mine drainage kinetics

ORIGIN PAPER

Pilot-scale tests to optimize the treatment of net-alkalinemine drainage

Min Jang • Hyunho Kwon

Received: 20 March 2010 / Accepted: 3 August 2010 / Published online: 3 November 2010

� Springer Science+Business Media B.V. 2010

Abstract A pilot-scale plant consisting of an oxi-

dation basin (OB), a neutralization basin (NB), a

reaction basin (RB), and a settling basin (SB) was

designed and built to conduct pilot-scale experiments.

With this system, the effects of aeration and pH on

ferrous oxidation and on precipitation of the oxidized

products were studied systemically. The results of

pilot-scale tests showed that aeration at 300 L/min

was optimum for oxidation of Fe(II) in the OB, and

the efficiency of oxidation of Fe(II) increased linearly

with increasing retention time. However, Fe(II) was

still present in the subsequent basins—NB, RB, and

SB. Results from pilot-scale tests in which neutral-

ization was excluded were used to obtain rate

constants for heterogeneous and homogeneous oxi-

dation. Oxidation of Fe(II) reached almost 100%

when the pH of the mine drainage was increased to

more than 7.5, and there was a linear relationship

between total rate constant, log (Ktotal), and pH.

Absorbance changes for samples from the NB under

different pH conditions were measured to determine

the precipitation properties of suspended solids in the

SB. Because ferrous remained in the inflow to the SB,

oxidation of Fe(II) was dominant initially, resulting

in increased absorbance, and the rate of precipitation

was slow. However, the absorbance of the suspension

in the SB rapidly dropped when pH was higher than

7.5.

Keywords Pilot-scale � Oxidation � Precipitation �Net alkalinity � Mine drainage

Introduction

In South Korea, there are approximately 2,500 mines,

including 906 metal mines, 379 coal mines, and 1,173

non-metal mines. Among these mines, approximately

150 abandoned surface and underground coal mines

are responsible for acid drainage discharges into

water streams, and are continuously contaminating

surface and groundwater. An estimated 100,000 tons

per day of mine drainage is being discharged from

these mines into water streams (CIPB 1995; Jung

2003). Acid mine drainage (AMD) from abandoned

metal mines not only pollutes the natural environ-

ment, for example surrounding soil, and surface and

ground water, but also has toxic effects on crops and

humans as a result of concentration of the contam-

ination (Jung 1994).

In general, drainage from coal mines is not only of

low pH but also contains high levels of SO42- and

heavy metals, including Fe, Al, and Mn (Sengupta

M. Jang (&) � H. Kwon

Institute of Mine Reclamation Technology, Korea Mine

Reclamation Corporation, Coal Center, 30 Chungjin-dong

Street (Susong-dong), Jongno-gu, Seoul 110-727,

South Korea

e-mail: [email protected]

123

Environ Geochem Health (2011) 33:91–101

DOI 10.1007/s10653-010-9353-3

1993). These properties of mine drainage disrupts

stream ecosystems and further aggravate the problem

by creating yellow or white sediments (Chon and

Hwang 2000). Thus, mine drainage treatment is

necessary to prevent pollution of watercourses with

iron/aluminum precipitates and acidity.

Basically, mine drainage treatment utilizes the

chemical properties of the main contaminants, for

example Fe, Al, and Mn. Two different conventional

methods are mainly used to treat mine drainage—

active or physicochemical treatment (which needs

energy and chemical input) and passive or biological

treatment (which does not need additional input of

energy or chemicals) (Johnson and Hallberg 2005;

Akcil and Koldas 2006). Most active treatment

involves pH adjustment, by adding an alkaline

material, and removal by precipitation as a result of

the formation of oxy/hydroxides. pH adjustment is

helpful for treatment of large quantities of AMD.

Although active treatment enhances treatment effi-

ciency by use of chemicals, there is a large economic

burden, because of the high cost of maintenance and

chemicals, and this process requires continuous

operation (Johnson and Hallberg 2005). Compared

with active treatment, passive systems are econom-

ical but require longer retention times and greater

space; they are, therefore, not appropriate for treat-

ment of large-scale mine drainage. Although passive

treatment has been implemented on full-scale sites in

several countries, treatment efficiency can be uncer-

tain, because of seasonal changes in flow rate and

temperature, and the systems are apt to fail during

long-term operation. As the primary mechanism of

metal removal from acidic effluents, passive treat-

ment uses the affinity properties of metal compounds

with sulfur or carbonate ions; metals are also

removed by adsorption or precipitation with organic

compounds (Gazea et al. 1996). Not only biological

treatment using microbial activity but also chemical

treatment using the sensitivity of metal species to

changes in pH is the main approach of the passive

system. Metal species occur as ions under acidic

conditions; however, they are converted to hydrox-

ides when pH increases. In South Korea, sites with

large-scale mine drainage flows occur in uneven and

narrow terrain. Passive treatment of these contami-

nated sites has the setback of needing large treatment

areas (Hedin 2008). Accordingly, considering the

special status of Korea and two different treatment

concepts with advantages and disadvantages, a com-

plementary new approach should be developed to

minimize maintenance and chemical input and

enhance treatment efficiency.

The objective of this study was to investigate the

applicability of aeration and pH adjustment for

treatment of mine drainage with net alkalinity on a

large scale by conducting pilot-scale tests. A pilot-

scale (30 m3/day) plant was designed and built to

conduct experiments on the large-scale treatment of

the mine drainage (about 6,000 m3/day). The main

themes were to determine the effects of aeration,

retention time, and pH change on the oxidation of

ferrous, and the effect of ferrous oxidation on

precipitation of the oxidized products. The results

of pilot-scale tests could be expected to contribute to

the design of a full-scale plant on the actual site by

constructing the optimized process on the uneven

ground of the site.

Materials and methods

Pilot-scale plant setup

With a scale of 30 m3/day, a pilot-scale treatment

facility was constructed and set up to treat AMD

which was continuously flowing from Jami (JM) adit

of Hamback coal mine. As shown in Fig. 1, the pilot-

scale plant mainly consisted of an oxidation basin

(OB, 1 m3), a neutralization basin (NB, 0.4 m3), a

reaction basin (RB, 0.4 m3), and a settling basin (SB,

1 m3). At the bottom of the OB, nine air diffusers

(disk-type) were set up at equal intervals to produce

air bubbles as homogeneously as possible. The rate of

aeration from all the diffusers ranged from 0 to

450 L/min and was controlled by a mechanical

flowmeter. The inlet flow rate was fixed by control-

ling the pumping pressure of a submersible pump and

a valve in an equalization basin (0.15 m3). An

agitational device (150–250 rpm) was set up in both

the NB and RB for homogeneous mixing. The round-

shaped SB has a circular sludge collector, including

scraper and sludge hopper, and a surface-loading rate

of 21 m3/m2 per day. The rotation speed of the

scraper was 0.4 rpm. The sludge could be discharged

by means of a sludge pump. To control the pH of

mine drainage in the NB, a pH controller was set up

with a pH probe and a metering pump. The target pH

92 Environ Geochem Health (2011) 33:91–101

123

of mine drainage was automatically adjusted with an

error of ±0.1. Lime (Ca(OH)2) was used as neutral-

izing agent. Lime slurry was prepared with a lime

concentration of 0.7 g/L and injected at different flow

rates (max. 0.25 L/min) into the NB.

In this study, pilot-scale tests were conducted to

estimate not only Fe(II) oxidation efficiency but also

the efficiency of precipitation of oxidation products.

As shown in Table 1, these tests were conducted

under different operating conditions. In the treatment

process, the main operating conditions were rate of

aeration of the OB, retention time, and pH of the NB.

Analysis and physicochemical properties

of mine drainage

Samples obtained from the inlet to the pilot plant and

of the effluent from each basin were analyzed for

physicochemical properties. The pH, ORP, TDS, and

EC of mine drainage and the samples were measured

with a portable pH/ORP and EC/TDS meter (Orion 3

star) and DO meter (Pocket calorimeter, HACH). To

evaluate the effectiveness on the oxidation of Fe(II)

and precipitation of oxidative products, and to

measure total recoverable and dissolved metals of

samples obtained before and after treatment, EPA

method 3005A (acid digestion procedure) was used

(EPA 1992). For analysis of total recoverable metals,

40 mL drainage sample was measured with a mea-

suring cylinder and transferred to a conical tube, and

0.8 mL conc. HNO3 and 2 mL conc. HCl were

injected. The samples were then digested by use of a

heating block at 90–95�C until the sample volume

reduced to 8–10 mL. The digested samples were

cooled to room temperature, and the volume was

measured and diluted to 40 mL with de-ionized

water, followed by filtration through a 0.45-lm pore

membrane filter. Concentrations of heavy metals in

the filtrate were measured by ICP-AES (720-ES,

Varian). The alkalinity of samples of pH [ 4.5, was

measured as follows:

addition of an indicator (bromocresol green methyl

red) to the sample;

titration of the sample with H2SO4 solution of

appropriate concentration, by use of a titrator adapted

with a cartridge, until the color of the solution

changed to light pink; then

measurement of the volume of H2SO4 solution

titrated to calculate the alkalinity.

To measure the Fe(II) concentration on-site, the

sample was filtered using a 0.45-lm pore membrane

filter, then ferrozine iron reagent was added into the

filtrate, and after 3 min Fe(II) was measured by UV–

visible absorption (DR 2800, Hach) at 510 nm. Table 1

shows the physicochemical properties of the mine

drainage. The mine drainage has a neutral pH (6.6) and

93 mg/L alkalinity. Use of Eq. (1) showed the mine

drainage had net alkalinity (45 mg/L).

Net alkalinity ¼ alkalinitymeasured � aciditycalculated

ð1ÞThe dissolved oxygen (7.8 mg/L) in the mine

drainage was also highly indicative of relative

oxidative condition. Total dissolved solids (TDS) is

Fig. 1 Schematic diagram

of pilot-scale mine drainage

treatment process

Environ Geochem Health (2011) 33:91–101 93

123

the total amount of cations and anions, and has a

large value when the concentration of sulfate in mine

drainage is high. Considering the analytical results

for DO, pH, Fe(II), and TDS, the mine drainage was

not highly contaminated, and had net alkalinity. The

analytical results for total recoverable metals in

mine drainage showed that the main metal species

([1 mg/L) in mine drainage were Fe, Mn, Si, Ca, and

Mg (Table 2). As the main metal species for the

occurrence of turbidity, the concentration of total

recoverable Fe was 25.3 mg/L. On-site measurement

for Fe(II) was 11.15 mg/L so the ratio of Fe(II) to

total recoverable Fe was low—44.1%. In other words,

the mine drainage sampled was highly oxidized.

Pilot-scale experimental setup

In this study, pilot-scale tests were conducted to

estimate not only Fe(II) oxidation efficiency but also

the efficiency of precipitation of the products of

oxidation. As shown in Table 3, these tests were

conducted under different operating conditions. In the

treatment process, the main operating conditions

were aeration rate, retention time, and pH.

Results and discussion

Kinetics of dissolved oxygen and Fe(II) oxidation

in the OB

Changes in dissolved oxygen (DO) and Fe(II) oxida-

tion were studied as a function of time. In the study of

the kinetics the aeration rate was set constant at

200 L/min and the flow rate of mine drainage was

0.23 L/s. In this test the retention time in the OB was

1.2 h. Figure 2 shows the change of pH, DO, and

ferrous oxidation efficiency with time in the OB.

About 1 h was taken to increase pH from 6.75 to 7.14

whereas only half hour was taken to increase DO

from 7 mg/L to 11 mg/L (Fig. 2a, c). Oxidation of

Fe(II) increased linearly to approximately 28% in 1 h

then decreased slightly. Accordingly, pH has the

dominant effect on the ferrous oxidation efficiency.

Therefore, because the ferrous oxidation efficiency

was rather low under these experimental conditions,

different rates of aeration was applied in the OB to

observe the efficiency of oxidation of Fe(II).

Effect of aeration rate on the efficiency

of oxidation of Fe(II)

To determine the effect of aeration rate on Fe(II)

oxidation efficiency, aeration rate was altered in the

range 0–450 L/min and the flow rate of mine drainage

and retention time were fixed at 0.23 L/min and 1.2 h

in the OB. Without aeration in the oxidation basin, the

efficiency of oxidation of the entire process was less

than 10% (Fig. 3). As the aeration rate increased to

300 L/min, the efficiency of oxidation of Fe(II)

increased to approximately 50 and 75% in the OB

and SB, respectively. However, the efficiency of

oxidation of Fe(II) in the OB did not increase further

when the rate of aeration was increased to more than

300 L/min. For all rates of aeration used in the OB,

Table 1 Physicochemical properties of mine drainage

Temp. (�C) pH ORP (RmV) TDS (mg/L) EC (ls/cm) Fe2? (mg/L) filtered Alkalinity (mg/L) DO (mg/L)

11.8 6.6 237.1 245 500 11.15 93 7.78

Table 2 Concentrations of total recoverable metals and anions in mine drainage (unit: mg/L)

Al As Ca Cd Co Cr Cu Fe K Sb

0.023 0.003 60.9 ND 0.007 0.005 ND 25.28 1.66 0.005

Li Mg Mn Mo Na Ni P Pb S Si

0.04 16.28 1.39 ND 0.91 0.01 0.022 ND 69.39 3.55

Sn Sr Ti U V Zn F- Cl- NO2- NO3

-

0.96 1.03 0.001 0.047 0.001 0.01 0.501 0.919 0.167 ND


123

the efficiency of oxidation of Fe(II) increased as mine

drainage flowed further into other basins. This might

be because of the increased retention time of mine

drainage (NB: 0.48 h, RB: 0.48 h, and SB: 1.2 h).

Among all cases, the optimum aeration rate for

achieving maximum Fe(II) oxidation in the SB was

300 L/min. Thus, the following pilot-scale tests were

conducted with 300 L/min aeration rate.

Effect of retention time on the efficiency

of oxidation of Fe(II)

The efficiency of oxidation of Fe(II) in each basin

was further observed with different retention times

(0.8, 1.2, 2.0 h) (Fig. 4). The aeration rate was fixed

at 300 L/min. The efficiency of oxidation of Fe(II) in

Table 3 Phases of pilot-scale tests

Phase Type of test Fixed conditions

I Kinetic test for Fe(II) oxidation efficiency and DO change Aeration: 200 L/min

Retention time: 1.2 h

II Dependence of Fe(II) oxidation efficiency in each basin on the aeration

rate (0–450 L/min) in the oxidation basin


III Dependence of Fe(II) oxidation efficiency on retention time (0.8, 1.2, and 2 h) Aeration rate: 300 L/min

IV Dependence of Fe(II) oxidation efficiency on the pH (no adjustment,

7.5, 8.0, and 8.5) in the neutralization basin

Aeration rate: 300 L/min


0

10

20

30

40

pH

6.8

7.0

7.2A

B

Time (min)0 20 40 60 80 100

Dis

solv

ed o

xyg

en (

mg

/L)

Oxi

dat

ion

eff

icie

ncy

of

ferr

ou

s (%

)

6

8

10

12C

Fig. 2 Effect of aeration in the oxidation basin on a dissolved

oxygen, b pH, and c Fe(II) oxidation efficiency (aeration rate:

200 L/min, retention time: 1.2 h)

Time (hr)

Oxi

dat

ion

eff

icie

ncy

of

ferr

ou

s (%

)

00 2 31

20

40

60

80

100

control

200 L/min

450 L/min

300 L/min

400 L/min

OB NB RB

SB

Fig. 3 Fe(II) oxidation efficiency in each basin according to

aeration rate in the oxidation basin


123

the OB and SB was approximately 22 and 38%,

respectively, for 0.8 h RT. As the RT increased to 1.2

and 2.0 h, the efficiency of oxidation of Fe(II)

increased linearly to approximately 50 and 72% in

the OB. This result might be mainly because the pH

increase as a result of aeration enhances the rate of

oxidation of Fe(II) for mine drainage with net alka-

linity. Although 100% oxidation efficiency was not

observed in this series of experiments, the maximum

efficiency of oxidation of Fe(II) (91%) in the SB was

obtained when the RT was 2 h. Considering the

treatment for large-scale mine drainage, the volume

of the oxidation reactor will increase in proportion to

increasing retention time. Thus, a long retention time

for aeration could be an economic disadvantage.

Subsequent tests were conducted with a constant

retention time (1.2 h).

Evaluation of homogeneous and heterogeneous

oxidation

Abiotic oxidation of Fe(II) takes place by homoge-

neous and heterogeneous processes simultaneously

(Singer and Stumm 1968; Millero et al. 1987). In a

suspended solution containing Fe(III) minerals, the

surface of Fe(III)(oxy)hydroxide minerals can act as a

catalyst for Fe(II) oxidation (Azher et al. 2008). In the

heterogeneous oxidation of Fe(II), the soluble Fe(II) is

adsorbed on the surface of Fe(III)(oxy)hydroxide

minerals and the adsorbed Fe(II) is oxidized by an

electron-transfer reaction enhanced by electron-poor

sites on the surface (Sung and Morgan 1980; Park and

Dempsey 2005). The kinetics of homogeneous and

heterogeneous oxidation of Fe(II) are highly interde-

pendent. As pH increases, the hydrolyzed iron species

(FeOH? and Fe(OH)2) predominate over Fe2? ions,

because of hydrolysis (Barnes et al. 2008). As

hydrolysis increases, the kinetic rate increases.

In the presence of dissolved oxygen (DO), the

homogeneous oxidation of Fe(II) can be expressed as

shown in Eq. (2). The oxidized species, Fe(III), is

instantly hydrolyzed by the supply of alkalinity to

Fe(III)(oxy)hydroxide compounds which eventually

precipitate (Eq. 3).

Fe(II) þ 1

4O2 þ

1

2H2O ¼ Fe(III)þ OH� ð2Þ

Fe(III) þ 3H2O ¼ Fe(OH)3ðs)þ 3Hþ ð3ÞIn the oxidation of Fe(II), oxidation of dissolved

Fe(II), including Fe2?, FeOH?, and Fe(OH)20, is

called as homogeneous oxidation and the oxidation of

adsorbed Fe(II) as heterogeneous oxidation.

The rate of homogeneous oxidation of Fe(II) was

expressed by Stumm and Lee (1961) as:

Homogeneous rate ¼ �d½FeðIIÞ�dt

¼ k1½Fe2þ�½O2�fHþg2

ð4Þ

where, k1 is the rate constant for homogeneous

oxidation of Fe(II).

Sung and Morgan (1980) and Tamura and Nagay-

ama (1976) expressed the rate of heterogeneous

oxidation in terms of [H?] (Eq. 4):

Heterogeneous rate¼�d½FeðIIÞ�dt

¼ k2½Fe3þ�½O2�fHþg ð5Þ

The homogeneous and heterogeneous oxidation

processes can be affected by pH in the range between

5 and 8, and the overall oxidation rates can be

expressed by Eq. 5.

Overall abiotic rate ðfor pH [ 5Þ ¼ �d½FeðIIÞ�dt

¼ ðk1 þ k2½Fe3þ�½Hþ�Þ½Fe2þ�½O2�fHþg2

ð6Þ

If all conditions are assumed to be constant except

the amount of Fe(II), this system follows pseudo-first

order reaction kinetics. The total rate constant, ktotal,

can be obtained by use of Eq. 7:

Time (hr)0.0 1.5 3.0 4.5 6.0

Oxi

dat

ion

eff

icie

ncy

of

ferr

ou

s (%

)

0

20

40

60

80

100

RT 0.8hr at OBRT 1.2hr at OBRT 2.0 hr at OB

OB

OB

OB

NB RB

SB

NB

SB

NB RB

SB

Fig. 4 Fe(II) oxidation efficiency in each basin with different

retention times


123

ktotal ¼½FeðIIÞinfluent � FeðIIÞeffluent�Qtotal

Fe(II)effluent Vreactor

ð7Þ

where, Vreactor, Qtotal, Fe(II)influent, and Fe(II)effluent are

the reactor volume, flow rate, Fe(II) concentration in

the influent, and Fe(II) concentration in the effluent,

respectively. By coupling with Eq. 6, Eq. 8 is

obtained.

�d½Fe(II)]dt

½Hþ�2

½Fe2þ�½O2�¼ k2½Fe3þ�½Hþ� þ k1 ð8Þ

Since ktotal ¼�d½FeðIIÞ�

dt½Fe2þ�

; the Eq: ð9Þ could be obtained

ktotal½Hþ�2

½O2�¼ k2½Fe3þ�½Hþ� þ k1 ð9Þ

By plotting [Fe3þ�½Hþ� againstktotal½Hþ�2

½O2�the slope

(k2) and intercept (k1) were obtained by linear

regression (Fig. 5).

The result of linear regression for these data showed

the rate constant for heterogeneous oxidation was

7 9 10-14 (mg/L)-1 s-1, which was lower than litera-

ture values (0.167 9 10-8 to 3.1 9 10-8 (mg/L)-1 s-1).

For homogeneous oxidation, the rate constant was

2 9 10-20 mol L-1 s-1, which was also lower than

literature values (1 9 10-14 to 1 9 10-12 mol L-1 s-1;

Tamura and Nagayama 1976; Sung and Morgan 1980;

Millero et al. 1987; Liang et al. 1993; Stumm and Morgan

1996; Ames 1998; Dempsey and Dietz 2001). These

discrepancies might be because the determination coef-

ficient (0.134) of linear regression was much lower than

0.7. In general, a determination coefficient (R2) of more

than 0.7 in linear regression analysis can be regarded as

indicative of a significant relationship (Bernick et al.

1995; Kalnicky and Singhvi 2001), and R2 = 1 is an ideal

case without any scattered data. Accordingly, this

analysis might not be a reliable means of obtaining rate

constants for homogeneous and heterogeneous oxidation.

Dempsey and Dietz (2001) measured the rate

constant of Fe(II) oxidation using a sludge return

method in which a broad range of Fe(III) (up to

2,000 mg L-1) was used to obtain the rate constants.

Therefore, the low reliability obtained in this study

might be because of the low concentration of Fe(III)

applied in the system. Thus, more tests will be

conducted using sludge to obtain more reliable and

exact rate constants.

Figure 6 shows relationships between pH, alkalin-

ity, and dissolved oxygen measured throughout pilot

tests in which neutralization was excluded. As shown

in Fig. 6, pH and alkalinity were in the range 6.7–6.8

and 92–105 mg/L, respectively (average. 93 mg/L as

CaCO3) for non-aeration. However, when the OB was

aerated, pH increased to 6.85–7.2 and alkalinity

decreased to 70–85 mg/L (avg. 75 mg/L as CaCO3).

Based on stoichiometric consideration of Eqs. 1 and

2, 1.8 mg/L alkalinity as CaCO3 could be removed by

1 mg/L Fe(II) oxidation and precipitation of the

oxidation products. Considering the oxidation of

approximately 90% of total Fe(II) which corresponds

to approximately 10 mg L-1 Fe(II), alkalinity should

be reduced by 18 mg/L. Thus, this result is a good

match with the stoichiometric concept.

Fe (III) conc. (mg/L)0 5 10 15 20 25 30

To

tal r

eact

ion

rat

e (K

tota

l, s-1

)

0.0000

0.0001

0.0002

0.0003

0.0004

[Fe (III)][H+]

1e-6 1e-6 2e-6

Kto

tal[H

+ ]2 /[O

2]

0

1e-19

2e-19

3e-19

y = 7E-14x + 2E-20R2 = 0.134

y = 1E-05x - 2E-05R2 = 0.3066

A

B

Fig. 5 Relationships between Fe(III) concentration and Ktotal

(a) and between Ktotal[H?]2/[O2] and [Fe(III)][H?] (b)


123

Effect of pH adjustment in the NB

on the efficiency of oxidation of Fe(II)

In this test, the efficiency of oxidation of Fe(II) in each

basin was observed when the pH in the NB was

adjusted to the predetermined pH by use of lime.

When the pH in the NB was increased to more than

7.5, the efficiency of oxidation of Fe(II) in the NB and

RB reached more than 95% and almost 100% in the

SB (Fig. 7). Accordingly, the oxidation speed of

Fe(II) became rapid because of the increase of the pH

of the mine drainage. On the basis of this result,

oxidation of Fe(II) in each basin was significantly

affected by pH. After obtaining log (Ktotal) at the

corresponding pH, fitting of the data by linear

regression analysis was conducted to determine the

correlation between pH and rate constant of Fe(II)

oxidation. The determination coefficient (0.8395)

obtained by this analysis was higher than 0.7,

indicating there is a significant relationship between

the overall rate constant and pH. The slope (2.1701) of

the plot fitting the data seems to match well with

literature reports that the rate constant increases by

two orders as pH increases by 1.0 unit (Sung and

Morgan 1980; Azher et al. 2008). In general, the

kinetic constant is very dependent on ionic strength

and dissolved counteranions, and decreases when

ionic strength and anion concentration increase (Em-

menegger et al. 2008). Accordingly, the alkalinity and

sulfate present at high levels in the mine drainage

might reduce the rate constant, as shown in Fig. 8.

Change in the absorbance properties

of suspensions in the NB with pH variation

To determine the precipitation properties of suspended

solids, approximately 10 mL suspension obtained in

the RB under different pH conditions was placed in a

cell and the absorbance of the sample was measured

kinetically by UV–visible absorbance (DR 2800,

Hach) at 560 nm (Lenter et al. 2002). Before mea-

surement of the suspension, the blank was the filtrate

obtained by use of a 0.45 lm-pore filter. Generally, the

absorbance increased with increasing suspended solids

Dissolved oxygen (mg/L)

6 8 10 12

pH

6.6

6.8

7.0

7.2

7.4

Alk

alin

ity

(mg

/L)

60

70

80

90

100

110

DO vs pH DO vs alkalinity

aeration

Fig. 6 Relationship between pH, alkalinity, and dissolved

oxygen measured throughout the pilot tests (neutralization

excluded)

Time (hr)3

Oxi

dat

ion

eff

icie

ncy

of

ferr

ou

s (%

)

010 2

20

40

60

80

100

pH not adjustedpH 7.5pH 8.0pH 8.5

OB

NBRB

SB

Fig. 7 Fe(II) oxidation efficiency in each basin for different

pH in the neutralization basin

pH6.5 7.0 7.5 8.0 8.5 9.0

log

(Kto

tal,

s-1)

-5

-4

-3

-2

-1

0

log(Ktotal

) = 2.1701 pH - 19.208

K = 0.0551x1014[OH-]2

(Azher et al., 2008)

K = 0.061x1014[OH-]2

(Sung and Morgan., 1980)

Fig. 8 Correlation between pH in the NB and log (Ktotal)

(R2 = 0.8395)


123

(SS) and decreased with precipitation of the SS. By

measuring SS and absorbance for specific suspensions

it was found they had a linear relationship (SS =

58.6 Absat 560nm - 1.08, R2 = 0.996). The objective

of this test was to observe not only precipitation of the

SS but also the effect of Fe(II) oxidation on precip-

itation (Fig. 9).

After measuring the absorbance of the suspension

kinetically, the agglomeration equation was used to

obtain the rate of removal of SS (Stumm and Morgan

1996). In the agglomeration equation, N represents

the number of particles in solution. In this study,

however, the rate of removal (ka) was obtained by

applying the absorbance instead of the particle

numbers, on the basis of the assumption that the

absorbance increases linearly as the number of

particles increases. The initial absorbance (N0) and

absorbance at different times (N) were used.

�dN

dt¼ kaN2 ð10Þ

1

N� 1

N0

¼ kat ð11Þ

The absorbance of suspension when the pH was not

adjusted in the NB increased from approximately 0.26

to 0.48 in 1.5 h. Although the absorbance decreased

with further increase of time, it was even higher (0.35)

at 4 h than initially. The absorbance decreased to 0.19

with 8 h of settling time. Using the data obtained from

1.5 to 8 h of settling time, the ka obtained was

0.5441 h abs-1. This might have happened because

oxidation of the remaining Fe(II) (only about 55% of

total Fe(II) was oxidized) was dominant at the initial

period and then precipitation of oxidized products

took place in the suspension. For suspensions with pH

adjusted to 7.5 and higher in the NB, the absorbance

dropped rapidly without the increase in absorbance.

The ka was obtained after 3 h of settling time. Among

suspensions with the pH adjusted to 7.5 and higher,

the removal rate was highest (7.897 h abs-1) for the

suspension adjusted to pH 7.5, and was approximately

14.5 times faster than for unneutralized suspension.

The absorbance of the suspension adjusted to pH 7.5

decreased to zero after 4 h of settling time.

Conclusions

The efficiency of Fe(II) oxidation, and precipitation

of the oxidized products, as a result of aeration of

the OB and pH variation in the NB was investigated

in this study. When aeration rates in the OB were

varied (0–450 L/min) and the RT was fixed (1.2 h),

pilot-scale tests on the oxidation of Fe(II) showed

that aeration at 300 L/min resulted in the opti-

mum efficiency—approximately 50% oxidation. With

higher aeration rates, oxidation efficiency did not

increase. With increasing RT in the OB, the efficiency

of oxidation of Fe(II) increased linearly and efficiency

was approximately 72% when the RT was 2 h.

However, Fe(II) still remained in subsequent basins

(neutralization, reaction, and settling basins). Because

of the limited volume of the oxidation basin and the

limited geological configuration of the site, it was not

possible to increase the volume of aeration unit or its

retention time in order to achieve 100% Fe(II)

oxidation. Results from pilot-scale tests in which

neutralization was excluded were used to obtain the

rate constants of heterogeneous and homogeneous

oxidation. The results showed the rate constants were

lower than literature values. Because a very low

determination coefficient resulted from the modeling,

however, the analysis was not suitable for obtaining

reliable values of the rate constants. Enhancement of

the oxidation of Fe(II) by changing the pH in the NB

was also investigated. Oxidation of Fe(II) reached

almost 100% when the pH of the mine drainage was

increased to more than 7.5, and there was a linear

Time (hr)9630 12

Ab

sorb

ance

at

560

nm

0.0

0.1

0.2

0.3

0.4

0.5

without pH adjustmentpH 7.5pH 8.0pH 8.5

Fig. 9 Kinetics of absorbance at 560 nm of effluents under

different pH conditions in the neutralization basin


123

relationship between log (Ktotal) and pH, the slope of

which was approximately 2.1. Thus, the rate of

oxidation of Fe(II) increased by two orders of

magnitude when the pH of the suspension was

increased by one unit.

Changes in the absorbance of samples obtained

from the NB under different pH conditions were

measured to determine the precipitation properties of

suspended solids. Consideration of all the results

revealed the trend of changes in absorbance with pH

was closely related to oxidation of Fe(II) and precip-

itation of the oxidized products. Because ferrous

remained in the inflow to the SB, oxidation of Fe(II)

was predominant initially, leading to increased absor-

bance, and the rate of precipitation was slow. How-

ever, the absorbance of the suspension in the SB

dropped rapidly, without a subsequent increase, when

the pH of the mine drainage was adjusted to more than

7.5. This might be because only precipitation

occurred, because all the Fe(II) was oxidized before

the settling basin. Compared with the system in which

Ca(OH)2 was used, the precipitation speed for the

control (suspension not treated with Ca(OH)2) was

much lower. Therefore, the density of floc was lower,

because there was no heterogeneous material, for

example Ca(OH)2, to enhance the floc density, and the

chance of polymerization and enlargement of the

oxidized products was reduced because of the low

collision frequency under the laminar conditions

characteristic of the settling basin, For suspensions

of pH higher than 7.5, however, the absorbance

decreased directly without an initial increase of

absorbance, because there was no Fe(II) left to

oxidize. As shown above, more enhanced treatment

could be designed to treat large-scale mine drainage

with net alkalinity if appropriate aeration and

subsequent pH adjustment could be found. These

results from pilot-scale tests could be expected to

contribute to the design of full-scale plant on the

actual site by constructing an optimized process on the

uneven ground of the site.

References

Akcil, A., & Koldas, S. (2006). Acid mine drainage (AMD):

Causes, treatment and case studies. Journal of CleanerProduction, 14, 1139–1145.

Ames, R. P. (1998). Iron oxidation, gas transfer, and solids

formation in passive treatment systems for mine drainage.

Pennsylvania: The Pennsylvania State University.

Azher, N. E., Gourich, B., Vialb, C., Soulami, M., & Ziyad, M.

(2008). Study of ferrous iron oxidation in Morocco

drinking water in an airlift reactor. Chemical Engineeringand Processing, 47, 1877–1886.

Barnes, A., Sapsford, D. J., Dey, M., & Williams, K. P. (2008).

Heterogeneous Fe(II) oxidation and zeta potential. Jour-nal of Geochemical Exploration, 100, 192–198.

Bernick, M. B., Getty, D., Prince, G., & Sprenger, M. (1995).

Statistical evaluation of field-portable X-ray fluorescence

soil preparation methods. Journal of Hazardous Materi-als, 43, 111–116.

Chon, H. T., & Hwang, J. H. (2000). Geochemical Charac-

teristics of the Acid Mine Drainage in the Water System

in the Vicinity of the Dogye Coal Mine in Korea. Envi-ronmental Geochemistry and Health, 22, 155–172.

CIPB (1995) Study on the environmental improvement in

mining areas (wastewater and waste rocks) (in Korean):

Coal Industry Promotion Board.

Dempsey, B. A., & Dietz, J. M. (2001). A demonstration of‘‘modified’’ passive abiotic treatment of mine drainage.

University Park: Pennsylvania State University.

Emmenegger, L., King, D. W., Sigg, L., & Sulzberger, B. (2008).

Oxidation kinetics of Fe(II) in a eutrophic Swiss lake.

Environmental Science and Technology, 32, 2990–2996.

EPA. (1992). Acid digestion of waters for total recoverable or

dissolved metals for analysis by FLAA or ICP spectros-

copy: SW846, method 3005A. Washington, DC: US EPA,

Office of Solid Waste and Emergency Response.

Gazea, B., Adam, K., & Kontopoulos, A. (1996). A review of

passive systems for the treatment of acid mine drainage.

Minerals Engineering, 9, 23–42.

Hedin, R. S. (2008). Iron removal by a passive system treating

alkaline coal mine drainage. Mine Water and the Envi-ronment, 7, 200–209.

Johnson, D. B., & Hallberg, K. B. (2005). Acid mine drainage

remediation options: A review. Science of the TotalEnvironment, 338, 3–14.

Jung, M. C. (1994). Sequential extraction of heavy metals in

soils and a case study. Korean Social Economic andEnvironmental Geology, 27, 469–477.

Jung, M. C. (2003). Environmental assessment for acid mine

drainage by past coal mining activities in the Youngwol,

Jungseon and Pyungchang areas, Korea (in Korean).

Economic and Environmental Geology, 36, 111–121.

Kalnicky, D. J., & Singhvi, R. (2001). Field portable XRF

analysis of environmental samples. Journal of HazardousMaterials, 83, 93–122.

Lenter, C. M., McDonald, J. L. M., Skousen, J. G., &

Ziemkiewicz, P. F. (2002). The effects of sulfate on the

physical and chemical properties of actively treated acid

mine drainage floc. Mine Water and the Environment, 21,

114–120.

Liang, L., Mcnabb, J. A., Paulk, J. M., Gu, B., & Mccarthy, J. F.

(1993). Kinetics of Fe(II) oxygenation at low partial

pressure of oxygen in the presence of natural organic

material. Environmental Science and Technology, 27,

1864–1870.


123

Millero, F. J., Sotolongo, S., & Izaguirre, M. (1987). The

oxidation kinetics of Fe(II) in seawater. Geochimicaet Cosmochimica Acta, 51, 793–801.

Park, B., & Dempsey, B. A. (2005). Heterogeneous oxidation

of Fe(II) on ferric oxide at neutral pH and a low partial

pressure of O2. Environmental Science and Technology,39, 6494–6650.

Sengupta, M. (1993). Environmental impacts of mining:Monitoring, restoration, and control. Boca Raton: Lewis

publishers.

Singer, P.C. and Stumm, W. (1968) Kinetics of the oxidation of

ferrous iron. In the Second Symposium of Coal Mine

Drainage. pp. 12-34.

Stumm, W., & Lee, G. F. (1961). Oxygenation of ferrous iron.

Industrial & Engineering Chemistry Research, 53,

143–146.

Stumm, W., & Morgan, J. J. (1996). Aquatic chemistry. New

York: Wiley.

Sung, W., & Morgan, J. J. (1980). Kinetics and products of

ferrous iron oxygenation in aqueous systems. Environ-mental Science and Technology, 14, 561–568.

Tamura, H., & Nagayama, M. (1976). Effect of anions on the

oxygenation of ferrous ion in neutral solutions. Journal ofInorganic and Nuclear Chemistry, 38, 113–117.


123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

mine drainage kinetics

Documents