2015 World of Coal Ash (WOCA) Conference in Nasvhille, TN - May 5-7, 2015

http://www.flyash.info/

Leaching from Roadways Stabilized with Fly

Ash: Data Assessment and Synthesis
Brigitte L. Brown1, Sabrina Bradshaw1, Tuncer B. Edil1, and Craig H.
Benson1
1University
of Wisconsin-Madison, Geological Engineering, Engineering Hall, 1415
Engineering Drive, Madison, WI 54706

CONFERENCE: 2015 World of Coal Ash – (www.worldofcoalash.org)

KEYWORDS: fly ash, coal combustion products, unbound roadway applications,

subgrade, base course, trace element leaching

ACKNOWLEDGEMENT

Financial support for this project has been provided by the Electric Power Research
Institute (EPRI). The Recycled Materials Resource Center (RMRC) at the University of
Wisconsin-Madison provided long-term data from projects in Wisconsin and Minnesota.
Their contributions are gratefully acknowledged.

INTRODUCTION

Approximately 37% of the electrical power used in the United States is generated by
coal-fired power plants1. Air pollution control systems installed on coal-fired power
plants collect solid byproducts of coal combustion, which are commonly referred to as
coal combustion products (CCPs). The common CCPs include fly ash, bottom ash,
boiler slag, and flue gas desulfurization (FGD) residuals. Disposing CCPs in landfills or
similar waste containment facilities is costly and land intensive, and many CCPs have
useful engineering properties. Consequently, CCPs are often used beneficially in other
products or applications, most notably as construction materials. Beneficial use of CCPs
has many positive benefits in the context of sustainability, including an annual reduction
in greenhouse gas emissions by 11 million tons, fossil fuel consumption by 17 TJ, and
water consumption by 121 GL, amounting to more than $11 billion (US) in total
economic benefit1,2,3,4,5,6.

Fly ash comprises 52% of the CCPs generated today and currently is reused at a non-
adjusted rate of 37% as reported by the Department of Energy (Fig. 1)7. Considering
ACAA extrapolations that adjust for missing data increases the reuse rate to 43% 2.
There is yet a great opportunity for increased beneficial reuse. Increasing the rate of fly
ash reuse will enhance sustainability while reducing disposal costs. Most fly ash is used
as a partial replacement for Portland cement in concrete, and most concrete
applications today include fly ash in the mixture8. Roadway applications such as
stabilization of subgrade and base course are less common uses for fly ash, and
present an opportunity for increasing reuse of CCPs. Research has shown that roadway
materials stabilized with fly ash have superior mechanical properties and
durability9,10,11,12,13,14,15, 16. Base and subgrade layers incorporating fly ash have
increased strength and stiffness, which results in roadways that last longer and need
less maintenance, reducing life cycle impacts and costs and consequently increasing
sustainability2,4,6,14,15. In many roadway cases, construction costs are lower and
construction is more expedient when CCPs are employed16. However, the perceived
risk of trace element leaching often prevents fly ash from being employed in unbound
applications (e.g., bases and subgrades).

Figure 1. Uses of CCPs as identified by EIA7.

In this study, the potential risk of contaminating ground water and surface water by
constituents leaching from fly ash used in roadways is being evaluated using field water
quality data collected from projects where fly ash has been used in roadway bases and
subgrades. Water quality data from the leachate of these projects were compared
directly to federal and state water quality standards to provide a conservative evaluation
of the risk of contaminating surface water and ground water from base and subgrade
applications using fly ash. More realistic analyses were also conducted by predicting
contaminant concentrations at points of compliance (POC) using the field data as input
to groundwater modeling software.

BACKGROUND

Fly ash is a powdery material removed from the flue gas from a coal-fired electric
generating plant that is comprised of silt-size spherical particles composed primarily of
silica and having the consistency of talcum powder18. Fly ash particles also contain
trace elements (e.g., heavy metals) that the coal contained prior to combustion 19. The
perceived risk is that trace elements will leach from fly ash when used in construction
applications exposed to precipitation or other sources of water.
Fly ash is a pozzolan that becomes cementitious when combined with water and an
activator (lime, Portland cement, or kiln dust)18, 20, 21. Some fly ashes are self-cementing
when hydrated. Fly ash is used as feed stock in production of Portland cement,
cementitious material in concrete in lieu of (or in addition to) Portland cement, mineral
filler in hot-mix asphalt, structural and embankment fill, stabilizer or solidifying agent for
soft materials, cementing and flow agent in flowable fill, and stabilizer for roadway
bases, sub-bases, and subgrades18.

AASHTO M and ASTM C 618 are often used to classify fly ash into Class C and Class F
ashes. Class C fly ashes contain at least 50% oxides, are produced form sub-
bituminous coal, and are typically brown and tan in color18, 20, 22, 23. Class F fly ashes
contain at least 70% oxides, are from bituminous and lignite coal, and are typically grey
and black in color20, 23, 24. Class F fly ash is more common than Class C fly ash
because use of sub-bituminous coal has been encouraged by the 1990 Amendments to
the Clean Air Act to help meet more stringent sulfur emission standards 5. Class C fly
ashes often are self-cementing. Class F fly ashes generally are not self-cementing.
There are also fly ashes not conforming to these specifications that are self-
cementing25.

MONITORING DATA

Lysimeter leachate water quality monitoring data from six field sites utilizing fly ash
stabilized subgrade or base layers in Minnesota and Wisconsin were obtained from the
RMRC for this study. Fig. 2 shows the location of the field sites, which included STH60,
US12-E, US12-W, Scenic Edge, Waseca, and MnROAD. Construction details for the
field sites are provided in Fig. 3. Field site characteristics are detailed in Table 1.
Figure 2. Locations of field sites were field data were available for use in this study,
which included STH60, US12-E, US12-W, Scenic Edge, Waseca, and
MnRoad26, 27, 28.

Soft soil subgrade was stabilized at the Wisconsin field sites with 10% to 12% fly ash by
weight to a depth of 300 mm. Fly ash was obtained from Columbia power plant from
Alliant Energy’s Columbia Power Station in Portage, WI. The STH60 field site is located
along a 0.1-km stretch of State Trunk Highway 60 (STH60) near Lodi, WI. The site
contained several test sections employing industrial byproducts in lieu of earthen
construction materials. The section of interest employed subgrade stabilized in place
with fly ash1717, 20. The US12 site is located along a 0.6-km section of US Highway 12
(US12) near Cambridge, Wisconsin. At the US12 field site one lysimeter was located at
the west end (US12-W) and the other at the east end (US12-E) of the site27, 28, 29. The
Scenic Edge field site is located along a 200-m stretch of residential street in the Scenic
Edge neighborhood in Cross Plains, Wisconsin9, 20, 27, 28, 29, 30.

Recycled pavement material was stabilized at the Minnesota field sites with 10% to 14%
fly ash by weight to a depth of 150 mm and 203 mm. Fly ash was obtained from Excel
Energy’s Riverside Power Plant. The MnROAD field site is located along a low-volume
loop at the Minnesota Department of Transportation (MnDOT) highway testing
laboratory located adjacent to Interstate 94 between Albertville and Monticello,
Minnesota28, 31, 32. The Waseca field site is located at the intersection at 7th Street and
7th Avenue in Waseca, Minnesota33, 27, 28. At both of these sites the RPM was
reclaimed on site and blended with fly ash using a road reclaimer.

Table 1. Characteristics of field sites.

Application

Stabilized Fly Ash

Field CCP Construction Sampling
Layer Percentage Control
Site Properties End Period
Thickness and Source

12% Class C in ASTM

Dolostone 9/14/00 to
STH60 300 mm Columbia C 618 and August 2000
Subbase 8/20/12
Plant AASHTO M 295
Stabilization

Un-
Subgrade

12% Class C in ASTM

Stabilized 11/10/05 to
US12 300 mm Columbia C 618 and October 2004
Subgrade 8/20/12
Plant AASHTO M 295
Soils
Soil

10% Class C in ASTM

Scenic 02/27/06 to
300 mm Columbia None C 618 and October 2000
Edge 03/12/10
Plant AASHTO M 295
14% Off-specification
09/11/07 to
Stabilization

MnROAD 203 mm Riverside 8 RPM Base ash (>5% carbon August 2007
06/30/12
Plant content)
RPM

10% Class C in ASTM

07/07/05 to
Waseca 150 mm Riverside 7 None C 618 and August 2004
06/20/08
Plant AASHTO M 295
A control lysimeter was also installed at some field sites where conventional
construction methods and materials were employed (i.e., no fly ash). STH60 field site
contained a control section composed of an 840-mm-thick layer of crushed dolostone
subbase on top of the subgrade to ensure adequate support for the pavement (Fig. 3).
A control lysimeter was also installed beneath the centerline of the road near the west
end, where unstabilized subgrade was used in lieu of subgrade stabilized with fly ash.
The MnROAD field site contained two control lysimeters: one control lysimeter was
installed beneath an identical roadway profile except the RPM base course was not
stabilized with fly ash and the other was installed beneath a similar profile where Class
5 crushed stone was used as base course but was not considered for this study.
Construction details for the control field sites are provided in Fig. 3.

Figure 3. Schematics of roadway profiles and roadway control profiles 9, 17, 20, 27, 28, 30.

Pan lysimeters ranging in size from 3.50 m x 4.75 m to 3.00 m x 3.00 m were installed
beneath each stabilized roadway section9, 17, 20 , 26, 27, 28, 29, 30, 31, 32, 33. Lysimeters were
lined with textured linear low density polyethylene geomembrane and overlain by a
geocomposite drainage layer comprised of a geonet sandwiched between two non-
woven geotextiles. Water collecting in the drainage layer in each lysimeter was routed
to 120-L collection tank via PVC pipe (Fig. 4).
Figure 4. Typical lysimeter detail based on designs reported in O’Donnell26 and Wen et
al.31.

DATA EVALUATION

The risk imposed by using unencapsulated fly ash in roadway construction was
evaluated through direct and indirect assessment according to the flow chart outlined in
Fig. 5. Direct assessments were made by comparing concentrations of the elements
being monitored in the lysimeters to water quality standards and concentrations
obtained from control lysimeters. Indirect assessments were made by modeling
concentrations at a point of compliance using maximum measured concentrations from
the lysimeters.
Figure 5. Flowchart depicting the process for categorizing the apparent risk posed by fly
ash use in roadway construction based on water quality (leachate) data.

Water quality limits were derived from drinking water limits and surface water limits.
Federal drinking water quality limits are referred to as maximum contaminant levels
(MCLs)34. Most states also define MCLs that are the same as, or lower than, the federal
MCLs. There are 13 elements that were assessed in this study that have MCLs. A
summary of the MCLs used in this study can be found in Table 2.
USEPA has established national recommended water quality criteria for different
surface water categories under Section 304(a)(1) of the Clean Water Act35. These non-
enforceable criteria for freshwater aquatic life34,36 were used to assess the field data in
the context of surface water quality from the perspective of federal criteria. State surface
water criteria in Minnesota and Wisconsin were also considered. Standards for Class 2a
waters were used for Minnesota37, and the acute toxicity criteria for cold waters were
used for Wisconsin38. There are 13 elements that were assessed in this study that have
surface water criteria. A summary of the surface water standards is also shown in
Table 2.

Table 2. Federal, Minnesota, and Wisconsin drinking water and surface water quality
limits. For trace elements where the surface water criterion depends on
hardness 100 mg/L CaCO3 was assumed37, 38, 39.
Trace
Ag Al As Ba Be Cd Co Cr Cu F Hg Ni Pb Sb Se Tl Zn
Element
Authority Drinking Water Limits (µg/L)

Federal - - 10 2000 4 5 - 100 1300* 4000 2 - 15* 6 50 2 -

Minnesota - - 10 2000 4 5 - 100 1300* 4000 2 100** 15* 6 50 2 -
Wisconsin - - 10 2000 4 5 - 100 1300* 4000 2 100** 15* 6 50 2 -

Authority Surface Water Standards (µg/L)

Federal 3.2 - 340 - - - - 570 - - - 470 65 - - - 120

Minnesota 2 748 360 - - 3.9 436 1737 18 - - 1418 82 90 20 64 117
Wisconsin - - 340 - - 4.4 - 1803 15 - - 469 107 - - - 120
-Denotes elements that do not have water quality limits for that authority.
*Federal MCLs have not been set for copper (Cu) and lead (Pb), but concentrations of these elements
are recommended not to exceed action levels (AL).
**Minnesota and Wisconsin drinking water limits include nickel (Ni).

DIRECT ASSESSMENT

Direct assessment of the field data consisted of comparing concentrations of each trace
element monitored in the lysimeter leachate to state and federal drinking water quality
standards and surface water quality standards indicated in Table 2 (generally referred to
henceforth as water quality limits). Comparisons were also made between
concentrations of leachate from roadway sections employing fly ash and roadway
sections constructed with conventional materials (controls).

Elements eluted from roadways constructed with fly ash (i.e. STH60, US12-W, US12-E,
Scenic Edge, MnROAD, and Waseca) were categorized as imposing “no risk” when the
concentration profiles of each trace element were entirely below the water quality limit.
Additionally, when elemental concentrations from a roadway constructed with fly ash
were not statistically different from concentrations eluted from control sections, as
determined by a paired-t test at the 5% significance level, the element was categorized
as imposing “no additional risk” relative to that imposed by a roadway constructed using
conventional materials. Elements falling into neither category required further
investigation and were evaluated via indirect analysis at the point of compliance using
WiscLEACH groundwater modeling software.

A summary of the outcomes of the direct assessment is shown in Tables 3 and 4 in

terms of field sites where fly ash was used. Most of the elements (11-13 out of 17
elements, depending on field site) fell into the no risk (Table 3) or no additional risk
(Table 4) categories. This implies that roadways employing fly ash in stabilizing
unbound materials were essentially no different for those elements, in terms of potential
impact on the environment, than roadways constructed with conventional unbound
construction materials.

Table 3. Elements for which CCP use in roadway applications posed no risk.
Category No Risk

Water Drinking Water Surface Water

Quality Limit Federal State Federal State
STH60 Ba, Be, Cr, Cu Ba, Be, Cr, Cu Al, As, Cr, Pb As, Cr, Pb
US12-W Ba, Be, Cd, Cr, Cu Ba, Be, Cd, Cr, Cu As, Cr, Ni As, Cd, Cr, Ni, Pb

US12-E Ba, Be, Cd, Cr, Cu Ba, Be, Cd, Cr, Cu Al, As, Cr, Ni, Pb As, Cd, Cr, Ni, Pb

Ba, Be, Cd, Cr, Cu, Ba, Be, Cd, Cr, Cu,
Scenic Edge Ag, Al, As, Cr, Ni, Pb As, Cd, Cr, Ni, Pb
Pb, Se, Tl Pb, Se, Tl
Al, Ag, Co, Cr, Cu, Ni,
MnROAD Ba, Cd, Cu, Pb Ba, Cd, Cu, Pb Al, Ag, Cr, Ni, Pb
Pb
Ba, Be, Cd, Cr, Cu, Ba, Be, Cd, Cr, Cu, Ag, As, Cd, Co, Cr, Cu,
Waseca Ag, As, Cr, Hg, Ni, Zn
Hg, Se Hg, Ni, Se Ni, Sb, Se, Tl, Zn

Table 4. Elements for which CCP use in roadway applications posed no additional risk relative
to controls. Scenic Edge and Waseca field sites did not contain control lysimeters.
Category No Additional Risk
Water Drinking Water Surface Water
Quality
Limit Federal State Federal State

STH60 As, Cu, Ni, Pb, Sb, Zn As, Cu, Ni, Pb, Sb, Zn As, Cu, Ni, Pb, Sb, Zn As, Cu, Ni, Pb, Sb, Zn

US12-W As, Be, Cd, Pb, Sb, Tl As, Be, Cd, Pb, Sb, Tl As, Be, Cd, Pb, Sb, Tl As, Be, Cd, Pb, Sb, Tl

US12-E Ag, Ba, Be, Cu, Tl Ag, Ba, Be, Cu, Tl Ag, Ba, Be, Cu, Tl Ag, Ba, Be, Cu, Tl
Scenic
No Control No Control No Control No Control
Edge
As, Cd, Cr, F, Fe, Mn, As, Cd, Cr, F, Fe, Mn, As, Cd, Cr, F, Fe, Mn, As, Cd, Cr, F, Fe, Mn,
MnROAD
Sb, Sn, Tl, V, Zn Sb, Sn, Tl, V, Zn Sb, Sn, Tl, V, Zn Sb, Sn, Tl, V, Zn
Waseca No Control No Control No Control No Control

INDIRECT ASSESSMENT

The direct assessment provides a conservative assessment of risk, as the water sample
collected at a lysimeter directly beneath a roadway profile is not available for human
consumption or for contact with biota in a surface water body. Dilution and attenuation
between the monitoring point and a receptor will substantially reduce concentrations
and risk imposed by using fly ash in roadway construction.

Direct assessment indicated that concentrations were not below water quality limits or
were statistically different from the control for 4-6 of 17 trace elements. Elements
requiring further investigation varied per field site, but overall included Ag, Al, As, Be,
Cd, Cu, Ni, Pb, Sb, Se, Tl, and Zn. An indirect analysis was conducted on these
elements to provide a representative assessment of concentrations that would be
realized at a receptor point in ground water. The regulatory point of compliance (POC)
for water quality for many roadway applications is the right-of-way of the roadway40, and
was used as the modeled receptor point.

The analysis was conducted using WiscLEACH software, which was developed
specifically for evaluating the potential for impacts to ground water by industrial
byproducts incorporated into a roadway33, 40. WiscLEACH follows the advective-
dispersive-reaction-equation (ADRE) in one dimension through the vadose zone and in
two dimensions through the saturated zone. The 2D column leach test simulation for
adsorption controlled release was used in WiscLEACH19, 30, 41. Input concentration, CO,
is applied evenly throughout the stabilized layer within the software, and was
conservatively taken as the maximum concentration documented for each element at
each field site.

Maximum concentration at POC, or CPOC, was output. The continuous injection type-2
boundary modeled at the boundary beneath the stabilized layer implies that a sustained
mass was leached, which is a conservative assumption that allowed the maximum CPOC
to be obtained from the model. The breakthrough curve for a typical field site in
Minnesota or Wisconsin was established to determine when the CPOC was reached. It
was found that regardless of trace element modeled, CPOC was reached by 5 years,
which is within the lifetime of a road (typically 20 to 40 years)2, 42. Values of CPOC for
each element were evaluated as above or below the drinking water or surface water
standard in a manner similar to the direct assessment.

The POC was taken as the right-of-way of the roadways (20 m), which is the regulatory
POC for water quality for many roadway applications40, defined from the center-line of
the road to the edge of the right-of-way. Scaling and retardation factors were
conservatively assumed to be one, i.e., no retardation. Published molecular diffusion
coefficients were input for each trace element or a low (conservative) molecular
diffusion coefficient of 0.005 m2/yr. was assumed for elements that had no published
values (As, Sn, Ti, and V)43. Dispersivities were taken as one-tenth the domain and
recommended grid parameters from Li et al. 2006 were used40. A summary of site-
specific WiscLEACH model inputs is provided in Table 5.
Table 5. Site-specific WiscLEACH model inputs.
Distance to
Point of Distance to
Width of Width of Top of
Field Sites Compliance Groundwater
Pavement (m) Shoulder (m) Stabilized
(POC) (m) Table (m)
Layer (m)
STH60 FA 20 10.4 1.5 >2.03 0.38
US12-W 20 10.4 1.5 1.52 0.457
US12-E 20 10.4 1.5 >2.03 0.457
Scenic Edge 20 10.4 1.5 >2.03 0.215
Waseca 20 10.4 1.5 1.09 0.075
MnROAD 20 10.4 1.5 1.09 0.102
Distance to
Hydraulic Regional
Bottom of Infiltration Porosity of
Field Sites Conductivity of Hydraulic
Stabilized Rate (m/yr.) Aquifer
Aquifer (m/yr.) Gradient
Layer (m)
STH60 FA 0.68 0.866 3156 0.3 0.001
US12-W 0.757 0.845 3156 0.3 0.001
US12-E 0.757 0.845 3156 0.3 0.001
Scenic Edge 0.515 0.839 3156 0.3 0.001
Waseca 0.225 0.871 3156 0.3 0.001
MnROAD 0.305 0.764 3156 0.3 0.001
Hydraulic Hydraulic Hydraulic
Hydraulic
Conductivity of Conductivity of Conductivity of Porosity of
Field Sites Conductivity of
Pavement Stabilized Subgrade Pavement
Base (m/yr.)
(m/yr.) Layer (m/yr.) (m/yr.)
STH60 FA 1.0 1.0 0.19 133.5 0.33
US12 WW 1.0 1.0 0.19 126.9 0.33
US12 EW 1.0 1.0 0.19 126.9 0.33
Scenic Edge 1.0 1.0 0.19 133.5 0.33
Waseca 1.0 1.0 757.4 126.9 0.33
MnROAD 1.0 1.0 757.4 133.5 0.33
Porosity of
Porosity of Porosity of Horizontal Vertical
Field Sites Stabilized
Base Subgrade Dispersion Dispersion
Layer

STH60 FA 0.33 0.41 0.16 2.0 0.20

US12 WW 0.33 0.27 0.10 2.0 0.15
US12 EW 0.33 0.27 0.10 2.0 0.20
Scenic Edge 0.33 0.39 0.17 2.0 0.20
Waseca 0.33 0.39 0.19 2.0 0.11
MnROAD 0.33 0.25 0.10 2.0 0.11
All 12 elements modeled (Ag, Al, As, Be, Cd, Cu, Ni, Pb, Sb, Se, Tl, Zn) with
WiscLEACH groundwater modeling software had CPOC below water quality limits, thus
were categorized as imposing “no predicted risk” (see Table 6). Trace elemental
concentrations had a minimum 97.5% concentration reduction, with an average
reduction of 98.4%. The direct assessment showed that fly ash used in base and
subgrade layers posed no predicted risk at the modeled receptor points and thus are
suitable beneficial reuse applications with respect to water quality.

Table 6. Elements for which fly ash use in roadway applications posed no predicted risk at
POC.
Category No Predicted Risk at Point of Compliance (POC)

Water Drinking Water Surface Water

Quality Limit Federal State Federal State
STH60 Cd, Se, Tl Cd, Se, Tl Ag Cd
US12-W Se Ni, Se Ag, Al, Zn Cu, Zn
US12-E As, Pb, Se As, Ni, Pb, Se Pb, Zn Zn
Scenic Edge As, Sb As, Sb Zn Cu, Zn
MnROAD Be, Se Be, Se Ag Ag, Se
Waseca As, Pb, Sb, Tl As, Pb, Sb, Tl Pb Pb

In order to determine if the CPOC was sensitive to the distance to the POC, the
concentration as a function of distance from the base of the fly ash layer was modeled
in WiscLEACH and is shown in Fig. 6. Concentration decreased by 61% in the vadose
zone directly underneath the pavement section and 96% as it reached the saturated
zone. Concentration reduced an additional 0.1% from the vadose zone to the POC,
demonstrating that most concentration reduction occurred at the vadose-saturated zone
boundary underneath the pavement section. This implies that the distance to the POC
from the pavement layer is not influential. However, as implied by Li et al. 33, 40 the
distance to the groundwater table would impact the distance the element travels through
the vadose zone, potentially affecting the magnitude of concentration reduction.
 C = 165 µg/L
O
70

GWT
Vadose Zone Saturated Zone
1D Flow 2D Flow
60

Drinking Water Limit = 50 µg/L

50
Maximum Concentration (µg/L)

40

30 Do = 0.066 (MAX)
Do = 0.005 (MIN)

20

10
C
POC

0
0 5 10 15 20 25 30 35

Flow Distance (m)

Figure 6. Concentration vs. flow distance for a typical field site in Minnesota and
Wisconsin for maximum and minimum molecular diffusion coefficients.
Concentration decreases in the vadose zone and drops dramatically after the
groundwater table. This example is for selenium and the drinking water limit
for selenium is market.

Using the WiscLEACH model, reduction factors (RF) were calculated by dividing CO by
CPOC to give a general sense of the amount of reduction that took place and to facilitate
the discussion and ranking of field sites. RFs spanned 45.7-87.9, with increasing RF
indicating that more reduction took place during water migration to POC.
Percentage of reduction that took place at field sites spanned 97.5% to 98.7%, the small
range of which does not facilitate discussion as well as the range of RF. Additionally,
RF allow for the quick estimation of CPOC given leachate characteristic data (e.g. CO) or
the back-calculation of an allowable CO given maximum allowable CPOC (e.g. water
quality limits).
RF represent the overall concentration reduction that occurred in field conditions and
are a function of the input parameters that influence CPOC, of which were most
influentially depth to groundwater table (ZGWT), stabilized layer thickness, width of
stabilized layer, hydraulic conductivity of the stabilized layer and aquifer, dispersivity,
and CO. ZGWT was the most sensitive parameter because it correlates to the length of
the flow path and thus the amount of dispersion that took place40. Thickness of the
stabilized layer was the second because it affected the total mass leached: higher
concentrations occurred at the POC when more mass was leached from the fly ash
stabilized layer40. Hydraulic conductivity of the stabilized layer typically controlled
seepage velocity in the vadose zone, and hydraulic conductivity of the aquifer controlled
seepage velocity in the saturated zone33, 40. Values of CO and CPOC varied greatly
between elements and between field sites, but RF were constant for each trace element
at a field site regardless of Co input into WiscLEACH because all other factors remained
constant and the ADRE used in WiscLEACH was linear.

The RFs for each trace element assessed at a given field site (points) and the average
RF per field site (lines) are shown in Fig. 7. The depth to groundwater table and
stabilized layer thickness are included for context in Fig. 7. Because all other
parameters were the same, the only influence on elemental RF per site was molecular
diffusion which differs between elements. Small margins of error (95% confidence
interval) suggest that the average of the RF for all elements at a given field site can
accurately represent all elements at that field site.

Waseca is the only field site that did not demonstrate a downward trend in RF with
decreasing depth to groundwater table (Fig. 7). The other field sites appear to be
controlled by depth to groundwater table. The average RF is higher at Waseca
because it had a thin (0.15 m) stabilized layer thickness compared to the other field
sites, which suggests Waseca was instead controlled by thickness of stabilized layer.
 100

80
POC

60
RF = C / C
O

40
Increasing Depth to Groundwater Table (m)
1.09 1.09 1.52 2.03 2.03 2.03

20
Generally Increasing Stabilized Layer Thickness (m)

0.20 0.15 0.30 0.30 0.30 0.30

0
MnROAD Waseca US12-W US12-E Scenic Edge STH60

Figure 7. RFs for elements at the field sites (points) and the average RF per field site
(line). Depth to groundwater table and stabilized layer thickness are included
for context.

RF statistics are presented in Table 7. RF ranged between 45.7 (MnROAD) and 87.9
(STH60). Overall, STH60 had the highest RFs due primarily to a relatively thicker
stabilized layer and deeper ZGWT while MnROAD had the lowest reducing power due to
a thinner stabilized layer and shallower ZGWT33, 40.

Using water quality limits as CPOC instead of the predicted CPOC to solve for RF
effectually back-calculated the minimum required RF that a field site would need to
obtain a CPOC below water quality limits. Even though STH60 had the maximum
average RF (77.2), a RF as low as 37.5 would still reduce Co to a CPOC below the water
quality limits. Similarly, RF could be as low as 31.2 at Scenic Edge, 3.5 at US12-E, 6.4
at US12-W, 28.0 at Waseca, and 19.6 at MnROAD. RF from one of the field sites could
potentially be used to quickly estimate CPOC given leachate characteristic data (CO) of a
similarly constructed and situated field site. RF could also potentially be used to back-
calculate an allowable CO given maximum allowable CPOC, such as water quality limits.

Table 7. Reduction factor (RF) statistics.

Minimum Maximum Standard 95% Margin Percent
Site Average
Value Value Deviation of Error Reduction

STH60 67.06 87.88 77.2 5.054 ±2.07 98.7%

US12-W 48.75 59.02 52.4 2.025 ±1.18 98.1%

US12-E 60.48 70.35 64.2 1.926 ±1.07 98.4%

Scenic Edge 71.70 85.79 76.8 2.718 ±1.41 98.7%

Waseca 59.62 67.03 62.3 2.092 ±0.89 98.4%

MnRD 45.74 51.44 47.9 1.156 ±0.64 97.9%

SUMMARY AND CONCLUSIONS

In this study, the potential risk of contaminating ground water and surface water by
constituents leaching from fly ash used in roadways was evaluated using field water
quality data collected from projects where fly ash has been used in roadway bases and
subgrades. Six field sites in Minnesota (MnROAD and Waseca) and Wisconsin (STH60,
US12-W, US12-E, and Scenic Edge) were evaluated, which consisted of fly ash
stabilizing recycled asphalt pavement base and soft soil subgrades, respectively.

The assessment evaluated field water quality data from leachate collected from field
sites where fly ash had been used in non-pavement applications in roadway bases and
subgrades. Water quality data were compared directly to federal and state water quality
standards to provide a conservative evaluation of the risk of contaminating surface
water and ground water from base and subgrade applications using fly ash. Elements
were categorized as imposing “no risk” when the concentration profiles of each trace
element were entirely below the water quality limit. Additionally, when elemental
concentrations from a roadway constructed with fly ash were not statistically different
from concentrations eluted from control sections, as determined by a paired-t test at the
5% significance level, the element was categorized as imposing “no additional risk”
relative to that imposed by a roadway constructed using conventional materials.

Indirect analyses were conducted to more realistically assess field conditions at a point
of compliance (POC) using the field data as input to WiscLEACH groundwater modeling
software (CO). When the modeled concentrations at the POC (CPOC) were below water
quality limits, the element was categorized as imposing “no predicted risk.” Reduction
factors (RF) were calculated by dividing the initial concentration by the predicted
concentration at the point of compliance to give a general sense of the amount of
reduction that took place and to facilitate the discussion and ranking of field sites.
Additionally, RF could allow for the quick estimation of CPOC given leachate
characteristic data (e.g. CO) or the back-calculation of an allowable CO given maximum
allowable CPOC (e.g. water quality limits).

Findings include:

1. The direct assessment demonstrated that concentrations at the base of the fly
ash stabilized layer of 11-13 of 17 trace elements were either below water quality
limits or were not statistically different from control field sites and were
consequently categorized as imposing “no risk” or “no additional risk.”
Essentially, the six roadways employing fly ash stabilization of unbound materials
were no different, in terms of potential impact on the environment for those
elements, than roadways constructed with conventional materials.

2. The direct assessment demonstrated that 4-6 of 17 trace elements exceeded

water quality limits per field site at the base of the fly ash stabilized layer but
subsequent indirect assessment via groundwater modeling determined that CPOC
were predicted to be below water quality limits and were characterized as “no
predicted risk”. Elements indirectly assessed varied per field site and included
Ag, Al, As, Be, Cd, Cu, Ni, Pb, Sb, Se, Tl, and Zn.

3. For all field sites, the initial concentration at the base of the fly ash stabilized
layer (CO) was reduced by an average of 98.4% at the point of compliance. Most
(96%) of the concentration reduction occurs just after vadose-saturated zone
boundary underneath the pavement section. This implies that a point of
compliance adjacent to the edge of the pavement would result in very little
difference in CPOC.

4. Reduction factors (RF) for all sites ranged from 45.7-87.9 and the average of all
trace elements was found to represent field sites. RF were controlled primarily
by depth to groundwater table and secondarily by thickness of stabilized layer.
More reduction occurred in field sites with a relatively thicker stabilized layer,
deeper ZGWT, lower seepage velocity in the vadose zone, and higher seepage
velocity in the saturated zone.

5. The RF actually required to decrease CO to water quality levels at POC were well
below the RF that field sites exhibited in this study. This implies that less
reduction could have taken place at the field sites and still imposed “no predicted
risk.”

Overall, this study demonstrated that there are no additional risks imposed by using
unbound fly ash in roadways than roadways constructed with conventional construction
materials, in terms of potential trace element impact on the environment. The unbound
fly ash applications described in this study have been concluded to be suitable
beneficial reuse applications with respect to water quality. Conclusions and RF can be
applied to similar field sites, however roadways with different field conditions than
evaluated in this study (esp. thicker stabilized layer or groundwater that is closer to the
base of the pavement structure) should be evaluated using the analytical procedure
provided herein to ensure eluted trace element concentrations meet the water quality
standards at a point of compliance.

REFERENCES

[1] U.S. Energy Information Administration (EIA) (2013). AEO2014 Early Release
Overview. EIA

[2] Carpenter, A., Gardner, K., Fopiano, J., Benson, C., and Edil, T. (2007), Life cycle
based risk assessment of recycled materials in roadway construction. Waste
management (New York, N.Y.), 27(10), 1458–1464.
doi:10.1016/j.wasman.2007.03.007.

[3] Environmental Protection Agency (2014, June 10). Beneficial Use. Retrieved
September 12, 2014, from http://www.epa.gov/osw/conserve/imr/ccps/benfuse.htm

[4] Lee, J. C., Edil, T. B., Tinjum, J. M., & Benson, C. H. (2010). Quantitative
assessment of environmental and economic benefits of using recycled construction
materials in highway construction. Transportation Research Board, 2158(-1), 1–11. doi:
10.3141/2158-17.

[5] Lee, J. C., Edil, T. B., Benson, C. H., & Tinjum, J. M. (2011). Evaluation of Variables
Affecting Sustainable Highway Design With BE2ST-in-Highways System.
Transportation Research Record: Journal of the Transportation Research Board, 2233(-
1), 178–186. doi: 10.3141/2233-21.

[6] Lee, J., Edil, T. B., Asce, D. M., Benson, C. H., Asce, F., Tinjum, J. M., & Asce, M.
(2013). Building Environmentally and Economically Sustainable Transportation
Infrastructure : Green Highway Rating System, 1–10. doi: 10.1061/ (ASCE) CO.1943-
7862.0000742.

[7] U.S. Department of Energy, the U.S. Energy Information Administration (2012).
Annual Environmental Information, Schedule 8. Part A. Annual Byproduct Disposition,
2013 Final Release. [Data set]. Retrieved from
http://www.eia.gov/electricity/data/eia923/>

[8] American Coal Ash Association. (2012). 2012 Coal Combustion Product (CCP)
Production & Use Survey Report. Farmington Hills: American Coal Ash Association
(ACAA), 2012. Web. 1 Jan. 2015. CCP Production & Use Survey Report.

[9] Bin-Shafique, S., Edil, T., and Benson, C. (2004), Incorporating a Fly Ash Stabilized
Layer into Pavement Design: Case Study, Geotechnical Engineering, 157(4), 239-249.
[10] Camargo, F., Edil, T., and Benson, C. (2013), Strength and Stiffness of Recycled
Pavement Materials Stabilized with Fly Ash: A Laboratory Study, Road Materials and
Pavement Design, 14(2) 1-13.

[11] Ebrahimi, A., Kootstra, B., Edil, T., and Benson, C. (2012), Practical Approach for
Designing Flexible Pavements Using Recycled Roadway Materials as Base Course,
Road Materials and Pavement Design, 13(3), 1-18.

[12] Edil, T. B., Asce, M., Acosta, H. A., & Benson, C. H. (2006). Stabilizing Soft Fine-
Grained Soils with Fly Ash, (April), 283–294.

[13] Ferguson, G., (1993). “Use of Self-Cementing Fly Ash as a Soil Stabilizing Agent,
“Geotechnical Special Publication No. 36, ASCE, New York, N.Y.

[14] Li, L., Benson, C., Edil, T., and Hatipoglu, B. (2008), Sustainable Construction Case
History: Fly Ash Stabilization of Recycled Asphalt Pavement Material, Geotechnical
and Geological Engineering, 26, 177-187.

[15] Senol, A., Edil, T., Bin-Shafique, S., Acosta, H., and Benson, C. (2006), Soft
Subgrade Stabilization Using Fly Ashes, Resources, Conservation and Recycling,
46(4), 365-376.

[16] Tastan, E. O., Edil, T. B., Benson, C. H., & Aydilek, A. H. (2011). Stabilization of
Organic Soils with Fly Ash. Journal of Geotechnical and Geoenvironmental Engineering,
137(9), 819–833. doi: 10.1061/ (ASCE) GT.1943-5606.0000502.

[17] Edil, T. B., Benson, C. H., Bin-Shafique, M. S., Tanyu, B. F., Kim, W., and Senol, A.
(2002). “Field Evaluation of Construction Alternatives for Roadway over Soft Subgrade,”
Transportation Research Record, 1786 pp. 36-48.

[18] Environmental Protection Agency. (2014, February 7). About Coal Combustion
Products | Coal Combustion Products | US EPA. Retrieved September 12, 2014, from
http://www.epa.gov/osw/conserve/imr/ccps/index.htm

[19] Yan, R., Gauthier, D., and Flamant, G. (2001). “Volatility and Chemistry of Trace
Metals in a Coal Combustor”, Fuel, Vol. 80 (15), pp. 2217-2226.

[20] Bin-Shafique, S., Benson, C. H., & Edil, T. B. (2002). Leaching of Heavy Metals
from Fly Ash Stabilized Soils Used in Highway Pavements. Geo Engineering Report,
University of Wisconsin-Madison, 02-14.

[21] FHWA. Fly Ash Facts for Highway Engineers. Federal Highway Administration,
US Department of Transportation, 1995, Report No. FHWA-SA-94-081.
[22] American Coal Ash Association. (2003). Fly Ash Facts for Highway Engineers.
Report No. FHWA-IF-03-019. Retrieved September 5, 2014, from
http://www.fhwa.dot.gov/Pavement/recycling/fafacts.pdf

[23] ASTM C 618. Standard Specification for Coal Fly Ash and Raw or Calcined Natural
Pozzolan for Use as a Mineral Admixture in Concrete (C 618-03), American Society for
Testing and Materials, West Conshohocken, PA, USA, 2003.

[24] Chu, S. C. and Kao, H. S. (1993). “A Study of Engineering Properties of a Clay

Modified by Fly Ash and Slag,” Fly Ash for Soil Improvement-Geotechnical Special
Publication, Vol. 36, pp. 89-99.

[25] ASTM D 5239. Standard Practice for Characterizing Fly Ash for Use in Soil
Stabilization. American Society for Testing and Materials, West Conshohocken, PA,
USA, 2003.

[26] O’Donnell, Jonathan. (2009). Leaching of Trace Elements from Roadway Materials
Stabilized with Fly Ash (Master's thesis). Retrieved from UW-Madison database.

[27] O’Donnell, J., Benson, C., Edil, T., and Bradshaw, S. (2010), Leaching of Trace
Elements from Pavement Materials Stabilized with Fly Ash, Proc. Green Streets and
Highways 2010, ASCE, Reston, VA, 272-279.

[28] O’Donnell, J., Benson, C., & Edil, T. (2010). “Trace Element Leaching from
Pavements with Fly Ash-Stabilized Base and Subgrades: Experience in the Midwestern
United States,” Proc. 2nd International Conference on Sustainable Construction
Materials and Technologies, Vol. 3, 2010, Università Politecnica delle Marche, Ancona,
Italy, pp/161-168.

[29] Li, L., Edil, T. B., & Benson, C. H. (2009). Mechanical Performance of Pavement
Geomaterials Stabilized with Fly Ash in Field Applications. Coal Combustion and
Gasification Products, 1(1), 43–49. doi:10.4177/CCGP-D-09-00018.1.

[30] Bin-Shafique, S., Benson, C., Edil, T., & Hwang, K. (2006). Leachate
Concentrations from Water Leach and Column Leach Tests on Fly Ash-Stabilized Soils.
Environmental Engineering Science, 23(1).

[31] Wen, H., Baugh, J., Edil, T., & Wang, J. (2011a). Cementitious High-Carbon Fly
Ash Used to Stabilize Recycled Pavement Materials as Course. Transportation
Research Record: Journal of the Transportation Research Board, (No. 2204), 110-113.

[32] Wen, H., Li, X., Edil, T., O’Donnell, J. (2011b). STTR: Utilize Cementitious High
Carbon Fly Ash (CHCFA) to Stabilize Cold In-Place Recycled (CIR) Asphalt Pavement
as Base Course. Phase II Report to the Department of Energy. Award Number: DE-
FG02-05EG86238.
[33] Li, L., Benson, C., Edil, T., and Hatipoglu, B. (2007), Groundwater Impacts from
Coal Ash in Highways, Waste and Resource Management, 159(4), 151-162.

[34] Environmental Protection Agency (2014, May 28). Water Quality Criteria. Retrieved
August 5, 2014, from http://water.epa.gov/scitech/swguidance/standards/criteria/

[35] Clean Water Act (CWA) and Federal Facilities. (2014, July 30). Clean Water Act
(CWA) and Federal Facilities. Retrieved August 5, 2014, from
http://www2.epa.gov/enforcement/clean-water-act-cwa-and-federal-facilities

[36] National Recommended Water Quality Criteria. (2012, May 1). Retrieved August 5,
2014, from http://water.epa.gov/scitech/swguidance/standards/current/index.cfm

[37] The Office of the Reviser of Statutes. (2012, January 24). Retrieved July 02, 2014,
from Minnesota Administrative Rules: https://www.revisor.mn.gov/rules/?id=7050.0222

[38] Wisconsin State Legislature. (2014, July 01). Water Quality Standards for
Wisconsin Surface Waters. Retrieved June 20, 2014, from Chapter NR 102:
http://docs.legis.wisconsin.gov/code/admin_code/nr/100/102

[39] United States Geological Survey. (2013, September 21). Water Hardness and
Alkalinity. Retrieved July 03, 2014, from http://water.usgs.gov/owq/hardness-
alkalinity.html

[40] Li, L., Benson, C., Edil, T., and Hatipoglu, B. (2006), WiscLEACH: A Model for
Predicting Ground Water Impacts from Fly-Ash Stabilized Layers in Roadways,
Geotechnical Engineering in the Information Technology Age, D. DeGroot, J. DeJong, J.
Frost, and L. Baise, eds., ASCE.

[41] Edil, T. B., Sandstorm, L. K. and Berthouex, P. M., (1992). “Interaction of Inorganic
Leachate with Compacted Pozzolanic Fly Ash,” J. of Geotech. Engrg., ASCE, New
York, Vol.118 (9), pp. 1410-1430.

[42] Transportation of the United States, 2006. National Atlas of the United States.
Retrieved April 4, 2006 from http://nationalatlas.gov/transportation.

[43] Haynes, W. M. ed., CRC Handbook of Chemistry and Physics, 95th Edition
(Internet Version 2015), CRC Press/Taylor and Francis, Boca Raton, FL. If a specific
table is cited, use the format: "Physical Constants of Organic Compounds," in CRC
Handbook of Chemistry and Physics, 95th Edition (Internet Version 2015), W. M.
Haynes, ed., CRC Press/Taylor and Francis, Boca Raton, FL.