Che Yiping^{1, 2}, DingXiaojing², ZhaoShan², Liu Pengyan¹
(¹ College of Chemistry and Environmental Science, Hebei University, Baoding 071002, China; ²Beijing Center for Disease Prevention and Control, Beijing 100013, China)

Abstract A new Reversed Phase High Performance Liquid Chromatographic method (RP-HPLC) has been established for the rapid determination of polyhexamethylenebiguanide (PHMB) in compound chemical disinfectants. PHMB is well separated from matrices on a Shiseido C₁₈ (150 × 4.6mm, 5mm) column with a temperature at 30ºC, using a mobile phase of A:B = 16:84 (A: acetonitrile, B: 20mmol/L ammonium acetate, adjusted with acetic acid to pH 4.0) and detected at 235nm. Good linearity was in the range from 100mg/L to 1000mg/L with relative standard deviation of 1.2% (n = 7). Recoveries at three different levels (100, 200, 400mg/L) were 99.06, 102.4 and 92.92%, respectively. The limit of detection was 50mg/L (S/N = 3). The present method is simple, rapid and suitable for the routine accurate assay of PHMB in compound chemical disinfectants.
Keywords RP-HPLC, polyhexamethylenebiguanide, compound chemical disinfectants

1 INTRODUCTION
Polyhexamethylenebiguanide (PHMB), a newly reported disinfectant in recent years, is a polymeric compound with low skin irritancy, low eye toxicity, fast speed of kill, stable and effective over wide pH range (2-11). It is un-soluble in water and usually prepared in the water-soluble form of hydrochloride salt at a concentration of 20% aqueous solution (w/v) with a molecular structure as shown in Fig.1.

Fig.1 The molecular structure of PHMB

    It has been found to be effective against a wide variety of bacteria and virus at low dose levels even in hard water. For example, PHMB at a concentration of 0.1% (w/v) can kill AIDS virus within 5 min. Most of all, PHMB is degradable and harmless. Therefore, it has been widely used for the formulation of compound chemical disinfectants and sanitizers, for use in industrial, agricultural, food and personal care product (i.e. ophthalmic solutions) disinfection applications. Its levels vary significantly in different application products, ranging from high concentrations (0.1% (w/v)) in several compound chemical disinfectants to a low concentration (0.0001% (w/v)) in ophthalmic solutions. If the concentration of PHMB is lower than the manufacturer's specification, the disinfectants will not effectively kill bacteria and virus. In order to supervise the quality of compound chemical disinfectants, the Ministry of Public Health of China has recommended that the accurate assay of PHMB in disinfectants is one of the effective methods for quality control purpose and instrumental methods are recommended for the first choice ^[1]. Therefore, instrumental methods with rapidity and simplicity for the accurate assay of PHMB in compound chemical disinfectants are required for product quality control.
    Reversed-Phase High Performance Liquid Chromatography (RP-HPLC) with Photo Diode Array detector (PDA) is the most frequently used method for routine determination. However, to the best of our knowledge, no RP-HPLC method for the analysis of PHMB in disinfectants has been reported. So, the aim of the present work is to establish a simple and rapid HPLC method for routine assay of PHMB in compound chemical disinfectants.

2. EXPERIMENTAL
2.1 Reagents and chemicals
The water for the preparation of all solutions was made by a Millipore Milli-QRG ultra-pure water system (Bedford, MA, USA). The PHMB (20%) standard was supplied by Lvsan Chemical Ltd. Corp. (Beijing, China). HPLC-grade acetonitrile was purchased from Dima Technology Inc. (USA). Analytical grade ammonium acetate and acetic acid were purchased from Beijing Chemical Reagents Company (Beijng, China). Stock standard solution of 2000mg/L PHMB was prepared by dilution of 1mL PHMB standard with 100mL 0.01mol/L HCl and stored at 4ºC. Working solutions (100, 200, 400, 800, 1000mg/L) were prepared by serial dilution of the stock solution with 10mmol/L HCl and stored at 4ºC.
2.2 Apparatus
A Waters 2695-996 high performance liquid chromatography (Milford, MA, USA) was employed with a Waters Empower Chromatography Manager Workstation for instrument control as well as data acquisition and processing. The chromatographic procedure employed was isocratic with a mobile of acetonitrile-20mmol/L ammonium acetate acidified with acetic acid, pH 4.0 (16:84, v/v). The column used was Shiseido C₁₈ (150×4.6mm, 5mm) (Shiseido Co. Ltd. Tokyo, Japan). Chromatography was carried out with the column at 30ºC. The flow rate was 1mL/min with injection volume of 20m L. UV chromatograms were extracted at a wavelength of 235nm.
    PB303-S balance was purchased from Mettler -Toledo Instruments Co. (Switzerland). JC-420pH/mV meter was purchased from Beijing Chuangye Instrumental Plant (Beijing, China).

3 RESULTS AND DISCUSSION
3.1 Choice of detection wavelength
In general, the absorbance detection is the first choice in HPLC analysis of PHMB owing to its simplicity and reliability. 20mL 200mg/L PHMB was directly injected into the HPLC-system without column and scanned from 200 to 400 nm. The ultraviolet absorption spectrum of PHMB in chromatographic mobile phase is illustrated in Fig.2. The maximum absorption is 235.7nm. Therefore, the detection wavelength was selected at 235nm.

Fig.2 The ultraviolet absorption spectrum of PHMB in chromatographic mobile phase. Chromatographic conditions are shown in section 2.2.

3.2 Choice of chromatographic separation conditions
In general, PHMB exists in cationic form in aqueous solution ^[2]. So, an anion-pair agent was added to form neutral ion-pairs to be retained by the C₁₈ stationary phase. To further improve the retention behavior of the ion-pair, inorganic salts such as formate, phosphate and acetate were also added to produce a salting-out effect, respectively. The experiment showed that acetate was a suitable ion-pair agent which results in a good peak shape of PHMB.
    The concentration of acetate in the mobile phase must be high (10mmol/L) enough to provide good buffer capacity. However, higher concentration will result in potential corrosive to pump seals. When 16% acetonitrile and the pH 4.0 of the mobile phase were kept unchanged, five concentration levels of NH₄AC at 12, 16, 20, 24 and 28mmol/L were added to the aqueous portion of the mobile phase, respectively. There were no significant effects on both the retention time and peak area of PHMB. As a compromise, 20mmol/L was the best choice.
    Acetonitrile is preferred to methanol in frequently used RP-HPLC because of its relatively higher elution strength as well as the lower column back pressure it creates. With the increase in acetonitrile concentration, the retention time of PHMB is reduced when 20mmol/L NH₄AC (pH = 4.0) remains unchanged in the mobile phase. Taking into the separation efficiency and time, the concentration of acetonitrile was set at 16%.
    The pH of the mobile phase is one of the key factors affecting both the retention behavior and peak shape of ionic analytes. When 16% acetonitrile and 20mmol/L NH₄AC was kept unchanged, better peak shape of PHMB could be obtained at pH 4.0 compared with pH 4.5 and 5.0, respectively. So, pH = 4.0 of the mobile phase was chosen.
    Under the above optimized conditions, the chromatogram of PHMB standard solution is shown in Fig.3.

Fig.3 Chromatogram of PHMB (200mg/L) in a standard solution Chromatographic conditions are shown in section 2.2
               [1. PHMB (t = 5.5min)]

    It was shown that PHMB is well separated and eluted at a reasonable time. The retention factors (k) for PHMB under isocratic separation conditions can be calculated as ^[3]
k = (t_R – t₀)/t₀                   (1)
    Where t_R and t₀ are the retention time of PHMB and the column dead time, respectively. The column dead volume for a 4.6 mm i.d. column can be estimated as
V_M0.1 L                             (2)
    Where V_M is the column dead volume in milliliters and L is the column length in centimeters. So, for current column, L is 150 mm or 15 cm, and 0.1×15 = 1.5mL. Convert V_M to t₀ by dividing the volume by the flow rate: 1.5mL/1.0mL/min = 1.5min. k value for PHMB is 2.7, which falls into the ideal range (2< k< 10) for separation.
    The theoretical plate height, H, is obtained from the chromatogram by:
H = (L /16) (W_t / t_R) ²             (3)
    Where L is the column length; W_t is the peak width at the baseline (in seconds); and t_R is the retention time (in seconds). The calculated H for PHMB was 0.046mm and the theoretical plate number is 3191.

3.3 Detection limit, precision and linear range
Under the optimized chromatographic conditions, 20mL of each PHMB working solutions was injected. The peak area (A) and the concentration (C, mg/L) of PHMB had good linear relationship. The regression equation was A = 0.0006C + 19.328 with a correlation coefficient r = 0.9997. The linear range was from 100 to 1000 mg/L. The detection limit (S/N = 3) was 50mg/L. The precision was evaluated by performing seven replicate analysis of a standard concentration of 200mg/L. The relative standard deviation for retention time and peak area were 0.7% and 1.2%, respectively.
3.4 Real sample analysis
3.4.1 Real sample determination
Three samples collected from the markets were diluted with ultra-pure water and determined in triplicates after filtered through a 0.45 mm filter membrane. The results were 0.21, 0.20 and 0.22% (w/v), respectively (see Table 1). Much attention should be paid that several drops of acetonitrile must be added to eliminate the foaming formed in the process of dilution.

Table 1 Analysis of real samples