A capillary zone electrophoresis (CE) method was developed for the determination of the biocide 2,2-dibromo-3-nitrilo-propionamide (DBNPA) in water used in cooling systems. The biocide is indirectly determined by CE measurement of the concentration of bromide ions produced by the reaction between the DBNPA and bisulfite. The relationship between the bromide peak areas and the DBNPA concentrations showed a good linearity and a coefficient of determination (R2) of 0.9997 in the evaluated concentration range of 0–75 μmol L−1. The detection and quantification limits for DBNPA were 0.23 and 0.75 μmol L−1, respectively. The proposed CE method was successfully applied for the analysis of samples of tap water and cooling water spiked with DBNPA. The intra-day and inter-day (intermediary) precisions were lower than 2.8 and 6.2%, respectively. The DBNPA concentrations measured by the CE method were compared to the values obtained by a spectrophotometric method and were found to agree well.

INTRODUCTION

The growth of micro-organisms such as bacteria, fungi, algae, and yeast in water used in industrial activities can result in slime and biofilm formation on the surfaces of equipment, particularly in cooling systems. This biofouling can cause major technical and economic problems, so a water treatment to prevent and control the growth of micro-organisms is usually required.

Industrial chemical biocides have been widely used to kill micro-organisms. According to the chemical properties and the mechanism of action of the biocides, they are usually classified into two major groups, oxidizing and non-oxidizing (Paulus 1993). Among the most common biocides, chlorine, chlorine dioxide, bromine, hydrogen peroxide, and ozone are considered oxidizing biocides because of their high oxidation power. In contrast, quaternary ammonium salts, carbamates, glutaraldehyde, organothiocyanates, biguanides, isothiazolins, tetrakis(hydroxymethyl) phosphonium sulfate, and 2,2-dibromo-3-nitrilo-propionamide (DBNPA) are examples of non-oxidizing biocides, because their mechanism of action does not involve oxidation reactions.

DBNPA is a halogenated amide (Figure 1) that is considered a non-oxidizing, quick kill, and broad spectrum biocide (Wolf et al. 1972; Paulus 1993; Huang et al. 2009). These features have caused DBNPA to be widely used in industrial water treatments, particularly in the paper industry (Lacorte et al. 2003). DBNPA has an electrophilic nature that makes it prone to react irreversibly with the nucleophilic thiol groups of the cytoplasmic proteins. Thus, the mechanism of action of this biocide is based on the inactivation of cellular proteins, particularly those containing the amino acids methionine and cysteine (Paulus 1993).

Figure 1

Chemical structure of DBNPA.

Figure 1

Chemical structure of DBNPA.

DBNPA hydrolyzes in aqueous solution, and the rate of this degradation increases under alkaline conditions (Paulus 1993). This biocide also reacts readily with sulfur-containing nucleophilic compounds, such as hydrogen sulfide and bisulfite.

DBNPA has been considered an environmentally friendly biocide because it degrades quickly in aqueous solution, producing non-toxic compounds (Exner et al. 1973a, b). Nevertheless, in a recent study (Chen 2012), DBNPA was found to cause chronic deleterious effects in aquatic organisms. The usual concentrations of DBNPA in water treatment, particularly cooling water can vary from 4 to 80 μmol L−1, depending on the level of contamination (2002; Bertheas et al. 2009).

The determination of DBNPA in water is important to establish its effective dose for treatment, monitor industrial effluents, and investigate the degradation of this biocide (Lacorte et al. 2003). Nevertheless, there are few methods of analysis reported in the literature (Abrantes et al. 1998; Rigol et al. 2002; Lacorte et al. 2003) for the determination of DBNPA. For example, Rigol et al. (2002) employed liquid chromatography-electrospray ionization-mass spectrometry (LC-ESI-MS) for determination of DBNPA in effluent waters. Abrantes et al. (1998) developed a micellar electrokinetic chromatography method for the detection of DBNPA and other biocides in paper for food packaging.

This paper describes the development and application of a novel analytical method for the determination of DBNPA in water. This method consists of the reaction of the biocide with bisulfite to produce bromide ions, which are then detected by capillary zone electrophoresis (CE). By using this method, five different samples of water spiked with DBNPA were analyzed. To the authors’ best knowledge, this work is the first use of CE for the determination of DBNPA in water.

METHODS

Reagents, samples, and solutions

All reagents were of analytical grade. DBNPA was purchased from Sigma–Aldrich (Saint Louis, MO, USA), and sodium bisulfite, sodium bromide, potassium iodide, potassium thiocyanate, and dihydrogen sodium phosphate were purchased from Synth (Diadema, Brazil). Ultra-pure water was obtained using a Direct-Q3 UV Water Purification System (Millipore, Molsheim, France). The background electrolyte solution (BGE) was a 10 mmol L−1 sodium phosphate solution (pH 6.5, adjusted with NaOH solution). A 10 mmol L−1 standard stock solution of NaBr was prepared by dissolving the required amount of the salt in ultra-pure water. This solution was diluted as required to prepare the working standard solutions used to obtain the analytical curves for bromide ions. Standard stock solutions of DBNPA (10 mmol L−1) and sodium bisulfite (10 mmol L−1) were prepared daily by dissolving the solid reagents in ultra-pure water. No adjustment of the pH (about 6.5) of the DBNPA solution was conducted and it was stable for at least 8 h at room temperature (20–25 °C). Potassium thiocyanate solution was added (250 μmol L−1) to the standard and sample solutions as an internal standard.

Tap water samples were obtained from our laboratory, and the other samples of water were collected from four different circulating cooling water systems located at the Institute of Chemistry of University of Campinas (IQ-Unicamp) and the Centro Nacional de Pesquisa em Energia e Materiais (CNPEM), Campinas, Brazil.

CE method

The CE method used for the determination of bromide ions was adapted from Pascali et al. (2006). An Agilent 7100 capillary CE system (Agilent, Waldbronn, Germany) equipped with a diode-array spectrophotometric detector was used. A bare fused silica capillary of 40.0 cm total length (31.5 cm effective) and 50 μm internal diameter was used. The sample solutions were hydrodynamically injected into the capillary using 50 mbar pressure for 15 s. The separation voltage was −25 kV, and the UV detection was performed at 200 nm with a data acquisition rate of 2.5 Hz. Before the first run of the day, the capillary was successively flushed for 5 min with each of the following: 1 mol L−1 NaOH aqueous solution, ultra-pure water, and finally BGE. After each experimental run, the capillary was flushed with BGE for 1 min.

Analytical curves for bromide were obtained by triplicate injections of seven working standard solutions containing 0, 25, 50, 75, 100, 125, and 150 μmol L−1 of bromide. The areas under the bromide peak in the electropherograms were obtained, and the analytical curves were then plotted as a ratio of the peak area of bromide to that of the internal standard vs. bromide concentration. Peak integration was performed by Chemstation software (Agilent), and statistical analyses were conducted using Origin 8.1 software (OriginLab, Northampton MA, USA).

Sample preparation

The water samples were spiked with the stock solution of DBNPA to achieve a final concentration of 40 μmol L−1, which is a typical concentration in treated cooling water. To 350 μL of the spiked water samples, sodium bisulfite solution was added to a final concentration of 2 mmol L−1. The internal standard (potassium thiocyanate) was then added (250 μmol L−1), and the samples were allowed to stand for at least 10 min before the injection into the CE system. Each spiked water sample was also analyzed to measure the blank concentration of bromide before the addition of bisulfite.

Spectrophotometric method

The accuracy of the CE method was evaluated by comparison with a spectrophotometric method published by the Dow Chemical Company (Dow 2002). This method is based on the reaction between the biocide and potassium iodide to form the triiodide complex (I3), which can be spectrophotometrically detected at 352 nm. Analytical curves were obtained using working standard solutions of DBNPA at concentration levels of 0, 10, 20, 30, 40, and 50 μmol L−1. These solutions were prepared by transferring aliquots of the biocide stock solution to 10 mL volumetric flasks, followed by the addition of 2.5 mL of 0.1 mol L−1 KI solution and 40 μL of 0.5 mol L−1 HCl solution. Finally, the flasks were filled up to the mark with ultra-pure water. The analyzed water samples were prepared in a similar way by adding the KI and HCl solutions to volumetric flasks that were filled up to the mark with the water sample. The samples and the working standard solutions were left to stand for 5–10 min before absorbance measurements, which were performed with a diode-array spectrophotometer HP 8453 (Hewlett-Packard, Palo Alto, CA, USA) using a quartz cuvette of 1 cm path length.

RESULTS AND DISCUSSION

CE separation

DBNPA is a neutral compound, so its direct determination by capillary zone CE is not favorable. However, in the proposed method, DBNPA was first reacted with sodium bisulfite (Figure 2) to produce 2 moles of bromide ions for each mole of the biocide. Using this stoichiometry and quantifying the concentration of the bromide ions produced, the DBNPA concentration could be indirectly measured. Figure 3 compares the electropherograms of standard solutions of NaBr (100 μmol L−1) and DBNPA (50 μmol L−1) after reaction with bisulfite. The peak areas for bromide were similar for both solutions, in accordance with the reaction shown in Figure 2.

Figure 2

Reaction between DBNPA and bisulfite.

Figure 2

Reaction between DBNPA and bisulfite.

Figure 3

Electropherograms of standard solutions of (a) NaBr (100 μmol L−1), and (b) DBNPA (50 μmol L−1) in the presence of bisulfite (2 mmol L−1). Peaks: (1) bromide, (2) thiocyanate (internal standard).

Figure 3

Electropherograms of standard solutions of (a) NaBr (100 μmol L−1), and (b) DBNPA (50 μmol L−1) in the presence of bisulfite (2 mmol L−1). Peaks: (1) bromide, (2) thiocyanate (internal standard).

The original CE method (Pascali et al. 2006) for bromide detection employed borate buffer (pH 9.2) as BGE, but DBNPA quickly degrades in alkaline medium. This degradation was a drawback for the measurement of the blank concentration of bromide in the samples. Nevertheless, using phosphate buffer solution at pH 6.5 as BGE, no degradation of the DBNPA was observed.

As in the original method, thiocyanide was maintained as the internal standard because its peak showed good resolution, and no detectable amount of this anion was found in the analyzed samples.

Figures of merit

Table 1 summarizes the main analytical parameters of the CE method, which were obtained in accordance with the recommendations of the literature (Ribani et al. 2004).

Table 1

Main figures of merit of the CE method

Migration time (min)a 2.18 ± 0.01 
Regression equationb y = 0.0215 × +0.0015 
Linear range (μmol L−10–75 
Determination coefficient (R20.9997 
Detection (μmol L−1)c 0.23 
QL (μmol L−1)d 0.75 
Intra-day instrumental precision (%)e 2.1 
Migration time (min)a 2.18 ± 0.01 
Regression equationb y = 0.0215 × +0.0015 
Linear range (μmol L−10–75 
Determination coefficient (R20.9997 
Detection (μmol L−1)c 0.23 
QL (μmol L−1)d 0.75 
Intra-day instrumental precision (%)e 2.1 

aFor bromide ion; mean and standard deviation for nine consecutive injections.

bx = concentration of DBNPA (μmol L−1); y = ratio of the peak area of the bromide to the peak area of the internal standard.

cS/N = 3.

dS/N = 10.

eRelative standard deviation (n = 9) for the ratio of the peak area of bromide to the peak area of internal standard.

The calculated quantification limit (QL) of the CE method was low enough for measurement of the lowest DBNPA concentration used in cooling water, i.e. 4 μmol L−1. The analytical curve over the evaluated concentration range showed a good coefficient of determination (R2), and the linearity was validated by the lack-of-fit test (Danzer et al. 1998) (data not shown). Analytical curves using bromide and DBNPA standard solutions were obtained and compared, but no statistical differences between their slope values were observed. Thus, the bromide standard solutions were chosen because they are more stable than DBNPA solutions. Moreover, the reagent (NaBr) is less expensive and does not require reaction with bisulfite before injection in the CE system. The intra-day instrumental precision was evaluated by nine consecutive injections of a standard solution of DBNPA (40 μmol L−1) containing bisulfite (2.0 mmol L−1) and thiocyanide (250 μmol L−1). The relative standard deviation in terms of the ratio of peak areas (bromide/thiocyanide) and migration time were 2.1 and 0.5%, respectively.

Application of the CE method

The CE method was applied for the determination of DBNPA in samples of tap water and cooling water spiked with this biocide. Figure 4(a)(b) shows the electropherograms of the tap water before and after the spiking with DBNPA. No differences in the electropherogram profiles were observed. However, after bisulfite was added to the DBNPA-spiked sample, a bromide peak was detected (Figure 4(c)), due to the decomposition of the biocide (Figure 2). No peak was observed for the sulfate ions produced by the reaction because this anion has low absorptivity at the detection wavelength used (200 nm). Nevertheless, for all samples, an unidentified peak between the bromide and the internal standard peaks was detected that did not interfere with the analysis. The blank concentration of bromide ions was determined for all samples and subtracted from that measured after the bisulfite addition. The accuracy of the proposed CE method was evaluated by comparison with the spectrophotometric method. The analysis results are summarized in Table 2.

Table 2

Results for the determination of DBNPA in water samples by the CE and the spectrophotometric methods

  Concentration measured  
  (mean ± SDb)/μmol L−1 Accuracy 
Sample CE Spectrophotometric Error/% Calculated t-valuec 
Tap water 39 ± 2 41.2 ± 0.4 −5.3 2.3 
CW1a 41 ± 1 40 ± 2 2.5 0.8 
CW2 38.3 ± 0.8 39.7 ± 0.4 −3.5 2.7 
CW3 38.8 ± 0.8 38.1 ± 0.5 1.8 1.3 
CW4 36 ± 1 40 ± 2 10.0 3.1 
  Concentration measured  
  (mean ± SDb)/μmol L−1 Accuracy 
Sample CE Spectrophotometric Error/% Calculated t-valuec 
Tap water 39 ± 2 41.2 ± 0.4 −5.3 2.3 
CW1a 41 ± 1 40 ± 2 2.5 0.8 
CW2 38.3 ± 0.8 39.7 ± 0.4 −3.5 2.7 
CW3 38.8 ± 0.8 38.1 ± 0.5 1.8 1.3 
CW4 36 ± 1 40 ± 2 10.0 3.1 

aCW = cooling water.

bSD = standard deviation (n = 3).

cCalculated Student's t-value; 2.78 is the tabulated t-value at a confidence level of 95% and 4 degrees of freedom.

Figure 4

Electropherograms of tap water sample (a) before, and (b) after DBNPA treatment (40 μmol L−1), (c) after bisulfite addition in the DBNPA-spiked sample. Peaks: (1) bromide, (*) unidentified peak, (2) thiocyanate (internal standard).

Figure 4

Electropherograms of tap water sample (a) before, and (b) after DBNPA treatment (40 μmol L−1), (c) after bisulfite addition in the DBNPA-spiked sample. Peaks: (1) bromide, (*) unidentified peak, (2) thiocyanate (internal standard).

The intra-day precisions (n = 3) for the CE method varied from 2.1 to 2.8% and were comparable to the values obtained by the spectrophotometric method (1–5%). The inter-day (intermediary) precision evaluated for the analysis of the tap water sample in three different days was 6.2%. Student's t-test was conducted at a confidence level of 95% and detected statistically significant differences only between the concentrations measured for the sample CW4. This difference can probably be ascribed to the high ionic content of this sample, which can affect the accuracy of the CE method using an external calibration method. This hypothesis is based on the fact that the concentrations of DBNPA found in the sample CW3 by both methods are statistically similar. However, this sample was obtained from the sample CW4 after undergoing an ion exchange process (softening). Thus, for analysis of samples with high ionic strength a standard addition method is recommended instead of external calibration.

The proposed CE method was faster and required less sample and reagent than the spectrophotometric method.

Degradation detection assay

The CE method was evaluated for the detection of the degradation of DBNPA in water. Figure 5 shows the electropherograms of the sample CW2 as soon as the treatment with DBNPA was conducted and 3 days after the treatment. A peak for bromide appears for the stored sample due to the DBNPA hydrolysis. These results suggested that the proposed CE method can also be useful for degradation studies of DBNPA in water.

Figure 5

Electropherograms of a sample of cooling water (a) after spiking with DBNPA (40 μmol L−1) and (b) 3 days after the spiking. Peak attribution as in Figure 4.

Figure 5

Electropherograms of a sample of cooling water (a) after spiking with DBNPA (40 μmol L−1) and (b) 3 days after the spiking. Peak attribution as in Figure 4.

CONCLUSIONS

Although, the biocide DBNPA is widely used in the treatment of industrial water, few analytical methods for the detection of this biocide have been reported in the literature. This paper demonstrated that an indirect determination of DBNPA could be performed by using CE to quantify the bromide ions produced in the reaction between the biocide and bisulfite. The proposed method was demonstrated to be simple and rapid, and it showed good precision and accuracy for the determination of DBNPA in samples of water used in cooling systems.

ACKNOWLEDGEMENTS

This work was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAFESP, grants no 2013/22485-6) and the Conselho Nacional de Desenvolvimento Científicoe Tecnológico (CNPq, grants no 305318/2012-8). The authors thank Leandro Y. Shiroma and the Centro Nacional de Pesquisa em Energia e Materiais (CNPEM) for providing the samples of cooling water.

REFERENCES

REFERENCES
Abrantes
S.
Philo
M.
Damant
A. P.
Castle
L.
1998
Determination of extractable biocides in paper food packaging materials using micellar electrokinetic chromatography
.
Journal of Microcolumn Separations
10
(
5
),
387
391
.
Bertheas
U.
Majamaa
K.
Arzu
A.
Pahnke
R.
2009
Use of DBNPA to control biofouling in RO systems
.
Desalination and Water Treatment
3
(
1–3
),
175
178
.
Dow
2002
Antimicrobial 7287 and DOW Antimicrobial 8536. The fast-acting, broad-spectrum biocides with low environmental impact Form No. 253-01464-06/18/02
,
Dow Chemical Company
,
Buffalo Grove, Illinois, USA
.
Exner
J. H.
Burk
G. A.
Kyriacou
D.
1973a
Degradable, new biocide – rates and products of decomposition of 2,2-dibromo-3-nitrilopropionamide
.
Abstracts of Papers of the American Chemical Society
(
26
),
67
.
Exner
J. H.
Burk
G. A.
Kyriacou
D.
1973b
Rates and products of decomposition of 2,2-dibromo-3-Nitrilopropionamide
.
Journal of Agricultural and Food Chemistry
21
(
5
),
838
842
.
Huang
C. Y.
Hsieh
S. P.
Kuo
P. A.
Jane
W. N.
Tu
J.
Wang
Y. N.
Ko
C. H.
2009
Impact of disinfectant and nutrient concentration on growth and biofilm formation for a pseudomonas strain and the mixed cultures from a fine papermachine system
.
International Biodeterioration and Biodegradation
63
(
8
),
998
1007
.
Lacorte
S.
Latorre
A.
Barcelo
D.
Rigol
A.
Malmqvist
A.
Welander
T.
2003
Organic compounds in paper-mill process waters and effluents
.
Trac-Trends in Analytical Chemistry
22
(
10
),
725
737
.
Pascali
J. P.
Trettene
M.
Bortolotti
F.
de Paoli
G.
Gottardo
R.
Tagliaro
F.
2006
Direct analysis of bromide in human serum by capillary electrophoresis
.
Journal of Chromatography B – Analytica Technologies in the Biomedical and Life Sciences
839
(
1–2
),
2
5
.
Paulus
W.
1993
Microbicides for the Protection of Materials – A Handbook
.
Chapman & Hall
,
London, UK
.
Ribani
M.
Bottoli
C. B. G.
Collins
C. H.
Jardim
I. C. S. F.
Melo
L. F. C.
2004
Validation for chromatographic and electrophoretic methods
.
Quimica Nova
27
(
5
),
771
780
.
Wolf
P. A.
Sterner
P. W.
1972
2,2-Dibromo-3-nitrilopropionamide, a compound with slimicidal activity
.
Applied Microbiology
24
(
4
),
581
.