Comparative study of Gram-negative bacteria response to solar photocatalytic inactivation


Solar photocatalytic inactivation of Gram-negative bacteria with immobilized TiO2-P25 in a fixed-bed reactor was modeled with simplified kinetic equations. The kinetic parameters are the following: the photocatalytic inactivation coefficient (kd,QUV), the initial bacterial reduction rate (A) in the contact with the disinfecting agent, and the threshold level of damage (n) were determined to report the effect of QUV/TiO2-P25 on bacterial cultivability and viability and to compare the response of bacterial strains to photocatalytic treatment. In addition, the integration of the reactivation coefficient (Cr) in the photocatalytic inactivation equation allowed evaluating the ability of bacterial reactivation after photocatalytic stress. Results showed different responses of the bacteria strains to photocatalytic stress and the ability of certain bacterial strains such as Escherichia coli ATCC25922 and Pseudomonas aeruginosa ATCC4114 to resuscitate after photocatalytic treatment.


In arid countries, re-using domestic wastewater for agricultural and industrial activities has been proposed as a strategy for suitable water use and resource conservation. However, the use of these effluents must be carefully considered, as they could contain high levels of pathogenic agents and harmful chemical compounds (Rengifo et al. 2010). Over the last 20 years, environmental remediation has emerged as a high national and international priority for the disinfection of contaminated water from hazardous pollutants.

Chlorine-based technologies have long been used as disinfection processes for drinking water supplies and also for the tertiary treatment of wastewater. However, these technologies are becoming of increasing concern due to the founding of recent studies indicating the formation of potentially harmful chloro-organic disinfection by-products (DBPs) during the treatment process (Nassar et al. 2017). For this reason, new disinfection technologies are currently under development.

Methods such as ozone and ultraviolet light are equally effective and less toxic. However, these techniques are expensive and often technically difficult to apply in less-favorable conditions. In this context, a semiconductor photocatalysis has emerged as a very attractive, environmentally friendly technology for water disinfection, especially considering the possibility of using solar light to drive the photocatalytic process.

Many authors consider TiO2 as the most effective material for catalytic photodegradation due to its low cost, high efficiency, and long-term photostability (Mills and Lee, 2002).

A titanium dioxide (TiO2) nanoparticle has attracted great interest in a wide range of applications such as photocatalysis, nanomedicine (Sapizah et al. 2012), sterilization (Mills et al. 2013), anti-fogging materials (Mills et al. 2013), lithography (MacFarlane et al. 2011), degradation of organic compounds (Hashimoto et al. 2005; Tatsuma et al. 2002), and prevention of metal corrosion (Zsilák et al. 2014; Vamathevan et al. 2002). In the last decade, TiO2 has been widely used as a self-cleaning and self-sterilizing material for coating many clinical tools including sanitary ware, food table and cooking ware, and items for use in hospitals (Rtimi et al. 2015; Rtimi 2017).

Once the TiO2 photocatalyst is activated by light with an energy equal or greater than the bandgap of the semiconductor, electron (e)/hole (h+) pairs are generated in the conduction and in the valence bands, respectively. Then, some of these pairs migrate to the photocatalyst surface. In the primary stage, e reacts with O2 adsorbed at the photocatalyst surface to generate superoxide O2·− radicals while h+ react with water to produce hydroxyl ·OH radicals. These hydroxyl radicals are able not only to mineralize organic pollutants but also to affect several microorganisms, including viruses, bacteria, spores, and protozoa (Wang et al. 2015).

The contact between bacteria and nanoparticles such as TiO2 is considered according to many researches as the key point of the bacterial inactivation (Gogniat et al. 2006). Therefore, TiO2 can induce a bactericidal effect in the dark when adsorbed on the surface of the bacterial cell (Huang et al. 2000). Other reactive oxygen species (ROS), such as superoxide ions (O2·− or HO2·), are less effective against bacteria, due to the negative charge which prevents them from penetrating bacteria cell membranes (Rengifo et al. 2010; Caballero et al. 2009).

The first research on the bactericidal effect of TiO2 photocatalytic reactions was conducted on Escherichia coli (Matsunaga et al. 1985) and subsequently has been intensively conducted on a wide spectrum of organisms including viruses, bacteria, and fungi (Silva et al. 2017).

Commercial TiO2 Degussa P-25 has been used in most previous studies (Chong et al. 2009). Most of the reports about photocatalytic treatment of water sources have used TiO2 slurries (Pablos et al. 2011). However, the difficulty of separation and reuse of nanostructures TiO2 (10–30 nm) from treated water often limits its real application (Caballero et al. 2009). Numerous material engineering solutions have been investigated in order to resolve this problem. One of the more prominent solutions is the immobilization of the TiO2 nanoparticles onto an inert carrier to retain its quantum effects in the form of a thin film, while allowing ease of separation from the treated water (Chonga et al. 2011). The thin-film fixed-bed reactor (TFFBR) is one of the most used solar photoreactor which has received an increasing interest as a suitable commercial application (Spasiano et al. 2015).

Therefore, these reactors can employ both direct and diffuse portions of solar radiation as a light source without the need to separate the photocatalysts from the purified water (Caballero et al. 2009). Indeed, the real application of this technology should avoid the separation step to allow a continuous water treatment. Indeed, employing the photocatalyst as a suspension or slurry makes the scaling up of the process difficult, as the TiO2 has to be removed from the decontaminated water to be reused several times. This problem presented a significant barrier to commercial application. The objective of this research was to develop a modeling approach to describe the kinetics of photocatalytic inactivation for solar photocatalytic reactor utilizing immobilized TiO2-P25. The overall goal was to build simple equations which could report the influence of the synergic effects of solar light radiation and TiO2-P25 on bacterial inactivation. The model will serve as a predictive tool to design disinfection systems, so that water can be disinfected quickly, efficiently, inexpensively, and without risk of recontamination with pathogenic bacteria.

Materials and methods


Degussa P-25 TiO2 commercial powder (Degussa, Germany) was used as a photocatalyst for all experiments. Its crystalline structure consists of anatase (80%) and rutile (20%) and has a specific surface area of 50 ± 15 m2/g and an average crystallite size of 30 nm.

Preparation of bacterial strains

Each bacterial strain (Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 4114, and Salmonella Typhi ATCC 560) was grown overnight in nutrient broth NB medium under aerobic conditions at 37 °C with constant shaking at 160 rpm. The bacterial growth was monitored by measurement of optical density at 600 nm. Then, for photocatalytic experiments, bacteria were cultured overnight (≈ 18 h), collected by centrifugation at 4000 rpm for 4 min and washed twice with sterile demineralized water. Cell pellets were finally re-suspended in sterile saline buffer (NaCl 0.85%) and the cell density was adjusted spectrometrically (OD600) to a final cell concentration of about 106 colony-forming units per milliliter (cfu/mL).

Deposition of TiO2-P25 on the glass plate

Commercial TiO2 (Degussa P25) powder was suspended in deionized water with a concentration of 10 g/L; a few drops of nitric acid (HNO3) are added up to pH 3 to positively charge the TiO2 nanoparticles and better disperse the suspension. Then the ordinary glass plate (surface area of 0.8 m2) was treated with a nitric acid bath (1 M) with a holding time of about half an hour and then rinsed with 20% solution of sodium hydroxide (NaOH). Alkaline treatment increases by one hand the density of surface OH groups, which react with the hydroxyls on the TiO2 P25 surface, and secondly the roughness of the support to facilitate the attachment of the photocatalyst particles (Ismail 2011). The suspension was sonicated for 10 min to avoid agglomerations. The mixture thus prepared is deposited on the glass plate by “spray” and heated in an oven at 100 °C. The process was repeated 10 times during a day until a homogeneous layer was obtained. Then the immobilized TiO2 layer was calcined at 475 °C for 7 h (Ismail 2011); subsequently, it was rinsed three times with deionized water to remove unfixed particles before the photocatalytic experiment.

Photocatalytic inactivation of bacterial strains

After immobilization of TiO2-P25 on the glass plate, the photocatalytic reactor was inclined at 30 °C (Fig. 1) (Bousselmi et al. 2002). Then 1 L of bacterial suspension (106 cfu/mL) was recycled continuously from the tank by a pump (flow rate 12 L/h) to the surface of the glass plate where it is irradiated by solar light and returned to the tank and recycled again (Fig. 1). During experiments, the average of global accumulated incident UV radiation energy QUV was equal to 134 kJ/L for 60 min of solar UV irradiation. A control test was conducted with TiO2-P25 in dark condition. The photocatalytic experiments with or without UV light were carried out under natural conditions with pH = 7 and a temperature of the day around 25 °C.

Fig. 1

Solar photocatalytic reactor with immobilized TiO2

Radiation evaluation

The incident solar irradiance in the photoreactor was measured with a pyranometer (Dr. Hönle UV-Meter Hand-held Technology (UV-A (330–400 nm)). Since those experiments were done in different days where solar intensity change from day to other, bacterial inactivation was represented as a function of the accumulated UV (QUV) in the photoreactor per unit of treated water volume for given periods of time during the experiment: QUV (kJ/L).

Incident UV energy (QUV) was determined by the following expression (Malato et al. 2000).

$$ {Q}_{\mathrm{UV},n}={Q}_{\mathrm{UV},n-1}+\kern0.37em \Delta {t}_n\ {\mathrm{UV}}_n\ {\mathrm{S}}_{\mathrm{r}}/{\mathrm{V}}_{\mathrm{t}};\Delta {t}_n={t}_n-{t}_{n-1} $$

where QUV is the accumulated energy received by the reactor per liter of solution for each sample taken during the experiment, tn is the experimental time for each sample, UVn is the average solar ultraviolet radiation measured during Δtn, Sr is the surface area (0.8 m2), and Vt is the total volume of the reactor (1 L). During experiments, the average of global accumulated incident UV radiation energy QUV was equal to 134 kJ/L for 60 min of solar irradiation.

Determination of viable and cultivable bacterial cells

Samples of bacterial suspension (1 mL) were taken at regular times of the photocatalytic experiments. Serial dilutions were performed in sterile saline buffer (NaCl 0.85%) and then spread on Petri dishes containing Luria-Bertani medium (LB agar Sigma-Aldrich) to count the viable cultivable bacteria. After incubation for 24 h at 37 °C, the bacterial cells were numbered. Three Petri dishes were used at each sampling time.

Bacterial reactivation in dark conditions

To investigate the ability of bacteria to resuscitate or reactivate after rest time in darkness, photo-irradiated samples were transferred into separate sterile Petri dishes covered with foil to examine the bacterial potential to dark repair at room temperature. This part of experiment was carried out to verify the ability of post-photo-irradiated bacteria in presence of fixed TiO2 to restore their solar light/TiO2-P25-induced lesions by molecular mechanisms and especially with dark repair systems such as excision-repair mechanisms and SOS system.

Results and discussion

Kinetics of photo-inactivation reactions

A series of experiments were carried out to highlight the effects of solar photocatalytic treatment on bacterial cultivability and viability. The number of surviving bacteria was expressed as the ratio (survival ratio) of the number of viable and cultivable bacteria remaining after exposure to experimental conditions (N) to the number of the initial viable bacteria (N0) (log N/N0). Figure 2 shows the photocatalytic Gram-negative bacilli: E. coli ATCC 25922, P. aeruginosa ATCC 4114, and S. typhi ATCC 560 inactivation with immobilized TiO2-P25.

Fig. 2

Monitoring of bacterial cultivability in contact with immobilized TiO2-P25 in the dark and with solar UV irradiation/TiO2-P25 conditions. aE. coli ATCC25922, bP. aeruginosa ATCC4114, cS. typhi ATCC560, and d comparison of the photocatalytic effect on the three strains

This figure shows that the bactericidal rates were enhanced when we have combined the effect of TiO2-P25 with solar UV and were increased by increasing irradiation time. The synergic effect of TiO2 and solar UV irradiation had a remarkable effect on bacterial cultivability. However, we cannot neglect the bactericidal effect of TiO2 without solar UV irradiation. In addition, the result also showed that fixed TiO2-P25 induces low bacterial inactivation in the dark (0.7 to 1 log unit)(Lifen et al. 2009) were reported that pure TiO2 had a high germicidal activity. Indeed, from microscopy observations, it was found that yeast cells on fresh TiO2 were deformed under specific experimental conditions of pH, TiO2 concentration, and suspended TiO2.

The bactericidal effect of this semiconductor was highlighted only in the presence of solar UV irradiation after of a contact time of 60 min (QUV = 134 kJ/L) with solar UV/ TiO2-P25.The bacterial strains E. coli, P. aeruginosa, and S. typhi showed a cultivability inactivation superior to 5 log10.

The antimicrobial photo-biological activity of TiO2 on bacterial cells using solar UV irradiation has been attributed to the generation of very active free radical species called ROS (reactive oxygen species) (Acevedo et al. 2011). In fact, hydroxyl radicals (HO·) and superoxide anions (O2·−) are considered as the main generated species in the anodic and cathodic pathways, respectively, of photocatalytic processes in the presence of oxygen (Fujishima et al. 2000). Both species are known to be highly reactive with biological samples. Other oxygen reactive species have been also proposed, including hydrogen peroxide (H2O2), hydroperoxyl radical (HO2·), and singlet oxygen (1O2) (Fujishima et al. 2000). Chromosomal aberration by DNA lesion caused by photoexcited TiO2 was also reported.

Moreover, during solar UV radiation, the combination of indirect effect of UVA (320–450 nm) radiation with direct germicidal action of UVB (290–320 nm) can be lethal to cells respectively, through the photosensitization of endogenous chromophores such as co-enzymes or cytochromes, which could damage essential enzymes to bacteria growth and the blockage of DNA replication and RNA transcription (Pigeot et al. 2012). However, low solar irradiation can easily affect solar disinfection causing bacterial regrowth by photorepair mechanisms (Schuch et al. 2009).

In particular, some studies have demonstrated that the decrease of cultivability results from alteration in the bacterial metabolism due to the deleterious action of solar UV light in combination with TiO2. Indeed, the cell membrane is the primary site of reactive photogenerated oxygen species attack, leading to lipid peroxidation (Kiwi and Nadtochenko, 2005). The combination of cell membrane damage and further oxidative attack of intracellular components ultimately lead to cell death (Rincon and Pulgarin, 2004). Other studies have suggested that the mode of action is the photooxidation of coenzyme A, leading to the inhibition of cell respiration and thus to cell death (Bonetta et al. 2013).

Knowing that the surface of TiO2 is amphoteric and consequently, the charge of surface is pH-dependent (Piscopo et al., 2001). For the TiO2-P25, the pH zpc is close to 6.3. Degussa P-25 has a negative charge at pH = 7. Thus, bacterial inactivation by the photocatalytic process may be due to free radicals in solution, not to radicals on the surface of the photocatalyst. This can be rationalized based on the electrostatic interaction between the TiO2 surface and the cell surface. In fact, at pH 7, the surface of TiO2 (isoelectric point, 6.3) and the outer membrane of bacteria are negatively charged (Cho et al. 2005) which lead to an electrostatic repulsion between the TiO2 surface and bacteria. Consequently, adsorption of bacteria onto the surfaces of the TiO2 particles is not favored at this pH, and the direct contact between the cells and the illuminated TiO2 surface should be minimal. Cho has reported that the photochemical removal of coliform bacteria was unaffected by the pH of the solution in the range of 6–8 pH units (Cho et al. 2005). Rahmani et al. have reported that the electrostatic interaction at the TiO2/water interface could be important in many cases of photocatalytic degradation of charged substrates (Rahmani et al. 2009); it is a relatively weak force, which can be dominated by other factors (Robertson et al. 2005). Thus, the initial pH of the water did not play an important role within a range of 5.5–8.5 pH units (Dheaya et al. 2009).

The performance of a disinfection system is directly relied on the knowledge of the inactivation rate of a target or indicator organism by the disinfectant. For photocatalysis, the synergistic effect of catalyst concentration and solar light intensity on the rate of the process determines the most efficient combination of contact time and the dose to be employed. Currently, most of these information were obtained from bench-scale studies and extrapolated with a series of empirical models which do not adequately describe photocatalytic disinfection.

The most common application is the Chick-Watson (C-W) model used primarily to fit inactivation data with first-order decay or modified for data with an initial lag, and whose general expression is given by this equation (Watson 1908):

$$ N/{N}_0=\mathrm{Exp}\ \left(-k\ {C}^n\ t\right) $$

where N/N0 is the reduction in the bacterial concentration, k is the disinfection kinetic constant, C is the disinfectant concentration with C = d × QUV, d is the catalyst concentration (g/L) (for our experimental condition (fixed-bed reactor), the disinfectant concentration (TiO2-P25) during photocatalysis was supposed to be constant), and QUV is the accumulated UV energy (kJ/L), t is the time (min), and n is the threshold level of series-event model; n is equal to 1 for the first-order C-W model.

The C-W model was modified to consider the initial reduction (A) in the contact with the disinfecting agent:

$$ N/{N}_0=A\ \mathrm{Exp}\ \left(-k\ {C}^nt\right) $$

where A is the initial bacterial reduction rate in the contact with the synergic action between cumulate solar UV radiation (QUV) and the catalyst TiO2-P25.

$$ A=\left({N}_{\mathrm{C}}+{N}_{\mathrm{d}}\right)/\mu $$

where NC is the number of CFUs after the first contact with solar UV radiation and catalyst TiO2-P25, Nd is the initial number of CFUs in the dark condition in the presence of semiconductor, and μt is the adsorption coefficient in the fixed bed of TiO2 that is assumed to be negligent due to the negative charge of TiO2-P25 at pH = 7.

In this C-W model, the reduction in enumerated bacteria was proportional to the contact time. This correlation, however, cannot be applied for every instance, due to various factors such as the reactor configuration, solar intensity change from day to other, bacterial response (dose/response), inactivation mode, and the bacterial resistance to disinfectant used which might cause severe non-linearity characteristics. Thereby, the incorporation of an additional empirical parameter was made to accurately account the photocatalytic inactivation mechanism.

$$ N/{N}_0=A\ \mathrm{Exp}\ \left(-k\ {C}^n\ {t}^m\right) $$

m is the empirical parameter that controls the deflection of the inactivation rate. If m is greater than 1, there is an increase in the inactivation rate and vice versa. This variation depends essentially of water quality and the photoreactor configuration.

$$ k={k}_{\Big(\mathrm{d}},{}_{\mathrm{Quv}\Big)}\ \left({k}_{\mathrm{d}}+{k}_{Q\mathrm{UV}}\right) $$

where k is the disinfection kinetic constant, k(d,QUV) is the global photocatalytic inactivation coefficient, kd and kQUV are the disinfection kinetic constant related separately to the catalytic action and solar UV effects, d is the catalytic concentration (g/L), and QUV is the accumulated UV energy (kJ/L).

In the studied strains, the time required for bacterial inactivation by photocatalysis depends of the type of microorganism. Indeed, the time required for the inactivation of nearly 99.99% of cultivable bacteria bacterial was estimated at 30 min of irradiation for the indicator of biological contamination bacteria, E. coli, and about 40 min for pathogenic bacteria, S. typhi, against 45 min of exposure time for nosocomial bacteria, P. aeruginosa.

Based on the kinetics parameters determined according to the C-W model (Fig. 1; Table 1), the photocatalytic inactivation coefficient (k) that represents the slope of inactivation curve was more important for E. coli ATCC 25922 (k = 0.19) strain than for S. typhi ATCC 560 (k = 0.17) and P. aeruginosa ATCC 4114 (k = 0.15). According to Pagan et al. (1999), the greater the coefficient k, the lower the bacterial resistance to the applied disinfectant. Thus, we can deduce the photocatalytic sensibility of E. coli followed by S. typhi and we can estimate from Fig. 2d that P. aeruginosa is the most tolerant to a solar UV with fixed TiO2-P25.

Table 1 Kinetic parameter for tested bacterial strains according to the first-order Chick-Watson model (C-W) with modification

The results represented in Table 1 show that the reduction of bacterial cultivability rate (A) in the initial contact with the disinfecting agent was more significant for P. aeruginosa. Thus, this strain was estimated to be the most sensitive bacteria in the retention of cultivability in the initial contact with the disinfectant. However, E. coli was qualified to be more resistant in the retention of cultivability, followed by S. typhi.

Thus, based on photocatalytic inactivation, P. aeruginosa was the most resistant or tolerant strain to photocatalytic activities. This difference in bacterial behavior or response was probably due to an inter-specific difference concerning the cell growth strategy, the response to the environmental stress, the metabolism activity, and the genetic regulation and flexibility.

To describe bacterial inactivation behavior, series-event (SE) kinetic model with modification was used in addition to C-W model with modification. The inactivation process was modeled as a progression of discrete damage levels. The microorganism was assumed to be inactivated at a threshold level of damage (Fig. 3, Table 2).

Fig. 3

Response of tested bacteria to solar photocatalytic treatment with immobilized TiO2-P25, and monitoring of bacterial cultivability according to Chick-Watson (C-W) and series-event (SE) models. aE. coli ATCC 25922.bP. aeruginosa ATCC 4114.cS. typhi ATCC 560

Table 2 Kinetic parameter for tested bacterial strain according to series-events model (SE) with modification

Each step is characterized by first-order kinetics. Each damage level Di has a kinetic constant ki and n is the threshold level of damage.

$$ {D}_0\overset{k1}{\to }{D}_1\overset{k2}{\to }{D}_2\overset{ki}{\to }{D}_3.\dots {D}_{n-1}\overset{kn}{\to }{D}_n $$

By assuming that the kinetic constant is the same at each level, the following generalized expression can be derived for the series-event model (Severin et al. 1984).

$$ \frac{N}{N0}=A\ \exp\ \left(-K\times {C}^n{t}^m\right)+\ln\ \frac{\left(1+{\sum}_{i=0}^{n-1}\times {C}^n{t}^m\right)i}{i!} $$

Table 2 illustrates the ability of the series-event model to report bacterial response behavior. The results indicated the absence of change in the kinetic constants (k and A) but the use of this model can report the series-events undergone by tested bacteria.

For example, to respond to solar UV with fixed TiO2-P25, E. coli was chosen to pass through different damage levels (D1 to D4), The accumulation of damages has been realized by keeping the cultivability in initial contact with disinfectant (solar UV/TiO2-P25). However, P. aeruginosa was adapted to another strategy. Indeed, this bacteria strain was decreased in the bacterial cultivability at the initial contact with the disinfectant (solar UV/TiO2-P25) as a response to stress. Thereby, the damage level determined for this bacteria was n = 1 (D1).

We can conclude that keeping or losing bacterial cultivability can be related to a bacterial adaptation to overcome the stress effects. Thus, the loss of cultivability is not constantly the synonym of sensibility and vulnerability and, in the same means, the retention of cultivability cannot reveal the resistance or tolerance of bacteria. The bacterial growth modality is directly related to the response of each species to the stress.

Post-irradiation events after the photocatalytic treatment

To study the efficiency of the photocatalytic process UV/TiO2 (Degussa P-25) fixed on a glass plate, we studied the post-irradiation events after the photocatalytic treatment.

To semi-quantify the reactivation level of post-irradiated bacteria in dark condition, a log ratio was determined according to a modified version of the Lindauer and Darby (1994) equation with modification:

$$ {C}_r={LogN}_r/{N}_{QUV/ TiO2} $$

where Cr is the coefficient of reactivation, Nr = number of viable and cultivable bacteria after a rest time in the darkness, UV irradiation and NUV/TiO2 is the number of viable and cultivable bacteria after photo-disinfection process with immobilized TiO2.

According to Lindauer and Darby (1994), the reported log values have ranged from 1 to 3.4. When the Cr was inferior to 1, there was no reactivation; if Cr ranged between 1 and 3.4, we can conclude that the reactivation occurred in the darkness or/and in visible light; when Cr was superior to 3.4, there were no disinfectant effects and the cells grew naturally without any environmental stress. In effluents from the studied photoreactor and according to the result of E. coli and P. aeruginosa, a slight recovery was observed after photocatalytic treatment during the subsequent 24 h in dark conditions. Indeed, Cr was equal to 1.07 for E. coli and 1.22 for P. aeruginosa (Fig. 4), where Cr is the coefficient of reactivation, Cr = Log Nr/NUV/TiO2; Nr = number of viable and cultivable bacteria after a rest time in the darkness; UV irradiation and NUV/TiO2 are the number of viable and cultivable bacteria after photo-disinfection process with immobilized TiO2.

Fig. 4

Coefficient of bacterial reactivation (Cr) determined after a rest time in dark condition

After solar UV/TiO2 treatment, the density of cultivable bacteria (E. coli and P. aeruginosa) was increased. This enhancement can be explained by the fact that biological systems have evolved some molecular mechanisms to appropriately respond to the environmental stresses which have damaged proteins and DNA (Marugán et al. 2008). In this case, to respond to UV damage, bacteria generally possess molecular mechanisms to restore DNA lesions. Indeed, bacteria can evolve four main mechanisms in the repair of UV radiation-damaged DNA, including photoreactivation, nucleotide excision repair (NER), mutagenic DNA repair (MDR), and recombinational DNA repair (Errol et al. 2006). These results can affect the effectiveness and the durability of solar photocatalytic treatment with fixed titanium dioxide device.

Bacterial resuscitation after the photocatalytic treatment could be explained by the fact that, under oxidative stress, their cells can enter into a viable but non-cultivable (VBNC) state.

The result of no cultivability after TiO2 photocatalytic treatment under solar light does not always represent a total bacterial death. In fact, the ROS produced during the photocatalytic process can induce oxidative stress on the microorganisms, causing the cells to enter a VBNC state.

Solar UV irradiation and attacks by oxidative species generate changes in the permeability of the lipid membrane and/or modify the bacterial DNA, leading to the loss of cultivability while viability remains unaltered (Ben Said et al. 2010). When the oxidative stress ends (dark conditions), the microorganisms recover their cultivability. In addition, the growth state of bacteria (exponential or stationary) varies in time. Depending on this parameter, some bacteria could persist under photocatalytic conditions and consequently, their recovery rate in the dark could also be influenced. This hypothesis can explain the recovery of E. coli and P. aeruginosa after a rest time in the darkness.

In addition, the decrease in the accumulated UV energy (QUV) and the fluctuation of the visible spectral composition of sunlight during exposed day can affect directly the solar photo-inactivation, and photoreactivation as well as the bacterial behavior in the subsequent dark period.

We can also observe a reactivation of bacteria even during photocatalytic treatment. Indeed, between t = 20 min and t = 30 min, P. aeruginosa presented a reactivation (Fig. 3). This bacteria resilience during this period can be explained by the cells’ ability to restore their UV-induced lesions by molecular photo-dependent mechanism called photoreactivation. This mechanism repair was thought to be an important component of the bacterial arsenal in the repair or reversal of UV-mediated DNA damage (Ben Said and Otaki, 2012). Photoreactivation in bacteria involves a single enzyme called photolyase which binds CPDs and, in a light-dependent step, monomerizes the CPD and dissociates from the repaired lesion (Ben Said and Masahiro, 2013). Indeed, UVA is essential for photoreactivation, although it also has lethal and sublethal effects on microorganisms (Oguma et al., 2001). Thus, when the value of log ratio (Cr) was inferior to 1, we can deduce that the lethal effects of QUV radiation take over its beneficial effects but when Cr was superior to 1, the inactivation effect of photocatalytic can be slowed down, and thereby, we can integrate this log ratio in the inactivation equation as well:

$$ N/{N}_0=A\ \left(1-1/{C}_{\mathrm{r}}\right)\ \left[\mathrm{Exp}\ \right(-k\ \left({C}^n\ {t}^m\right)\Big] $$

Figure 5 shows the bacterial kinetic of P. aeruginosa and E. coli after integration of Cr in the kinetic model. We can note a modification of the initial bacterial reduction rate in the contact with solar UV radiation and TiO2-P25 for both tested bacteria that showed a reactivation after photocatalytic treatment where, for E. coli, A′ was equal to 0.082 instead of 0.14 determined with SE model and for P. aeruginosa, A′ was equal to 0.38 instead of 0.94 determined with C-W model. The difference between the two kinetic values (A–A′) represents the potential bacteria rate able to resuscitate after photocatalytic treatment.

Fig. 5

Bacterial kinetic of aP. aeruginosa ATCC4114 and bE. coli ATCC25922 after integration of coefficient of reactivation (Cr) in the kinetic models: Chick-Watson model (C-W) and series-events model (SE)

The purpose of Cr rate integration in the photocatalytic inactivation equation was to consider an eventual reactivation after photocatalytic treatment in real scale and to find a solution to overcome the inconvenient related to this interesting technology.

According to Fig. 4, no recovery was observed for the tested S. typhi after 24 h in the darkness (Cr equal to 0.52). In this case, only S. typhi completed loss of viability and cultivability; other tested bacteria were not affected. Thus, the behavior in the dark of S. typhi suggests that during photocatalytic disinfection, radicals and other oxidative species produced by illuminated TiO2-P25 induced damage that can in certain cases get worse in the dark, generating a “residual effect” of the photocatalytic treatment. For this reason in post-treated S. typhi suspension, bacterial cells continued to decrease in the dark. In this case, the DNA repair mechanism became less active rendering the irreversible S. typhi inactivation. In other words, the bacterial injury generated by photocatalytic treatment continued to enhance in the dark.

For this delayed process, we have applied the term of “residual effect” but it is not necessarily induced by the residual presence of any active oxidative compound. In a previous article, we have reported that this “residual effect” in the dark was dependent on the light intensity previously applied (Marugán et al. 2008).


The photocatalytic solar UV/TiO2 system was tested for its bactericide action against three bacterial strains:

E. coli ATCC 25922, P. aeruginosa ATCC 4114, and S. typhi ATCC 560. The comparison of different kinetic parameters determined according to the C-W model and series-events model (k, A) for the studied bacteria showed a difference in each tested bacterial response to a disinfectant agent. This difference in bacterial behavior or response was probably due to an inter-specific difference concerning the cell growth strategy, the response to the environmental stress, the metabolism activity, the genetic regulation and flexibility, etc.

After solar UV/TiO2 treatment, bacterial strains (E. coli and P. aeruginosa) showed an ability to overcome photocatalytic stress, determined by the reactivation coefficient (Cr). This reactivation can be related to the fact that the biological systems have evolved mechanisms to appropriately respond to environmental stresses. These results can affect the effectiveness and the durability of solar photocatalytic treatment with fixed titanium dioxide device. However, no recovery was observed for tested S. typhi after 24 h in the darkness. In this case, only S. typhi complete loss of viability and cultivability. Thus, the behavior in the dark of S. typhi suggest that during photocatalytic disinfection, radicals and other oxidative species produced by illuminated TiO2 induce damage that can in certain cases get worse in the dark, generating a “residual effect” of the photocatalytic treatment.

Enhancing the photocatalytic process efficiency remains a challenge and a subject of extensive research. In the end of this paper, we can propose many solutions such as (i) the increase of retention time, (ii) the change of photoreactor conception, and (iii) the addition of supplement step applied before or after solar UV-TiO2.


  1. Acevedo A, Carpio EA, Rodríguez J, Manzano MA (2011) Disinfection of natural water by solar photocatalysis using immobilized TiO2 devices: efficiency in eliminating indicator bacteria and operating life of the system. J Sol Energy Eng 134:011008.

    Article  CAS  Google Scholar 

  2. Ben Said M, Masahiro O (2013) Enhancement of ultraviolet water disinfection process. Afr J Biotechnol 20:2932–2938.

    CAS  Article  Google Scholar 

  3. Ben Said M, Otaki M (2012) Development of a DNA-dosimeter system for monitoring the effects of pulsed ultraviolet radiation. Ann Microbiol 62:1339–1344.

    Article  CAS  Google Scholar 

  4. Ben Said M, Masahiro O, Hassen A (2010) Detection of viable but non cultivable Escherichia coli after UV irradiation using a lytic Qβ phage. Ann Microbiol 60:121–127.

    Article  CAS  Google Scholar 

  5. Bonetta S, Bonetta S, Motta F, Strini A, Carraro E (2013) Photocatalytic bacterial inactivation by TiO2-coated surfaces. AMB Express 3(1):59.

    Article  CAS  Google Scholar 

  6. Bousselmi L, Ghrabi A, Ghozzi K, Zayani G, Ennabli M (2002) Solar photocatalytic treatment of textile wastewater: possibilities and limitations in the Tunisian context. Proceedings of International Symposium on Environmental Pollution Control and Waste Management (EPCOWM’2002) 804–812

  7. Caballero L, Whitehead KA, Allen NS, Verran J (2009) Inactivation of Escherichia coli on immobilized TiO2 using fluorescent light. J Photochem Photobiol A-Chem 202:92–98.

    Article  CAS  Google Scholar 

  8. Cho M, Chung H, Choi W, Yoon J (2005) Different inactivation behaviors of MS-2 phage and Escherichia coli in TiO2 photocatalytic disinfection. Appl Environ Microbiol 71(1):270–275.

    Article  CAS  Google Scholar 

  9. Chong MN, Lei S, Jin B, Saint C, Chow CWK (2009) Optimisation of an annular photoreactor process for degradation of Congo Red using a newly synthesized titania impregnated kaolinite nano-photocatalyst. Sep Purif Technol 67:355–363.

    Article  CAS  Google Scholar 

  10. Chonga MN, Jinb B, Saint CP (2011) Bacterial inactivation kinetics of a photo-disinfection system using novel titania-impregnated kaolinite photocatalyst. Chem Eng J 171:16–23.

    Article  CAS  Google Scholar 

  11. Dheaya MAA, Patrick SMD, McMurray TA, Byrne JA (2009) Photocatalytic inactivation of E. coli in surface water using immobilised nanoparticle TiO2 films. Water Res 43:47–54.

    Article  CAS  Google Scholar 

  12. Errol CF , Graham CW, Wolfram S, Richard DW, Roger AS , Tom E (2006) DNA repair and mutagenesis. ASM Pres 687–698.

  13. Fujishima A, Rao TN, Tryk DA (2000) Titanium dioxide photocatalysis. J Photochem Photobiol. C: Photochem Rev 1:1–21.

    Article  CAS  Google Scholar 

  14. Gogniat G, Thyssen M, Denis M, Pulgarin C, Dukan S (2006) The bactericidal effect of TiO2 photocatalysis involves adsorption onto catalyst and the loss of membrane integrity. FEMS Microbiol Lett 258:18–24.

    Article  CAS  Google Scholar 

  15. Hashimoto K, Irie H, Fujishima A (2005) TiO2 photocatalysis: a historical overview and future prospects. J Appl Phys 44:8269–8285.

    Article  CAS  Google Scholar 

  16. Huang Z, Maness PC, Blake DM, Wolfrum EJ, Smolinski SL, Jacoby WA (2000) Bactericidal mode of titanium dioxide photocatalysis. J Photoch Photobio A 130:163–170.

    Article  CAS  Google Scholar 

  17. Ismail M (2011) Preparation et caracterisation de nouveaux materiaux pourles reactions de depollution photocatalytique de l’eau dans le visible. Thèse à l’Institut National Polytechnique de Lorraine, ENSIC Nancy France

  18. Kiwi J, Nadtochenko V (2005) Evidence for the mechanism of photocatalytic degradation of the bacterial wall membrane at the TiO2 interface by ATR-FTIR and laser kinetic spectroscopy. Langmuir 21:4631–4641.

    Article  CAS  Google Scholar 

  19. Lifen L, Barford J, King Lun Y (2009) Non-UV germicidal activity of fresh TiO2 and Ag/TiO2. J Environ Sci 21:700–706.

    Article  CAS  Google Scholar 

  20. Lindauer KG, Darby J (1994) Ultraviolet disinfection of waste water: effect of dose on subsequent photoreactivation. Water Res 28:805–817.

    Article  Google Scholar 

  21. MacFarlane JW, Jenkinson HF, Scott TB (2011) Sterilization of microorganisms on jet spray formed titanium dioxide surfaces. Appl Catal B Environ 106:181–185.

    CAS  Article  Google Scholar 

  22. Malato S, Blanco J, Richter C, Fernández P, Maldonado MI (2000) Solar photocatalytic mineralization of commercial pesticides: oxamyl. Sol Energ Mat Sol C 64:1–14.

    Article  CAS  Google Scholar 

  23. Marugán JO, Grieken RV, Sordo C, Cruz C (2008) Kinetics of the photocatalytic disinfection of Escherichia coli suspensions. Appl Catal B: Environ 82:27–36.

    Article  CAS  Google Scholar 

  24. Matsunaga R, Tomodam T, Wake NH (1985) Photoelectrochemical sterilization of microbial cells by semiconductor powders. FEMS Microbiol Lett 29:211–214.

    Article  CAS  Google Scholar 

  25. Mills A, Lee SK (2002) A web-based overview of semiconductor photochemistry-based current commercial applications. J Photoch Photobio A 152:233–247.

    Article  CAS  Google Scholar 

  26. Mills A, Hepburn J, Hazafy D, O'Rourke C, Krysa J, Baudys M, Zlamal M, Bartkova H, Hill CE, Winn KR, Simonsen ME, Søgaard EG, Pillai SC, Leyland NS, Fagan R, Neumann F, Lampe C, Graumann T (2013) A simple, inexpensive method for the rapid testing of the photocatalytic activity of self-cleaning surfaces. J Photoch Photobio A Chem 272:18–20.

    Article  CAS  Google Scholar 

  27. Nassar R, Mokh S, Rifai A, Chamas F, Hoteit M, AlIskandarani M (2017) Transformation of sulfaquinoxaline by chlorine and UV light in water: kinetics and by-product identification. Environ Sci Pollut Res:1–10.

  28. Oguma K, Katayama H, Ohgaki S (2001) Photoreactivation of Escherichia coli after low- or medium-pressure UV disinfection determined by an endonuclease sensitive site assay. Appl Environ Microbiol 68:6029–6035.

    Article  CAS  Google Scholar 

  29. Pablos C, VanGrieken R, Marugán J, Moreno B (2011) Photocatalytic inactivation of bacteria in a fixed-bed reactor: mechanistic insights by epifluorescence microscopy. Catal Today 161:133–139

    Article  CAS  Google Scholar 

  30. Pagan R, Manas P, Raso J, Condon S (1999) Bacterial resistance to ultrasonic waves under pressure at nonlethal (manosonication) and lethal (manothermosonication) temperatures. Appl Environ Microbiol 65:297–300

    CAS  Google Scholar 

  31. Pigeot-Rémy S, Simonet F, Atlan D, Lazzaroni JC, Guillard C (2012) Bactericidal efficiency and mode of action: a comparative study of photochemistry and photocatalysis. Water Res 46:3208–3218.

    Article  CAS  Google Scholar 

  32. Piscopo A, Robert D, Weber JV (2001) Influence of pH and chloride anion on the photocatalytic degradation of organic compounds Part I. Effect on the benzamide and para-hydroxybenzoic acid in TiO2 aqueous solution. Appl Catal B: Environ 35:117–124

    Article  CAS  Google Scholar 

  33. Rahmani AR, Samarghandi MR, Samadi MT, Nazemi F (2009) Photocatalytic disinfection of coliform bacteria using UV/TiO2. J Res Health Sci 9:1–6

    CAS  Google Scholar 

  34. Rengifo HJA, Pulgarin C, Machuca F, Sanabria J (2010) TiO2 photocatalytic inactivation under simulated solar light of bacterial consortia in domestic wastewaters previously treated by uasb, duckweed and facultative ponds. Quim Nova 33:1636–1639.

    Article  Google Scholar 

  35. Rincon AG, Pulgarin C (2004) Field solar E. coli inactivation in the absence and presence of TiO2: is UV solar dose an appropriate parameter for standardization of water solar disinfection. Sol Energy 77:635–648.

    Article  CAS  Google Scholar 

  36. Robertson JMC, Robertson PKJ, Lawton LA (2005) A comparison of the effectiveness of TiO2 photocatalysis and UVA photolysis for the destruction of three pathogenic micro-organisms. J Photoch Photobio A 175:51–56.

    Article  CAS  Google Scholar 

  37. Rtimi S (2017) Indoor light enhanced photocatalytic ultra-thin films on flexible non-heat resistant substrates reducing bacterial infection risks. Catalysts 7:57.

    Article  CAS  Google Scholar 

  38. Rtimi S, Pulgarin C, Sanjines R, Kiwi J (2015) Kinetics and mechanism for transparent polyethylene-TiO2 films mediated self-cleaning leading to MB dye discoloration under sunlight irradiation. Appl Catal B Environ 162:236–244.

    Article  CAS  Google Scholar 

  39. Sapizah R, Shahidan R, Ainon H (2012) Inactivation of Escherichia coli under fluorescent lamp using TiO2 nanoparticles synthesized via sol gel method. Sains Malays 41(2):219–224

    Google Scholar 

  40. Schuch AP, Galhardo RS, Lima-Bessa KM, Schuch NJ, Menck CFM (2009) Development of a DNA-dosimeter system for monitoring the effects of solar-ultraviolet radiation. Photochem Photobiol Sci 8:111–120.

    Article  CAS  Google Scholar 

  41. Severin BF, Suidan MT, Engelbrecht RS (1984) Series-event kinetic model for chemical disinfection. J Environ Eng 110:430–439.

    Article  CAS  Google Scholar 

  42. Silva CR, Miranda SM, Lopes FVS, Silva M, Dezotti M, Silva AM, Faria JL, Boaventura RAR, Vilar VJP, Pinto E (2017) Bacteria and fungi inactivation by photocatalysis under UVA irradiation: liquid and gas phase. Environ Sci Pollut Res 24:6372–6381.

    Article  CAS  Google Scholar 

  43. Spasiano D, Marotta R, Malato S, Fernandez-Ibanez P, Di Somma I (2015) Solar photocatalysis: materials, reactors, some commercial, and pre-industrialized applications. A comprehensive approach. Appl Catal B: Environ 170-171:90–123.

    Article  CAS  Google Scholar 

  44. Tatsuma T, Kubo W, Fujishima A (2002) Patterning of solid surfaces by photocatalytic lithography based on the remote oxidation effect of TiO2. Langmuir 18:9632–9634.

    Article  CAS  Google Scholar 

  45. Vamathevan V, Amal R, Beydoun D, Low G, McEvoy S (2002) Photocatalytic oxidation of organics in water using pure and silver-modified titanium dioxide particles. J Photoch Photobio A Chem 148:227–233.

    Article  CAS  Google Scholar 

  46. Wang W, Huang G, Yu JC, Wong PK (2015) Advances in photocatalytic disinfection of bacteria: development of photocatalysts and mechanisms.34:232–247.

  47. Watson HE (1908) A note on the variation of the rate of disinfection with change in the concentration of the disinfectant. J Hyg 8:536–542

    CAS  Google Scholar 

  48. Zsilák Z, Szabó-Bárdos E, Fónagy O, Horváth O, Horváth K, Hajós P (2014) Degradation of benzenesulfonate by heterogeneous photocatalysis combined with ozonation. CatalToday 230:55–60.

    CAS  Article  Google Scholar 

Download references


This work is partially supported by the Tunisian- French project PHC Utique CMCU 14G0821.

Author information



Corresponding author

Correspondence to Faouzi Achouri.

Additional information

Responsible editor: Suresh Pillai

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Achouri, F., BenSaid, M., Bousselmi, L. et al. Comparative study of Gram-negative bacteria response to solar photocatalytic inactivation. Environ Sci Pollut Res 26, 18961–18970 (2019).

Download citation


  • Photocatalysis
  • TiO2-P25
  • Solar irradiation
  • Inactivation kinetic
  • Reactivation