Insights into solid phase characteristics and release of heavy metals and arsenic from industrial sludge via combined chemical, mineralogical, and microanalysis

RESEARCH ARTICLEInsights into solid phase characteristics and release of heavy metals and arsenic from industrial sludge via combined chemical, mineralogical, and microanalysis Received:

RESEARCH ARTICLE

Insights into solid phase characteristics and release of heavy

metals and arsenic from industrial sludge via combined chemical, mineralogical, and microanalysis

Received: 16 May 2014 / Accepted: 10 August 2014 / Published online: 31 August 2014

# Springer-Verlag Berlin Heidelberg 2014

Abstract This study investigates the solid phase

characteris-tics and release of heavy metals (i.e., Cd, Co, Cu, Cr, Mo, Ni,

Pb, and Zn) and arsenic (As) from sludge samples derived

from industrial wastewater treatment plants The emphasis is

determining the influence of acidification on element

mobili-zation based on a multidisciplinary approach that combines

characteri-zation through X-ray diffraction (XRD), field emission gun

electron probe micro analysis (FEG-EPMA), and

thermody-namic modeling (Visual MinteQ 3.0) Solid phase

character-ization and thermodynamic modeling results allow prediction

of Ni and Zn leachabilities FEG-EPMA is useful for direct

solid phase characterization because it provides information

on additional phases including specific element associations

leaching test results indicate that disposal of improperly

treat-ed sludges at landfills may lead to extreme environmental

risks due to high leachable concentrations of Zn, Ni, Cu, Cr,

and Pb However, high leachabilities under acid conditions of

provide a potential opportunity for acid extraction recovery of

Ni and Zn from such sludges

Keywords Heavy metals and arsenic Industrial sludge

FEG-EPMA

Introduction All over the world, urbanization and industrialization have

espe-cially in developing countries such as Vietnam Sludge gen-erated from wastewater treatment plants is usually treated as solid waste and classified as hazardous or non-hazardous

In Vietnam, management options for sludge include land-fill disposal, stabilization/solidification, and incineration

of the sludge and leachable concentrations of contaminants

ap-ply only to sludge generated at factories registered at the National Environmental Protection Agency, the source of

65 % by mass of total sludge production (LBCD & Experco

illegal dump sites Depending on its origin, sludge composi-tion can vary from highly organic (domestic sludge) to

Industrial sludge that contains high quantities of heavy metals (HMs) and As can contaminate soil and groundwater, even

due to the increasing volume of industrially derived sludge

Responsible editor: Philippe Garrigues

T T T Dung:A Golreihan:E Vassilieva:V Cappuyns:

R Swennen

Department of Earth and Environmental Sciences, KU Leuven,

Celestijnenlaan 200E, 3001 Leuven, Belgium

V Cappuyns

Faculty of Business and Economics, KU Leuven, Warmoesberg 26,

1000 Brussels, Belgium

T T T Dung ( *)

University of Science, Faculty of Environment, Vietnam National

University Ho Chi Minh City, 227 Nguyen Van Cu St., W4, D5, Ho

Chi Minh City, Vietnam

e-mail: tttdung@hcmus.edu.vn

N K Phung

Department of Science and Technology, 244 Dien Bien Phu St., W7,

D3, Ho Chi Minh City, Vietnam

DOI 10.1007/s11356-014-3438-y

and associated environmental concerns, HMs and arsenic in

sludge have become a popular research topic (Perez-Cid et al

bulk composition and leaching properties (Orescanin et al

immo-bilize HMs and As in Portland cement, or bio-stabilization via

Understanding chemical and mineralogical properties of

in-dustrial sludge and the factors that may influence release of

HMs and As provides essential information for defining

man-agement options The present study investigates leaching of

HMs (Cd, Co, Cu, Cr, Mo, Ni, Pb, and Zn) and As from

The focus is determining the effect of acidification on element

mobilization Given that the leaching behavior of solid

mate-rials is largely dependent on the mineralogical characteristics

involves identification of possible mineralogical phases in

sludge via X-ray diffraction (XRD) and field emission gun

electron probe micro analysis (FEG-EPMA) Mineralogical

composition can be related to leaching of particular elements

Furthermore, the environmental risk of HMs and As is

strong-ly dependent on chemical speciation We present predictions

of leachate HM and As species distributions based on Visual

MinteQ 3.0 thermodynamic models In this study, the term

“heavy metals” denotes the elements Cd, Cr, Co, Cu, Ni, Pb,

and Zn Arsenic (As), which is actually a metalloid, is

men-tioned separately

Material and methods

Sampling and sample pretreatment

Three industrial sludge samples with different chemical

com-position originating from three wastewater treatment plants

were collected in Binh Duong and Dong Nai provinces

(southern Vietnam) in February 2013 These provinces host

chemical, garment, shoe and leather, metal plating, iron and

of the industrial sludge samples were collected from central-ized wastewater treatment plants in the industrial parks of Dong Nai (sample SI1) and Binh Duong (sample SI2) that accept the effluent from a variety of nearby industrial facto-ries The third industrial sludge (sample SE) originated from

an electroplating wastewater treatment plant in Binh Duong Electroplating sludge is regarded as a hazardous waste in

samples were placed in sealed plastic bags and transported

to KU Leuven, Belgium, for further analysis The moisture contents of samples SI1 and SE were, respectively, determined from the weight difference between the wet and the dry samples (105 °C) The moisture content was 90 % for SI1 and 79 % for SE

Each sample was divided into two portions One portion was then dried in an oven at 60 °C until constant weight The other portion was air-dried The oven-dried samples were used for bulk chemical and mineralogical sample characterization, while the air-dried samples were used for leaching tests Once dry, the samples were ground and homogenized in a porcelain mortar and sieved through a 2-mm sieve

Chemical characterization

geo-chemical data collected using a three-acid digestion method,

analytical grade Total concentrations of the elements Al, Ca,

Fe, K, Mg, P, S, As, Cd, Co, Cr, Cu, Mn, Mo, Ni, Pb, and Zn were then determined via ICP-OES (Varian 720ES) In order

to evaluate the quality of the analytical method, a certified reference material (NIST2782, industrial sludge) and sample duplicates were analyzed The comparison between the mea-sured concentrations with the certified data in the certified

was measured (pH Hamilton single-pore electrode, calibrated

at pH 4 and 7) in a suspension solution of 5.0 g of sludge in 25.0 mL water, following shaking for 2 h The content of

Table 1 Comparison between the measured concentrations with the certified data in the certified reference material (NIST2782, industrial sludge) (average±standard deviation of two replicates)

As (mg/kg)

Cd (mg/kg)

Co (mg/kg)

Cr (mg/kg)

Cu (mg/kg)

Mo (mg/kg)

Ni (mg/kg)

Pb (mg/kg)

Zn (mg/kg)

organic matter was determined by the Walkley and Black

Solid phase characterization

We performed XRD analysis on the original samples and on

the mineralogical composition of the solid phase during the

and graphite monochromator was used Data from

FEG-EPMA analysis were used to complement the XRD

charac-terization For FEG-EPMA analysis, sludge samples were

embedded in a resin and prepared as polished thin sections

sections were examined with a Jeol JXA8530F machine, with

energy dispersive spectrometer (EDS) mode (in spot analysis)

or wavelength dispersive spectrometer mode (WDS) (in

map-ping mode)

Leaching tests

Cascade leaching test

assess the extent of leaching as a function of liquid/solid (L/S)

ratio This is a serial batch test in which material is

succes-sively extracted five times, resulting in L/S ratios of 20, 40, 60,

80, and 100 (L/kg) This test was replaced by compliance test

extractions at liquid-to-solid ratio 2 and 8 L/kg dry matter

However, because our interest is leaching characteristics

rath-er than compliance with environmental standards, we chose to use CLT in the present study Extractions were carried out in duplicate in acid-rinsed 50 mL polyethylene centrifuge tubes with screw caps Thirty milliliters of Milli-Q water, acidified

sample, shaken (for 24 h), centrifuged at 3,000 rpm for

Macherey) No pH adjustment was performed during the test

a pH of 4 to assess the influence of acidic conditions on the release of HMs and As from sludge The test was based on an automatic multititration system (Titro-Wico Multititrator, Wittenfield and Cornelius, Bochum, Germany) Eighty grams

of dried sample was put in an erlenmeyer flask together with

800 mL of Milli-Q water (L/S ratio=10 L/kg) We continually monitored the pH and adjusted it by automatic addition of

maintain the predefined pH (pH 4) for sample SE Therefore,

different samples (SI1 and SI2 samples, 1 mol/L; SE sample, 2.5 mol/L) At regular time intervals (0, 1, 3, 6, 12, 24, 48, 72,

96 h), a sample of the suspension (10 mL) was taken over a

GmbH & Co KG, Germany) by means of a syringe attached

to a flexible tube

Fig 1 XRD patterns of original sample and sample after the pH test (pH=4, sample SI2)

Analysis of leachates

Al, Ca, Fe, K, Mg, P, S, As, Cd, Co, Cr, Cu, Mn, Mo, Ni, Pb,

and Zn by ICP-OES whereas the other part was not acidified,

(IC-Dionex ICS-2000) We used the colorimetric

deter-mine the Cr(VI) content in the leachates Cr measured by

ICP-OES constitutes total Cr Cr(III) was obtained by

subtrac-tion of Cr(VI) from total Cr

Aqueous speciation based on modeling by Visual MinteQ,

version 3.0

We performed calculations based on Visual MinteQ version

3.0 to determine speciation of HMs in the leachate Input data

include concentrations of elements (Al, K, Ca, Mg, Fe, Co, Cr,

used in the calculation The specified redox couples were

Results

General sludge characteristics

Bulk chemical composition

The three sludge samples have a nearly neutral to slightly

P, S, and organic matter (percentage level) were found in all

samples These elements might derive from the use of

chem-ical precipitants and coagulants (e.g., lime, ferrous sulfate,

ferrous phosphate, and poly aluminum chloride (PAC)) in

the chemical remediation step in wastewater treatment (Hsieh

Pb (8,130 mg/kg), likely associated with the battery and steel

processing factory in the industrial park The electroplating

sludge (sample SE) has the highest concentrations of As and

HMs (Co, Cu, Mo, and Zn) and extremely high concentrations

of Cr and Ni (13,208 and 55,732 mg/kg, respectively)

and a comparison with published data from electroplating

composition of industrial sludge may vary widely, depending

on industry type As mentioned, sample SI1 and SI2 originate from centralized industrial wastewater treatment plants of industrials parks that actually treat mixed wastewaters from factories within the industrials parks, making comparison to published data difficult

Regarding the electroplating sludge sample (SE), concen-trations of major elements (Al, Ca, and Fe) and Zn are higher and Mg, Mn, As, Cd, and Cu are lower compared to those reported by other workers Zn concentration is 75 times higher

Cr, Ni, and Pb lie within the same range

Solid phase characterization Understanding the correlation between solid phase properties and leaching behavior of materials requires a detailed knowl-edge of the minerals present together with their chemical

of quartz and gypsum in sample SE is in accordance with

reported the presence of quartz and gypsum as common phases in metal plating sludge Calcite is also a common by-product of the neutralization step by lime in wastewater treat-ment systems due to its precipitation from high calcium

Some phosphate and sulfate minerals are also present:

Mineral phases detected by XRD analysis are consistent with high contents of Al, Ca, Fe, P, and S in all samples in ICP-OES data Apart from Pb, no HMs as discrete mineral phases was detected by XRD We applied FEG-EPMA as a check on the consistency of XRD results and to identify the

the chemical composition from selected EDS spot analysis by FEG-EPMA

In sample SI1, 62 spots were analyzed by FEG-EPMA to examine elemental compositions of the phases Results reveal

phosphorus (spots 3, 4, 5, 6, 7, 12, and 13), and sulfur species Although some HMs (Cr, Ni, and Zn) were detected with FEG-EPMA, no Cd or Co was detected As was only detected

A different distribution of Cr, Ni, and Zn was observed

9, 10, and 11) or in a matrix with Zn, Ni, P and small amount

of Ca, Al, and Fe (spot 12) Ni was found in Zn-rich spots containing P, Al, Fe (spots 12 and 13), and Si (spot 14) Most

) Al (%)

Ca (%)

Fe (%)

P (%)

S (%)

K (mg/kg)

Mg (mg/kg)

Mn (mg/kg)

OC (%)

IR (%)

pH (H

As (mg/kg)

Cd (mg/kg)

Co (mg/kg)

Cr (mg/kg)

Cu (mg

Mo (mg/kg)

Ni (mg/kg)

Pb (mg/kg)

Zn (mg/kg)

Zn-rich spots occurred within P-Fe-rich spots (spots 12, 13

and 14) or in a complex matrix of Na, Al, Si, and Ca with

small amounts of S and Cl (data not shown) WDS mapping

was also used to deduce the micro-scale elemental

phases, namely phases with high Zn-Ni-P-Fe and phases with

high Al-Si content The coexistence of Zn, Ni, P, and Fe, as

are important host phases for Zn and Ni The coexistence of

also identified by XRD

In sample SI2, 42 spots were analyzed in combination with

WDS mapping of selected areas Representative EDS point

consists of a large variety of elements that are mostly P, Si,

Al, and S, most likely coinciding with quartz, aluminum

phosphate and Si, Al-rich phases (spots 15, 16, 17, and 18)

Pb was the only HM that was detected with FEG-EPMA, most

likely because of its elevated concentration (8,130 mg/kg Pb

in SI2) and the heterogeneous distribution of other HMs in this sludge sample In spot analyses, traces of Pb were observed in Ti-rich spots with trace amount of P, Fe, Al, and Si (spot 21) Pb-rich spots were also found in Al-Si-rich spots (spots 19 and 20) and in Al-, P-, S-rich spots containing small amounts of

Fe, Si, and Ca (spot 22) Mapping of selected areas revealed the presence of quartz and lead sulfide by the coexistence of Si-O and Pb-S in the selected area

With regard to sample SE, 61 spot analyses were per-formed Major phases in SE comprise calcite (spots 23 and 24) and metal-rich spots with Fe, Ni, Sn, and Zn (spots 29 and 30) Results reveal an elevated content of some HMs, includ-ing Ni, Zn, Cr, and Sn Cd, Co, and Mo were not detected Zn-rich spots were identified in 30 of the 61 analyzed spot which

is in accordance with the fact that Zn is the most abundant HM

varying contents of Al, Fe and Sn and, in some cases, traces of

Table 3 Chemical composition (wt%) of selected spots by FEG-EPMA, the spot numbers refer to the numbers given in Fig 2

Cr (spots 25, 26, and 28), S (spots 25 and 26), and As and Pb

(data not shown) were found Arsenic was also observed in a

Ca-Mg-O matrix (spots 23 and 27)

Leaching tests

Although many different elements were measured, the

follow-ing discussion mainly focuses on As, Cd, Co, Cr, Cu, Ni, Mo,

Pb, and Zn because of their potential toxicity Major elements

mentioned because of their relevance for interpretation of

release mechanisms of elements of interest

Cascade leaching tests

leachates are below detection limit in all three samples In the

CLT, the final pH of the leachates varied from step to step

extent of pH change mainly depends on the acid neutralization

capacity (ANC) of the samples From this result, we deduce

that sample SE has the highest ANC because its pH change

the cumulative leachability (sum of five extractions steps), expressed in percent of an element leached relative to its total content in a sample

In sample SI1, the leachability is in the following order: S

(2 %)>Zn (1 %)>Cu≈P (0.1–0.2 %) The leachability of Al,

Fe, and Cr is negligible (<0.02 %) Even though a relatively low cumulative leachability of Ni and Zn was observed (as %

of their total content in the sludge sample), special attention should be paid to these elements because of the high absolute concentrations (Ni, 322 mg/kg; Zn, 861 mg/kg) cumulatively released during the CLT Speciation analysis of Cr indicates that the leachates of sample SI2 did not contain Cr(VI) The concentrations of Pb in the leachates were below detec-tion limit

In sample SI2, we observed a considerable release for almost all analyzed elements Among HMs, Mo showed the highest leachability (60 %) The lowest leachability was mea-sured for Cu and Cr (2 %) In descending order, leachability for the other HMs is as follows: Ni (14 %)>Co (9 %)>Zn (4 %) Speciation analysis of Cr indicates that the leachates of sample SI2 do not contain Cr(VI)

In general, sample SE displays a lower leachability of elements compared to SI1 and SI2 Similar to SI2, Mo showed

Fig 2 Graph of selected spots for EDS analysis and area for WDS mapping analyzed by FEG-EPMA From left to right: a SI1; b WDS mapping of selected area revealed the coexistence of Ni, P, and Fe in SI1; c SI2; d SE The spot numbers refer to the numbers given in Table 3

the highest release among the HMs (23 % of the total content

was released) The leachability of Zn, Ni, and Cu is negligible

(<0.1 %) while leachabilities of Cr and Co are in the range 1–

2 % The concentration of Pb in the leachates of sample SE is

below detection limit The release of Cr appears to correspond

mainly to Cr(VI) because Cr(VI) is the main species measured

in the leachate Therefore, Cr should be taken into

consider-ation because of the high absolute cumulative concentrconsider-ation

leached (122 mg/kg) and the occurrence of Cr(VI) in the

leachate The analysis of Cr speciation shows that cumulative

concentration leached Cr from sample SE was dominantly

Cr(VI) (108 mg/kg)

the following section, leachability refers to leachable

relative to total concentration

Acid neutralization capacity, major elements, and anions

We calculated ANC based on the quantity of acid added to

the highest ANC (at 96 h) was observed in sample SE

(5,627 mmol/kg), followed by samples SI1 (910 mmol/kg) and

SE most likely contributes to the high ANC of this sample This result agrees with the CLT result that sample SE has the highest ANC In all three sludge samples, despite the high total concentration, Al and Fe exhibit very low leachability (<1 %) compared to other major elements, such as Ca, K, Mg, Mn, and

S (>6 %) This suggests that no significant dissolution of

Sample SI1 shows the lowest leachability of major elements

samples show the following order for released amount: SI2>

Heavy metals and As

leachabil-ity is high (9–31 % of the total content) This is consistent with

the acid-soluble fraction of industrial sludge samples, suggest-ing release at low pH

Cu exhibits a medium leachability (1–5 %), while Cr and

Pb show a low leachability (<1 %) Cr(VI) occurs as an anion, which may account for its relatively stronger bonding to positively charged components (e.g., surfaces of Fe(hydr)oxides, organic matter) Speciation analysis on Cr

Fig 3 Cumulative amount leached of HMs including Cr(VI) (sample SE), and SO4−, and pH change during the cascade leaching test

shows that most dissolved Cr in the leachates in SE is Cr(VI)

(up to 78–98 % relative to total dissolved Cr) We did not

detect Cr(VI) in the leachates of sample SI1 and SI2

small amounts (0.2–0.5 mg/kg) in sample SE No As was

detected in samples SI1 and SI2 Mo concentrations were also

below detection limit for the three samples

From an environmental point of view, if such types

of sludge are disposed of landfills without proper

treat-ment, considerable release of Cu, Ni, and Zn may be

expected Acidification of sludge implies extreme

envi-ronmental risk, especially for sludges similar to samples

Aqueous speciation based on modeling by Visual MinteQ

version 3.0

Thermodynamic modeling results reveal a great variety of

chemical species depending on the composition, pH, and

redox potential of leachates Most elements form free metal

ions and complexes with hydroxides, sulfates, chlorides,

form the dominant species types in the leachates of all three samples Pb-sulfate is the main species in sample SI2, whereas Pb-nitrate complex was observed as dominant species in sample SE

Cr is dominantly present as trivalent in all three samples

main species in sample SI1 and SE, while Cr(III)-sulfate is the dominant soluble species (45 %) in sample SI2

sample SE (2.8 mg/l) and accounted for up to 98 % of total Cr measured by chemical analysis, thermodynamic modeling predicts a negligible amount of Cr(VI) (<0.001 %)

Thermodynamic modeling results suggest that Cd mainly

where-as Cd-thiosulfate is the most abundant species in sample SI2

Discussion Solid phase characterization

We performed XRD phase analysis on residual sludge

Fig 4 Release of major elements, ANC and anions in pHstatleaching test at pH 4 Some components were multiplied by a factor of 10, 100, or 1,000 for

a better visualization

changes to major solid phases Changes include the

such as gypsum dissolve easily from solid waste material in

the presence of water and acidic conditions

disso-lution and then used Pb concentrations to calculate the mass of

lanarkite in the samples Based on this assumption, we

per-formed a Visual MinteQ calculation of lanarkite solubility at

pH 4 and found that lanarkite dissolves completely at pH 4

Nevertheless, XRD analysis detected lanarkite at pH 4,

show-ing that dissolution was partial

XRD analysis alone is insufficient for understanding the

mineral characteristics of the sludge samples because XRD

generally detects only abundant (>2 % of dry sample weight

with Rietveld corrections) minerals with well-defined

and As adsorbed on host minerals such as oxides, carbonates,

sulfates, and phosphates or as inclusions in amorphous phases

these deficiencies As mentioned in the previous section

identification of phosphosiderite; we used EPMA to identify

phase for Zn and Ni (sample SI1) Similar leaching patterns of

hypothesis that they come from the same host phase Their rapid release on exposure of the samples to acid

incorporated within phosphosiderite but were adsorbed

on its surface (sample SI1)

We observed lead sulfide via FEG-EPMA, but not by XRD, most likely because its amount is below the detection limit (sample SI2) A large number of Zn-rich spots were observed by FEG-EPMA, although no mineral related to Zn was identified by XRD, suggest-ing that sample SE contained a large quantity of poorly crystalline Zn phases This conclusion is supported by the fact that a high leachability of Zn was observed in a

with amorphous or poorly crystallized phases and easily leached from industrial sludge Our results also show that FEG-EPMA can be applied to complement XRD analysis by identification of phases that are not

Fig 5 Release of HMs in pHstatleaching test at pH 4 The concentration of some elements was multiplied by a factor of 10, 100, or 1,000 for a better visualization