Kim Y.J. (Ed.) Advanced Environmental Monitoring

Подождите немного. Документ загружается.

21 An Advanced Monitoring and Control System 273

21.2.2 Hydrogen Peroxide Analysis System

The hydrogen peroxide analyzing unit was based on a fluorometric detection

method using the reaction of p-hydroxyphenyl acetic acid and residual hydro-

gen peroxide in the presence of a peroxidase enzyme (Lazrus et al. 1985).

Figure 21.2(a) shows a scheme for the on-line residual peroxide detection sys-

tem, which was carefully devised to exclude interference from ozone. Three DC

drive pumps (Cole-Parmer Peristaltic pump, USA) continuously injected the

following reagents; fluorometric reagent (0.35 M potassium hydrogen phthalate

(KHP), 8.0×10

−3

M p-hydroxyphenyl acetic acid (POPHA), 2 purpurogallin

units of peroxidase/ml reagent), buffered solution (0.5 M NaOH) and the sample,

into the hydrogen peroxide analyzer, with tubing connected to both the inlet

and outlet of the instrument. The POPHA dimer formed in the reaction module

Fig. 21.1 (a) Schematic diagram of the ozone analysis system. (b) Typical ozone decomposition

pattern of ozone in raw water obtained from this setup.

Fig. 21.2 (a) Schematic diagram of the hydrogen peroxide analysis system. (b) Signal response.

274 J.-W. Kang et al.

was stabilized raising pH >10, and analyzed using fluorescence detection

(SOMA, Japan), with excitation and emission at 320 and 400 nm, respectively.

Data signals were converted into actual hydrogen peroxide concentration units

(µg/l), and recorded by a computer system via an interface card (Labview,

USA). Figure 21.2(b) shows the stability and sensitivity of data signal obtained

with this method, which enabled detection to the ∼ppt level (MDL < 1 µg/l)

acquiring two signals per second.

21.2.3 Bromate Analysis System

The bromate analysis was based on the EPA 326.0 method, using a suppressor

acidified post-column reagent (Salhi and Gunten 1999). As shown in Fig. 21.3, the

bromate analyzing system was composed of two main parts: ion chromatograph

Fig. 21.3 (a) Schematic diagram of the bromate analysis system. (b) Shapes of the bromate peaks

(MDL <0.3 µg/l)

21 An Advanced Monitoring and Control System 275

Fig. 21.4 Effect of hydrogen perox-

ide on Ibuprofen removal efficiency.

Initial Ibuprofen: 2 µM. Ozone dose:

2 mg/l, hydrogen peroxide:ozone

ratio = 0 ∼ 0.8 (w/w)

and post-column reaction (PCR). The ion chromatograph system (DX-500, Dionex)

was comprised of a guard column (AS9-HC, Dionex), an analytical column (AG9-

HC, Dionex), a suppressor devices (ASRS-I) and a conductivity detector. The

apparatus for the PCR were composed of a post- column reagent delivery system

(PC-10, Dionex), a heated post-column reaction coil (Dionex) and an ultraviolet/

visible (UV/Vis) detector (Prostar325, Varian). The effluent of bromate ions sepa-

rated by the ion chromatograph mixed with an acidic solution of potassium iodide

containing an ammonium molybdate and heated at 80 ˚C

to form tri-iodide ions,

which were measured by UV/Vis detection at 352 nm. With this setup, bromate

could be precisely detected at levels as low as 0.3 µg/l.

21.3 Results and Discussion

21.3.1 Peroxone Process for the Removal

of Pharmaceuticals

The peroxone process is one of the most effective ways of producing OH• for the

oxidation of micropollutants. The stoichiometric optimum for this reaction is a

hydrogen peroxide:ozone ratio of 0.5 (mole: mole) or 0.35 (weight: weight), but

since other ozone-consuming species are usually present in water, the required dose

ratio is usually less than this value. In fact, the actual ratio is to be directly

determined with experimental studies. Figure 21.4 shows the effect of hydrogen

peroxide on the percentage of the substrate removed; the optimum hydrogen perox-

ide:ozone ratio was 0.3 (w/w).

276 J.-W. Kang et al.

21.3.2 Bromate Formation Characteristics in Ozone Alone

For the future application of ozone, the characteristics of bromate formation during

ozonation were studied by testing water samples collected from six different water

treatment plants located in Korea. With ozone doses of 1 ∼ 2 mg/l, the formation of

bromate ranged between 2 and 12 µg/l, which in some cases exceeded the MCL

(= 10 µg/l). The level of bromide in the water samples tested were relatively low (11

∼ 42 µg/l), but the conversion rate of bromide to bromate appears to be high under

certain conditions. The key factors that could increase the production of bromate

are the ozone doses (Fig. 21.5), bromide level (Fig. 21.6), and pH. Increasing

the ozone dose from 1 to 3 mg/l and the contact time from 2 to 10 min, the level of

Fig. 21.5 Effect of ozone dose and contact time on the formation of bromate. pH: 7.2. Br

−

: 40 µg/l

Fig. 21.6 Effect of contact time and bromide level on the formation of bromate. (a) Ozone dose:

2 mg/l. pH: 7.0. Initial Br

−

: 40 µg/l. (b) Ozone dose: 2 mg/l. pH: 7.6. initial Br

−

: 14 µg/l

21 An Advanced Monitoring and Control System 277

bromate formation was found to increase from 0.4 to 5.2 µg/l (Fig. 21.5). Figure 21.6

compares the effect of the bromide level on the formation of bromate in the two

different raw water sources with that of raw waters collected from S and M water

treatment plants. In each raw water sample, 50 µg/l of bromide was intentionally

added and ozonated. In results, about twofold increase of bromate formation was

observed in the bromide spiked water as compared to the case in raw water. The

effect of pH was found to be the most significant factor affecting the formation of

bromate. Increasing the pH from 7 to 8.5, the conversion yield (%) of bromate was

increased from 6.8 to 58%, far exceeding the MCL guideline (data not shown).

21.3.3 Characteristics of Bromate Formation

in the Peroxone Process

The peroxone process has been proved to enhance the degradation of various

micropollutants by converting aqueous ozone into the more reactive OH•. However,

the formation of bromate cannot be avoided as OH• is also strongly involved in the

peroxone process, bromate could also be formed, as the amount of molecular ozone

would be small, but would still be involved in the bromate formation pathways. In

this process, less ozone would be utilized in the formation of bromate due to its

competitive reaction with the conjugate base of H

+ HO

−

→ O

−•

+ HO

•

);

therefore, less bromate production would expected in the peroxone process.

However, the opposite was observed in our experiments. With regard to the effect

of hydrogen peroxide on the formation of bromate in the peroxone process, the

results reported in a reference survey were inconsistent (Gunten et al. 1988).

Furthermore, it is hard to assess the effect of hydrogen peroxide on the formation

of bromate without performing experimentation. Figure 21.7 shows the bromate

formation profiles vs. contact time for a raw water initially spiked with bromide

(= 100 µg/l) by varying the hydrogen peroxide dose between 0 and 0.6 (w/w), with

Fig. 21.7 Effect of hydrogen perox-

ide on the rate of bromate formation.

Ozone: 2 mg/l. pH: 7.0. Br

−

: 100 µg/l

278 J.-W. Kang et al.

the ozone dose fixed at 2 mg/l. On the addition of hydrogen peroxide, the observed

level of bromate production compared to ozone alone was much higher. The highest

level of bromate production was observed with a hydrogen peroxide:ozone ratio

(w/w) of 0.4, but the rate of bromate production decreased with a ratio of 0.5, which

is the desired peroxide dose for the control of bromate.

21.3.4 Scheme for Oxidant Dosage Control in the Ozone

and Peroxone Processes for the Control of Bromate Formation

A real-time method for the effective control of ozone and hydrogen peroxide dosages

has been developed, based on a specifically devised analytical setup, as previously

shown (Fig. 21.1 and 2), and an optimum ozone dosage control method proposed

(Oh et al. 2003; Oh et al. 2005). This control method achieved both the required dis-

infection and oxidation objectives. For the purpose of disinfection, ozone was dosed

to the corresponding CT (C: residual disinfectant concentration (mg/l); T: contact

time [min]) criteria required to meet the target inactivation goal. With respect to the

oxidation objectives, which have recently increased in emphasis due to the contami-

nation of water by various toxic chemicals, ozone was dosed to take into account the

yield of the more reactive species, OH•, by measuring the ozone decay rate constant,

, and R

value (ratio of the OH• to residual ozone). For this purpose, the kCT-control

method, which optimizes the ozone dosage on a real-time basis (Oh et al. 2003), is to

be further developed and utilized to achieve the target water treatment objectives, the

oxidation and/or disinfection goal. The kCT is defined as

kCT O dt=

∫

[]

where, kCT is the area under the curve for the time varied residual ozone concen-

tration (Fig. 21.1(b) to be precisely measured by the setup, as shown in Fig. 21.1

(a). The k

value varies according to variations in the water quality, but could be

continuously monitored by the automated real time monitoring/control system,

and the amount of ozone adjusted to the set kCT value. The recommended target

value, which was tested in a prior study on H river water, was 2 ∼ 3 mg/l-min

(Oh et al. 2003).

Since the oxidation of toxic micropollutants, such as pesticides and pharmaceuti-

cals, and taste and order causing compounds, has now become a more centered issue

in water treatment, several municipal drinking water treatment plants that used ozone

have now been converted to the peroxone process by the addition of a hydrogen per-

oxide feeding system. In this process, the control of the optimum hydrogen peroxide

dose is regarded as an important factor in the optimization of the process. The stoi-

chiometric optimum for the reaction between ozone and hydrogen peroxide is a

hydrogen peroxide:ozone ratio of 0.35 (w/w), but due to other demands on ozone or

reacted compounds usually present in water, the required dose ratio would usually be

less, but this can be determined experimentally. The kCT analytical devise enables

21 An Advanced Monitoring and Control System 279

calculation of the optimum hydrogen peroxide dosage based on the measured resid-

ual ozone, which is controlled by the set kCT value. Figure 21.8 shows the residual

ozone output obtained with the input kCT set at 2 mg/l-min vs. the calculated hydro-

gen peroxide dosage. The residual ozone output response to the set kCT was ∼0.35.

In a previous study, the recommended hydrogen peroxide dose was a hydrogen per-

oxide:ozone (w/w) ratio of 0.5, which corresponds to a hydrogen peroxide dose of

Fig. 21.8 Optimum oxidant dose control for the peroxone process using the kCT control method

Fig. 21.9 Optimum process control logic to meet bromate MCL in ozone and peroxone AOPs

280 J.-W. Kang et al.

0.16 ∼ 0.18 mg/l, as shown in Fig. 21.7. With this kCT based peroxide dose control,

the bromate production could be successfully suppressed below the MCL guide line

(10 µg/l). Figure 21.9 shows the flow chart of the kCT based control logic for the

determination of the dosages of ozone and hydrogen peroxide to meet the bromate

MCL guideline. The overall control system incorporated with the specifically-

conceived analytical devices and a control unit was able to attain the best water

treatment objectives for minimizing the formation of bromate.

21.4 Conclusion

The following conclusions were drawn from this study:

1) Online monitoring devices for the analyses of residual ozone and hydrogen

peroxide were devised, not only for real time monitoring, but also for optimum

dosage control.

2) The online monitoring device for the key ozone related by-product, bromate, was

also devised, and was able to detect bromate at a level as low as 0.3 µg/l.

3) For the future application of ozone and ozone related AOPs, the characteristics

of bromate formation have been studied. Even at low bromide levels, the conver-

sion rate of bromide to bromate could be high, but this will depend on fluctua-

tion in the water quality.

4) In the peroxone process, the formation of bromate was sensitive to the change in

the hydrogen peroxide:ozone ratio. At a ratio of 0.5 (w/w), the minimum

bromate formation was attained.

5) The overall control system, incorporated with the specifically conceived analyti-

cal devices and control logic, could attain the best water treatment objectives,

with the suppression of the formation of the harmful by-product, bromate.

Acknowdegements This subject was supported by Ministry of Environment as “The Eco-technopia

21 project,” and also by “The Brain Korea 21 project (BK21)” of Ministry of Education & Human

Resource Development.

References

Amtsblatt der Euopäischen Gemeinschaften L 330: Richtlinie 98/83/EG, 1998.

Buffle M-O., Galli S., and von Gunten U. (2004), Enhanced bromate control during ozonation:

The chlorine-ammonia process, Environ. Sci. Technol., 38, 5187–5195.

Duguet J.P., Brodard E., Dussert B., and Mallevialle J. (1985), Improvement in the effectiveness

of ozonation of drinking water through the use of hydrogen peroxide, Ozone Sci. Eng.,

7, 241–258.

Hoigné J. and Bader H. (1981), Determination of ozone in water by the indigo method, Water Res.,

15, 449–456.

21 An Advanced Monitoring and Control System 281

Lazrus A.L., Kok G.L., Gitlin S.N., Lind J.A., and Mclaren S.E. (1985), Automated fluorometric

method for hydrogen peroxide in atmospheric precipitation, Anal. Chem., 57, 917–922.

Oh B.S., Kim K.S., Kang M.G., Oh H.J., and Kang J.W. (2005), Kinetic study and optimum con-

trol of the ozone/UV process measuring hydrogen peroxide formed in-situ, Ozone Sci. Eng.,

27, 421–430.

Oh H.J., Kim W.J., Choi J.S. Gee C.S. Hwang T.M., Kang J.G. and Kang J.W. (2003), Optimization

and control of ozonation plant using raw water characterization method, Ozone Sci. Eng.,

25, 383–392.

Park H.S. Hwang T.M., Kang J.W., Choi H.C., and Oh H.J. (2000), Characterization of raw water

for the ozone application measuring ozone consumption rate, Water Res., 35, 2607–2614.

Staehelin J. and Hoigné J. (1982), Decomposition of ozone in water: Rate of initiation by

hydroxide ions and hydrogen peroxide, Environ. Sci. Technol., 16, 676–681.

Salhi E. and von Gunten U. (1999), Simultaneous determination of bromide, bromate and nitrate

in low µgl

−1

levels by ion chromatography without sample pretreatment, Water Res.,

33, 3239–3244.

USEPA, Fed. Resist. (1994) National primary drinking water regulations; disinfectants and disin-

fection byproducts; proposed rule59:145:38668.

von Gunten U., Bruchet A., and Costentin E. (1988), Bromate formation in conventional and

advanced oxidation processes (O

): Theoretical and empirical evaluation of full-scale

experiments, J. Am. Water Works Assoc., 88, 53–65.

Chapter 22

Monitoring of Dissolved Organic Carbon

(DOC) in a Water Treatment Process

by UV-Laser Induced Fluorescence

Uwe Wachsmuth

, Matthias Niederkrüger

, Gerd Marowsky

Norbert Konradt

, and Hans-Peter Rohns

Abstract Results of online investigations of water quality during a water treatment

process by ultraviolet-laser-induced ﬂ uorescence (UV-LIF) are presented. In the

ﬁ rst part the integrated ﬂ uorescence intensity is correlated to the classically deter-

mined concentration of dissolved organic carbon (DOC). Decision, detection and

determination limits are evaluated for this procedure and online DOC measure-

ments conducted with the presented LIF system are compared to reference analysis.

In the second part the ozone demand in the water treatment process is derived from

the LIF-signal directly. The calibration was done by correlating LIF-signals with

the ozone doses for different conditions and a deﬁ nite residual ozone concentration

in the processed water.

Keywords: Dissolved organic carbon (DOC), drinking water, laser induced

fluorescence, online process control, real-time measurements

22.1 Introduction

Large cities situated at big streams often fulfill their demand of drinking water from

these streams. Often the water is not obtained from the river directly, but after natu-

ral filtering from the river banks. This raw water is still enriched with aromatic

organic substances which have to be removed to avoid the growth of bacteria in the

water supply pipe systems and for health reasons. In water treatment processes, this

is often realized by oxidation of the aromatic components and by additional

filtering of the hydrophilic reaction derivates on active carbon. The amount of the

oxidant added to the raw water is controlled by the residual concentration of this

substance after the process.

Laser-Laboratorium Göttingen GmbH, Hans-Adolf-Krebs-Weg 1, 37077 Göttingen, Germany

Laser-Laboratorium Göttingen e.V., Hans-Adolf-Krebs-Weg 1, 37077 Göttingen, Germany

Stadtwerke Düsseldorf AG, Qualitätsüberwachung Wasser (OE 423), Postfach 101136, 40002

Düsseldorf, Germany

282

Y.J. Kim and U. Platt (eds.), Advanced Environmental Monitoring,

282–294.