Bacteria and Cyanobacteria Inactivation Using UV-C, UV-C/H₂O₂, and Solar/H₂O₂ Processes: A Comparative Study

Abstract

Effective disinfection processes have been investigated to provide pathogen-free drinking water. Due to growing concern about the potential negative effects of cyanobacteria in portable water, their treatment has gained more attention recently. This study aims to compare the inhibition efficiencies of Gram-negative bacteria (Escherichia coli; E. coli), Gram-positive bacteria (Bacillus subtilis; B. subtilis), and cyanobacteria (Microcystis aeruginosa; M. aeruginosa) using UV-C and solar irradiation, and their combination process with H₂O₂. Over 6 log removal value (LRV) of E. coli and B. subtilis was achieved within 1 min of UV-C irradiation (0.76 ± 0.02 mW/cm²). The solar and solar/H₂O₂ (50 mg/L) processes effectively reduced (>99%) both bacteria after 20 min. E. coli was more sensitive to hydroxyl radicals (•OH) compared to the B. subtilis due to a different cell wall structure, resulting in a 0.18–0.62 higher LRV than B. subtilis. However, solar-based processes did not effectively inhibit M. aeruginosa (>52.23%). The UV-C/H₂O₂ (50 mg/L) process showed the highest inhibition rate for M. aeruginosa (77.83%) due to the generation of •OH, leading to oxidative damage to cells. Additionally, chlorophyll-a (Chl-a) was measured to indicate cell lysis of M. aeruginosa. The removal rate of Chl-a extracted by viable M. aeruginosa was higher using the UV-C process (93.03%) rather than the UV-C/H₂O₂ process (80.95%), because UV-C irradiation could be most effective in damaging Chl-a.

1. Introduction

Safe and clean drinking water is essential for human health and society; however, it is still not sufficiently provided to 1.1 million people worldwide [1,2]. Waterborne bacteria such as Escherichia coli (E. coli) and Bacillus subtilis (B. subtilis) can cause human diseases if ingested by drinking water [3]. These waterborne bacteria can cause diseases such as diarrhea, cholera, and typhoid fever, contributing to an estimated 505,000 deaths annually from diarrhea in developing countries [4,5]. In addition, harmful cyanobacterial blooms (HCBs) negatively impact the production of safe and clean drinking water due to the deterioration of drinking water quality [6]. HCBs frequently occur during the hot season with the production of cyanotoxins and Microcystis aeruginosa (M. aeruginosa) is a representative type of cyanobacteria in freshwater systems during summer in South Korea [6,7]. M. aeruginosa is one of the most ecologically problematic species because it produces taste and odor compounds as well as microcystins, which pose serious risks to human health and the environment [8,9]. The quality of drinking water sources is affected by the occurrence of waterborne bacteria and HCBs; thus, water treatment processes should be developed to effectively reduce them. If the disinfection process is inappropriately applied, the produced drinking water can be re-contaminated by microorganisms during transportation through water pipes [10]. Therefore, the development of effective processes for microbial reduction is necessary for public hygiene and human health.

The major disinfection process of tap water utilizes chlorine-based disinfectants to produce hypochlorous acid and hypochlorite, and advanced oxidation processes (AOPs) such as ozone treatment have been recently used [11]. Chlorination and ozonation treatments are effective for disinfection; however, they can produce harmful by-products and the chemicals may pose health risks to operators in water treatment plants [12]. Alternatively, disinfection using ultraviolet (UV) light and sunlight has gained more attention as a chemical-free process and can effectively reduce bacteria and viruses [13,14,15]. For the removal of cyanobacteria, several studies have been performed using chlorination [16], coagulation and precipitation [17], ultrasound [18], and UV-C irradiation [19]. UV-C irradiation (200–280 nm) can effectively inactivate photosynthesis in M. aeruginosa and suppress its growth [19]. Additionally, a few studies have shown the potential use of solar irradiation for inactivation of pathogenic microorganisms, because UV light involves the generation of reactive oxygen species (ROS) in water [20]. Bolye et al. [21] reported complete inactivation of Campylobacter jejuni, Yersinia enterocolitica, E. coli, Staphylococcus epidemides, and B. subtilis by natural sunlight. Moreover, Eleren et al. [22] found that addition of H₂O₂ during solar irradiation significantly increases the inactivation of E. coli, because photolysis of H₂O₂ occurs by photons (wavelength < 300 nm) as shown in Equation (1):

$$H_2{O}_2\stackrel{hv}{\to } 2•\mathrm{OH}$$

(1)

The generated •OH radicals from the photolysis of H₂O₂ have a high oxidation potential and can further reduce bacteria and cyanobacteria [9,20]. Therefore, the combination of UV-C or solar irradiation with H₂O₂ as an oxidizing agent can promote microbial reduction within a shorter reaction time [20,22].

The aim of this study is to compare the inhibition of bacteria and cyanobacteria using UV-C, UV/H₂O₂, solar, and solar/H₂O₂ processes. Gram-negative bacteria (E. coli), Gram-positive bacteria (B. subtilis), and cyanobacteria (M. aeruginosa) were used to investigate the inhibition rates. In addition, chlorophyll-a (Chl-a) was measured to confirm the cell lysis of M. aeruginosa.

2. Materials and Methods

2.1. Preparation of Bacteria Solution

The two types of bacteria used in the study were Gram-negative bacteria, E. coli (ATCC 11105), and Gram-positive bacteria, B. subtilis (ATCC 6633), obtained from the Korean Culture Center of Microorganisms (KCCM). Luria-Bertani (LB) broth (25 g/L) was used for growth medium for both bacteria and each bacterial strain was inoculated, respectively. The LB broth inoculated with E. coli was cultured in a shaking incubator at 37 °C and 200 rpm for 18 h, while the medium inoculated with B. subtilis was cultured at 30 °C and 200 rpm for 24 h. The cultured E. coli and B. Subtilis were centrifuged using a centrifuge (1236R, Labogene, Allerød, Denmark) at 5000 and 8000 rpm for 15 min at 4 °C. After removing the supernatant, the bacteria cells were collected and washed with sterilized deionized water (Millipore, Burlington, MA, USA) thoroughly to remove residual medium. Each bacteria solution, E. coli and B. Subtilis, was prepared as an initial concentration of ~10⁶ CFU (colony forming unit)/mL (optical density at 600 nm (OD600) = 0.5) in autoclaved deionized water. The optical density of the solution was measured using a UV-visible spectrophotometer (GENESYS 150, Thermo Scientific, Lenexa, KS, USA).

2.2. Preparation of Cyanobacteria Solution

Cyanobacteria, M. aeruginosa (FBCC-A68), was obtained from the Freshwater Bioresources Culture Collection in South Korea (FBCC). The culture medium was prepared by diluting BG-11 (×50) using deionized water and then autoclaving (at 121 °C for 15 min) (ST-65G, Jeio Tech, Daejeon, Korea) it before use. M. aeruginosa was cultured at a temperature of 25 °C, light intensity of 2100 lux, and a light–dark cycle of 14:10 h (light:dark). Sub-culturing was performed weekly by inoculating 10% (v/v) of the pre-cultured solution into fresh BG-11 medium. The initial concentration of M. aeruginosa was adjusted as ~10⁶ cells/mL by diluting in autoclaved deionized water corresponding to the concentration of HABs [23].

2.3. UV-C, Solar, UV-C/H₂O₂, and Solar/H₂O₂ Processes

Two irradiation sources were employed: a UV reactor equipped with a low-pressure mercury lamp and a solar simulator (10500, ABET Technologies, Milford, CT, USA). The UV-C irradiation intensity, measured using the KI/KIO₃ actinometry method [24] was determined to be 0.76 ± 0.02 mW/cm². The solar simulator was calibrated to provide an output equivalent to one sun (AM 1.5G spectrum). Microbial suspensions (40 mL) were placed and stirred by a magnetic bar in a cylindrical quartz reactor and exposed to UV-C or sunlight. For the UV/H₂O₂ and solar/H₂O₂ process, hydrogen peroxide (30%, Daejung Chemicals, Siheung, South Korea) was diluted in deionized water to achieve initial concentrations of 20 and 50 mg/L, respectively. The hydrogen peroxide concentration was estimated using a hydrogen peroxide test kit (HYP-1, Hach, Loveland, USA).

The exposure time for UV-C irradiation, UV-C/H₂O₂, and solar/H₂O₂ process to inactivate E. coli and B. subtilis were set at 0, 1, 3, 5, 10, 20, and 30 min, while only the sunlight exposure process was conducted for 60 min. Aliquots (1 mL) were extracted using micropipettes at each respective time for inactivation analysis of E. coli and B. subtilis. For M. aeruginosa, all the processes (UV-C, solar, UV-C/H₂O₂, and solar/H₂O₂) were performed for 60 min and 40 mL of M. aeruginosa solution was entirely collected with respect to reaction time (0, 5, 10, 20, 30, and 60 min) for further analysis. To prevent additional oxidation effects after the reaction time, residual H₂O₂ in extracted samples were immediately quenched by the addition of sodium thiosulfate solution (2.8–7 g/L; Na₂S₂O₃·5H₂O; Daejung Chemicals, Siheung, South Korea) [25]. From the control experiment, no inhibition of E. coli and B. subtilis was observed under dark conditions. All experiments were conducted at room temperature (23 ± 2 °C) in duplicate. Statistical analysis was performed by the Mann–Whitney test in IBM SPSS (Ver. 26) and the statistical significance was set at p < 0.05.

2.4. Enumeration of Bacteria and Cyanobacteria

The number of bacteria (E. coli and B. subtilis) and cyanobacteria (M. aeruginosa) was enumerated by colony and cell count assays, respectively. For the enumeration of E. coli and B. subtilis, the withdrawn 1 mL samples at each reaction time were subjected to serial tenfold dilutions using sterile phosphate-buffered saline (PBS). Then, 0.1 mL of dilutions were spread on LB agar plates and incubated at 37 °C for 24 h for E. coli and at 30 °C for 48 h for B. subtilis, respectively. After incubations, the plates containing 30–300 colonies were selected to enumerate the concentration of bacteria as CFU/mL. The number of M. aeruginosa cells was directly counted using a Sedgewick-Rafter counting chamber with an optical microscope (Leica DM500, Leica, Wetzlar, Germany). Prior to counting cells, samples were sonicated with an ultrasonicator bath (POWERSONIC 510, Hwashin Tech Co., Gwangju, Korea) to separate them into single cells. The cell counts were performed on at least ten randomly selected fields, and the results were averaged to determine the cell concentration per mL. Inhibition rates (%) and log reduction value (LRV) were calculated as the following Equations (2) and (3), respectively:

$$\mathrm{Inhibition} \mathrm{rates} (\%)={(\mathrm{N}}_0-{\mathrm{N}}_{\mathrm{t}})\times 100/{\mathrm{N}}_0$$

(2)

$$\mathrm{LRV}=-{log}_{10}\left({\mathrm{N}}_{\mathrm{t}}/{\mathrm{N}}_0\right)$$

(3)

where N₀ represents the initial concentration of bacteria/cyanobacteria, N_t is the concentration of bacteria/cyanobacteria after reaction time of the processes, and t is the reaction time. The rate constants for the first-order reaction (k, /min) were estimated by the following Equation (4):

$$\ln \left({\mathrm{N}}_{\mathrm{t}}/{\mathrm{N}}_0\right)=-\mathrm{kt}$$

(4)

Chl-a concentration was additionally measured by following the standard water pollution process test method [26] using UV–vis spectrophotometer with the wavelengths of 630, 663, 645, and 750 nm. Briefly, samples were filtered through glass microfiber filters (GF/C, pore size = 1.2 μm; Cytiva, Marlborough, MA, USA) and Chl-a in viable cells on the filter was extracted by acetone in the refrigerator overnight at 4 °C. Then, supernatant was measured after centrifugation for 20 min at 2100 rpm. Concentrations of Chl-a was estimated by Equation (5).

$$\mathrm{Chl}-\mathrm{a} (\mathrm{mg}/{\mathrm{m}}^3)=[11.64\mathrm{X}1-2.16\mathrm{X}2+0.10\mathrm{X}3]\mathrm{V}1/\mathrm{V}2$$

(5)

where X₁ is OD₆₆₃ − OD₇₅₀, X₂ is OD₆₄₅ − OD₇₅₀, X₃ is OD₆₃₀ − OD₇₅₀, V₁ is the volume of supernatant (mL), and V₂ is volume of sample filtered (L).

3. Results and Discussion

3.1. Inhibition of E. coli and B. Subtilis under UV-C, Solar, and Solar/H₂O₂ Processes

Figure 1 illustrates the inhibition rates of E. coli and B. subtilis using UV-C, solar, and solar/H₂O₂ processes. The only UV-C process achieved complete inhibition within 1 min, meaning that not a single colony of E. coli and B. subtilis was observed. This indicates a reduction of E. coli and B. subtilis over 6 LRV considering the initial bacterial concentration (~10⁶ CFU/mL). UV-C showed statistically significant inhibition efficiency compared to the solar and solar/H₂O₂ (50 mg/L) processes (p-value = 0.02). Nyangaresi et al. [27] and Lui et al. [28] found that UV-C light can effectively reduce various microorganisms by chemically altering their DNA and RNA, leading to cell death. Meanwhile, the solar and solar/H₂O₂ (50 mg/L) processes reached over 99% of E. coli and B. subtilis inhibition after 20 min. Sunlight is less efficient in reducing microbial growth and causing cell death compared to the UV-C light, because sunlight mostly consists of visible light with a small portion of UV radiation [29]. Higher inhibition rates of E. coli and B. subtilis were observed using the solar/H₂O₂ (50 mg/L) process compared to only solar irradiation.

Figure 2 shows the LRV of E. coli and B. subtilis using solar and solar/H₂O₂ processes according to different H₂O₂ concentrations (20 and 50 mg/L). LRV for both E. coli and B. subtilis decreased in the following order: solar/H₂O₂ (50 mg/L) > solar/H₂O₂ (20 mg/L) > solar. The data for H₂O₂ (20 and 50 mg/L) processes were not shown, because H₂O₂ alone in the dark condition showed no significant impact on the inhibition of E. coli and B. subtilis. Specifically, after 30 min of reaction, the LRV for E. coli were 3.9 log for solar irradiation, 4.6 log for the solar/H₂O₂ (20 mg/L) process, and 5.1 log for the solar/H₂O₂ (50 mg/L) process (Figure 2a). Similarly, the LRVs for B. subtilis were 3.7 for solar, 4.0 for solar/H₂O₂ (20 mg/L) and 4.8 for solar/H₂O₂ (50 mg/L) at 30 min of reaction (Figure 2b). Although the results were not statistically significant (p-value > 0.05), the higher H₂O₂ concentrations could increase the inhibition of both E. coli and B. subtilis. The observed bacterial inhibition by solar irradiation alone is attributed to the increase of extracellular reactive oxygen species (ROS) such as •OH and O₂^⦁− generated by the UV light in sunlight, causing oxidative damage and reducing microbial populations [30]. When H₂O₂ is added, photolysis under UV light in sunlight generates additional •OH radicals. The generated •OH radicals could oxidize the cell walls, cell membranes, and even DNA and RNA of E. coli and B. subtilis by diffusing into the cell, resulting in cell lysis and disruption of cellular functions [20,30]. Therefore, the higher LRV of E. coli and B. subtilis observed in the solar/H₂O₂ (50 mg/L) disinfection process can be attributed to the higher concentration of •OH radicals. The effect of H₂O₂ addition was relatively insignificant to E. coli inhibition (Figure 2a), because E. coli might possess the H₂O₂ scavenging enzymes, which could catalyze H₂O₂ into water and oxygen, leading to a delay of intracellular oxidation [31].

Compared to LRV (Figure 3), the inhibition of B. subtilis was slightly less effective than that of E. coli for all solar and solar/H₂O₂ processes, indicating that B. subtilis could be more resistant than E. coli. Additionally, a photo-induced electron converts H₂O₂ to form •OH radicals, which could improve the inhibition of E. coli and B. subtilis. The total LRV was higher for E. coli than B. subtilis. Generally, Gram-negative bacteria such as E. coli are more sensitive to •OH radicals compared to Gram-positive bacteria including B. subtilis [32]. This could be explained by the thicker and more rigid peptidoglycan cell wall of Gram-positive bacteria, which makes it less susceptible to reactive oxidation species [33]. Furthermore, the relatively higher resistance of B. subtilis might be attributed to the mrgA gene, which encodes to form highly stable protein–DNA complexes and helps to protect the cells from oxidative damage caused by H₂O₂ [34].

3.2. Inhibition of M. aeruginosa Using UV-C, UV-C/H₂O₂, Solar, and Solar/H₂O₂ Processes

The inhibition rates of M. aeruginosa of UV-C, UV-C/H₂O₂, solar, and solar/H₂O₂ processes, measured by cell count assay, are shown in Figure 4a. Comparing the inhibition rate after 10 min reaction of each process, the highest value was achieved by UV-C/H₂O₂ (50 mg/L), followed by UV-C/H₂O₂ (20 mg/L) > UV-C >> solar/H₂O₂ (20 mg/L) > solar. This result indicated that the UV-C light source was more effective in reducing M. aeruginosa cells than solar light. Short-wavelength (<280 nm) and high-intensity of UV-C irradiation is highly effective in suppressing cyanobacterial growth by inhibiting gene expression [19]. In

Table 1. The rate constants for the first-order reaction of M. aeruginosa cell number and Chlorophyll a.

Process	M. aeruginosa		Chl-a
	k (1/min)	R2	k (1/min)	R2
UV-C/H2O2 (50 mg/L)	0.053	0.959	0.022	0.855
UV-C/H2O2 (20 mg/L)	0.046	0.989	0.027	0.908
UV-C	0.041	0.973	0.031	0.958
Solar/H2O2 (20 mg/L)	0.011	0.937	0.017	0.783
Solar	0.005	0.989	-	-

, the processes with H₂O₂ showed higher reaction rate constants (k) compared to the UV-C and solar irradiation alone processes. The reaction rate constant (k) for UV-C/H₂O₂ (50 mg/L) was 0.053 /min, which showed the highest reduction of M. aeruginosa. This value was 1.3 times higher than that of the UV-C process (k = 0.041 min). Solar and solar/H₂O₂ (20 mg/L) processes showed low k values of 0.005 and 0.011 min, respectively. This suggests that the type of light could have a more significant impact on the total M. aeruginosa reduction than the addition of H₂O₂. This is evidenced by the fact that the inhibition rates of M. aeruginosa among the three UV-based processes and between the two solar-based processes were respectively similar after 30 min of reaction. However, the addition of H₂O₂ could affect the initial stage of the reaction (0–20 min). The combination of H₂O₂ with UV-C irradiation could quickly generate ROS by decomposition of H₂O₂, mainly •OH, which could be further involved in inhibition of M. aeruginosa due to oxidative damage to cell membranes [19,35,36].

The Chl-a concentration was additionally analyzed as an indicator of M. aeruginosa cell lysis, and the results are presented in Figure 4b. The reaction rate constants of Chl-a are shown in Table 1. The reaction rate constants decreased in the order of UV-C > UV-C/H₂O₂ (20 mg/L) > UV-C/H₂O₂ (50 mg/L) >> solar/H₂O₂ (20 mg/L) > solar processes. Although the rigidity of M. aeruginosa cell membrane was partially damaged during the solar irradiation process, total Chl-a concentration was maintained. In the solar/H₂O₂ (20 mg/L) process, Chl-a removal reached 21.5% within the first 5 min of reaction, with no further significant removal of Chl-a for the next 60 min. It appears that UV light during solar irradiation caused partial cell damage to M. aeruginosa, leading to a reduction in cell counts; however, it was ineffective in removing Chl-a. However, significant removal of Chl-a was observed during UV-C irradiation (k = 0.031 min). UV-C exposure damages phycocyanin, proteins, and Chl-a of M. aeruginosa in sequence, ultimately causing significant damage to the photosystem, leading to cell lysis and death, which results in a color change. [35] Upon visual inspection of the samples before and after UV-C irradation, green-colored M. aeruginosa cells transformed to light yellow due to the damage of Chl-a from M. aeruginosa (Figure 5a–d). Observations by microscope showed that the intact cell membranes of M. aeruginosa were damaged after UV-C exposure, resulting in the leakage of cell contents and further removed by UV-C irradiation (Figure 5d). Among the UV-C based processes, the UV-C/H₂O₂ (50 mg/L) process showed the lowest removal rate of Chl-a, while achieving the highest inhibition rate of M. aeruginosa. After 60 min of UV-C process, 93.0% of Chl-a was removed, while the UV-C/H₂O₂ (50 mg/L) and UV-C/H₂O₂ (20 mg/L) processes showed slightly lower removal rates of 80.16% and 81.73%, respectively. This might be explained by that the higher concentration of hydrogen peroxide would rather absorb UV-C, resulting in a reduction of the average light intensity available to remove Chl-a [37]. Therefore, UV-C is effective in removing Chl-a, while increasing the concentration of H₂O₂ might reduce its availability to remove Chl-a.

4. Conclusions

This study successfully compared the effect of light sources (UV-C and solar) and their combination process with H₂O₂ on the treatment of bacteria (E. coli and B. subtilis) and cyanobacteria (M. aeruginosa). Although the UV-C-based processes demonstrated higher inhibition rates than the solar-based process, solar irradiation was found to be a promising alternative light source to treat microorganisms because it partly includes UV range light sources and is safe to use. In addition, increasing the concentration of H₂O₂ led to a more effective reduction of bacteria and cyanobacteria due to the generation of •OH, which could increase cell membrane damage and cell lysis. Although effective disinfection could be achieved using solar light with longer irradiation time than UV-C light, the reaction time would be shortened by combination with H₂O₂. On the other hand, an increase in H₂O₂ concentration could competitively absorb UV-C light, potentially interfering the cell damage caused by UV-C irradiation. This study highlights the possible application of solar light for disinfection processes in water treatment and the importance of selecting appropriate light sources and optimal H₂O₂ concentrations for the effective disinfection of bacteria and cyanobacteria.