You are on page 1of 7

Chemosphere 154 (2016) 276e282

Contents lists available at ScienceDirect

journal homepage:

Potential for leaching of arsenic from excavated rock after different

drying treatments
Jining Li a, Tomoya Kosugi a, Shohei Riya a, Yohey Hashimoto b, Hong Hou c,
Akihiko Terada a, Masaaki Hosomi a, *
Department of Chemical Engineering, Tokyo University of Agriculture and Technology, 2-24-16 Naka, Koganei, Tokyo 184-8588, Japan
Department of Bioapplications and Systems Engineering (BASE), Tokyo University of Agriculture and Technology, 2-24-16 Naka, Koganei, Tokyo 184-8588,
State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Dayangfang 8, Beijing 100012,
PR China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 Considerable total As was water-

leached from an excavated rock.
 Air-, 40  C-, and freeze-drying treat- Rock Different drying methods

ments increased the As leaching.

 Freeze-drying increased As leaching
by the largest extent.

 Dissolution of Fe, Al, and dissolved

organic carbon correlated with As
Sequential leaching tests Rapid small-scale column tests
 Pollution potential indices are pro-
Potential for leaching of arsenic
posed to evaluate the pollution threat
of As.

a r t i c l e i n f o a b s t r a c t

Article history: Leaching of arsenic (As) from excavated rock subjected to different drying methods is compared using
Received 21 January 2016 sequential leaching tests and rapid small-scale column tests combined with a sequential extraction
Received in revised form procedure. Although the total As content in the rock was low (8.81 mg kg1), its resulting concentration
19 March 2016
in the leachate when leached at a liquid-to-solid ratio of 10 L kg1 exceeded the environmental standard
Accepted 28 March 2016
(10 mg L1). As existed mainly in dissolved forms in the leachates. All of the drying procedures applied in
this study increased the leaching of As, with freeze-drying leading to the largest increase. Water
Handling Editor: X. Cao extraction of As using the two tests showed different leaching behaviors as a function of the liquid-to-
solid ratio, and achieved average extractions of up to 35.7% and 25.8% total As, respectively. Dissolu-
Keywords: tion of As from the mineral surfaces and subsequent re-adsorption controlled the short-term release of
Arsenic As; dissolution of Fe, Al, and dissolved organic carbon played important roles in long-term As leaching.
Excavated rock Results of the sequential extraction procedure showed that use of 0.05 M (NH4)2SO4 underestimates the
Leaching readily soluble As. Long-term water extraction removed almost all of the non-specically sorbed As and
most of the specically sorbed As. The concept of pollution potential indices, which are easily determined
by the sequential leaching test, is proposed in this study and is considered for possible use in assessing
efcacy of treatment of excavated rocks.
2016 Elsevier Ltd. All rights reserved.

* Corresponding author.
E-mail address: (M. Hosomi).
0045-6535/ 2016 Elsevier Ltd. All rights reserved.
J. Li et al. / Chemosphere 154 (2016) 276e282 277

1. Introduction Japan, in February 2015. After collection, all samples were placed in
airtight polyethylene bags and transported to the laboratory. The
Arsenic (As), a metalloid belonging to Group 15 of the Periodic collected rock was pretreated by four common drying methods: (1)
Table, is ubiquitously present in the natural environment at trace air-drying at room temperature for 7 days (AD); (2) oven-drying at
levels (Wilson et al., 2010). According to a report of the World 40  C for 24 h (40D); (3) freeze-drying at 50  C for 24 h (FD); and
Health Organization (WHO), As is one of ten chemicals of major (4) no drying (ND). After being subjected to the selected pretreat-
public health concern (WHO, 2012). Long-term As exposure from ment, the samples were gently crushed and sieved through a 2-mm
food and drinking water can cause cancer and skin lesions, as well sieve. A portion of the sample with a particle size of less than 2 mm
as developmental effects, cardiovascular disease, neurotoxicity, and was used for the leaching experiments. Further description of
diabetes (WHO, 2012). Because of these negative impacts on hu- sample characterization is provided in the Supplementary Material.
man health, the WHO recommended a provisional standard of Selected physicochemical properties are shown in Table S1. The
10 mg L1 for arsenic in drinking water (WHO, 1994). total As concentration in this rock was 8.81 mg kg1.
Many construction projects require excavation of large quantities
of deep rocks. Although the amounts of As in these rocks are often
very low, the concentrations in their resultant leach solutions can 2.2. Leaching experiments
exceed environmental standards: for example, Tabelin et al. (2014a)
found that the As concentrations in leachates from some excavated 2.2.1. Sequential leaching tests
rocks were higher than the Japanese environmental standard Sequential leaching tests (SLT) are sometimes used to study
(10 mg L1). This underscores the need to properly dispose of such long-term leaching of contaminants from solids into water
rocks. In fact, in Japan, efforts are now directed at the potential risks (Grathwohl and Susset, 2009). In this study, referring to Notica-
of excavated rocks, and the Soil Contamination Countermeasures tion No. 46 of the Ministry of Environment (Japan), the SLT was
Law of Japan (Ministry of the Environment, 2011) was amended to conducted under ambient conditions. A 3-g sample (dry weight
extend the interpretation to cover these solids. To assess the po- basis; 4.44 g for sample with ND) and 30 mL of deionized water
tential for contamination of As from such rocks and to develop (18 MU cm) were mixed in a 50-mL centrifuge tube. To determine
effective and economically feasible remediation technologies, more an appropriate shaking time, we performed preliminary batch
detailed knowledge about As behavior in excavated rocks is required leaching experiments with different shaking periods (0.5 h, 1 h, 3 h,
to better understand its release to the surrounding environment. 6 h, 12 h, and 24 h). The mixture was shaken at a rate of 200 rpm at
To date, the risk of pollutants leaching from a variety of solids room temperature and pressure. The slurry was centrifuged at
has been determined by laboratory leaching tests, such as batch about 4700 g for 15 min and the supernatant was ltered through a
and column tests (Grathwohl and Susset, 2009). Batch tests provide 0.45-mm membrane lter before analysis. These procedures were
a snapshot of a particular liquid-to-solid (L/S) ratio (Maszkowska repeated until the As concentration in the leachate was below its
et al., 2013), but do not simulate actual eld conditions because detection limit (0.05 mg L1). The residual sample material was
of friction between the particles during mechanical agitation, dried using the methods described above and then digested for As
centrifugation, and ltration of the sample prior to analysis. Col- measurement.
umn test conditions are more similar to eld conditions and more After the 0.45-mm ltration, As in leachates is commonly clas-
representative than batch tests because there is no mechanical sied into dissolved and colloidal fractions (Nystrand et al., 2012).
friction, the packed material has self-ltration capacity, and In our study, part of the 0.45-mm ltered leachate after 24-h
observation of time-dependent leaching behavior is possible shaking was further ltered using a 0.1-mm lter and the As in
(Krger et al., 2012). Although some studies have compared batch this ltrate was considered to be present in dissolved form. The
and column tests with respect to As release from soils or wastes (Al- difference between the 0.45-mm and 0.1-mm lterable As was
Abed et al., 2008; Kim and Hyun, 2015), no study has quantitatively attributed to colloidal form. Fe and Al in the leachates were also
compared these techniques for investigating the potential for As classied by the same method.
release from excavated rocks.
An appropriate methodology for storage of excavated rocks still
requires development. As it acts on a soil sample, drying treatment 2.2.2. Rapid small-scale column tests
plausibly enables a rock sample to be stored for longer time and Rapid small-scale column tests (RSSCT), which allow break-
contributes to development of a standardized methodology, which through curves as functions of time and volume of permeates to be
allows one to compare results for As release from different rocks. generated, were originally developed to simulate the removal of
However, drying may change the status and properties of trace el- organic pollutants and natural organic matter by granular activated
ements in the rocks and raises the question of whether As release carbon (Crittenden et al., 1987, 1991). They are appropriate for use
from dried samples is representative of eld conditions (Supriatin with the excavated rock in this study because of its high perme-
et al., 2015). To date, the effect of different drying methods on As ability. A series of stand-alone RSSCT columns (internal diameter,
extraction from soils has been studied (Huang et al., 2015), however, 1.7 cm; length, 24 cm) were constructed and an up-ow congu-
the inuence of drying on As leaching behavior from excavated ration was employed, which differs from the regular RSSCT test. The
rocks is largely unknown. The objectives of this study are, therefore: experimental set-up is shown in the Supplementary Material. To
(1) to investigate the effects of drying methods on the As leaching readily compare these As leaching results with those of the SLT, a
from an excavated rock; (2) to compare the potential for As leaching 3.0-g sample (4.44 g for sample with ND) was packed in each col-
under batch and column conditions; and (3) to identify the domi- umn. The column was rst slowly saturated with water from bot-
nant processes governing As leaching from this excavated rock. tom to top; water was then continuously supplied at a ow rate of
1.0 mL min1. Fifteen column efuent samples of 30 mL each
2. Materials and methods (equivalent to 10 L kg1 of L/S) were collected (sample numbers
1e15); thereafter, samples were collected every 50 mL (equivalent
2.1. Sample collection and characterization to every 16.7 L kg1) and numbered from 16 onward. Sampling
continued until the As concentration was below its detection limit.
The excavated rock used in our study was obtained from Tokyo, The residual material was dried and digested for As measurement.
278 J. Li et al. / Chemosphere 154 (2016) 276e282

2.2.3. Chemical analysis release; freeze-drying most enhanced As release (Fig. S1 (b)). After
pH and Eh of the samples were measured by pH (Navi F-52; the second ltration (0.1 mm) step, the As concentrations showed
HORIBA, Kyoto, Japan) and oxidationereduction potential meters no obvious decrease in comparison with those of leachates passed
(D-53, HORIBA, Kyoto, Japan), respectively. As, Fe, Al, and Mn con- though the 0.45-mm lters (Fig. S2 (a)); however, the Fe and Al
tents in the leachates were measured by inductively coupled concentrations decreased signicantly (close to zero) after passing
plasma mass spectrometry (ICPeMS, 7500c, Agilent, Tokyo, Japan), through the 0.1-mm lters (Fig. S2 (b) and (c)). This nding indicates
after appropriate dilution as necessary. Dissolved organic carbon that most Fe and Al existed in colloidal forms in the leachates, while
(DOC) in the leachate was measured using a TOC-VCSH/CSN As mainly existed in dissolved forms.
analyzer (Shimadzu, Kyoto, Japan).
3.2. Arsenic leaching in sequential leaching and rapid small-scale
2.3. Pollution potential indices column tests

In this study, we propose the concept of pollution potential The mean As concentrations (the ratio of cumulative As mass to
indices (PPI). At a certain L/S value, the ratio of cumulative As to cumulative leachate volume) during the SLT and RSSCT were
cumulative volume will reach the environmental standard plotted as a function of L/S ratio (Fig. 1). In most cases, the As
(10 mg L1) (see equation (1)). The value of L/S at this time is concentration dropped sharply at the beginning of the test and then
considered as the PPI. A higher PPI value indicates a higher pollu- slowly converged to a stable value; however, As concentrations in
tion threat. The key point for this method is obtaining the cumu- the SLT leachates generated from samples subjected to the 40D, AD,
lative amount of As released. Two equations were used to t the and ND treatments (Fig. 1 (a)) increased at rst and then gradually
cumulative release of As in the SLT and RSSCT with increasing of L/ decreased. The cumulative As releases are shown in Table 1. The
S: equation (2), based on Cappuyns and Swennen (2008), and percentages of cumulative leached As in the SLT were 1.4 times
equation (3), described as a two-site rst-order leaching model higher on average than those of the RSSCT. In addition, compared
according to Kim and Hyun (2015). In the comparison, two L/S data with the rock with ND, drying treatments increased the As leaching
sets (10e50 and 10e500 L kg1) were tted separately. and freeze-drying resulted in the largest increase.
For both SLT and RSSCT, the As releases under two L/S ranges
1000 CPPI =PPI 10; (1) (10e50 and 10e500 L kg1) tted well to equations (2) and (3),
with correlation coefcient (R2) values ranging from 0.935 to 0.999
Cx k lnx a; (2) (Table 2). As shown in Table 2, the PPIs calculated based on equa-
tions (2) and (3) with L/S from 10 to 500 were generally in agree-
Cx CT 1  f1  expk1 x  f2  exp  k2 x; (3) ment with the values measured in our experiments, especially for

where x is the L/S ratio (L kg1); Cx is the concentration of As

released at L/S of x (mg kg1); k and a are constants; CT is the total (a) 100
As concentration in the rock (mg kg1); f1 and f2 are the rapid and
Mean As concentration (g L-1)

slow leaching fractions, respectively, (the sum of f1 and f2 is 1); k1 80

and k2 are the corresponding leaching rate constants. The approx-
imate PPI in equation (1) was solved using the Goal seek function 60 SLT
and equations (2) and (3) were solved using Curve tting and Solver
in Microsoft Excel (2007), respectively.

2.4. Sequential extraction procedure 20

The fractionations of As in the rock before and after water

extraction (for both SLT and RSSCT) were investigated by a ve-step 0 150 300 450 600 750
sequential extraction procedure (SEP) based on the method L/S (L kg-1)
described by Wenzel et al. (2001). Non-specically sorbed As (F1), (b) 140
specically sorbed As (F2), amorphous oxide-associated As (F3),
120 40 D AD FD ND
Mean As concentration (g L-1)

crystalline oxide-associated As (F4), and residual As were sequen-

tially extracted. The detailed conditions are provided in the Sup- 100
plementary Material.
3. Results 60

3.1. Batch leaching experiments with variation of contact time 40

Fig. S1 shows the pH and released concentrations of As, Fe, Al,
and DOC in the leachates after different shaking times. All the 0
leachate pH values were alkaline, regardless of the drying method 0 200 400 600 800 1000
and contact time. Irrespective of drying method, the trend of As L/S (L kg-1)
concentrations was consistent: either reaching its highest value or
stabilizing after 6 h. This indicated that 6 h was an appropriate time Fig. 1. Mean As concentrations (ratio of cumulative As mass to cumulative leachate
volume) as a function of L/S ratio during (a) sequential leaching tests (SLT) and (b)
after which to proceed to the next sequential leaching test. rapid small-scale column tests (RSSCT). 40D: oven-drying at 40  C; AD: air-drying at
Compared with that of wet rock without water evaporation, the room temperature; FD: freeze-drying at 50  C; ND: no drying. Experiments used 3 g
dryingerewetting process generally promoted short-term As dried samples (4.44 g for sample with ND).
J. Li et al. / Chemosphere 154 (2016) 276e282 279

Table 1
Cumulative releases of As in the SLT and RSSCT.
Released As Residual As Accountability (%)

mg a % b
mg c
SLT 40D 9.40 35.6 15.3 93.5
AD 9.10 34.4 16.6 97.2
FD 10.9 41.3 13.2 91.2
ND 8.28 31.3 e e
Average 9.42 35.6 15.0 94.0
RSSCT 40D 6.44 24.4 18.0 92.6
AD 6.39 24.2 16.5 86.7
FD 8.35 31.6 16.1 92.4
ND 6.11 23.1 e e
Average 6.82 25.8 16.9 90.6
Cumulative mass of As released from 3 g sample (dry weight basis; 4.44 g for sample with ND).
Calculated by 100  (cumulative mass of As released from sample/total mass of As in sample).
Residual mass of As in the dried residual material.
Calculated by 100  [(cumulative mass of released As residual mass of As)/total mass of As]. SLT: sequential leaching tests; RSSCT: rapid small-scale column tests; 40D:
oven-drying at 40  C; AD: air-drying at room temperature; FD: freeze-drying at 50  C; ND: no drying; : not measured.

those of the SLT. When a L/S from 10 to 50 was used, however, SLT were generally higher than those from the RSSCT, resulting in
inconsistency existed between the measured PPIs and those higher PPI values and greater amounts of total leached As in the SLT.
calculated from equation (3); PPIs calculated from equation (2) in Fig. 2 also compares the Fe, Al, and DOC concentrations in the
SLT agreed well with the measured PPIs (Table 2). leachates generated by the SLT and RSSCT. The trend of DOC release
was similar to that of As, with higher DOC release in the RSSCT at an
3.3. Arsenic fractionations L/S of 10 and then generally higher DOC release in SLT when the L/S
exceeded 10. Such correlations were not observed with the Fe and
As shown in Table 3, before water extraction, arsenic was pre- Al concentrations: these elements were hardly detected in the
sent mainly in the residual fractions (F5), followed by the specif- RSSCT leachates. Along with the results in Fig. S1, this trend reveals
ically sorbed fractions (F2) and the amorphous oxide-associated that As mobility appears to differ from that of Fe and Al colloids.
fractions (F3), with the other fractions present in minor amounts As release may crucially depend on the dynamics of the mass-
(<4.5%). After water extraction, arsenic also mostly existed in the transfer processes, and equilibrium exists when the mass-transfer
residual fractions (F5), but in decreased amount; the amounts of rates between solid and water are equal. At an L/S ratio of 10 in
the amorphous oxide-associated fractions (F3), specically sorbed the RSSCT, the release of As was the highest (Fig. 1 (b)). Equilibria
fractions (F2), and non-specically sorbed fractions (F1) also commonly occur near the beginning of the test because concen-
decreased, whereas the crystalline oxide-associated fractions (F4) tration gradients between the solid and aqueous phases at this time
increased to a certain extent. are extremely steep, resulting in high mass-transfer rates. In a dy-
namic column test, water is continuously replaced, which main-
4. Discussion tains the driving force for leaching (Cappuyns and Swennen, 2008);
however, the equilibria are postponed (L/S 20) after the 40D, AD,
4.1. Comparison of sequential leaching and rapid-small-scale and ND treatments in the SLT (Fig. 1 (a)). The equilibria can be
column tests inuenced by the L/S ratio; pH and redox conditions are also very
important (Grathwohl and Susset, 2009). Lower pH and higher Eh
Releases of As (L/S from 10 to 150) during the SLTs compared values were obtained in the SLT, compared with those in the RSSCT
with those from the RSSCTs are plotted in Fig. 2. Plots below the (Fig. S3). Although the variations were small, this was consistent
diagonal line indicate higher release recorded by the RSSCT than by with the later-emerging equilibria under the SLT conditions.
the SLT. At an L/S ratio of 10, As release was higher in the RSSCT than Moreover, agitation of solids in the SLT could mobilize ne particles
in the SLT, which indicated that As release was underestimated by created by particle fracturing. This process may promote the release
the batch test compared with the column test. On the contrary, of elements from the solid materials (Krger et al., 2012); a greater
when the L/S ratios exceeded 20, the As concentrations from the number of ne particles may also increase contaminant re-

Table 2
Parameters for the tted cumulative release of As and determination of pollution potential indices.

Cx k lnx a; Cx CT 1  f1  expk1 x  f2  exp  k2 x; PPIM

L/S (10e50 L kg1) L/S (10e500 L kg1) L/S (10e50 L kg1) L/S (10e500 L kg1)

k a R2 PPIC k a R2 PPIC f1 k1 k2 R2 PPIC f1 k1 k2 R2 PPIC

SLT 40 D 0.74 1.2 0.9991 297 0.65 0.86 0.9860 278 0.25 0.028 5.6E-07 0.9942 222 0.27 0.021 2.4E-04 0.9932 279 280e290
AD 0.70 1.2 0.9973 272 0.65 1.0 0.9895 264 0.26 0.023 1.0E-14 0.9938 225 0.26 0.019 2.4E-04 0.9949 266 260e270
FD 0.74 0.77 0.9987 358 0.66 0.41 0.9837 342 0.19 0.069 1.5E-03 0.9992 589 0.31 0.028 3.1E-04 0.9841 336 340e350
ND 0.63 1.3 0.9830 208 0.65 1.2 0.9877 228 0.11 0.013 2.2E-03 0.9944 716 0.24 0.015 1.8E-04 0.9976 236 230e240
RSSCT 40 D 0.30 0.067 0.9996 157 0.48 0.53 0.9801 200 0.093 0.17 1.1E-03 0.9989 304 0.17 0.036 3.1E-04 0.9610 192 183e200
AD 0.26 0.095 0.9999 136 0.53 0.87 0.9557 189 0.087 0.16 8.7E-04 0.9992 211 0.14 0.039 4.0E-04 0.9598 178 167e183
FD 0.37 0.46 0.9997 250 0.46 0.21 0.9885 279 0.16 0.20 1.5E-03 0.9992 544 0.25 0.054 3.1E-04 0.9610 271 283e300
ND 0.31 0.30 0.9999 120 0.43 0.68 0.9935 162 0.074 0.081 7.0E-04 0.9999 143 0.14 0.027 2.3E-04 0.9352 142 140e150

PPIC: Calculated PPI; PPIM: Measured PPI.

280 J. Li et al. / Chemosphere 154 (2016) 276e282

Table 3
Fractionations of As before and after water extraction.

Fractions a, mg kg1 (%) b

F1 F2 F3 F4 F5

Before extraction 40D 0.213 (2.41) 2.12 (24.0) 1.17 (13.3) 0.326 (3.70) 4.98 (56.6)
AD 0.228 (2.59) 1.89 (21.4) 1.30 (14.8) 0.395 (4.48) 5.00 (56.8)
FD 0.224 (2.54) 1.89 (21.5) 1.12 (12.7) 0.310 (3.52) 5.27 (59.8)
Average 0.222 0.008 (2.52 0.090) 1.96 0.132 (22.3 1.50) 1.20 0.095 (13.6 1.08) 0.344 0.045 (3.90 0.512) 5.08 0.159 (57.7 1.81)
After SLT 40D 0.000 (0.000) 0.383 (6.62) 0.967 (16.7) 0.555 (9.60) 3.88 (67.1)
AD 0.000 (0.000) 0.427 (6.63) 1.33 (20.7) 0.903 (14.0) 3.77 (58.6)
FD 0.000 (0.000) 0.326 (6.80) 0.865 (18.0) 0.501 (10.4) 3.11 (64.8)
Average 0.000 0.000 (0.000 0.000) 0.379 0.050 (6.69 0.099) 1.06 0.247 (18.5 2.04) 0.653 0.218 (11.4 2.35) 3.59 0.416 (63.5 4.36)
After RSSCT 40D 0.017 (0.27) 0.571 (9.05) 0.875 (13.9) 0.378 (5.98) 4.47 (70.8)
AD 0.011 (0.18) 0.604 (10.0) 0.964 (16.0) 0.473 (7.84) 3.98 (66.0)
FD 0.008 (0.15) 0.443 (8.04) 0.796 (14.5) 0.377 (6.84) 3.88 (70.5)
Average 0.012 0.005 (0.198 0.064) 0.539 0.085 (9.04 0.992) 0.878 0.084 (14.8 1.09) 0.409 0.055 (6.89 0.931) 0.222 0.008 (69.1 2.72)
F1: non-specically sorbed As; F2: specically sorbed As; F3: amorphous oxide-associated As; F4: crystalline oxide-associated As; F5: residual As.
Values in parentheses: percentage of each fraction (%).

(a) (b) 20

DOC concentration in SLT (mg L-1)

As concentration in SLT (g L-1)



L/S = 10 8

/S = 10

1 0
1 10 100 0 4 8 12 16 20
As concentration in RSSCT (g L-1) DOC concentration in RSSCT (mg L-1)
(c) (d) 8
Fe concentration in SLT (mg L-1)

Al concentration in SLT (mg L-1)

2 4

0 0
0 2 4 0 2 4 6 8
Fe concentration in RSSCT (mg L-1) Al concentration in RSSCT (mg L-1)

Fig. 2. Comparison of (a) As (mg L1), (b) DOC (mg L1), (c) Fe (mg L1), and (d) Al (mg L1) in leachates (sample numbers 1e15 with cumulative L/S from 10 to 150) produced in SLT
versus RSSCT.

adsorption because of the increased surface area (Maszkowska leachates were also lower than those of the water extraction
et al., 2013). This re-adsorption probably decreased the As con- leachates (Table S2). This may account for the lower release of As in
centration in the SLT (L/S 10), but the effect may be reversible and F1 extraction because of greater adsorption of As onto clay minerals
adsorbed As would be released again during the later shaking or precipitates (Tabelin et al., 2014a). The Fe and Al concentrations
experiments. in the F2 extraction leachates were also very low (Table S2), which
means that re-adsorption of As probably occurs during this step;
4.2. Potential for As leaching however, it is reported that phosphate and arsenate show similar
environmental behaviors (Smedley and Kinniburgh, 2002). Given
The results revealed that water extraction at an L/S ratio of 10 the fact that phosphate is able to extract not only the easily
released more As than the non-specically sorbed As obtained by adsorbed (outer-sphere complexed) As but also strongly adsorbed
the SEP (Table S2). The Fe and Al concentrations in the leachates of (inner-sphere) As (Cheng et al., 2009), the presence of phosphate
F1 extraction were much lower than those of water extraction potentially competes with arsenate for adsorption sites, leading to
(Table S2), probably because more Fe-/Al-oxyhydroxide precipitate greater As release. This indicates that the use of 0.05 M (NH4)2SO4
was formed during F1 extraction. The pH values of the F1 extraction as the leaching medium in the SEP may result in underestimation of
J. Li et al. / Chemosphere 154 (2016) 276e282 281

the readily soluble fraction. After the SLT and RSSCT were values determined using the other two drying methods.
completed, it was found that water extraction could remove almost
all of the non-specically sorbed As and most of the specically 4.4. Possible application of pollution potential indices in the
sorbed As (Table 3). These two fractions are suggested as the pri- disposal of excavated rocks
mary sources of As release. Furthermore, removing labile species
could disturb the chemical equilibria among different species, The leaching of As from excavated rocks can pose a risk to the
resulting in transformation between different fractions (Wen et al., environment (Tabelin et al., 2014b). More efcient and safe ways of
2013). A previous study conrmed that less-crystalline iron oxides managing hazardous excavated rocks are needed. Tatsuhara et al.
may undergo phase transformations into more crystalline phases (2012) suggested a combined neutralizationeadsorption system
(Cornell and Schwertmann, 2003), which results in the incorpora- for the disposal of hydrothermally altered excavated rock, which
tion of arsenic into the structure of the crystalline product mainly uses an adsorption layer at the base to immobilize leached
(Pedersen et al., 2006). This may explain the increase of the crys- As; however, this approach requires identication of suitable
talline oxide-associated As. The total water-extractable As in the adsorbent dosages based not only on the adsorption capacity of the
SLT after freeze-drying (41.3%) was slightly higher than the sum of adsorbent, but also on the potential for As leaching. Tabelin et al.
the rst four fractions in SEP (40.2%), demonstrating that long-term (2013) employed column tests to predict the potential for As
water extraction can dissolve all or part of the ve As fractions. leaching from excavated rocks. Based on this study, the PPI ob-
It has been reported that As release is controlled by the disso- tained from an SLT may be used to estimate the potential for As
lution of Fe or Al oxides (Xu et al., 2011). The relationships between leaching. We consider this method more practical, economical, and
the mean As and Fe or Al concentrations in the SLT are shown in conservative compared with a column test; however, its feasibility
Fig. S4. Interestingly, Fe or Al release was delayed compared with As and range of application warrant future investigation.
release at the beginning, which is corroborated by the results in
Fig. S1. This is probably because some dissolved Fe and Al colloids 5. Conclusions
precipitated at rst and were then re-released to the leachates in
later steps. Correlation plots between As and Fe or Al (Fig. S5) This study investigated the potential for leaching of As from an
suggest that the Fe or Al oxides could provide a mechanism for As excavated rock into water using SLT and RSSCT. The As leaching (L/
leaching. Taking the SEP results into account, it is inferred that As S 10) was higher in RSSCT than SLT at the beginning, while the
was rst mainly released from the mineral surfaces, as previously amount of As leaching in SLT prevailed over that in RSSCT in the
reported (Wenzel et al., 2001), and then the dissolution of Fe or Al later steps, resulting in higher cumulative As release in the SLT
oxides contributed to the concomitant release of As. DOC plays an (average: 35.6%) than in the RSSCT (average: 25.8%). This study also
important role in controlling As mobility and the complexation of showed that drying treatments (air-drying, oven-drying, and
As by DOC may increase its leaching from soils (Luo et al., 2008). A freeze-drying) lead to higher As leaching compared with that of
good correlation between the mean As and DOC concentrations in fresh rock; freeze-drying increased As leaching to the greatest
the RSSCT (Fig. S6) indicates that DOC also controls the As leaching extent. These results underscore the importance of drying treat-
from this rock. ment selection during the contamination assessment. This study
proposes the concept of PPI to evaluate the pollution threat of As
4.3. Inuence of drying methods on As leaching leaching from excavated rock.

Compared with that of wet rock, the higher release of As after Appendix A. Supplementary data
drying (Fig. 1 and Table 1) was in accord with higher release of Fe,
Al, and DOC (Fig. S1). Drying may generate considerable negative Supplementary data related to this article can be found at http://
pore water pressures between soil particles and therefore result in
greater generation of colloids (Majdalani et al., 2008; Mohanty
et al., 2015)dthe same effect observed for the rock used in this
study. Drying treatments can also increase DOC concentrations in
soil extractions (Kaiser et al., 2001; Merckx et al., 2001), which is Al-Abed, S.R., Jegadeesan, G., Purandare, J., Allen, D., 2008. Leaching behavior of
consistent with our results; however, a larger amount of As was mineral processing waste: comparison of batch and column investigations.
released after FD treatment, despite smaller Fe, Al, and DOC re- J. Hazard. Mater. 153, 1088e1092.
Cappuyns, V., Swennen, R., 2008. The use of leaching tests to study the potential
leases compared with AD and 40D treatments (Fig. S1). Freezing mobilization of heavy metals from soils and sediments: a comparison. Water
causes expansion of water to ice in the rock pores, which are similar Air Soil Pollut. 191, 95e111.
to pores in soils; this expansion may break aggregates and form Cheng, H.F., Hu, Y.N., Luo, J., Xu, B., Zhao, J.F., 2009. Geochemical processes con-
trolling fate and transport of arsenic in acid mine drainage (AMD) and natural
colloids (Oztas and Fayetorbay, 2003). Moreover, as the ice in the systems. J. Hazard. Mater. 165, 13e26.
rock pores sublimates, water movement from nearby pores towards Cornell, R.M., Schwertmann, U., 2003. The Iron Oxides. Structure, Properties, Re-
the freezing fronts lead to new fractures and increases the number action, Occurrence and Uses, second ed. Wiley-VCH, New York.
Crittenden, J.C., Berrigan, J.K., Hand, D.W., Lykins, B., 1987. Design of rapid xed-bed
of active fracture sites (Mohanty et al., 2014). The water-extractable adsorption tests for nonconstant diffusivities. J. Environ. Eng. ASCE 113,
As is mainly associated with surfaces of solid phases, which may 243e259.
explain the higher As release under these conditions. Crittenden, J.C., Reddy, P.S., Arora, H., Trynoski, J., Hand, D.W., Perram, D.L.,
Summers, R.S., 1991. Predicting gac performance with rapid small-scale column
The above results demonstrate that the different drying tests. J. Am. Water Works Assoc. 83, 77e87.
methods potentially lead to under- or overestimation of As leaching Grathwohl, P., Susset, B., 2009. Comparison of percolation to batch and sequential
from an excavated rock. For example, use of fresh samples with no leaching tests: theory and data. Waste Manag. 29, 2681e2688.
Huang, G.X., Chen, Z.Y., Sun, J.C., Liu, F., Wang, J., Zhang, Y., 2015. Effect of sample
pretreatment has been previously suggested for determining As
pretreatment on the fractionation of arsenic in anoxic soils. Environ. Sci. Pollut.
extraction from soils (Sz akova  et al., 2009); however, the As R. 22, 8367e8374.
leaching results from the excavated rock with no pretreatment in Kaiser, K., Kaupenjohann, M., Zech, W., 2001. Sorption of dissolved organic carbon in
this study resulted in underestimation. Moreover, the As contam- soils: effects of soil sample storage, soil-to-solution ratio, and temperature.
Geoderma 99, 317e328.
ination assessment of this rock with respect to pollution level was Kim, J., Hyun, S., 2015. Nonequilibrium leaching behavior of metallic elements (Cu,
easily overestimated if freeze-drying was used, compared with Zn, As, Cd, and Pb) from soils collected from long-term abandoned mine sites.
282 J. Li et al. / Chemosphere 154 (2016) 276e282

Chemosphere 134, 150e158. distribution of arsenic in natural waters. Appl. Geochem. 17, 517e568.
Krger, O., Kalbe, U., Berger, W., Nordhauss, K., Christoph, G., Walzel, H.P., 2012. Supriatin, S., Terrones, C.A., Bussink, W., Weng, L.P., 2015. Drying effects on selenium
Comparison of batch and column tests for the elution of articial turf system and copper in 0.01 M calcium chloride soil extractions. Geoderma 255, 104e114.
components. Environ. Sci. Technol. 46, 13085e13092. Sz
akov a, J., Tlustos, P., Goessler, W., Frkova
, Z., Najmanov a, J., 2009. Mobility of
Luo, W., Lu, Y.L., Wang, G., Shi, Y.J., Wang, T.Y., Giesy, J.P., 2008. Distribution and arsenic and its compounds in soil and soil solution: the effect of soil pretreat-
availability of arsenic in soils from the industrialized urban area of Beijing, ment and extraction methods. J. Hazard. Mater. 172, 1244e1251.
China. Chemosphere 72, 797e802. Tabelin, C.B., Igarashi, T., Yoneda, T., Tamamura, S., 2013. Utilization of natural and
Majdalani, S., Michel, E., Di-Pietro, L., Angulo-Jaramillo, R., 2008. Effects of wetting articial adsorbents in the mitigation of arsenic leached from hydrothermally
and drying cycles on in situ soil particle mobilization. Eur. J. Soil Sci. 59, altered rock. Eng. Geol. 156, 58e67.
147e155. Tabelin, C.B., Hashimoto, A., Igarashi, T., Yoneda, T., 2014a. Leaching of boron, arsenic
Maszkowska, J., Koodziejska, M., Biak-Bielin  ska, A., Mrozik, W., Kumirska, J., and selenium from sedimentary rocks: II. pH dependence, speciation and
Stepnowski, P., Palavinskas, R., Krger, O., Kalbe, U., 2013. Column and batch mechanisms of release. Sci. Total Environ. 473, 244e253.
tests of sulfonamide leaching from different types of soil. J. Hazard. Mater. 260, Tabelin, C.B., Hashimoto, A., Igarashi, T., Yoneda, T., 2014b. Leaching of boron,
468e474. arsenic and selenium from sedimentary rocks: I. Effects of contact time, mixing
Merckx, R., Brans, K., Smolders, E., 2001. Decomposition of dissolved organic carbon speed and liquid-to-solid ratio. Sci. Total Environ. 472, 620e629.
after soil drying and rewetting as an indicator of metal toxicity in soils. Soil Biol. Tatsuhara, T., Arima, T., Igarashi, T., Tabelin, C.B., 2012. Combined neutralization-
Biochem. 33, 235e240. adsorption system for the disposal of hydrothermally altered excavated rock
Ministry of the Environment, 2011. Soil Contamination Countermeasures Law (in producing acidic leachate with hazardous elements. Eng. Geol. 139, 76e84.
Japanese). Wen, F., Hou, H., Yao, N., Yan, Z.G., Bai, L.P., Li, F.S., 2013. Effects of simulated acid
taisyo.pdf. rain, EDTA, or their combination, on migration and chemical fraction distri-
Mohanty, S.K., Saiers, J.E., Ryan, J.N., 2014. Colloid-facilitated mobilization of metals bution of extraneous metals in Ferrosol. Chemosphere 90, 349e357.
by freeze-thaw cycles. Environ. Sci. Technol. 48, 977e984. Wenzel, W.W., Kirchbaumer, N., Prohaska, T., Stingeder, G., Lombi, E., Adriano, D.C.,
Mohanty, S.K., Saiers, J.E., Ryan, J.N., 2015. Colloid mobilization in a fractured soil 2001. Arsenic fractionation in soils using an improved sequential extraction
during dry-wet cycles: role of drying duration and ow path permeability. procedure. Anal. Chim. Acta 436, 309e323.
Environ. Sci. Technol. 49, 9100e9106. Wilson, S.C., Lockwood, P.V., Ashley, P.M., Tighe, M., 2010. The chemistry and
Nystrand, M.I., Osterholm, P., Nyberg, M.E., Gustafsson, J.P., 2012. Metal speciation in behaviour of antimony in the soil environment with comparisons to arsenic: a
rivers affected by enhanced soil erosion and acidity. Appl. Geochem 27, critical review. Environ. Pollut. 158, 1169e1181.
906e916. World Health Organization, 2012. Arsenic. World Health Organization, Geneva.
Oztas, T., Fayetorbay, F., 2003. Effect of freezing and thawing processes on soil
aggregate stability. Catena 52, 1e8. World Health Organization, 1994. Guidelines for Drinking Water Quality, second ed.,
Pedersen, H.D., Postma, D., Jakobsen, R., 2006. Release of arsenic associated with the vol. 1. WHO, Geneva.
reduction and transformation of iron oxides. Geochim. Cosmochim. Acta 70, Xu, W., Wang, H.J., Liu, R.P., Zhao, X., Qu, J.H., 2011. Arsenic release from arsenic-
4116e4129. bearing Fe-Mn binary oxide: effects of Eh condition. Chemosphere 83,
Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behaviour and 1020e1027.