Bacteriostasis and cleaning effect of trace ozone replacing personal care products

ABSTRACT Ozone is widely used to inactivate bacteria, fungi, and viruses. In recent years, the treatment of itchy skin diseases (eczema and atopic dermatitis) using trace ozone has also received attention. However, the feasibility of using trace ozone to replace personal care products (PCPs) has rarely been analyzed. In this study, the applicability of trace ozone was evaluated in terms of its efficiency for microbial inactivation in three types of skin microbiomes, cleaning performance on simulated human hair and epidermis, safety for simulated human hair, and contribution to emission reduction. The results revealed that at a 10:1 ratio of ozonated water to bacterial suspension, the inactivation ratios of Malassezia, C. albicans, and S. epidermidis reached 99.63%, 83.47%, and 100%, respectively. In addition, the cleaning performance of an ozone solution (0.4 mg/L) for simulated human skin contaminated with carbon black and sebum could reach 95.89% and 95.63%, respectively, with 5 min of washing. The average scores were 0.40 and 0.37 after 5 min and 10 min of ozone treatments, respectively, indicating that trace ozone does not significantly damage simulated human hair. Results also revealed that the total emissions of COD, TP, and TN would be reduced by 1.29×106, 3.55×103, and 3.63×103 mg/ (household · year), respectively, if PCPs are replaced by trace ozone. In short, our findings indicate that trace ozone is a potential alternative to PCPs. By replacing PCPs with trace ozone, the use of synthetic chemical products can be reduced and carbon emissions from oil extraction can be countered. GRAPHICAL ABSTRACT


Introduction
The human skin is inhabited by abundant and diverse microbial communities, including bacteria, fungi, and viruses. These microorganisms colonize different parts of the human skin according to their specific growth requirements. For example, lipophilic microorganisms such as Propionibacterium, Malassezia, and mites mainly colonize sebaceous areas [1][2][3]. Staphylococcus, Corynebacterium, and foot fungi prefer glabrous, humid skin such as the toe web [1,4]. Many common skin diseases are closely related to microbial populations. Previous studies have shown that Malassezia has a direct causal relationship with symptomatic skin diseases such as folliculitis, dandruff, and seborrheic dermatitis [5]. Candida albicans are usually asymptomatic in the skin and oral areas of the human body. However, alterations in the host microbiota or variations in the colonization environment can lead to the overgrowth of C. albicans, which causes thrush and diaper rash [6]. Staphylococcus epidermidis is a common symbiotic bacterium of the skin [7], and it has been reported to be the most common pathogen in indwelling medical devices, causing infections of prosthetic joints and vascular grafts [8]. Therefore, it is very important to find an effective method for controlling the excessive colonization of skin microbes and maintain the balance between skin microbes and the host.
Personal care products (PCPs) are widely used in human daily activities such as cleaning and washing. Existing studies have shown that PCPs are mainly composed of synthetic musky surfactants, disinfectants, and preservatives [9], which can remove dirt (sebum and solid particles), inhibit microbial growth, and effectively control or kill bacteria, fungi, viruses, and other microorganisms [10]. For example, excessive colonization of Staphylococcus aureus on skin with atopic dermatitis can be avoided by using an ultra-mild body wash with lipids [11]. Body wash containing triclosan and bamboo salt have an antibacterial effect on Corynebacterium glutamicum and Corynebacterium xerosis [12]. However, the human skin and ecological environment have been damaged by the long-term use of PCPs. Surfactants in PCPs directly interact with the human skin and remove the oily protective layer on the skin surface, resulting in the rupture of dry skin and causing irritant dermatitis [10]. Some studies have pointed out that PCPs disrupt endocrine functions [13] and may cause health problems such as obesity [14]. Many studies have also detected PCPs in water, sludge, and the atmosphere, reflecting a certain degree of environmental pollution [15,16]. For example, Peng et al. [17] detected nonylphenol, triclosan, and methylparaben in PCPs in the urban riverine water of Guangzhou, with the highest concentrations of 33231 ng/L, 1023 ng/L, and 1062 ng/L, respectively. Triclosan is most frequently detected in municipal solids, at concentrations of 0.5-15.6 µg/g [18]. With the improvement of people's living standards, the per capita usage of PCPs will continue to increase. Therefore, the potential threat of PCPs to human and environmental health should not be underestimated.
Ozone is produced by air or oxygen through high voltage discharge or intense ultraviolet radiation. As a strong oxidant with highly effective bactericidal functions, ozone is widely used for treating drinking water and effectively applied as a disinfectant in hospital environments [19,20], through which the deposition of pathogens on surfaces can be effectively reduced and the spread of pathogens via aerosols prevented [21]. However, the inhalation of ozone gas is extremely harmful to human respiratory health, and workers should perform disinfection activities in environments with ozone concentrations less than 0.16 ppm, and the exposure time should be less than 8 h [22]. Currently, trace ozone can be widely used as a bactericidal agent. For instance, the use of 0.5 mg/L ozonized oil instead of chemical drugs was effective in the treatment of skin diseases caused by Microsporum gypseum and Microsporum canis [23]. Sadatullah et al. [24] also illustrated that exposure to ozonated water at a concentration of 0.1 mg/ L for 30 s could effectively reduce microbial load on dental plaque.
Nevertheless, the replacement of PCPs by trace ozone has rarely been investigated thus far. Reasonable utilization of ozone concentration can not only inactivate pathogenic bacteria but also significantly contribute to environmental protection. Considering these possibilities, the following objectives were set in this study: (1) investigate the efficiency of trace ozone for microbial inactivation in three types of skin microbiomes; (2) analyze its cleaning performance in simulated human hair and epidermis; (3) evaluate its damage potential to simulated human hair; (4) estimate emission reductions attributable to the replacement of PCPs. The findings of this study can guide efforts towards reducing chemical pollution, and provide an experimental basis for the application of trace ozone in personal care activities.

Ozone inactivation experiments
The ozonated water was prepared by an ozone generator (M-6, Confosin Technology Co., Ltd., Dongguan, China). Sterile water was poured in and drained until the ozone concentration in the water became stable, as described in Figure 1. The volume ratios of ozonated water and bacterial suspension (O 3 /BS) were 1:1, 10:1, and 20:1, respectively. At an O 3 /BS ratio of 1:1, the initial concentration of bacterial suspension used in the inactivation experiment were 2.07×10 3 , 1.79×10 4 , 1.99×10 5 , and 1.39×10 6 CFU/mL, respectively. When the O 3 /BS ratio of 10:1, the initial concentration of bacterial suspension used in the inactivation experiment were 2.60×10 3 , 2.37×10 4 , 1.58×10 5 , and 2.00×10 6 CFU/ mL, respectively. As the O 3 /BS ratio of 20:1, the initial concentration of bacterial suspension used in the inactivation experiment were 2.21×10 3 , 1.94×10 4 , 1.38×10 5 , and 1.10×10 6 CFU/mL, respectively. After measuring the ozone concentration (0.4 mg/L) in the water, the corresponding volume of ozonated water was added to the bacterial suspension. Samples were collected at predetermined time points of 2 min, 5 min, 8 min, 10 min, 12 min and 15 min. The heterotrophic plate counting method (HPC) was used to determine the cultivable number of bacteria and fungi, and recorded in colony forming unit per milliliter (CFU/mL).

Kinetic data analysis
The results of disinfectant concentration and time were fitted according to the classic first-order reaction kinetic equation where C 0 and C represent the concentration of ozone at times 0 and t, respectively (mg/L), and k' represents the first order decay rate constant of ozone (min −1 ). In this study, under the condition of disinfectant attenuation, four disinfection kinetic models were established, as listed in Table 1. Origin (2021b) software was used to conduct nonlinear fitting analysis on the experimental data, and the optimal fitting values of each parameter in the model were obtained. C 0 represent the initial concentration of ozone (mg/L). k' represents the first order decay rate constant of ozone (min −1 ). Ln represents natural logarithm. C avg represents the arithmetic square root of the product of disinfectant concentrations at times 0 and t. N 0 represents the initial bacteria concentration (CFU/mL). N represents the bacteria concentration after contact time t (CFU/mL). k represents the inactivation rate constant. t represents the contact time (min). n and m represent kinetic parameters, where m is the Hom index. x represents microbial resistance to disinfectants. k 1 represents sensitivity of microorganisms to disinfectants. k 2 represents sensitivity of initial concentration of microorganism to disinfectant.

Measurement and calculation of washing performance
The simulated human epidermis was placed flat, and the reflectivity was measured with a whiteness metre (WSB-1, Hangzhou Qiwei Instrument Co., Ltd., China). Then, the simulated human epidermis was dyed evenly in prepared carbon black and sebum contamination solution, and the simulated human skin was removed with tweezers to dry naturally. The samples were washed in ozonated water for 5 min. The reflectivity of the simulated human epidermis soiled with sebum or carbon black before and after trace ozone treatment was measured with a whiteness metre [25]. The wash ratio was used to represent the washing performance of trace ozone. The wash ratio can be expressed as follows: where D r represents the wash ratio of trace ozone to the simulated human epidermis (%), R w represents the reflectivity of the simulated human epidermis soiled with sebum or carbon black after washing (%), R u represents the reflectivity of the simulated human epidermis soiled with sebum or carbon black before washing (%), and R 0 represents the reflectivity of the simulated human epidermis (%).

Morphology observation by scanning electron microscope (SEM)
After being treated with trace ozone, S. epidermidis were washed with PBS and fixed in 2.5% glutaraldehyde at 4°C overnight. Subsequently, the cells were washed with PBS twice for 10 min, further dehydrated with different ratios of ethanol solution (30%, 50%, 70%, 80%, 90%, 100%) for 10 min each, and the mixtures were lyophilized. The morphology of S. epidermidis was observed using a Hitachi S-4800 scanning electron microscope (Hitachi Instruments Inc., Japan). The simulated human hair was naturally dried after treatment with trace ozone for 5 min and 10 min, respectively. The sample was then pasted on the sample platform and coated with gold palladium. SEM was used to observe the surface morphology of simulated human hair. At least 3 images of non-overlapping regions were obtained for each of 28 hair samples. The control sample was observed as described above.

Evaluation of simulated hair damage
Four scores of damage to simulated hair were established, as shown in Figure 2. The first score corresponded to normal simulated human hair (0-0.25). Normal hair has cuticles resembling fish scales horizontally arranged on the surface of the hair shaft in an orderly manner ( Figure  1(a)). The second score corresponded to slightly damaged simulated human hair (0.25-0.5); in this case, the scales are deformed (Figure 1(b)). The third score corresponded to seriously damaged simulated human hair (0.5-0.75); in this case, the scales are lifted and the surface is rougher (Figure 1(c)). The fourth score corresponded to completely damaged simulated human hair (0.75-1); in this case, some scales are undiscernible (Figure 1(d)).

Measurement of physicochemical parameters in PCPs
1 g shampoo or body wash and 500 mL ultra-pure water were thoroughly stirred and used as the water sample to Ln N/N 0 = -k C 0 n /n k'×(1-e -k't ) k , n The C avg Hom model Ln N/N 0 = -k C avg n t m , C avg = C 0 ×√-e -k't k, n, m The Biphasic model Ln N/N 0 = ln[x×(e k be tested. Multi-parameter water quality tester (5B-3B-V8, Beijing Lianhua Yongxing Science and Technology Development Co., Ltd., China) was used to measure the concentrations of chemical oxygen demand (COD) and total phosphorus (TP). Total nitrogen (TN) was measured by an alkaline potassium persulfate digestion-UV spectrophotometric method [26].

Results and discussion
2.1. Impact of trace ozone treatment on the culturability of S. epidermidis As outlined in Figure 3(a-c), different initial concentrations of S. epidermidis were exposed to various ratios of O 3   increased and a tailing zone appeared during 2-15 min of ozone exposure. This observation could be explained by the aggregation of microorganisms increasing their tolerance to disinfectants [27,28]. Additionally, the inactivation efficiency of shampoos, body wash, and trace ozone on S. epidermidis are shown in Figure 3(d). At shampoo and body wash to bacterial suspension ratios of 1:1, the inactivation ratios were 100% and 95.48%, respectively. A similar inactivation efficiency was observed at an O 3 /BS ratio of 1:1. This result indicates that there is no obvious difference between trace ozone and PCPs (shampoo, body wash) in antibacterial efficiency regarding S. epidermidis.

Morphological changes in S. epidermidis
SEM images of S. epidermidis before and after ozone exposure are shown in Figure 4. Before inactivation, S. epidermidis had spherical or elliptic shapes and a smooth surface (Figure 4(a)). After ozone inactivation, the shape and surface characteristics of S. epidermidis were altered. At an O 3 /BS ratio of 1:1, the size of S. epidermidis was reduced, and the cell wall was destroyed (Figure 4(b)). As the proportion increased, the surface of S. epidermidis shrank visibly and the bacteria became rough (Figure 4(c,d)). The cell structure was destroyed more prominently at high ozone concentrations. Wen et al. [29] and Ding et al. [20] also found that the structure of the bacteria was destroyed completely and the appearance of compounds around the cells was altered at ozone concentrations of 2.0 mg/L and 3.0 mg/L, respectively.

Effect of ozone on inactivation of bacteria and fungi
Trace ozone was found to be effective for the inactivation of S. epidermidis, but the sensitivity of different microorganisms to ozone varied. Figure 5 shows the inactivation ratios of fungi (Malassezia, C. albicans) and bacteria (S. epidermidis) at an ozone concentration of 0.4 mg/L for 0-15 min of exposure, when the ratio of O 3 /BS was 1:1. The HPC of Malassezia, C. albicans, and S. epidermidis decreased from 6.83×10 2 CFU/mL to 3.0×10 1 CFU/mL, 1.85×10 2 CFU/mL to 0 CFU/mL, and 8.20×10 2 CFU/mL to 0 CFU/mL, respectively, within 5 min. The corresponding inactivation efficiencies were 95.77%, 100%, and 100%, respectively, and there was no significant difference (P = 0.131-1.000). The inactivation ratios of Malassezia (4.40×10 3 CFU/mL), C. albicans (1.95×10 3 CFU/mL), and S. epidermidis (2.60×10 3 CFU/mL) reached 99.63%, 83.47%, and 100%, respectively, indicating that Malassezia and C. albicans have higher resistance to ozone than S. epidermidis. Fungi have been reported to be very different from bacteria in cell size and structure, making them more resistant than bacteria [30]. Glycoproteins and chitin in fungal cell walls cross-link with each other to form complex networks, imparting them resistance against ozone inactivation, whereas bacterial cell walls mainly comprise peptidoglycan [31]. In addition, fungi have complex cell structure and large cell size, while bacterial cells are short and simple, and their genetic material is present in the cytoplasm, increasing their vulnerability to inactivation by ozone [30,32]. In this study, the inactivation rate of the three types of skin microorganisms could reach 100% after treatment with trace ozone for 10 min. Therefore, for the recommended time for bathing is 10 min or more to prevent excessive colonization of fungi and bacteria on human skin when using trace ozone to replace PCPs.

Models of ozone decay and disinfection dynamics
As outlined in Table 2, in the process of disinfection, ozone was rapidly consumed in the first 2 min, but the attenuation of ozone slowed down and remained almost unchanged after 2 min. Ozone depletion could not be avoided even if high concentrations were used [33], and the first-order kinetic model could be used to describe the attenuation of ozone. Table 3 shows the first-order decay rate constant of ozone in the process of ozone inactivation of S. epidermidis. R 2 in all fitting curves were above 0.9, indicating that the first-order kinetic model fits the experimental data well, and the first order disinfectant decay rate (k') could be used to fit the disinfection dynamics model. Table 4 shows the parameter estimates fitted by the Chick, Chick-Watson, Biphasic, and C avg Hom disinfection dynamics models. Chick's law uses the first-order reaction to express the inactivation ratio of microorganisms, and the Chick-Watson model is a first-order reaction kinetics model with changes in disinfectants based on the improvement of Chick model data. As shown in Table 4, R 2 of the Chick and Chick-Watson model was less than 0.658, indicating that the fitting results of all inactivation curves were poor. This may be attributable to the prominent tailing of the survival curve at low ozone concentration. The Chick and Chick-Watson model could not describe deviations from first order   dynamics models, and the trailing phenomenon of microbial inactivation could not be explained [34,35]. The inactivation kinetics data of S. epidermidis were also fitted to the biphasic and C avg Hom models and evaluated for goodness of fit. The biphasic model is a continuum of two different dynamics models, comprising a rapid initial dynamics phase and a tailing of the disinfection curve [36]. As presented in Table 4, R 2 of the biphasic model was greater than 0.9 only when the ratios of ozonated water to streams of cell suspension was 10:1. Results showed that the biphasic model was also not a good fit for the experimental data analyzed.
The Hom model can describe 'shoulders' or 'tailing off' in inactivation, and the C avg Hom model is an improvement of the Hom model, considering the firstorder decay of disinfectants [37]. As shown in Table 4, R 2 in the C avg Hom model was above 0.85 for all inactivation curves. Analysis of kinetics data revealed that the C avg Hom model was the best fit among the four applied models. In the C avg Hom model, n and m are important parameter indexes of the survival curve, where n is the coefficient of dilution and m is the empirical constant [38]. The value of n < 1 in the fitting curve of the ratio of 20:1 indicates that contact time has a stronger effect on the inactivation of S. epidermidis than ozone concentration. The value of n > 1 in other inactivation curves implies that ozone concentration has a stronger effect than contact time. The bactericidal effect is dependent on contact time at sufficiently high ozone concentrations, but it is dependent on ozone concentration at low concentrations. In addition, the value of m < 1 in all the inactivation curves indicates a tailing effect [37]. Previous studies have reported that the tailing of the survival curve is largely ascribed to a   variety of factors, including the consumption of disinfectants over time, shielding of microorganisms by particulate matter, and aggregation of microorganisms [39,40].

Analysis of washing efficiency
The wash ratio is an important indicator of cleaning performance in washing processes, and it is commonly estimated by measuring the reflectance of test samples before and after washing [41]. As presented in Figure 6, in a treatment with 4.14 g of PCPs for 5 min, the wash ratios of the simulated human epidermis soiled with sebum and carbon black were 92.77% and 96.87%, respectively. Replacing the PCPs with trace ozone, the wash ratios of simulated human epidermis soiled with sebum and carbon black were 95.89% and 95.63%, respectively. The results indicated that there was no significant difference (P = 0.765) in the cleaning performance of the simulated human epidermis between trace ozone and PCPs. Photomicrographs of the simulated human hair were obtained through scanning electron microscope. As shown in Figure 7(a,d), the surface of the simulated human hair was thickly coated with sebum and carbon black. It is almost impossible to observe the original morphology of simulated human hair. After treatment with an ozone concentration of 0.4 mg/L for 5 min, sebum and carbon black were basically removed. Hair cuticles were clean and neatly covered the hair shaft without evident curling or peeling (Figure 7(b,e)). Similar results were observed in SEM images of the simulated human hair soiled with sebum and carbon black treated by PCPs for 5 min (Figure 7(c,f)). This suggests that trace ozone and PCPs have equivalent cleaning efficiency for the simulated human hair.
2.6. Safety analysis of trace ozone on simulated human hair Table 5 shows SEM images of the simulated human hair without treatment and after ozone treatment (0.4 mg/L) for 5 min and 10 min. The cuticle crimp degree and cuticle edge integrity of the simulated human hair were scored. Four different grades of simulated human hair damage scores were established to evaluate the degree of damage caused by trace ozone. In the original hair, the hair cuticle overlaps the hair shaft, forming a layered structure around the central cortex that protects the integrity of the hair [42]. However, adverse environmental factors (solar radiation, wind, pollution, perm, dyeing) can damage the morphological structure of hair (warps, cracks, holes) and change its surface properties [43]. As presented in Table S1, before ozone treatment and after ozone treatment (0.4 mg/L) for 5 min and 10 min, the average scores of simulated human hair injury were 0.36, 0.40, and 0.37, respectively, corresponding to the second grade (0.25-0.5). Accordingly, no significant difference (P = 0.820-0.932) was observed between the control group and the treatment group. In this study, some broken edges, warps, and holes were observed on all simulated human hair, likely because the hair fibres had been exposed to adverse environmental conditions prior to ozone treatment. Consequently, the hair morphology and structure were already damaged. The results showed that trace ozone had minimal impact on the fibre quality of simulated human hair. Simulated human hair could be efficiently cleaned using trace ozone for 5 min with no apparent damage. The literature indicates that although prolonged exposure to toxic levels of ozone dose can alter or oxidize the lipid and protein composition of the skin, brief exposure at low and controllable ozone concentrations may be non-toxic [44,45]. This is because ozone immediately reacts with polyunsaturated fatty acids and water in the cuticle of the skin after brief exposure, producing reactive oxygen species (ROS) and lipopeptides (LOP) that are degraded by antioxidants in the skin [45]. Ozone is also an ideal drug at the right concentration, for example, short-term use of ozonated water or ozonated oil can effectively treat skin wound infections [46]. Previous literature has reported the development of an ozone generator for patients with atopic dermatitis that produces the optimal dose of ozonated water during treatment by controlling and measuring accurate ozone concentrations in real time to bathe or soak skin lesions, providing a better option for those who cannot tolerate the adverse effects of medication [47,48]. Therefore, trace ozone may become a new bath tool with great development prospect.

Analysis of trace ozone emission reduction
In this study, the feasibility of replacing PCPs with trace ozone in terms of cleaning efficiency and safety were evaluated. The application of trace ozone to replace PCPs in personal care activities could significantly contribute to the reduction of the emission of organic matter, TN, and TP. In the study, the emission reduction potential of trace ozone was also calculated. Table S2 shows the results of physicochemical parameters (TN, TP, COD) associated with PCPs. The contents of COD, TP, and TN in PCPs were 1179.5, 0.333, and 0.331 mg/ g, respectively. If each person in a family of three uses 10 g PCPs per day, the total emissions of COD, TP, and TN would be 1.29×10 6 , 3.55×10 3 , and 3.63×10 3 mg/ (household · year), respectively (Table S3). Additionally, the widespread use of PCPs also leads to the discharge of more surfactants into the wastewater system [49], many of which are extracted from petrochemicals mined from fossil carbon in the Earth's crust. Carbon in organic matter in sewage can be converted into CO 2 and CH 4 , and released into the environment from sewage treatment plants [50], which can impact the existing carbon cycle in the atmosphere. Therefore, trace ozone can be used as an alternative to PCPs used in daily bathing in order to reduce carbon emissions.

Conclusion
This study demonstrated that trace ozone is an effective disinfectant against S. epidermidis, C. albicans, and Malassezia. The inactivation process of S. epidermidis exhibited an initial rapid inactivation stage followed by a slow tail effect stage. Four disinfection dynamics models (the Chick, Chick-Watson, Biphasic, and C avg Hom models) were used to fit the inactivation curve of S. epidermidis, and analysis of kinetics data showed that the inactivation curve best fitted the C avg Hom model. Trace ozone and PCPs exhibited equivalent cleaning efficiencies for simulated human hair and epidermis. No significant damage to the simulated human hair was observed after ozone exposure. Analysis of the emission reduction potential of ozone treatment revealed that the replacement of PCPs by trace ozone could help in reducing the emission of organic matter, TN, and TP, respectively. Therefore, the replacement of PCPs will reduce chemical pollution and help promote low-carbon lifestyles. These findings suggest that trace ozone is a promising alternative to PCPs, and provide further knowledge of the applicability of ozone to personal care activities.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.