A reverse-flow injection analysis method for the determination of dissolved oxygen in fresh and...

A reverse-flow injection analysis method for the determinationof dissolved oxygen in fresh and marine waters

Saowapha Muangkaew a, Ian D. McKelvie b,*, Michael R. Grace b,Mongkon Rayanakorn a, Kate Grudpan a, Jaroon Jakmunee a,

Duangjai Nacapricha c

a Department of Chemistry, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailandb Water Studies Centre, School of Chemistry, PO Box 23, Monash University, Vic. 3800, Australia

c Department of Chemistry, Faculty of Science, Mahidol University, Bangkok 10240, Thailand

Received 4 March 2002; received in revised form 12 March 2002; accepted 13 March 2002

Abstract

A reverse flow injection method (rFIA) based on the Winkler titration chemistry, is reported for the determination of

dissolved oxygen (DO) in natural waters. Manganese(II) sulfate is injected into a continuously flowing stream of sample

and subsequently merges with a reagent stream of sodium hydroxide and sodium iodide. Manganese(II) hydroxide that

is formed reacts with DO in the sample to form an oxidized manganese hydroxyoxide floc. Addition of 10% sulfuric

acid dissolves this floc, and under acidic conditions, the triiodide ion formed is detected by photometry in a flow

through cell at a wavelength of 440 nm. The method is rapid (48 measurements per h), repeatable (R.S.D. ca. 3%, n�/

3), and has a calculated detection limit of 0.25 mg l�1 (P�/0.001). No interference from nitrite or ferrous ions was

observed at concentrations typically found in natural waters. The method has been successfully applied to on-line

measurement of DO in sediment respiration reactors.

# 2002 Elsevier Science B.V. All rights reserved.

Keywords: Reverse-FIA; Winkler; Iodine; Photometric detection

1. Introduction

Natural waters in equilibrium with the atmo-

sphere typically contain dissolved oxygen (DO)

concentrations in the range from 5 to 15 mg l�1 O2

depending on the water temperature, salinity and

altitude [1]. The DO concentration present in

water reflects atmospheric dissolution, as well as

autotrophic and heterotrophic processes that pro-

duce and consume oxygen, respectively.

Analysis of DO is extremely important in

determining water quality. It provides information

on the biological and biochemical reactions occur-

ring in a water body, and is, therefore, an

important indicator of stream metabolism [2]. In

situ measurements of this parameter can be used as

a primary indicator of water quality, and regula-

* Corresponding author. Tel./fax: �/61-3-990-54558

E-mail address: [email protected] (I.D.

McKelvie).

Talanta 58 (2002) 1285�/1291

www.elsevier.com/locate/talanta

0039-9140/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.

PII: S 0 0 3 9 - 9 1 4 0 ( 0 2 ) 0 0 4 1 4 - 9

ÅÔ¢ÊÔ·¸Ô ìÁËÒÇÔ·ÂÒÅÑÂàªÕÂ§ãËÁèCopyright by Chiang Mai UniversityA l l r i g h t s r e s e r v e d


mailto:[email protected]

tory agencies recommend a minimum DO require-ment for maintenance of fish populations, e.g.

DO]/6 mg l�1 or 80�/90% saturation measured

over at least one diurnal cycle [3]. Measurement of

DO is also the basis of many respirometric tests,

including the 5-day Biochemical Oxygen Demand

(BOD5) test.

The classical method for DO determination is

the Winkler titration. In this method, DO oxidizesmanganese(II) hydroxide floc produced in situ, to

form a brown colored, oxidized manganese hydro-

xyoxide [4]. Addition of sulfuric acid in the

presence of excess iodide dissolves this hydroxy-

oxide floc, producing triiodide that is then titrated

with standardized thiosulfate using a starch in-

dicator.

Measurement of DO for BOD, respirometrictests, and in-field studies is more commonly and

conveniently performed using an electrometric

technique, based on either galvanic or voltam-

metric membrane sensors. However, these probes

must be frequently calibrated, often against the

Winkler titration [5], may be insensitive to changes

in oxygen concentrations of less than 1 mg l�1 [2],

and require a sample volume of several millilitersand active sample transport across the gas diffu-

sion membrane for successful measurement. For

these reasons measurement by non-electrometric

techniques may be preferred in some instances to

the use of the DO electrode.

While the Winkler titration, and its various

modifications, offers a robust means of determin-

ing DO [1], the precision of the classical titrationmay be inadequate for the purposes of accurate

respirometric measurements, especially for low

concentrations. In an attempt to address this,

Hartwig and Michael [6] used photometric detec-

tion at 350 nm of the triiodide complex formed in

the Winkler reaction to perform photometric

titrations and achieved precision of better than

0.4% R.S.D. for shipboard measurements.Various suggestions have been made for auto-

mating either the total determination of DO, or

detection of the iodine formed by the Winkler

reaction, based on flow injection analysis. For

example, Novic [5] used a flow injection system

with amperometric detection, and Sakai et al. [7]

used the fluorescence quenching of 2-thionaphthol

to determine I2 produced by the Winkler reaction.In both these methods, the formation of iodine

was performed in batch, off-line, thus avoiding

likely problems associated with the formation of

manganese(II) hydroxide precipitate in the flow

lines of an FIA system. However, because of this

off-line step, neither method would be suitable for

the fully automated determination of DO. Few

FIA methods have been described that measureDO directly. Of these, two involve the reaction of

molecular oxygen with the redox indicator, leuco-

methylene blue and photometric detection of the

product species [8,9], while another is based on the

quenching of the room temperature phosphores-

cence of a sol�/gel metal�/chelate by oxygen [10].

In this paper, we describe the development of a

flow injection system for DO measurement basedon the spectrophotometric detection of triiodide

ion from the Winkler reaction. To avoid the

problem of flow tube blockage by manganese(II)

hydroxide that occurs if Mn2� is continually

merged with sodium hydroxide, we have adopted

a reverse flow (reagent) injection approach in

which Mn2� is injected into a continuous flowing

stream of sample before merging with a mixedreagent stream of NaOH and KI. The oxidized

manganese hydroxide floc that forms is readily

transported through the flow tubing, and is

completely dissolved after merging with the sulfu-

ric acid stream. Triiodide produced is measured

photometrically at 440 nm. The method is both

rapid and reproducible, can be performed using

any simple laboratory or field FIA system withphotometric detection capability, and is eminently

suitable for on-line monitoring applications.

2. Experimental

2.1. Chemicals

All reagents were of analytical-reagent grade.The 5% (w/v) solution of manganese(II) sulfate

was prepared by dissolving 36.2 g of mangane-

se(II) sulfate-4-hydrate in 500 ml ultra pure water

(Continental Water systems). Sulfuric acid (10% v/

v) was prepared by dissolving 56 ml of 96%

sulfuric acid in 500-ml ultra pure water. Sodium

S. Muangkaew et al. / Talanta 58 (2002) 1285�/12911286



hydroxide (0.5% (w/v)) and potassium iodide(0.2% (w/v)) were prepared by dissolving 2.5 g

sodium hydroxide and 1.0 g of potassium iodide in

500 ml ultra pure water. An acidic iodide flushing

solution was prepared by mixing 10% (v/v) sulfuric

acid solution and the combined sodium hydroxide

and potassium iodide solution in a 3:1 ratio.

2.2. Preparation of calibration standards of varying

dissolved oxygen concentrations

Solutions containing various DO concentrations

within the range of 0�/14 mg l�1 O2 were prepared

in sealed 1 l bottles, the caps of which were fitted

with a DO probe connected to a YSI Model 55

handheld DO meter, a thermometer, and a tubefor air or nitrogen gas sparging and a sampling

line. Using these bottles, 1 l volumes of ultra pure

water at temperatures from 0 to 50 8C were

sparged with either air or nitrogen gas to obtain

a ranges of different concentrations of DO, as

indicated by the in situ oxygen electrode. Samples

of each water prepared in this way were withdrawn

and the DO concentration determined by Winklertitration [1], which was used as the principal

reference method.

Ideally the FIA DO method should give a linear

response over a wide range of oxygen concentra-

tion values. As this was the case, a simple two

point calibration based on the difference in the

theoretical oxygen saturation values at different

salinities and temperatures would be feasible, inmuch the same manner that theoretical oxygen

saturation values at specified temperatures are

used to calibrate DO electrodes [1]. This would

avoid the need to calibrate against the Winkler

titration value.

2.3. Flow injection system

The flow injection system used for development

of the DO method is shown in Fig. 1. The FIA

system was equipped with two peristaltic pumps

(Ismatec CA4E, 20 rpm) and an electrically

actuated rotary six-port valve (Rheodyne, Model

5301). A three way solenoid switching valve (SV)

(LEE, Model No LFAA 1201618H) was used to

introduce an acidic iodide wash solution between

samples. Valve switching and data acquisition was

performed using Flow Control Software (A-Chem

Technologies, Melbourne) and was run on a 486

computer. A variable wavelength spectrophoto-

metric (Applied Biosystems, Model 759A) with a 6

ml flow cell was used at 440 nm. PTFE (0.8-mm

i.d.) tubing was used for knitted reaction coils

(M1, 2 and 3) and all flow tubing. All reagents

were pumped at a flow rate of 1.4-ml min�1.

3. Results and discussion

Two criteria were applied in determining the

optimum conditions for the flow injection mani-

fold, viz; highest sensitivity and shortest practic-

able analysis time. However, because the method

also involves the formation of an hydroxide floc,

there is the potential for blockage problems within

the flow lines, and this becomes a practical

constraint in the choice of reagents, injection

volume and flow rates.

3.1. Wavelength selection

The maximum absorption wavelength of aqu-

eous iodine solution was measured at 380 nm. At

this wavelength the absorbance of a sample with

8.5 mg l�1 O2 was approximately 1.9, and as this

was near the upper end of the absorbance range of

the detector used (0�/2), it represents the practical

upper limit for determination of DO. Since the

desirable linear detection range for O2 in natural

waters is 5�/15 mg l�1 O2, the sensitivity at 380 nm

was deemed to be too great. Therefore, a wave-

length of 440 nm was selected because under this

condition of reduced sensitivity, measurement of

waters with concentrations as high as 15 mg l�1

O2 could be accommodated. However, to max-

imize the sensitivity and resolution of waters with

low DO concentrations, a wavelength of 380 nm

should be used.

S. Muangkaew et al. / Talanta 58 (2002) 1285�/1291 1287



3.2. Effect of injection volume of manganese(II)

sulfate solution on peak height and shape

Injection volumes from 25 to 350 ml were used to

determine the optimum value, i.e. the volume that

gave the largest peak response and a single peak.

Fig. 2 shows that the greatest peak height was

obtained at an injection volume of 50 ml. Doublet

peaks were observed for all injection volumes

larger than this, suggesting that for all but the

smallest injections (25 and 50 ml) of manganese(II)

sulfate, the reaction between DO in the sample

stream and the bolus of manganese(II) hydroxide

that forms, was complete at the leading and

trailing edges, and incomplete in the mid portion

of the injected zone because of a paucity of DO in

this region. While it may be possible to overcome

doublet peak formation by more efficient mixing,

say by the use of a longer mixing coil at M1, this

would be counterproductive because the increased

dispersion would offset any gains in sensitivity.

For these reasons, a 50-ml injection volume was

chosen for all subsequent measurements.

3.3. Effect of mixing coil lengths

Using a fixed injection volume of 50 ml, theinfluence of independently varying the mixing coils

M1, 2 and 3 (Fig. 1) on final signal response and

analysis time was investigated. Increasing the

length of the mixing coil, M1, from 0�/200 cm,

caused further dispersion of the injected volume of

manganese(II) sulfate resulting in a decreasing

peak height (Fig. 3).

Not surprisingly, increasing the length of M2,immediately after the addition of the sodium

Fig. 1. Reverse FIA manifold for DO determination. P1, P2: peristaltic pumps, M1, M2 and M3: knitted reaction coils of specified

lengths, V: injection valve, SV: 3-way solenoid switching value and D; flow-through photometric detector.

Fig. 2. Peak heights obtained from the determination of DO in

air-saturated ultra pure water at 22 8C (8.5 mg l�1 O2) by the

proposed rFIA method at various injection volumes from 25 to

350 ml. Error bars are 9/sn�1 for n�/3.

Fig. 3. Peak heights obtained from the determination of DO in

air-saturated ultra pure water at 22 8C by the proposed rFIA

method with various mixing coil lengths for M1(j), M2 (")

and M3 (').




hydroxide/iodide reagent stream results in anincreased peak response because additional mixing

promotes the reaction between DO and the

manganese(II) hydroxide.

The dissolution of the oxidized manganese

hydroxide floc following the addition of sulfuric

acid is rapid, and increasing the length of mixing

coil M3 (Fig. 1) had no effect other than to

decrease the peak response and increase the peakwidth (Fig. 3).

On the basis of these observations, mixing coils

M1 and 3 were omitted entirely from the final

version of the flow injection manifold. A length of

100 cm was chosen for M2, and this represents a

compromise between maximizing the sensitivity

and maintaining an acceptable sample throughput.

One additional modification was made to themanifold. Oxidized manganese hydroxyxide floc is

formed on-line after the merging of the sodium

hydroxide/iodide reagent stream with the injected

Mn(II) and sample stream. While the bulk of the

oxidized manganese hydroxyoxide floc is trans-

ported through the manifold tubing, and subse-

quently dissolved by the addition of sulfuric acid,

it was observed that a small residual amount ofbrown precipitate remained in mixing coil M2, and

that this gradually accumulated over time. While

this process had no effect on the reproducibility of

the analyses, if ignored, gradual accumulation

would result in the complete occlusion of the

flow tubing completely. This potential problem

was overcome by incorporating a solenoid SV, in

the NaOH/KI reagent line (Fig. 1), and using thisto introduce an acidic iodide wash solution after

every third injection. This washing process was

controlled by the instrument software, and pro-

grammed such that it occurred during the aspira-

tion of a new sample into the FIA system. An

improvement on this configuration would be to

move the SV to the sample line, which would mean

that sample switching could be achieved withoutthe need to stop the pump, while manifold washing

was occurring.

3.4. Analytical figures of merit

The DO calibration data obtained using the

proposed rFIA system for a DO concentration

range of approximately 2�/13 mg l�1 O2 fitteda linear plot, according to the equation:

Peak Height (Abs)

�0:0414[DO]�0:4784;

for which r2�0:9945 (n�6)

Linearity over the calibration range means that

a two point calibration method can be used, i.e.

using high and low DO standards, whose concen-

trations are defined by the saturation concentra-

tion at particular temperature and salinity values.

This offers the most feasible option for calibration

in the field.It should be noted that there is a substantial

blank signal that originates from DO in the

reagents, and while this could be reduced by

sparging the reagents with nitrogen before use,

this is generally not practicable.

The calculated detection limit was 0.25 mg l�1

O2 (P�/0.001). The relative standard deviation for

the proposed rFIA method was typically approxi-mately 3% for triplicate injections, and a

sample throughput of 48 injections per h was

possible.

3.5. Effect of salinity on the determination of

dissolved oxygen concentration by the proposed

rFIA method

As salinity increases, there is a decrease in the

oxygen solubility. Increasing ionic strength causes

no interference in the classical Winkler titration,

but could conceivably have a kinetic effect on thereactions in a flow injection method. To investi-

gate whether the proposed method is tolerant of

changes in salinity, air-saturated ultra pure water

at 20�/21 8C with a range of different salinities

was analyzed by the proposed method, and

compared with measured values using a DO meter

and the theoretical saturation value for the same

conditions of temperature and salinity. The results(Table 1) show excellent agreement between the

rFIA results, those from the DO meter

(salinity corrected readings) and theoretical

values [1], and indicate that this method is suitable

for DO measurements over a wide range of

salinities.




3.6. Effect of potential interferents on the

determination of dissolved oxygen concentration by

the proposed rFIA method

Nitrite at concentrations greater than 50 mg l�1

NO2�1 is a potential interferent in the Winkler

titration because in acid medium it reacts with

excess iodide to produce iodine and N2O2. Further

exposure to air will produce more nitrite and hence

more iodine [2], and this cyclic process can lead to

significant overestimation of the true DO concen-

tration. In natural waters the nitrite concentration

would seldom be higher than 200 mg l�1. Potential

interference of nitrite on the rFIA DO determina-

tion was studied by adding nitrite in the concen-

tration range 0�/600 mg l�1 NO2� to samples of

oxgenated water. Table 2 shows that the proposed

method is tolerant to nitrite up to a concentration

of 600 mg l�1. This tolerance for relatively high

concentrations of nitrite is not surprising given

that the reactions are all performed on-line in a

closed system that offers no opportunity for

reoxygenation of the sample after final acidifica-tion and subsequent production of triiodide.

Iron(II) at concentrations greater than 1 mg l�1

is also known to interfere in the DO determina-

tion[11], and while the Fe(II) concentration in

most natural waters is generally lower than 1 mg

l�1, we chose to test for interference from this

species over the concentration range of 0�/1.5 mg

l�1. Table 2 shows that DO concentrationsdetermined using the proposed rFIA method

were unaffected by increasing Fe(II) concentra-

tions, demonstrating that there was no interfering

effect of Fe(II) for DO measurements of natural

waters.

3.7. Application to real samples

A method comparison between this proposed

rFIA method and Winkler titration values was

performed on six natural water samples, and astrong correlation of the developed rFIA and

Winkler titration methods was obtained, with a

linear regression equation of:

DOrFIA�0:9136[DO]winkler�0:5849;

r2�0:9926 (n�6)

To evaluate the longer term reliability of the

rFIA method developed, the sample intake line of

the FIA system was connected directly to a stirred

reactor containing tap water that had been deox-ygenated by nitrogen sparging. A DO probe and

the proposed rFIA method were used to determine

DO in the water sample simultaneously over a 7 h

period. After initial deoxygenation, the water in

the reactor was allowed to reoxygenate up until

1500 h, and, thereafter, the temperature of the

Table 1

DO concentrations of air-saturated ultra pure water at 20�/

21 8C with various salinities (0�/35 ppt)

Salinity (ppt) Dissolved oxygen concentration (mg l�1)

Theoretical RFIA (%R.S.D.) DO meter

0 9.04 8.90 (2.2) 8.94

5 8.83 8.63 (1.2) 8.39

10 8.57 8.25 (2.4) 8.18

20 7.93 7.87 (1.2) 7.74

30 7.48 7.16 (1.4) 7.02

35 7.26 7.16 (4.2) 7.04

Table 2

Effect of added nitrite and Fe(II) on DO measurements (mean9/S.D., n�/3)

Nitrite (mg l�1) DO concentration (mg l�1) Fe(II) (mg l�1) DO concentration (mg l�1)

0 8.719/0.31 0.0 7.899/0.27

100 8.669/0.70 0.2 7.859/0.34

200 8.549/0.48 0.4 7.599/0.20

300 8.529/0.24 0.8 7.919/0.46

400 8.649/0.28 1.0 7.699/0.32

500 8.729/0.29 1.2 7.879/0.17

600 8.639/0.26 1.5 8.059/0.05




reactor was manipulated to obtain a range of DOvalues. Fig. 4 shows that DO concentrations

obtained from the proposed rFIA method closely

as followed the theoretical values over the latter 4

h period of time (DOrFIA�/0.998DOTheoretical�/

0.195, r2�/0.9642, n�/3). However, the DO con-

centrations obtained from the DO probe were

lower than the theoretical values, and essentially

constant during the last 3 h 30 min, an effect that isprobably due to inadequate liquid transport across

the surface of the DO electrode membrane.

Subsequent to this evaluation, the proposed

rFIA DO method has been successfully applied

in the measurement of oxygen uptake kinetics in a

series of sediment respiration studies in both

laboratory and field situations.

Enhancement of the sensitivity, and hence theresolution, of DO over a narrower range of values,

e.g. in anaerobic waters, could readily be achieved

by performing measurements at a wavelength

closer to the lmax of 380 nm.

4. Conclusion

A reverse flow injection analysis technique for

determination of DO in natural water has been

successfully developed based on the Winkler titra-

tion method. The proposed reagent injection

method overcomes manifold blockage problems

of precipitated manganese(II) hydroxide thatwould occur if conventional sample injection

FIA were used, and the method is very tolerant

of the common interferents Fe2� and NO2�. The

DO concentrations obtained using this rFIA

method were in close agreement with the theore-

tical saturation and Winkler titration DO values of

a number of real samples. The method has been

successfully deployed in tracking DO variations in

a stirred reactor over a period of several hours,

and is now being routinely used for this purpose in

our laboratory and in the field.

Acknowledgements

Saowapha Muangkaew thanks the Thai Minis-

try of University Affairs and PERCH (Postgrad-

uate Education and Research Program in

Chemistry) for their support of her research inAustralia and Thailand. Technical assistance by

Mr Peter Ellis is gratefully acknowledged.

References

[1] APHA-AWWA-WEF, Standard methods for the Exam-

ination of Water and Wastewater, 20th ed., American

Public Health Association, Washington, DC, 1998.

[2] R.G. Wetzel, G. Likens, Limnological Analysis, second

ed., Springer, New York, 1991, p. 391.

[3] ANZECC, Australian Water Quality Guidelines for Fresh

and Marine Waters, Australian and New Zealand Envir-

onment and Conservation Council, 1992.

[4] M.L. Lopez, J.A. Santaballa, Education in Chemistry,

1999, pp. 162.

[5] M. Novic, B. Pihlar, M. Dular, Fresenius’ Z. Anal. Chem.

332 (1988) 750.

[6] E.O. Hartwig, J.A. Michael, Environ. Sci. Technol 12

(1978) 712.

[7] T. Sakai, H. Takio, N. Teshima, H. Nishikawa, Anal.

Chim. Acta 438 (2001) 117.

[8] A. Sanz-Martinez, A. Rios, M. Valcarcel, Anal. Chim.

Acta 284 (1993) 189.

[9] V.V. Kuznetsov, M.V. Murasheva, J. Anal. Chem. 51

(1996) 993.

[10] J.M. Costa-Fernandez, M.E. Diaz-Garcıa, A. Sanz-Medel,

Anal. Chim. Acta 360 (1998) 17.

[11] H.L. de Medina, Chemical parameters, in: L.M.L. Nollet

(Ed.), Handbook of Water Analysis, Marcel Dekker, New

York, Basel, 2000, p. 921.

Fig. 4. DO concentrations in reactor measured on-line over a 7

h period (1230�/1930) using the rFIA method and a DO

electrode. Theoretical DO saturation values are shown for

comparison.




https://www.researchgate.net/publication/274920072_Limnological_Analyses?el=1_x_8&enrichId=rgreq-457fc1b236fb0973c52392201a54a408-XXX&enrichSource=Y292ZXJQYWdlOzIzNDM2NTA3O0FTOjEwMTI5MDA1MDg1MDgyNEAxNDAxMTYwODM4NjE2

https://www.researchgate.net/publication/274920072_Limnological_Analyses?el=1_x_8&enrichId=rgreq-457fc1b236fb0973c52392201a54a408-XXX&enrichSource=Y292ZXJQYWdlOzIzNDM2NTA3O0FTOjEwMTI5MDA1MDg1MDgyNEAxNDAxMTYwODM4NjE2

VITA

Name: Miss Saowapha Muangkaew

Date of Birth: August 31, 1974

Academic status:

- B.Sc. (Education), Second Class Honors, Prince of Songkla University, 1995

- M.S. (Applied Analytical and Inorganic Chemistry), Mahidol University,

1999

- Ph.D. student in analytical chemistry, Chiang Mai University

Practical Experience:

- Visiting Graduate Student of Water Studies Centers, Monash University,

Australia, 2001

Scholarships

The Ministry of University Affairs (PSU, Surathani Campus)

The Postgraduate Education and Research Program in Chemistry (PERCH)

University Mobility in Asia and the Pacific (UMAP)

List of Publications:

International Journals

1. S. Muangkaew, I.D. McKelvie, M.R. Grace, M. Rayanakon, K. Grudpan,

J. Jakmunee and D. Nacapricha, “A reverse-flow injection analysis method

for the determination of dissolved oxygen in fresh and marine waters”

Talanta. 58, 1285-1291, 2002.



104

International Conferences

1. S. Muangkaew, I. D. McKelvie, M. R. Grace, M. Rayanakorn, K. Grudpan,

J. Jakmunee, and D. Nacapricha, “The Development of A Reversed FIA

Method for the Determination of Dissolved Oxygen In Fresh and Marine

Water” The 11th International Conference on Flow Injection Analysis and The

38th Semi-annual Meeting of the Japanese Association for Flow Injection

Analysis, Chiang Mai, 2001.

National Conferences

1. K. Grudpan, R. Chantiwas, S. Muangkaew, P. Aumpan, W. Praditwiengcome,

L. Hamer, J. Jakmunee and M. Rayanakorn “Determination of Some

Anions in Water Sample by Flow injection Dialysis-Ion Chromatography”

25th Congress on Science and Technology, Pitsanuloke, 1999.

2. S. Muangkaew, M. Rayanakorn, K. Grudpan, J. Jakmunee, and D. Nacapricha

“Derivatization of Iodide Prior to HPLC Determination” RGJ Seminar

NO.2 Analytical Chemistry and Chemistry In The North, Chiang Mai, 2000.


“Development of High Performance Liquid Chromatography For The Determination

of Iodide at Trace Levels” 26th Congress on Science and Technology, Bangkok, 2000.


“Effect of Ionic Strength on the Determination of Iodide by Flow Injection

Potentiometry” 27th Congress on Science and Technology, Songkla, 2001.



https://www.researchgate.net/publication/23436507_A_reverse-flow_injection_analysis_method_for_the_determination_of_dissolved_oxygen_in_fresh_and_marine_waters?el=1_x_8&enrichId=rgreq-457fc1b236fb0973c52392201a54a408-XXX&enrichSource=Y292ZXJQYWdlOzIzNDM2NTA3O0FTOjEwMTI5MDA1MDg1MDgyNEAxNDAxMTYwODM4NjE2










105

5. S. Muangkaew, I. D. McKelvie, M. R. Grace, M. Rayanakorn, K. Grudpan,

J. Jakmunee, and D. Nacapricha, “Performance Test of the rFIA System

Developed For Dissolved Oxygen Determination” 28th Congress on

Science and Technology, Bangkok, 2002.

6. S. Muangkaew, J. Jakmunee, S. Lapanantnoppakhun, M. Rayanakorn,

D. Nacapricha, and K. Grudpan, “Determination of Iodide by Flow Injection

In-Valve Derivatization with High Performance Liquid Chromatography”

29th Congress on Science and Technology, Khon Kaen, 2003.



A reverse-flow injection analysis method for the determination of dissolved oxygen in fresh and...

Documents

Transcript of A reverse-flow injection analysis method for the determination of dissolved oxygen in fresh and...