Environmental Science and Pollution Research

https://doi.org/10.1007/s11356-018-4027-2

RESEARCH ARTICLE

Gas thermal remediation of an organic contaminated site: field trial

Jian Xu 1,2 & Fei Wang 3 & Cheng Sun 2 & Xueliang Zhang 4 & Yunhui Zhang 5

Received: 3 September 2018 / Accepted: 17 December 2018

# Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
This paper investigated the performance of in situ gas thermal remediation (GTR) technique for the remediation of organic
contaminated soil and groundwater. The pollutants of the site are mainly benzene, chlorobenzene, and petroleum. To evaluate the
effectiveness of GTR, 100-m2 site soil with a depth of up to 18 m were installed with heating wells and extraction wells. The
temperature in the GTR system was 100~200 °C, and the pressure was 0.5~2.0 kPa. It ran for 34 days. Later, soil and
groundwater were sampled 7 days after treatment. The results show that the temperatures of the central remediation area and
the heating well surrounding area were higher than that of the edge position. The average temperature of the surface soil was the
lowest. The highest temperatures were found at depths of − 3 m and − 6 m. In addition, the removal rates of benzene, chloro-
benzene, and petroleum in soil were up to 99.81%, 99.72%, and 98.23%, respectively. Meanwhile, those for groundwater were
up to 98.77%, 97.70%, and 99.99%, respectively. The removal rate of pollutants in soil is affected by temperature and the nature
of the soil. Although many GTR projects have been applied in the USA, this project is the first engineering application project of
GTR technology in China. The results present the good in situ organic removal efficiency of GTR system in this study and
therefore provide a guidance and technical support for similar projects in China.

Keywords Gas thermal remediation . Field trial . Organic pollutants . Groundwater

Introduction
With the rapid development of China’s economy and society
and the implementation of Bcompanies should retreat into in-
dustry parks^ policy, a large number of enterprises were
forced to move, transform, or shut down. In this case, many
contaminated sites are left behind in cities and their surround-
ing areas, which pose a great risk to the environment and
restrict the construction and development of cities (Yang
et al. 2014b). Most sites have been designated as highly pol-
luted and high-risk. The pollutants in these site soils are com-
plex and varied, even accompanied by groundwater pollution

A paper submitted for possible publication as technical reports to

Environmental Science and Pollution Research.
Responsible editor: Zhihong Xu

* Fei Wang 1
Nanjing Institute of Environmental Sciences, Ministry of
101012020@seu.edu.cn Environmental Protection, State Environmental Protection Key
Laboratory of Soil Environmental Management and Pollution
Jian Xu Control, Nanjing 210042, People’s Republic of China
xujian@nies.org 2
State Key Laboratory of Pollution Control and Resource Reuse,
School of the Environment, Nanjing University, Nanjing 210023,
Cheng Sun People’s Republic of China
envidean@nju.edu.cn 3
Institute of Geotechnical Engineering, School of Transportation,
Southeast University, Nanjing 210096, People’s Republic of China
Xueliang Zhang 4
2016103039@njau.edu.cn College of Resources and Environmental Sciences, Nanjing
Agricultural University, Nanjing 210095, People’s Republic of China
5
Yunhui Zhang Department of Engineering, University of Cambridge,
yz485@cam.ac.uk Cambridge CB2 1PZ, UK

(Geng et al. 2010; Chen 2010a, b). It is vital and urgent to
remediate them with suitable techniques so as to protect hu-
man health and the eco-environment.
Many remediation technologies have been applied to treat
organic contaminated soil during past years. Thermal remedi-
ation technique is one of the most efficient and low-risk
methods to remove organic pollutants and some inorganic
pollutants (Falciglia et al. 2011; Bonnard et al. 2010; Yang
et al. 2014a). The organic pollutants are heated up by direct
or indirect heat exchange in the soil, and they are separated
from the soil when a sufficiently high temperature reaches
(O’Brien et al. 2018). It is especially suitable for soils contam-
inated petroleum, chlorinated solvents, volatile organic com-
pounds, and semi-volatile organic compounds (USEPA
2004). Recently, there is a strong sustainable driver from gov-
ernments for the application of in situ treatment (Wang et al.
2015). The in situ gas thermal remediation (GTR) technique in
this study has been widely applied to the remediation of high
concentration organic contaminated sites in American and
European countries (USEPA 2006; Aresta et al. 2008;
Biache et al. 2008; Sakaguchi et al. 2015). It offers many
advantages including great efficiency, relatively short time
frame, advanced equipment, avoidance of second pollutions,
and its suitability for removing a wide range of contaminants
(Kawala and Atamanczuk 1998; Chien 2012).
The decontamination processes have been recognized to be
affected by soil temperature, treatment time, and heating rate
(Bonnard et al. 2010; Saito et al. 1998; Pina et al. 2002;
Mechati et al. 2004). Mechati et al. (2004) studied the influence
of temperature, flow rate, and pressure on the decontamination
and found that the operating temperature was the determining
parameter followed by the gas flow rate. Pina et al. (2002)
reported that the quantity and composition of volatile
compounds generated during thermal treatment were strongly
influenced by soil chemical composition and treatment
temperature. Bucala et al. (1994) heated the fuel oil artificially
contaminated soil with high temperatures and rapid heating
rates, and found out the effect of the heating rate, final temper-
ature, and degree of contamination on the removal efficiency.
In spite of significant advances in understanding the effect of
temperature on soil thermal decontamination, most studies
were based on laboratory studies and almost no field trials of
GTR have been carried out to our best knowledge. It is mean-
ingful to investigate the heat distribution in situ due to the fact
that, in practice, heat transfer and removal mechanisms for in
situ GTR system vary at different depths and spatially depend
on the proximity to heat wells.
Limited studies have been conducted on soil characteris-
tics, another key factor of thermal remedial processes.
Falciglia et al. (2011) investigated the influence of tempera-
ture and soil texture on diesel removal kinetics, and their re-
sults claimed that a temperature of 175 °C is sufficient to
remedy diesel polluted sandy and silty soils, while 250 °C is
Environ Sci Pollut Res
needed for clays. However, Baker and Kuhlman (2002) found
that the thermal conductivity is not affected significantly by
soil types, and the heterogeneity in soil structure or
contaminant dispersal was found to have minimal effect.
Chien (2012) reported that the moisture of soil may play an
important role in the distribution of heat and soil characteris-
tics are closely linked to the heat transfer. Therefore, there is a
need to find out the dormitory soil characteristics and explore
the relationships among the nature of the soil, the heat trans-
port, and the removal mechanisms.
The aim of this study is to enable a better understanding of
the performance of GTR system and its effectiveness in re-
moving organics from site soil and groundwater. The specific
objectives of this trial are (1) to study the heat distribution in
both the vertical and the horizontal directions, (2) to investi-
gate the pollutants removal efficiency of GTR system at dif-
ferent depths, (3) to assess the impacts of temperature and soil
characteristics on the removal efficiency of GTR system, and
(4) to provide a technical support for the future application of
GTR technique in similar contaminated sites.
Site, materials, and methods
The site used to be a solvent treating plant in Suzhou, China.
Suzhou is situated on the lower reaches of the Yangtze River
and the shores of Lake Tai and belongs to the Yangtze River
Delta Region. This place is mainly dominated by cohesive soil
and shallow underground water level. It has a four-season hu-
mid subtropical climate with hot, humid summers, and cool,
cloudy, damp winters with occasional snowfall. The pilot trial
is 10 m × 10 m with a depth of 18 m. The soil strata consist of
the made ground (~ 0–3 m), clay/silty clay (~ 3–9 m), silty
clay/silt (~ 9–18 m) below ground level. The average ground-
water level is reported at ~ 8.74 m below ground level. The
products of this factory were plasticizers, diphenyl ether, hy-
drogenated triphenyl, and other chemical products. The soil
contaminants are mainly benzene, chlorobenzene, and petro-
leum. Their concentrations at different depths in the field soil
from two sampling points (KB-S01 and KB-S02 are found in
Fig. 2) and in groundwater are detailed in Table 1. A detailed
summary of contaminated site soil’s physical parameters, ther-
mal physical parameters, resistivity, penetration, and particle
composition at different depths can be found in Table 2.
The GTR process is mainly composed of two parts: thermal
desorption and steam extraction. High temperature gases are
obtained by the full combustion of liquefied petroleum gas.
They were injected into the heating well, flowed down
through the inner tube of the heating well first, and then
flowed upward through the outer tube. The target soil temper-
ature was achieved through heat transfers of these tubes. It
was found that the difference in heat conductivity of different
types of soil is very small (Baker and Kuhlman 2002).
Environ Sci Pollut Res

Table 1 The concentrations of target pollutants in groundwater and site soil at different depths (mg/kg)

Pollutants Ground water (μg/L) Sampling point − 0.5 m −3 m −6 m −9 m − 12 m − 15 m − 18 m

Benzene 373,000 KB-S01 58.30 10.30 5.83 12.9 17.9 4.04 7.20
KB-S02 14.0 0.55 0.74 26.70 5.16 1.79 0.55
Chlorobenzene 1,560,000 KB-S01 1560 1470 37.20 173 228 91.8 78.9
KB-S02 64.8 18.7 9.57 704 65.9 51.3 13.8
Petroleum 400,000 KB-S01 330 191 124 203 114 144 127
KB-S02 180 35 194 164 44 37 81

Therefore, heat conduction is a very stable process in soil. A groundwater by vacuum. When heating wells run stably, dual-
computer simulation of the process of soil temperature rise phase extraction wells were switched on, while vapor extrac-
was used to optimize the process design and operation. The tion wells were switched off. In order to prevent the incom-
numerical heating modeling is based on radial heat flow; plete extraction of pollutants and leakage, horizontal gas-
influencing factors such as soil, groundwater, underground phase extraction pipe was laid at a place of 0.5 m underground
gas flow, geology, and hydrology of the restored area were for capturing the dispersed gas and preventing soil pollution
considered in the modeling process. Its control equation is as steam from entering the atmosphere. In addition, pressure
follows: monitoring points were also used to ensure that the under-
ground part was in a negative phase that soil vapors were
∂2 T k ∂T not emitted into the air. The whole GTR system runs for
¼−
∂x2 ρc ∂t 34 days and was monitored by experienced operators. They
were also responsible for adjusting the heating circuit,
where T is the temperature (°C), x is the distance between the
balancing the injection and extraction flow and pressure, and
point and the heat source (m), t is the time (s), ρ is the density
monitoring and adjusting the operation of subsequent process-
of the soil (g/cm3), k is the thermal diffusivity (m2/s), and c is
ing equipment to maintain an optimal treating performance.
the heat capacity (J/(kg/K)). Soil vapor extraction was then
The combustor in GTR was controlled and monitored by a
used to extract vapor containing contaminants to the surface
Programmable Logic Controller (PLC) control system.
for treatment.
Four soil sampling points (DB-S01, DB-S02, DB-S03, and
Before the installation of various wells, large chunks of
DB-S04) and one groundwater sampling point (DB-W) were
construction waste were picked out and a cut-off wall was
installed after treatment. Geoprobe with a Mitsubishi drilling
constructed along the trial site (the depth of which is 18 m
rig (steel pipe) was used to sample the site soil, and Geoprobe
and the circumference of which is 44 m). The groundwater
Direct Push sampling system was used to collect groundwater
was then drained within the cut-off wall to prevent the loss of
sample at each sampling point. The sampling points are found
heat during the heating process. A total of 33 heating wells, 17
in Fig. 2. Four soil sampling points were set near the extrac-
extraction wells (D1~D17), 11 temperature monitoring points
tion well, the well with good heating effect, the edge of the test
(T1~T11), and 5 pressure monitoring points (P1~P5) were
area, and the well with poor heating effect, respectively. The
installed, and the layout is found in Fig. 1. The hexagon points
depth of soil sampling went down to 18 m, and soil samples
show the distribution of heating wells, the triangle points rep-
were collected at depths of − 0.5 m, − 3 m, − 6 m, − 9 m, −
resent the location of extraction wells, and the square points
12 m, − 15 m, and − 18 m. Groundwater samples were col-
represent the location of temperature monitoring points, while
lected at a depth of 9–18 m. They were kept at 0–4 °C fridge
the oval-shaped points show the location of pressure monitor-
and tested in 48 h. EPA Method 8260C (2007) (SW-846) was
ing points. The design of these wells was based on the
applied to measure the concentrations of benzene and chloro-
hydrogeological conditions of the site and computer-aided
benzene, and the infrared photometric method was used for
simulation information. The depths of heating wells varied
the measurement of oil contents (GB/T 16488 1996).
according to the soil contamination, and the intervals among
wells were 2–3 m (equilateral triangle distribution) so that
each heating equipment was able to support one to two
heating wells. The steam extraction system is composed of Results and discussion
vapor extraction well, horizontal vapor extraction pipe, and
dual-phase extraction well. The extraction wells were located Heat distribution
up to ~ 9 m below ground level. Vapor extraction wells were
close to heating wells so as to extract organic pollutants and Figure 3 shows the average temperature distribution of each
point after 14, 21, 28, and 33 days heating treatment. The data
 Environ Sci Pollut Res

2.0~0.5 0.5~0.25 0.25~0.075 0.075~0.005 < 0.005

were collected from 11 temperature monitoring points (each

11.2
28.7
385.3
31.6
25.6
25.9

33.9
Clay
point has four monitoring sensors at depths of − 1 m, − 6 m, −

%
12 m, and − 18 m). From Fig. 3, the average temperature of the
surrounding area of each heating well went up to 100 °C after
14 days treatment and the areas with an average temperature

56.4

78.3
59.8
67.8
73.5

64.9
Silt

72
%
higher than 100 °C expanded after 21 days. Subsequently,
temperatures reached 100~120 °C in the majority part of the
site and reached 150 °C in the part of the heating well sur-

4.8
2.0
0.7
0.9

10.5
2.1

1.2
rounding area after 28 days treatment. At 33 days, in spite of
%

the outlying area, the average temperature of the whole site

reached 100~150 °C; meanwhile, temperatures at the heating
Grain composition (mm)

well surrounding area were even higher at 150~200 °C.

7.1
%

The GTR system stopped after the treatment for 34 days.

The temperatures of four monitoring points at different depths
Gravel Sand

9.2
%

were tested and are shown in Fig. 4. The average temperature

was maintained at ~ 60 °C, and the temperatures of surface
4.9 × 10−5 12.2

soil were found lower than those of deep soil. This is because
>2

the insulation layer of this project did not function well. Later,
1.2 × 10−7
9.1 × 10−7
3.0 × 10−6
5.2 × 10−6

1.7 × 10−5 5.4 × 10−5

1.5 × 10−6 3.6 × 10−6

similar projects confirmed that the temperature of the surface

Osmotic coefficient
Physical parameters, thermal physical parameters, penetration, and particle composition of site soil at different depths

cm/s

soil is actually rising the fastest. The highest temperatures

KH

were found at depths of − 3 m and − 6 m. This is due to the

7.1 × 10−5
8.7 × 10−8
4.3 × 10−7
8.5 × 10−7
2.0 × 10−6

fact that the combined action of the gas and water was extract-
(20 °C)

ed from a depth of − 9 m and the low permeability of clay/silty

cm/s
KV

clay at depths of ~ 3–9 m below ground level. It is worth

conductivity diffusivity

mentioning that the low speed of temperature that rises in

0.001565
0.002033

0.001891
0.001856

0.002016
0.001621
0.002128
Thermal

groundwater surrounding soil is because the groundwater it-

m2/h
Thermal physical

self needs calorific value. In addition, the temperatures of DB-

S02 at each depth were found lower than those of other points
parameters

which are consistent with the conclusion in Fig. 2 that the

Thermal

W/m/K

outlying area of the site was not radiated well by heat.

1.34
0.92
1.32
1.26

1.17
1.40

1.47

Influencing factors
Plastic
limit

19.8
29.1
21.8
24.1
19.5

21.2
20.9
Wp

The moisture content, heat capacity, particle size of the soil,

Liquid

and the heat transfer are critical factors determining the final
limit

31.7
47.6
41.8

0.888 35.2
36.2
31.6

0.792 30.4
WL

soil remediation temperature (Troxler et al. 1993; Abu-

Hamdeh and Reeder 2000). Figure 5 shows the effects of soil
0.902
1.204
0.766

0.855
0.846
Void
ratio

moisture, void ratio, and thermal conductivity on the average

γ (kN/m3) e

temperatures of soil at different depths. From Fig. 5a, it is

Moisture Volume
weight

found that there is a linear correlation between moisture con-

18.9

19.3
19.7

19.1
17.9

19.1
19.2

tent and temperature by fitting, and the correlation coefficient

R2 is 0.9108. Specifically, soil temperatures generally de-
crease with the rise of the moisture content in soil. The nega-
w (%)

32.4

27.2
26.5

32.2
40.1

30.5
30.1

tive correlation between moisture content and temperature can

4) Silty clay (− 11.29~− 8.60 m)

be interpreted by the fact that high soil moisture content re-

4) Silty clay with thin layer silty
1) Filling-up soil (− 1.7–3.3 m)

3) Silty clay (− 8.29~− 6.60 m)

4) Silty (− 17.27~− 16.11 m)

quires more energy to heat the soil; meanwhile, it also affects

soil (− 16.28~− 12.41 m)

the physical handling properties of clay soil. That is the reason

5) Silty clay (− 18 m)

why cut-off walls constructed and ground water was drained

Soil code and name

before the GTR system began. However, Paul et al. (2006)

2) Clay (− 3 m)

studied the contaminant removal from dry and wet sands by

thermal desorption and claimed that the presence of small
Table 2

moisture content can help the removal process by providing

an additional convective heat transfer mechanism. Zhang et al.
Environ Sci Pollut Res
Fig. 1 The plan view of wells and
monitoring points in the field
(2012) reported that nitrobenzene desorption requires an ade-
quate moisture content at 15% to reach a maximal thermal
desorption efficiency of nitrobenzene. Both higher and lower
soil moistures result in poor treatment effect (Zhang et al.
2012; Xu et al. 2013). This verified that site soil with higher
or lower water content has negative effects on the thermal
desorption efficiency in this study.
Figure 5b presents the relationship between soil tempera-
ture and void ratio and shows that the trend of the void ratio is
similar to that of moisture. Generally, the temperature de-
creases with the increasing porosity. A homologous linear
correlation between void ratio and temperature (R2 = 0.9214)
Fig. 2 The plan view of sampling
points in the field
is observed. These findings are consistent with previous stud-
ies that more heat is extracted through bigger voids. For ex-
ample, Troxler et al. (1993) claimed that material cohesion
characteristics can influence vapor transfer. Qi et al. (2014)
found that coarse particles presented lower removal efficiency
and destruction efficiency than fine soil particles, suggesting
that mass transfer influences the desorption of organic
pollutants from coarse particles. Wang et al. (2011) also re-
ported that particle size (three levels 0.25–0.85 mm, 0.15–
0.25 mm, and 0.15 mm) had a considerable impact on the
DDT removal rates and the increase of soil particle size can
improve the desorption of DDT. The effect of void ratio on

a b
c d
Fig. 3 Temperature distribution after 14, 21, 28, and 33 days treatment
soil temperature maintenance indicates that GTR is a suitable
technique in treating contaminated clay soil in situ.
In Fig. 5c, due to the lowest thermal conductivity (0.92 W/
m/K) and lack of insulation layer, the temperature of surface
soil is the lowest at 52.4 °C. The thermal conductivities of soil
at − 3~− 18 m are in the range of 1.17–1.47 W/m/K; the tem-
peratures of them are in the range of ~ 61.4–66.3 °C. No
significant correlation (R2 = 0.7411) can be found between
soil temperature and thermal conductivity in Fig. 5c. This
may be due to the fact that the variability of thermal conduc-
tivity is not big enough, indicating that temperature, moisture
content, reaction time, and void ratio play dominant roles
rather than the slight changes of thermal conductivity in de-
termining the GTR treating efficiency in this study.
Additionally, it is important to note that Fig. 5a, b helped in
explaining the reason why temperatures were obtained higher
at depths of − 3 m, − 6 m, and − 9 m than other soil layers.
Environ Sci Pollut Res
Nevertheless, the two slight deviations of − 18 m point are
observed in Fig. 5a, b, which may be ascribed to the relatively
low thermal conductivity compared to other soil layers (Abu-
Hamdeh and Reeder 2000).
Pollutants removal efficiency
Figure 6 presents the remaining concentrations of organic pol-
lutants in the soil after treatment. In Fig. 6a, the remaining
benzene concentrations of soil tested from DB-S01 and DB-
S04 at most depths were lower than the standard of Soil
Environmental Quality Risk Control Standard for Soil
Contamination of Development Land (Risk Control
Standard) in China (GB36600 2018) at 4 mg/kg. This was
followed by DB-S03, and the benzene concentrations left in
soil at depths of − 0.5, − 3, − 15, and − 18 m were lower than
4 mg/kg after GTR treatment. Due to the relatively low
Environ Sci Pollut Res

Fig. 4 Temperature of monitoring points at different depths after 1 week

(°C)

temperature, the benzene concentrations in DB-S02 at a depth

of − 18 m were the only one reaching the Risk Control
Standard (GB36600 2018). This indicates the predominance
of temperature in GTR technique (Qi et al. 2014). The trends
of the remaining chlorobenzene concentrations of DB-S01–
04 at different depths are similar to those of benzene as men-
tioned above. The remaining chlorobenzene concentrations at
all depths in DB-S01–03 could meet its requirement of Risk
Control Standard (GB36600 2018) at 270 mg/kg. Figure 6c
shows that the GTR removal efficiency of surface soil was
lower than those of deep soil. This is due to the fact that the
heat dissipates quickly in the surface soil, and therefore, the
temperature is relatively low; another possible reason is that
pollutants can not only enter the extraction well but also
spread to the soil near the surface after heating. This indicates
that surface insulation layer and horizontal extraction wells are
must-haves for later projects. Although the GTR removal ef-
ficiency of surface soil was lower compared with those of
deep soil, petroleum at all depths was well removed with
residual concentrations far below 4500 mg/kg (GB36600
2018). The high removal efficiency of petroleum at deep soil
is because the petroleum in the site soil was mainly low chain
petroleum hydrocarbon. Hence, the majority of the petroleum
in soil was able to be removed at ~ 100 °C, through (1) pro-
viding saturated vapor pressure of the pollutant gas phase, (2)
reducing viscosity coefficient, and (3) decreasing soil adsorp-
tion coefficient.
In addition, the removal efficiencies of benzene, chloroben- Fig. 5 The relationship between soil characteristics and soil temperature
at different depths. a Moisture content. b Void ratio. c Thermal
zene, and petroleum varied greatly despite the same tempera- conductivity
ture. Table 3 summarizes the removal efficiencies of pollut-
ants from − 0.5 to − 18 m. The removal efficiency was calcu-
lated by dividing the final pollutant concentration of each petroleum in most points were satisfactory. The highest re-
point in soil with the average concentration of pollutants of moval rates of benzene, chlorobenzene, and petroleum were
two soil sampling points before remediation. It shows that the at 99.81%, 99.72%, and 98.23%, respectively. Among four
removal efficiencies of benzene, chlorobenzene, and monitoring points, the removal rates of benzene and
 Environ Sci Pollut Res

Table 3 The removal rates (%) of pollutants in soil at the depth of −

0.5~− 18 m

Contaminants DB-S01 DB-S02 DB-S03 DB-S04

Benzene 19.03–99.81 12.63–95.61 29.33–99.25 23.84–99.54

Chlorobenzene 41.86–99.68 30.95–89.80 47.17–98.94 23.10–99.72
Petroleum 46.96–98.11 65.38–98.23 51.90–91.28 65.82–92.45

In summary, the removal efficiency order of the four mon-

itoring points is DB-S04 ≈ DB-S01 > DB-S03 > DB-S02. In
addition, in spite of temperature and the nature of the soil,
sampling disturbance may also influence the GTR removal
efficiency of organic pollutants. Before remediation, the pol-
lutants were mainly distributed at depths of − 0.5 m and − 9 m
(the bottom of clay/silty clay where the pollutants accumulat-
ed); it was later detected at depths of − 3 m and − 12 m. This
may be due to the drilling process, which migrated the pollut-
ants to deeper soil. This can also explain the high temperature
and low pollutants removal efficiency at depths of − 3 to −
12 m.
In Table 4, it can be found that benzene, chlorobenzene,
and petroleum have been thoroughly removed from the
groundwater with the removal rates of 98.77%, 97.70%, and
99.99%, respectively. The concentrations of benzene, chloro-
benzene, and petroleum tested after treatment were at
4590 μg/L, 35,900 μg/L, and 18.6 μg/L, respectively.
Compared to pollutant removal efficiency in the soil, although
the initial concentrations of pollutants in groundwater were
very high, the performance of GTR system in removing pol-
lutants from groundwater was more significant.

Conclusions

In this paper, the performance of the in situ GTR-treated site

soil and groundwater after 34 days treatment has been inves-
tigated. The main findings of this study are summarized as
follows:

1. Soil temperatures in the middle of the remediation area,

especially these near heating wells, are found higher than
those at the edge of the site.
Fig. 6 The pollutants concentrations left in soil. a Benzene. b
Chlorobenzene. c Petroleum
Table 4 Concentration and removal rate of pollutants in groundwater
chlorobenzene at different depths in DB-S01 and DB-S04
were the highest; this is because DB-S01 was the closest to Pollutants KB-S01 (before DB-S01 (after Removal
treatment) (μg/L) treatment) (μg/L) rate (%)
a heating well and DB-S04 was the one nearest to an extrac-
tion well. This was followed by DB-S03, and the removal Benzene 373,000 4590 98.77
rates of pollutants in DB-S02 (at the edge of the field) was Chlorobenzene 1,560,000 35,900 97.70
the lowest. The surface soil treated by GTR system had a less Petroleum 400,000 18.6 99.99
efficient treatment of petroleum compared with deep soil.
Environ Sci Pollut Res
2. One week after the GTR system stopped, the soil temper-
ature of most site area was at ~ 70 °C. The distribution of
soil temperature in the vertical direction is not even, and
the average temperature of the surface soil was the lowest.
The highest temperatures were found at depths of − 3 m
and − 6 m.
3. Soil temperatures decrease with the increase of moisture
content and porosity.
4. The removal efficiency order of the four monitoring
points is DB-S04 ≈ DB-S01 > DB-S03 > DB-S02.
5. After in situ treatment by GTR technology, the concentra-
tions of benzene, chlorobenzene, and petroleum pollut-
ants in the soil and groundwater in the test area decreased
significantly, and the pollutant removal efficiency in
groundwater was found higher than that in site soil.
6. The highest removal rates of benzene, chlorobenzene, and
petroleum in soil were 99.81%, 99.72%, and 98.23%,
respectively, and those in groundwater were 98.77%,
97.70%, and 99.99%, respectively.
Funding information This work was partly supported by the National
Natural Science Foundation of China (No. 51608113) and the
Fundamental Research Funds for the Central Universities (No.
2242018K40006).
Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.
References
Abu-Hamdeh NH, Reeder RC (2000) Soil thermal conductivity: effects
of density, moisture, salt concentration, and organic matter. Soil Sci
Soc Am J 64:1285–1290
Aresta M, Dibenedetto A, Fragale C, Giannoccaro P, Pastore C,
Zammiello D, Ferragina C (2008) Thermal desorption of
polychlorobiphenyls from contaminated soils and their
hydrodechlorination using Pd and Rh supported catalysts.
Chemophere 70:1052–1058
Baker RS, Kuhlman M (2002) A description of the mechanisms of in-situ
thermaldestruction (ISTD) reactions. In: Al-Ekabi H (ed) Current
practices in oxidation and reduction Technologies for Soil and
Groundwater (available on CD). Presented at the 2nd international
Conf. On oxidation and reduction Technologies for Soil and
Groundwater, ORTs-2, Toronto, Ontario, Nov. 17–21, 2002
Biache C, Mansuy-Huault L, Faure P, Munier-Lamy C, Leyval C (2008)
Effects of thermal desorption on the composition of two coking
plant soils: impact on solvent extractable organic compounds and
metal biovailability. Environ Pollut 156:671–677
Bonnard M, Devin S, Leyval C, Morel JL, Vasseur P (2010) The influ-
ence of thermal desorption on genotoxicity of muti-polluted soil.
Ecotoxicol Environ Saf 73:955–960
Bucala V, Saito H, Howard JB, Peters WA (1994) Thermal treatment of
fuel oil-contaminated soils under rapid heating conditions. Environ
Eng Sci 28:1801–1807
Chen M (2010a) Analytical integration procedures for the derivation of
risk-based generic assessment criteria for soil. Hum Ecol Risk
Assess 16:1295–1317
Chen M (2010b) Alternative integration procedures in combining multi-
ple exposure routes for the deviation of generic assessment criteria
within the CLEA model. Hum Ecol Risk Assess 18:135–150
Chien YC (2012) Field study of in situ remediation of petroleum hydro-
carbon contaminated soil on site using microwave energy. J Hazard
Mater 199:457–461
Falciglia PP, Giustra MG, Vagliasindi FGA (2011) Low-temperature ther-
mal desorption of diesel-polluted soil: influence of temperature and
soil texture on contaminant removal kinetics. J Hazard Mater 185:
392–400
GB/T 16488 (1996) Determination of petroleum, animal and plant oils.
Infrared photometric method: GB/T 16488-1996, PRC SEPA. http://
websearch.mee.gov.cn/was5. Accessed 26 July 2018
GB36600 (2018) Soil environmental quality: Agricultural Land Pollution
Risk Management Standards: GB36600–2018, PRC SEPA. http://
kjs.mee.gov.cn/hjbhbz/bzwb/trhj/trhjzlbz/201807/t20180703_
446027.shtml. Accessed 14 June 2018
Geng CN, Luo QS, Chen MF, Li ZY, Zhang CB (2010) Quantitative risk
assessment of trichloroethylene for a former chemical works in
Shanghai, China. Hum Ecol Risk Assess 16:429–443
Kawala K, Atamanczuk T (1998) Microwave-enhanced thermal decon-
tamination of soil. Environ Sci Technol 32:2602–2607
Mechati F, Roth E, Renault V, Risoul V, Trouve G, Gilot P (2004) Pilot
scale and theoretical study of thermal remediation of soils. Environ
Eng Sci 21:361–370
O’Brien PL, Desutter TM, Casey FXM, Khan E, Wick AF (2018)
Thermal remediation alters soil properties-a review. J Environ
Manag 206:826–835
Paul A, Stosser R, Zehl A (2006) Contaminant removal from dry and wet
sands by thermal desorption. Int J Environ Waste Manag 10:39–48
Pina J, Merino J, Errazu AF, Bucala V (2002) Thermal treatment of soils
contaminated with gas oil: influence of soil composition and treat-
ment temperature. J Hazard Mater 94:73–290
Qi ZF, Chen T, Bai SH, Yan M, Lu SY, Buekens A, Yan JH, Bulmau C, Li
XD (2014) Effect of temperature and particle size on the thermal
desorption of PCBs from contaminated soil. Environ Sci Pollut Res
21:4697–4704
Saito H, Howard JB, Peters WA, Bucalá V (1998) Soil thermal decon-
tamination: fundamentals. In: Meyers RA (ed) Encyclopedia of en-
vironmental analysis and remediation. Wiley, New York, pp 4554–
4589
Sakaguchi I, Inoue Y, Nakamura S, Kojima Y, Sasai R, Sawada K, Suzuki
K, Takenaka C, Katayama A (2015) Assessment of soil remediation
technologies by comparing health risk reduction and potential im-
pacts using unified index, disability-adjusted life years. Clean
Technol Environ Policy 17:1663–1670
Troxler WL, Cudahy JJ, Yezzi JJ, Zink RP, Rosenthal SL (1993)
Treatment of nonhazardous petroleum-contaminated soils by ther-
mal desorption technologies. J Air Waste Manage Assoc 43:1512–
1525
USEPA (2004) In situ thermal treatment of chlorinated solvents: funda-
mental and field applications. Office of solid waste and emergency
response, Washington, DC https://www.epa.gov/remedytech.
Accessed 14 June 2018
USEPA (2006) In situ thermal desorption for treatment of POPs in soils
and sediment. Office of solid waste and emergency response,
Washington, DC https://nepis.epa.gov/Exe/ZyNET.exe/. Accessed
14 June 2018
USEPA (2007) Method 8260C: volatile organic compounds by gas
chromatography/mass spectroscopy (GC/MS). https://www.epa.
gov/remedytech/. Accessed 14 June 2018
Wang Y, Yang L, Huang QF (2011) Effects of different pollutant concen-
trations and soil particle size on thermal desorption efficiency of
DDT-contaminated soil. Res Environ Sci 24:1016–1022

Wang F, Wang H, Jin F, Al-Tabbaa A (2015) The performance of blended
conventional and novel binders in the in-situ stabilisation/
solidification of a contaminated site soil. J Hazard Mater 285:46–52
Xu DP, He YL, Zhang XN, Gu QB (2013) Desorption kinetics of DDTs
from contaminated soil during processes of thermal desorption. Res
Environ Sci 26:202–207
Yang B, Xue ND, Ding Q, Vogt RD, Zhou LL, Li FS, Wu GL,
Zhang SL, Zhou DD, Liu B, Yan YZ (2014a) Polychlorinated
Environ Sci Pollut Res
biphenyls removal from contaminated soil using a transportable
indirect thermal dryer unit: implications for emissions.
Chemosphere 114:84–92
Yang H, Huang XJ, Thompson JR, Flower RJ (2014b) China’s soil pol-
lution: urban brown fields. Science 344:691–692
Zhang P, Gao YZ, Kong HL (2012) Thermal desorption of nitrobenzene
in contaminated soil. Soils 44:801–806