elevated temperatures
were employed in this work since the mineral phases can change during the high-temperature
MC analysis
• Drywall, including the gypsum core of the drywall board, contains a small amount of organic
carbon. The average total carbon and sulfur content of the gypsum samples using a combustion
methodology were 0.87 and 17.56%), respectively (air-dry weight basis). Formaldehyde was
detected at a concentration range from 500 to 8500 |Lxg kg"1, with a median of 1800 and an
average of 3700 |Lxg kg"1. Tributyltin (TBT) was also detected in some samples, namely, those
products manufactured for greater mold control. Polynuclear aromatic hydrocarbons (PAHs)
were also detected in some samples, and the presence of PAHs was attributed to the paper
fraction of the drywall product.
• The dominant mineral in the drywall products was gypsum, accompanied by small amounts of
hemihydrate and anhydrite. Carbonate and silica were also detected.
• The total acid extractable sulfur and metal concentrations of the gypsum core of the drywall
samples were investigated using different methods. The average sulfur contents of the gypsum
samples using EPA M3051A, 0.25 M HC1 extraction (24 h), and 10%> HNO3 at 90 °C were
13.67, 17.65, and 18.34%>, respectively. The average calcium contents by these methods were
18.57, 24.50, and 24.03%), respectively. The average strontium content was 140, 175, and 189
mg kg"1, respectively. Re-precipitation is a common occurrence after microwave digestion of
materials with high amounts of calcium sulfate minerals, and the re-precipitation was confirmed
using digestion experiments in which dilution was conducted at different temperatures. The
results suggest that analysts should be cautious of measuring elemental concentrations on
gypsum materials using EPA M3051A; use of this method might significantly underestimate the
content of sulfur, calcium, strontium, and other compositions. A new procedure of acid
extraction (10%> HNO3 at sub-boiling temperatures (90 °C) for 16 h (overnight) was developed in
this work and is recommended for future work. The results of extractable sulfur in this new
procedure were like the total sulfur concentration measured using the combustion technique and
85
-------
significantly higher than the total sulfur measured using EPA M3051A. The results for calcium
and strontium using the new procedure were also significantly higher than the results obtained
from measurement by EPA M3051A.
• Water-extractable sulfur and inorganic element concentrations were studied by repeating a water
extraction procedure four times. Very high cumulative water-extractable sulfate and calcium
were observed in the gypsum samples tested. The other detectable elements in the water extracts
were Sr, Ba, Mg, Fe, P, and Si. The average cumulative water-extractable SO4, Ca, and Sr
concentrations from the gypsum samples were 54.4±1.5%, 22.1±0.5%, and 193±211 mg kg"1,
respectively. Based on the total sulfur content by combustion, 98.2% of the water-extractable
sulfur was in the form of sulfate (SO4). The average water-extractable calcium and strontium
content in the gypsum samples was 90%, and 95%, respectively, when the cumulative water
extraction concentrations were compared to those measured using the new acid extraction
procedure.
5.2. Leaching Behavior of Drywall
• Kinetic leaching experiments were conducted for periods up to 2 months using five of the
drywall products at a fixed liquid-to-solid ratio and room temperature. The chemical
concentrations, pH, and conductivity in the leachates were measured and based on the chemical
measurements and MINTEQ modeling, the kinetics of the saturation index (SI) of the controlling
minerals were assessed. Chemical equilibrium is a dynamic process, and there is no universal
time at which chemical equilibrium is reached for all constituents in the leachate. In many cases,
a constituent concentration (e.g., calcium) was not controlled by a single mineral phase, and the
changes in leachate concentration over time were related to changing mineral phases. An
equilibrium time of one week was found appropriate for the work conducted here.
• Liquid-solid partitioning of inorganic constituents from the drywall samples was examined on
five drywalls using a modified EPA Method 1316 with ten different L/S ratios (from 2.5 to 400).
The linear dependence of logarithmic constituent concentration as a function of the logarithmic
L/S ratio was observed and found to be dependent on the saturation index (SI) of the minerals
controlling the constituent equilibrium. When the controlling mineral for a leached constituent
was in an unsaturated status (SI<0), the linear dependence was found to be valid for all the
samples studied. These relationships were further used to estimate the constituent concentrations
in the pore water. The estimated average of pore water concentrations of Sr, B, Ba, Zn, Cu, Mn,
Ni, Co, Mo, Cd, and Se were measured at 10, 11.5, 0.29, 5.6, 6.3, 4.1, 0.2, 0.2, 0.23, 0.02, and
0.6 mg L"1, respectively. The linear dependence relationship was demonstrated if the constituent
was released from a single mineral phase.
• Monolithic leaching tests were conducted using EPA Method 1315 for five different drywall
products. A linear relationship was observed between logarithmic cumulative released
constituent concentrations and logarithmic total leaching time. The slopes of the linear equation
indicated that the leaching process was not controlled by diffusion. The dominant species
(sulfate, Ca), as well as Sr, leached following a dissolution mechanism. A surface wash-off
pattern, a delayed-release pattern, and a depletion pattern were also observed for the other minor
constituents depending on the mineral source and composition of the gypsum. The diffusivity of
the leached constituents, as well as the leachability index, was further calculated. The average
weighted arithmetic mean (De) of S, Ca, Sr, Zn, Mn, Mg, P, Ba, Si, Fe and DOC from the
samples was 60.2, 55.4, 41.9, 18.1, 6.5, 5.9, 0.45, 0.38, 0.30, 0.21, and 0.02 xlO"8 cm2 s"1,
respectively. The more highly leachable constituents were Zn, Sr, SO4, and Ca; the moderately
86
-------
leachable constituents were Mg and Mn; and the relatively slow release constituents were P, Fe,
Si, Ba, and DOC. The leachability index of most constituents was between 8 and 9.
87
-------
6. References
Alhajri, I.H., Alarifi, I.M., Asadi, A., Nguyen, H.M. and Moayedi, H., 2020. A general model for
prediction of BaSC>4 and SrSC>4 solubility in aqueous electrolyte solutions over a wide range of
temperatures and pressures. Journal of Molecular Liquids, 299:112142.
Anders, O.U., 1978. Determination of the leachability of solids. Analytical Chemistry 50: 564-569.
ANS, 1986. Measurements of the Leachability of Solidified Low-Level Radioactive Wastes, ANSI/ANS
16.1 American Nuclear Society, La Grandge Park, IL,.
ASTM-International, 2016. Standard Test Methods for Chemical Analysis of Gypsum and Gypsum
Products (Metric). ASTM International.
ASTM-International, 2010. Standard Test Methods for Laboratory Determination of Water (Moisture)
Content of Soil and Rock by Mass. ASTM International.
Boulding, J.R., 1996. EPA Environmental Assessment Sourcebook. CRC Press, Taylor & Francis.
Canada, E., 1991. Proposed Evaluation Protocol for Cement-Based Solidified Wastes, Environmental
Protection Series.
CDRA, 2015. Drywall Recycling. Construction & Demolition Recycling Association, website of
CDRA, Milwaukee, WI.
Chung, F.H., 1974. Quantitative interpretation of X-ray diffraction patterns, I. Matrix-flushing method
of quantitative multicomponent analysis. Journal of Applied Crystallography 7: 513-519.
Crank, J., 1975. The Mathematics of Diffusion. Oxford University Press, London, UK.
da Silva, E.B., de Oliveira, L.M., Wilkie, A.C., Liu, Y. and Ma, L.Q., 2018. Arsenic removal from As-
hyperaccumulator Pteris vittata biomass: Coupling extraction with precipitation. Chemosphere, 193:288-
294.
De Groot G.J., 1993. Detailed description of the leaching behaviour of secondary construction products.
ECN-93-085 (in Dutch).
De Groot, G.J., van der Sloot, H.A., 1992. Determination of leaching characteristics of waste materials
leading to environmental product certification. 2nd International Symposium on Stabilization and
Solidification of Hazardous, Radioactive, and Mixed Wastes. ASTM, Philadelphia, PA, United States,
149-170.
Dermatas, D., Moon, D.H., Menounou, N., Meng, X., Hires, R., 2004. An evaluation of arsenic release
from monolithic solids using a modified semi-dynamic leaching test. Journal of Hazardous Materials
116: 25-38.
Founie, A., 2006. Gypsum. U.S. Geological Survey, Mineral Commodity Summaries,, USGS.
Freyer, D., Voigt, W., 2003. Crystallization and phase stability of CaSC>4 and CaSC>4- based salts.
Monatshefte fiir Chemie 134: 693-719.
Galan, I., Glasser, F.P., Andrade, C., 2013. Calcium carbonate decomposition. Journal of Thermal
Analysis and Calorimetry 111: 1197-1202.
Garland, S.E., Greene, M.A., 2009. Statistical Analysis of the Chemical Screening of a Small Sample of
Unused Chinese and non-Chinese Drywall. Division of Hazard Analysis, U.S. Consumer Product Safety
Commission p. 52.
88
-------
Godbee, H.W., Compere, E.L., Joy, D.S., Kibbey, A.H., Moore, J.G., Nestor Jr, C.W., Anders, O.U.,
Neilson Jr, R.M., 1980. Application of mass transport theory to the leaching of radionuclides from waste
solids. Nuclear and Chemical Waste Management 1: 29-35.
Gorski, C.A. and Fantle, M.S., 2017. Stable mineral recrystallization in low temperature aqueous
systems: A critical review. Geochimica et Cosmochimica Acta, 198:439-465.
Griffin, B.A., Jurinak, J.J., 1973. Estimation of activity coefficients from the electrical conductivity of
natural aquatic systems and soil extracts. Soil Science 116: 26-30.
G Gustafsson, J.P., 2016. Visual MINTEQ, version 3.1; KTH. SEED: Stockholm, Sweden.
Heebink, L.V., Hassett, D.J., 2005. Mercury release from FGD. Fuel 84:1372-1377.
Heiri, O., Lotter, A.F., Lemcke, G., 2001. Loss on ignition as a method for estimating organic and
carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology
25:101-110.
Hulett, G.A., 1902. The solubility of gypsum. Journal of The American Chemical Society 24: 667-679.
Jimenez Rivero, A., Sathre, R., Garcia Navarro, J., 2016. Life cycle energy and material flow
implications of gypsum plasterboard recycling in the European Union. Resources Conservation &
Recycling. 108: 171-181.
Kosson, D.S., Van Der Sloot, H.A., Sanchez, F., Garrabrants, A.C., 2002. An integrated framework for
evaluating leaching in waste management and utilization of secondary materials. Environmental
Engineering Science 19: 159-204.
Krone, C.A., Brown, D.W., Burrows, D.G., Bogar, R.G., Sin-Lam, C., Varanasi, U., 1989. A method for
analysis of butyltin species and measurement of butyltins in sediment and English sole livers from Puget
Sound. Marine Environmental Research 27: 1-18.
LECO, 2005. CS230 Carbon/Sulfur Determinator INSTRUCTION MANUAL. LECO Corporation, p.
420.
Li, Z., Demopoulos, G.P., 2005. Solubility of CaSC>4 Phases in aqueous HC1 + CaCh solutions from 283
K to 353 K. Journal of Chemical & Engineering Data 50: 1971-1982.
Li, Z., Demopoulos, G.P., 2006. Development of an improved chemical model for the estimation of
CaSC>4 solubilities in the HCl-CaCh-FhO system up to 100°C. Industrial and Engineering Chemistry
Research 45: 2914-2922.
Marshall, W.L., Jones, E.V., 1966. Second dissociation constant of sulfuric acid from 25 to 350°
evaluated from solubilities of calcium sulfate in sulfuric acid solutions. Journal of Physical Chemistry
70: 4028-4040.
Moon, D.H., Dermatas, D., 2007. Arsenic and lead release from fly ash stabilized/solidified soils under
modified semi-dynamic leaching conditions. Journal of Hazardous Materials 141: 388-394.
Musson, S.E., Xu, Q., Townsend, T.G., 2008. Measuring the gypsum content of C&D debris fines.
Waste Management 28: 2091-2096.
Olson, D.W., 2001. Gypsum. US Geological survey minerals yearbook, 35, pp.1-10..W., 2001.
GYPSUM. U.S. Department of the Interior | U.S. Geological Survey, p. 10.
Pedreno-Rojas, M.A., Flores-Colen, I., De Brito, J. and Rodriguez-Linan, C., 2019. Influence of the
heating process on the use of gypsum wastes in plasters: Mechanical, thermal and environmental
analysis. Journal of Cleaner Production, 215: pp.444-457.
89
-------
Reidy, L., Williams, R., Bussan, D., Brewer, S., Cizdziel, J.V., 2014. Elemental fingerprinting of
gypsum drywall using sector field ICP-MS and multivariate statistics. International Journal of
Environmental Analytical Chemistry, 94: 1273-1287.
Roy, W.R., Krapac, I.G., J, Chou, S.F., Griffin, R.A., 1992. Batch-type procedures for estimating soil
adsorption of chemicals U.S. Environmental Protection Agency, Washington, D.C., EPA/530/SW-
87/006F.
Shukla, J., Mohandas, Y.P., Kumar, A., 2008. Effect of pH on the solubility of CaS04 2H2O in aqueous
NaCl solutions and physicochemical solution properties at 35°C. Journal of Chemical & Engineering
Data 53: 2797-2800.
Somasundaram, S., Jeon, T.W., Kang, Y.Y., Kim, W.I.L., Jeong, S.K., Kim, Y.J., Yeon, J.M., Shin,
S.K., 2014. Characterization of wastes from construction and demolition sector. Environmental
Monitoring & Assessment. 187: 4200.
Sparks, D.L., 2013. Kinetics of Soil Chemical Processes. Academic Press.
Sun, W., Barlaz, M.A., 2015. Measurement of chemical leaching potential of sulfate from landfill
disposed sulfate containing wastes. Waste Management. 36: 191-196.
Tiruta-Barna, L., Imyim, A., Barna, R., 2004. Long-term prediction of the leaching behavior of
pollutants from solidified wastes. Advances in Environmental Research. 8: 697-711.
Townsend, T., Jambeck, J., Jang, Y., Plaza, C., Xu, Q. and Clark, C., 2004. C&D Waste Landfill in
Florida: Assessment of True Impact and Exploration of Innovative Control Techniques. Florida Center
for Solid and Hazardous Waste Management, Gainesville, FL, USA..
USEPA, 1996a. EPA Method 8315A (SW-846): Determination of Carbonyl Compounds by High
Performance Liquid Chromatography (HPLC). Test Methods for Evaluating Solid Waste:
Physical/Chemical Methods. USEPA, Washington, DC.
USEPA, 1996b. SW-846 Test Method 8260B: Volatile Organic Compounds by Gas
Chromatography/Mass Spectrometry (GC/MS). Test Methods for Evaluating Solid Waste:
Physical/Chemical Methods. USEPA, Washington, DC.
USEPA, 1998a. Characterization of building related construction and demolition debris in the United
States. Construction and Demolition Debris. USEPA.
USEPA, 1998b. EPA Method 7010 (SW-846): Graphite Furnace Atomic Absorption
Spectrophotometry. Test Methods for Evaluating Solid Waste: Physical/Chemical Methods. USEPA,
Washington, DC.
USEPA, 2006. Management Practices to Prevent and Control Hydrogen Sulfide Gas Emissions at C&D
Debris Landfills Which Dispose of Pulverized Gypsum Debris in Ohio. USEPA, REGION 5,
SUPERFUND DIVISION, USEPA.
USEPA, 2007a. EPA Method 6010C (SW-846): Inductively Coupled Plasma - Atomic Emission
Spectrometry. Test Methods for Evaluating Solid Waste: Physical/Chemical Methods. USEPA,
Washington, DC.
USEPA, 2007b. EPA Method 8082A (SW-846): Polychlorinated Biphenyls (PCBs) by Gas
Chromatography. Test Methods for Evaluating Solid Waste: Physical/Chemical Methods. USEPA,
Washington, DC.
USEPA, 2007c. SW-846 Test Method 3051 A: Microwave Assisted Acid Digestion of Sediments,
Sludges, Soils, and Oils. Test Methods for Evaluating Solid Waste: Physical/Chemical Methods.
USEPA, Washington, DC.
90
-------
USEPA, 2007d. SW-846 Test Method 3546: Microwave Extraction. Test Methods for Evaluating Solid
Waste: Physical/Chemical Methods. USEPA, Washington, DC.
USEPA, 2014a. Tolaymat, T., Carson, B. Best Management Practices to Prevent and Control Hydrogen
Sulfide and Reduced Sulfur Compound Emissions at Landfills That Dispose of Gypsum Drywall. U.S.
Environmental Protection Agency, Washington, D.C., EPA/600/R-14/039, 2015.
USEPA, 2014b. SW-846 Test Method 8270D: Semivolatile Organic Compounds by Gas
Chromatography/Mass Spectrometry. Test Methods for Evaluating Solid Waste: Physical/Chemical
Methods. USEPA, Washington, DC.
USEPA, 2014c. Coal Combustion Residual Beneficial Use Evaluation : Fly Ash Concrete and FGD
Gypsum Wallboard - EPA530-R-14-001 - Office of Solid Waste and Emergency Response - Office of
Resource Conservation and Recovery, United States Environ. Prot. Agency. (2014) 1-91.
USEPA, 2015. Drywall, WARM Version 13 USEPA.
Van Driessche, A.E.S., Stawski, T.M. and Kellermeier, M., 2019. Calcium sulfate precipitation
pathways in natural and engineering environments. Chemical Geology: 119274.
Van Praagh, M., Persson, K.M., 2008. Metal releases from a municipal solid waste incineration air
pollution control residue mixed with compost. Waste Management and Research 26: 377-388.
Venta, G., 1997. Life Cycle Analysis of Gypsum Board and Associated Finishing Products. Athena
Sustainable Materials Institute, Ottawa, Canada.
Wang, Q., Li, Y., Wang, Y., 2011. Optimizing the weight loss-on-ignition methodology to quantify
organic and carbonate carbon of sediments from diverse sources. Environmental Monitoring &
Assessment, 174: 241-257.
Wang, W., Zeng, D., Chen, Q., Yin, X., 2013. Experimental determination and modeling of gypsum and
insoluble anhydrite solubility in the system CaSO^FhSO/^FhO. Chemical Engineering Science, 101:
120-129.
Wang, W., Zeng, D., Zhou, H., Wu, X., Yin, X., 2015. Solubility Isotherms of gypsum, hemihydrate,
and anhydrite in the ternary systems CaS04 + MSO4 + H2O (M = Mn, Co, Ni, Cu, Zn) at T = 298.1 K to
373.1 K. Journal of Chemical & Engineering Data 60: 3024-3032.
Wollmann, G., Voigt, W., 2008. Solubility of gypsum in MSO4 solutions (M = Mg, Mn, Co, Ni, Cu, Zn)
at 298.15 K and 313.15 K. Journal of Chemical & Engineering Data 53: 1375-1380.
Zeng, D., Wang, W., 2011. Solubility phenomena involving CaS04 in hydrometallurgical processes
concerning heavy metals. Pure and Applied Chemistry 83: 1045-1061.
91
-------
oEPA
United States
Environmental Protection
Agency
PRESORTED
STANDARD
POSTAGE & FEES
PAID EPA
PERMIT NO. G-35
Office of Research and
Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |