Determination of trace amounts of mercury ion by using a modified Belousov-Zhabotinskii oscillating chemical reaction

Gao Jinzhang, Wang Wenbin, Ren Jie, Guo Miao, Jin Lili, Chen Xiaodong, Wang Jie, Yang Wu
(College of Chemistry & Chemical Engineering, Northwest Normal University, Lanzhou 730070, China)

Abstract A highly sensitive method for the determination of trace amounts of mercury ion by using the iodide-modified Belousov-Zhabotinskii (B-Z) oscillating chemical reaction was proposed. The results showed that the change ratio of oscillating period (PR) is directly proportional to the negative logarithm of concentrations of mercury ion in the range of 3.98¡Á10-10 — 6.31¡Á10-12 mol L-1 (n=10, r = 0.9985) with a detection limit of 1.68¡Á10-12 mol L-1. The proposed method has been used in the mercury determination of tap water and surface water and the satisfactory results were obtained.

    Since the first paper concerning the determination of ruthenium published in 1978, the study of oscillating chemical reaction in analytical chemistry has made a long progress [1-3], in which the combination of analyte pulse perturbation technique (APP) and continuous-flow stirred tank reactor (CSTR) could be considered as an important turning-point[4]. The basic principle is that when a stable regular oscillating chemical system was perturbed by analyte added, the response of system is correlative to the amounts of analyte added. That is to say, the changes of response-signal are directly proportional to the amounts of analyte. Commonly, the changes both amplitude and/or period were recorded to be response-signal and the stable regular oscillating profile as baseline. So far, many inorganic and organic substances have been successfully determined by using B-Z oscillating reaction in the range from 10-3 to 10-6 mol L-1 with a lower detection limit (ca. 10-6 — 10-8 mol L-1). To improve further the sensitivity of B-Z oscillating reaction, Strizhak et al [5] reported the determination of thallium by using the largest Lyapunov exponent in the transient chaotic regime and proposed a new method with very high sensitivity (detection limit £ 10-12 mol L-1). Vukojevic and Pejic et al [6-8] studied the characteristics of a non-equilibrium stationary state close to the bifurcation point and created successfully a novel kinetic method for the determination of organic compounds and inorganic ions. Gao et al [9] found that the B-Z system modified by sulfide become to be very sensitive to the some metal ions. As we know, B-Z oscillating chemical system is a far-from equilibrium complex redox reaction containing multiple steady states. The boundary conditions of different steady states are not the same so that the sensitivities of systems to the surroundings are also not the same. The reason that sulfide can modify the B-Z system from one steady state to another one may be considered to be owing to the multi-valence of sulfur appearing the redox process. If so, iodide should be also a modified regent for B-Z reaction due to iodine being a multi-valence element just like sulfur in the redox behavior, too. Following this clue, we used the KI solution to modify the classical B-Z oscillating reaction and proposed a high sensitivity method for determination of Hg2+ ion in water. All facts indicated that the sensitivity of determination using oscillating chemical reaction could be further improved if some of kinetic parameters were changed.
    2.1 Apparatus
    The experimental assembly consisted of an oscillation reactor (ca. 50 mL) and a potential measuring system. The reactor was coupled with a Model ML-902 magnetic stirrer and a Model 501 thermostat to maintain the temperature of the reaction sytem at 303 ¡À 0.1 K. A CHI-832 (CHI, USA) analyzer was directly connected to the reactor through two Pt electrodes in which one as working electrode and the other as counter electrode and a Hg2SO4 reference electrode against which the potential changes was recorded. A micro-injector was used to inject sample solution.
    2.2 Reagents
    All chemicals used to create the B-Z oscillating system such as KBrO3, CH2(COOH)2, H2SO4, 1,10-phenanthroline monohydrate, FeSO4¡Á 7H2O and KI were of analytical grade and used as received. Stock solution of mercury ion (sulfate salt) was prepared with doubly distilled-deionized water and standardized with EDTA. Working solutions with lower concentration were temporarily diluted with doubly distilled-deionized water just before use.
    2.3 Procedure
    A mixture solution containing 2.50¡Á10-3mol L-1 [Fe(Phen)3]2+, 0.20mol L-1 CH2(COOH)2, 0.60mol L-1 H2SO4 and 6.40¡Á10-4mol L-1 KI were added into the reactor in turn and finally 0.05mol L-1 KBrO3 solution was added. Then the magnetic stirrer was started and the temperature kept at 303 ¡À0.1 K. Meanwhile, the signal of potential was recorded (as showing in Figure 1). When the system reached the steady oscillating state, a few micro-liters of working solution containing various amounts of mercury ion were injected and the changes of the potential were recorded (as showing in Figure 2).

    Figure 1. a: The classical B-Z oscillating system profiles; b: The modified B-Z oscillating reaction profiles.
    Common conditions: [Fe(Phen)3]2+=2.50´ 10-3mol L-1, [KBrO3]= 5.00´ 10-2mol L-1, [CH2(COOH)2]= 2.00´ 10-1mol L-1, [H2SO4]= 0.60mol L-1 ,T=303K¡À0.1 K;
    a: [KI]= 0 mol L-1, b: [KI]= 6.40¡Á10-4mol L-1.

From Figure 1 it can be seen that there is a little difference between the modified oscillating profile and the classical oscillating profile. However, due to the presence of I, the modified oscillating profile is easier to be disturbed by Hg2+. As Figure 1b shown that the period increases in the first 3 cycles gently, and then, keeps constant from the 3rd to 18th period. That is, a stable regular profile was obtained as baseline. So Hg2+ sample should be injected during 3rd to 18th cycle, for example, the top of the 10th cycle. Figure 2 showed a comparison of the oscillating profiles perturbed by adding Hg2+ in the presence and absence of I at the same injection period. The concentration of Hg2+ injected in the modified system was 2.50¡Á10-11 mol L-1 while in the no modified system was only 2.50¡Á10-8 mol L-1. Results indicated clearly that the modified system is more sensitive than the no modified one for determinong Hg2+. If the periods before and after adding samples were defined as Po and P respectively, the change ratio of period PR could be calculated by the equation: PR = P/Po.

Figure 2. Perturbation of Hg2+on the classical B-Z oscillating system and the modified B-Z oscillating system. (The other conditions are as same as those in Figure 1)

    3.1 Optimization of experimental conditions
    As we known that the B-Z oscillating reaction was carried out in acidic medium. If the concentration of H2SO4 was too high or too low, regular oscillations would not be observed in the system. That is, the oscillating profiles become irregular. An acceptable scope was adopted ranging from 0.40 to 0.75 mol L-1 (see Figure 3a). In this study, 0.60mol L-1 of H2SO4 was used.
    Because both CH2(COOH)2 and KBrO3 are the essential substrates to create the B-Z oscillating chemical reaction, to keep the oscillating profile constant their variable concentration ranges are selected as 0.10-0.45 mol L-1 for the former and as 0.03-0.07 mol L-1 for the later, respectively (see Figure 3b and 3c). To determine Hg2+, the former is chosen as 0.20 mol L-1 and the later as 0.05 mol L-1.
    Here, the [Fe(Phen)3]2+ ion is considered as a catalyst, which was studied ranging from 1.50¡Á10-3 to 4.50¡Á10-3 mol L-1. Figure 3d indicates that 2.50¡Á10-3 mol L-1 was the optimum concentration for determining Hg2+.
    As shown in Figure 3e, the concentration of KI is very important for determining Hg2+ with high sensitivity using the modified B-Z oscillating system. Results tested in repeat indicated that 6.40¡Á10-4 mol L-1 of KI was the suitable concentration for determining Hg2+.
    In general, temperature affects the oscillating period. The lower temperature, the longer period of oscillating is. So 303K were chosen for all experiments.

Figure 3. Optimization of experimental conditions: T=303K, a: [H2SO4]; b: MA; c: KBrO3; d: [Fe(Phen)3]2+; e: KI.

3.2 Determination of Hg2+ ion
Under the optimum conditions described above, 0.2 mL different concentrations of Hg2+ were injected into the oscillation system to perturb the regular oscillating profile. A plot of PR against -logC (where C is the concentration of Hg2+) was made as shown in Figure 4. A larger linear range from 3.98¡Á10-10 to 6.31¡Á10-12 mol L-1 was obtained with a lower detection limit down to 1.68¡Á10-12 mol L-1. The linear relationship responsible can be described by the following equations:
PR= 0.9174 ¨C0.0786 [¨ClogC (Hg2+)]                  (n = 10, R = 0.9985)

Figure 4 Calibration curve for the determination of Hg2+.

3.3 Effect of foreign species on the determination of Hg2+
It is known that the oscillating chemical reaction is very vulnerable to the foreign species. To evaluate the selectivity for determination of Hg2+, some foreign inorganic ions and organic compounds were investigated and the results were shown in Table 1. Generally, most of metal ions have little influence on the determination. The tolerable ratio for organic compounds with small molecular weight (eg. methanol, ethanol, and formic acid etc) and Cu2+, Ag+, Au3+ ions was more than 50-fold. Anions such as Cl and Br have a little.

Table 1. Effect of foreign species on the determination of Hg2+ (2.50 ¡Á 10-11 mol L-1 )

Foreign species Tolerable ratio (foreign /Hg2+)
Na+, K+, Zn2+, Mg2+

Tl3+, Pb2+, Sn2+, Ni2+

Cu2+, Ag+, Au3+

Methanol, Ethanol, Formic acid





Cl, Br 10

3.4 Detection of water samples
The validity and practicability of proposed method for the determination of mercury ion was exemplified by assaying tap water and surface water samples. The water samples were filtered through 0.45¦Ìm filter to remove suspended substance and adjusted with diluted H2SO4 to pH=6.0 and then extracted with CCl4 to remove the organic substances. The results were shown in Table 2 and Table 3.

Table 2. The determination results of Hg2+ in tap water sample

1 2 3 4 5 6
(m g L-1)
7.74¡Á10-4 7.61¡Á10-4 7.79¡Á10-4 7.63¡Á10-4 7.76¡Á10-4 7.67¡Á10-4

Table 3. The determination results of Hg2+ in surface water sample

1 2 3 4 5 6
(m g L-1)
5.41¡Á10-4 5.47¡Á10-4 5.59¡Á10-4 5.33¡Á10-4 5.36¡Á10-4 5.47¡Á10-4
    Although many papers concerning the determination of metal ions by using the oscillating chemical reaction have been reported, highly sensitive methods are still very few. In this paper we proposed a modified B-Z oscillating chemical system by using KI solution which has been used for the determination of mercury ion with a lower detection limit down to 1.68¡Á10-12 mol L-1. The analysis results of real samples indicated that the proposed method has very high sensitivity, good reproducibility and acceptable selectivity. It has been used successfully for the determination of trance amount of mercury ion in water.

This work was supported in part by the Project of International Cooperation between China and Ukraine (043-05), the National Natural Science Foundation (20873101) and the Invention Project of Science & Technology (KJCXGC-01, NWNU), China.