Introduction

The changing needs of every human being in different countries and continents pose challenges for the environment. The diversity of surfactant applications is astonishing. Soap has been used by humankind since ancient times and was probably already known by the Sumerians (Wu et al. 2021, Coiffard, Couteau 2020). Today, during the COVID-19 pandemic, thorough and frequent handwashing with surfactants (soap) is one of the main protective measures at the core of the contamination prevention strategy (Wu et al. 2021). The present ongoing development of the industry relates to the growing importance of surfactants, which are used in the emerging technologies including: food industry (Kralova and Sjöblom 2009; Nitschke and Sousa e Silva 2018), chemical technology and biotechnology (Teixeira et al. 2017), metallurgy, mining, paint and pigment technology (Haramagatti et al. 2018), nanotechnology, pharmacy (Rajput et al., 2018), and medicine (Sar et al. 2019). The increasing production and consumption of surfactants results in a need for continuous modernization and improvement of the wastewater treatment methods with respect to removal of these substances (Palmer, Hatley 2018). Surfactants can be classified into many categories (impact on the environment, possible applications, or the raw material used for their production). However, the most popular classification is based on their chemical structure (ionic: cationic, anionic; amphoteric, non-ionic) (Müllertz et al. 2010). This classification allows to distinguish which particular groups are used, what is their impact on the environment, or which analytical procedures should be applied to detect them in environmental samples (Olkowska et al. 2011). The characteristic feature of surfactants is the presence of hydrophobic (usually long-chain hydrocarbons) and hydrophilic (ionic or highly polar groups) groups in their structure (Cirelli et al. 2009). This phenomenon enables the formation of nano-domains that are spontaneously self-assembling into spherical micelles and aggregates in water (Kim et al. 2017). In fact, this is a key feature of surfactants which translates into a wide spectrum of their applications, closely related to their chemical structure. Anionic surfactants are the largest and most popular group, which account for approx. 60% of the total global production of surfactants (Khaled et al., 2015). They are mostly based on alkylbenzene sulfonates and are used in the production of detergents or cosmetics for personal care. The next group are non-ionic surfactants, which rank second after anionic surfactants in terms of production and consumption. Their production accounts for ~ 30% of the total market share. In contrast to other groups, the solubility of the non-ionic surface-active agents decrease with the increase in the temperature. They are mostly used as emulsifiers or detergents in pulp and paper, textile, and oil industry (Kneper et al., 2003; Siyal 2020; Liwarska-Bizukojc et al. 2005; Lindman et al. 2016). It is worth to mention that non-ionic surfactants are less toxic in comparison to other surfactants. Moreover, non-ionic surfactants also reduce the toxic impact of anionic surface-active agents (Cserhati et al., 2002). Cationic surfactants are another group, the production of which accounts for ~ 10% of all surfactant production. Nevertheless, they offer additional benefits compared to other surfactants. In addition to surface activity, they also exhibit antibacterial and antistatic properties (Zhou and Wang 2019). In terms of the chemical structure, they are usually quaternary ammonium salts and are used mostly as disinfectants (Zhou and Wang 2019) as well as textile softeners (Siyal 2020; Mata 2005; Kneper et al., 2003).

The consumption of surfactants has increased since the 1980s from 1.7 million tons in 1984 to 15.93 million tons in 2014, and it is estimated that in 2023 the production will exceed 24.12 million tons per year. It should also be mentioned that in 2014, approx. 40% of the production of all surfactants in the world was intended for the fabrication of detergents and cleaning agents for households (Palmer and Hatley 2018). Similar to all chemical compounds, surfactants should also be constantly monitored because they are not completely neutral to the environment. Due to their amphiphilic nature, surfactants can adsorb on the surface of solid particles. They can also precipitate from solutions in the presence of metal ions. After use, surfactants are released into the sewage system and then into the sewage treatment plant, where they are completely or incompletely removed by, for example, adsorption and biodegradation. Unfortunately, high concentrations of surfactants may not be completely eliminated during the purification processes and surfactants enter drinking water resources, posing a threat to humans, animals and plants (Scott et al., 2000; Gonzalez et al., 2007; Siyal 2020). They can disrupt the hormone balance by mimicking the activity of hormones and thus disrupt chemical signaling and the development of the host organism. The ecological awareness of the society has increased significantly in recent years and care for the environment is considered a priority, therefore the production of a new surfactant requires data regarding its toxicity and biodegradability (Britton 1998; Rosal et al. 2010). In consequence, the development of effective techniques for neutralization of these compounds from wastewater is of vast importance. Among various recently developed techniques, photocatalytic degradation of surfactants is relatively poorly described in the literature. Most of the scientific attention is shifted toward photocatalytic degradation of phenols (Mohamed et al. 2020), dyes (Lee et al. 2020), drugs (Da Silva et al. 2015), hormones (Padovan et al. 2021) and pesticides (Miguel et al. 2012). Photodegradation of surfactants has been known for many years and many reports regarding this topic have been described by Hidaka and Zhao (1992) and Hidaka et al. (1992). The research has mainly focused on the use of TiO2 as a catalyst and it has been confirmed that such a treatment method is effective. Nevertheless, there has been considerable progress since these studies have been published. Our recent studies confirm that photocatalysis might be effectively applied for UV–Vis degradation of non-ionic surfactants (Huszla et al. 2022). Most preceding studies have been dedicated to photocatalytic degradation of surfactant’s model solutions with concentrations up to 10 mg/L. While the industrial, highly concentrated (up to 1000 mg/L (Klimonda & Kowalska 2021), wastewater has the environmentally most negative impact. Therefore, the ultimate aim of this study was the comprehensive comparative analysis of the catalytic photodegradation of representatives of three-surfactants groups at concentrations 10 mg/L, 100 mg/L, and 1000 mg/L. Additionally, current state of the art regarding photocatalytic surfactant degradation severely lacks the evaluation of ecotoxicity of the photodegradation products. Photodegradation can potentially lead to the formation of more toxic decomposition products. Therefore, the overriding assurance of work was the identification of degradation products and the assessment of their ecotoxicity (Huszla et al. 2022).

Materials and methods

Reagents and chemicals

All cationic surfactants (benzalkonium chloride, hexadecyltrimethylammonium bromide—CTAB, dioctyldimethylammonium chloride—DDAC) and all anionic surfactants (sodium dodecyl sulfate—SDS, sodium dodecylbenzenesulfonate—SDBS, dioctyl sulfosuccinate sodium salt—Manoxol) as well as two non-ionic surfactants (dodecylalcohol ethoxylate—C12E10 and nonylphenol ethoxylate—TX-100) were obtained from Merck (Germany). Rokopol 30p27 (Block copolymer of propylene oxide (30 units) and ethylene oxide (27 units) were provided by PCC EXOL (Poland). All reagents used in spectrophotometric analysis (bismuth(III) nitrate, glacial acetic acid, potassium iodide, barium chloride, anhydrous sodium hypophosphite (NaH2PO2), thiourea, nitric acid, sulfuric (VI) acid, methylene blue, disulphine blue, ethanol, and anhydrous sodium acetate) were obtained from Avantor Performance Materials Poland S.A. (Gliwice, Poland). Zinc acetate dihydrate used for synthesis of ZnO nanoparticles was obtained from Merck (Germany).

LC/MS-grade methanol and ammonium acetate were purchased from Sigma–Aldrich (Saint Louis, MO, USA). HPLC-grade water was prepared by reverse osmosis in a Demiwa system from Watek (Ledec and Sazavou, The Czech Republic) followed by double distillation from a quartz apparatus.

Synthesis of ZnO nanoparticles

The preparation of ZnO was performed accordingly to the method published previously by Tian et al. (2012) with a slight modification (Huszla et al. 2022). The crucible containing 25 g of solid zinc acetate dihydrate was placed in the furnace (Nabertherm, Germany) and annealed to 600 °C with a heating rate of 5 °C/min in the air atmosphere and kept at the maximum temperature for 120 min. Afterward, the sample was naturally cooled down to ambient temperature, 9 g of the gray precipitate was collected and used for further studies. The detailed physicochemical characterization of ZnO nanoparticles as catalyst is described in our previous publication (Huszla et al. 2022). Formed photocatalyst has a hexagonal crystalline structure with preferred (101) orientation and crystallites size ~ 40 nm and specific surface area ~ 10 m2/g (Huszla et al. 2022).

Photocatalytic degradation experiment

Photoactivity tests were performed in a UV-RS2 laboratory reactor (Heraeus, Germany) equipped with a 150-W medium-pressure mercury lamp as a UV light source, surrounded by a water-cooling quartz jacket, at room temperature. In the experiment, the photocatalyst (700 mg) was suspended in 700 mL of an aqueous solution of a selected surfactant at initial concentrations of 10 mg/L; 100 mg/L or 1000 mg/L. Firstly, the surfactant solution and photocatalyst were magnetically stirred using an R05 IKAMAG magnetic stirrer (IKA Werke GmbH, Germany) in darkness for 30 min to reach the adsorption/desorption equilibrium prior to UV–Vis irradiation. After equilibrium was reached, the adsorbed concentration of surfactants was measured. The radiation was then turned on to initiate the photocatalytic reaction. The distance of lamp to the reactor was 10 cm. Experiments where performed at pH 7. In the next step, the irradiated mixtures (8 mL) were collected from the reactor at specific time intervals (1, 5, 10, 20, 30, 40, 50, and 60 min) and centrifuged (12,000 rpm, MPW-352, MPW MED. Instruments, Poland) to remove the photocatalyst. The perfect mixing was assumed in this process. Thus, sampling does not change the concentration of photocatalyst in the reaction system.

Spectrophotometric determination of surfactant content

Cationic surfactants determination

Cationic surfactants were determined using the disulphine blue active substances (DBAS) method (Uning et al. 2018 and Kurrey et al. 2019). This method is based on the reaction of cationic surfactants with the anionic dye (disulphine blue) which forms ion associates that are extracted using chloroform. Afterward, the absorbance is measured spectrophotometrically. In brief, 0,16 g of disulphine blue dye was dissolved in 20 mL of a 10% (v/v) ethanol solution, then the solution was transferred into a 50-mL conical flask and filled with demineralized water. For the analysis, a given aliquot of the analyzed sample and 15 mL of distilled water were placed in a 50-mL separating funnel. Then, 2 mL of acetate buffer (pH 5), 1 mL of disulphine blue solution (Sigma Chemical, USA) and 5 mL of chloroform were added to the funnel and the obtained mixture was vigorously shaken for 3 min. The separating funnel was set aside for 5 min to allow for the complete separation of the phases. The chloroform phase was carefully transferred into a 1-cm quartz cell and the light absorbance was measured at λ = 627 nm using a UV–Vis V-530 spectrophotometer (Jasco, Japan).

Anionic surfactants determination

Anionic surfactant determination was performed using the methylene blue active substances (MBAS) method. Firstly, the acidic methylene blue solution was prepared by dissolving 350 mg of methylene blue and 6.5 mL of concentrated sulfuric acid in 1 L of demineralized water. An aliquot of the analyzed sample was placed in a 100-mL separating funnel together with 45 mL of de water, 5 mL of acidic methylene blue solution and 10 mL of chloroform, and obtained mixture was vigorously shaken for 3 min. Then, the separating funnel was set aside for 5 min to allow for the complete separation of the phases. In next step, the chloroform phase was filtered through the qualitative filter paper directly to a 1-cm quartz cell and the light absorbance was measured at λ = 652 nm using a UV–Vis V-530 spectrophotometer.

Non-ionic surfactants determination

The concentration of non-ionic surfactants after adsorption and UV irradiation was determined by means of the iodobismuthate method with final measurement of absorbance of the bismuth-thiourea complex (BiAS-thio), based on the precipitation of oxyethylate with the Dragendorff's reagent. The precipitate was dissolved with the simultaneous formation of the bismuth-thiourea complex, the absorbance of which was proportional to the surfactant content of the tested sample. The absorbance of the solution at λ = 468 nm was measured using a UV–Vis V-530 spectrophotometer.

Liquid chromatography-mass spectrometry identification of photodegradation residues

Identification of compounds formed during the degradation of surfactants was performed using the UltiMate 3000 HPLC (Dionex, Sunnyvale, CA, USA) with the 4000 QTRAP mass spectrometer (ABSciex, Foster City, CA, USA). A sample volume of 5 μL was injected into the analytical column Gemini-NX C18 (100 mm; 2.0 mm I.D.; 3 μm) purchased from Phenomenex (Torrance, CA, USA) which was thermostated at 35 °C. The analysis was conducted in a gradient mode with mobile phase A (5 mM ammonium acetate in water) and mobile phase B (methanol) flowing at 0.3 mL min−1. In the applied gradient, methanol concentration was changing from 70 to 95% during 3 min and next isocratic elution was maintained for 3 min. The electrospray (ESI) ion source operated in a negative or positive mode depending on the surfactant. The negative mode was used for SDS, SDBS and Manoxol, while the positive mode was used for Triton X-100 and dodecyl alcohol ethoxylate. Nitrogen was used in both the source and the mass spectrometer. The following analytical parameters were used: curtain gas pressure 20 psi, nebulization gas pressure 40 psi, auxiliary gas pressure 40 psi, source temperature 350 °C, ESI voltage ± 4500 V, declustering potential 50 V, entrance potential 10 V. Data were collected in the Q1 mode in the 100–1200 m/z range.

Evaluation of phytotoxicity after photodegradation

To analyze the potential toxicity of residual compounds after photodegradation, the phytotoxicity toward the model plant was determined before and after the purification process. Sorghum (Sorghum saccharatum) was selected as a model monocotyledon plant used for the research due to its common application as a widespread crop. This nutritional plant is cultivated in various regions of the world and can serve both as an ingredient of human diet and animal fodder.

The experiments were carried out using commercial Phytotoxkit containers (Tigret, Belgium) filled with 100 ± 0.1 g of agricultural soil which contained the studied surfactants. The soil was collected from a field near the city of Konin (Poland) at a depth of 5–10 cm and sieved using a 0.1-mm mesh sieve in order to remove any residual debris. Prior to the experiments, two sets of solutions of the studied surfactants were prepared at a concentration of 1000 mg/L. One set of solutions was used for the experiments without further modifications, whereas the second set was subjected to photocatalytic degradation for 60 min as described in Sect. "Photocatalytic degradation experiment".

In order to prepare the samples, 20 mL of a given set of solutions were introduced into the prepared soil and manually mixed in order to achieve homogeneity. In case of control samples, 20 mL of distilled water was used instead. After the soil was placed in the Phytotoxkit containers, 10 sorghum seeds were placed on the soil and covered with a protective lid. All samples were prepared in triplicate. The kits were subsequently incubated in the dark at 25 °C for 7 days. After the incubation, the number of germinated seeds in each sample was counted and length of their shoots and roots was measured.

On the basis of the collected measurements, the germination index was calculated in accordance with the following formula:

$$GI=\frac{Gs}{Gc}*\frac{Ls}{Lc}*100 \left[\%\right]$$

where: Gs and Gc are the number of germinated seeds in the sample and the control, respectively; while Ls and Lc are the length of the shoot in the sample and the control, respectively.

Results and discussion

Degradation of surfactants

The efficiency of photodegradation of the studied surfactants was evaluated in two separate series of experiments: without the use of a catalyst and during photocatalytic degradation with the use of ZnO. In the latter case, the overall removal of the surfactant was divided into adsorption and degradation. The results are summarized in Table 1.

Table 1 Effectiveness of removing various types of surfactants from the aquatic environment for different concentration levels without a photocatalyst and with the use of a ZnO photocatalyst after 60 min
Full size table

At the relatively low concentration (10 mg/L), the photodegradation efficiency was quite satisfactory in the rage 55–99% for all types of surfactants. Only one exception was registered for anionic surfactant Manoxol, for which the photodegradation efficiency was equal to only 23%, but the addition of a catalyst increased the removal efficiency up to 100%.

At a concentration of 100 mg/L without the catalyst, the degradation efficiency of all anionic surfactants was in the range of 24–30%. For non-ionic C12E10 and Rokopol 30p27, the range was equal to 37–47%, however for TX-100 the degradation rate reached 60%. In case of cationic surfactants, the range for CTAB and DDAC was equal to 27–28%, but for benzalkonium chloride, the degradation rate was equal to 51%. After addition of the catalyst, the photodegradation efficiency for all surfactants increased accordingly to a range of 41–50% for anionic surfactants, 64–70% for non-ionic compounds and to 38–43% for cationic CTAB and DDAC, while in case of benzalkonium chloride the efficiency reached 94%.

Differences in the reduction of surfactants from solution, indicating varying degradation levels, can be attributed to several factors. Firstly, adsorption capacity—the inherent ability of a specific surfactant to adsorb onto a catalyst can influence its degradation rate. Surfactant structure plays also pivotal role. There is an evidence suggesting that surfactants containing benzene rings photodegrade more readily, our research reveals other nuances. For instance, linear surfactant structures, like C12E10 at 1000 mg/L, displayed a 65% reduction, suggesting easier photodegradation. Conversely, branched structures, such as Manoxol at 1000 mg/L, showed only a 28% reduction, indicating resistance to photodegradation.

For very high concentrations of surfactants (1000 mg/L) in water systems, the biodegradation processes practically disappear, and it is impossible to reduce this type of contamination by microorganisms, but in the tests performed previously in our laboratory, the efficiency of photodegradation was still present although only in the range of 9 ÷ 48% (for anionic surfactants: 9 ÷ 20%, for non-ionic surfactants: 27 ÷ 35%, for cationic surfactants: 16 ÷ 48%). Biological methods are ineffective against such high concentrations of surfactants found in concentrated industrial wastewater. Introducing a large load of surfactants to a biological wastewater treatment plant will result in the loss of active sludge and a significant decrease in the efficiency of the entire treatment plant (Wyrwas et al. 2011).

The addition of the ZnO photocatalyst improved the photodegradation for each of the tested surfactants. The lowest effect of the photocatalyst on photodegradation (concentration of 1000 mg/L) was observed for anionic surfactants: SDBS (13.9%) and Manoxol (17.2%). Surfactants belong to a specific group of organic compounds, therefore their removal rate from the aquatic environment is highly influenced by the adsorption process on the photocatalyst. Cationic surfactants stand out in this respect, reaching as much as 30 ÷ 65% of adsorption at a concentration of 10 mg/L. The presence of a benzene ring in the surfactant structure facilitates the photodegradation process. On the other hand, it seems that the efficiency of the photodegradation processes decreases with increased branching of the surfactant structure. Cationic surfactants are characterized by the highest toxicity among the tested types of surfactants (Kaczerewska et al. 2020); therefore, they can be resistant to biodegradation, especially at high concentrations of 1000 mg/L. However, the use of a photocatalyst during photodegradation enables their reduction in the aquatic samples by 30–66%. By using the ZnO photocatalyst, it is possible to completely remove the surfactant from the solution within 1 h for concentrations up to 10 mg/L. The higher the surfactant concentration, the lower the efficiency in relation to the introduced dose, but even for a high concentration of 1000 mg/L, the removal efficiency may reach 27 ÷ 66% (for SDBS only 15%). The use of ZnO not only significantly increased the photodegradation efficiency, but also significantly accelerates the process itself. Photodegradation with the use of ZnO can be a promising alternative to the processes of bio-purification and an effective solution for eliminating the presence of surfactants in the aquatic environment, especially in case of compounds characterized by high toxicity or present at high concentrations.

The results obtained from two separate experimental runs served as the basis for the determination of reaction kinetic parameters. It was proposed that a system in which only UV was used to induce the reaction run is described by equilibrium reaction kinetics since the reaction proceeds in volume and, which is particularly evident at higher concentrations, tends toward a respective final concentration different from zero. The reaction in the system catalyzed by ZnO was carried out in two consecutive steps: the first step without UV exposure of the sample and the second step with the additional introduction of UV exposure. The two stages are defined by distinguishing a separate phase without and with UV.

For the reaction without catalyst using UV sample excitation, the kinetic model is as follows:

$$ r = - \frac{dC}{{dt}} = - k_{1} C_{t} + \frac{{k_{1} }}{{K_{eq} }}(C_{init} - C_{t} ) $$
(1)

In the above equation, the kinetic constant k1 and the equilibrium constant Keq are quantities determined from experimental results on the time dependence of concentrations.

The reaction catalyzed by ZnO nanoparticles can be described by a pseudo first-order (2) or pseudo second-order (3) reaction. Both kinetic models have the forms:

$$ r_{I} = \frac{{dq_{t} }}{dt} = k_{q1} \left( {q_{e} - q_{t} } \right) $$
(2)
$$ r_{II} = \frac{{dq_{t} }}{dt} = k_{q2} \left( {q_{e} - q_{t} } \right)^{2} $$
(3)

The procedure for the fitting of kinetic parameters was based on the application of a nonlinear least squares minimization algorithm. Appropriate kinetic models were fitted to appropriate experimental data sets describing the time series. The proposed algorithm consists of two steps:

a) reformulation the models (1), (2) and (3) into integral forms (4), (5) and (6),

b) fitting the parameters of models (1), (2) or (3), respectively, by using their corresponding integral representations.

The closed form solution for integrals is found for all three models. For the model without catalyst which is given by the Eq. (1), it is found to be:

$$ C_{t} = \frac{{C_{{{\text{init}}}} \left( {1 + {\text{exp}}\left( { - \frac{{k_{1} \left( {K_{eq} + 1} \right)t}}{{K_{eq} }}} \right)K_{eq} } \right)}}{{K_{eq} + 1}} $$
(4)

For the model given by Eq. (2), the integral form reads:

$$ q_{t} = q_{e} + \left( {q_{init} - q_{e} } \right){\text{exp(}} - k_{q1} t{)} $$
(5)

and of the model (3) for the second-order kinetics it is:

$$ q_{t} = \frac{{\left( {q_{e}^{2} - q_{e} q_{init} } \right)k_{q2} t + q_{init} }}{{1 + \left( {q_{e} - q_{init} } \right)k_{q2} t}} $$
(6)

The integration is performed with an initial condition selected as the values of Cinit and qinit. The integral formulation was then used in the fitting procedure by means of the least squares algorithm.

The equilibrium analysis was performed based on experimental points obtained at end time interval assuming their equilibrium character. The concentrations were recalculated to obtain the sorption capacities qt at each time point using Eq. (7). For each component concentration used, the corresponding equilibrium capacity was obtained, qe.

$$ q_{t} = \frac{{(C_{0} - C_{t} )V}}{m} $$
(7)

In the above, C0 is the initial concentration of component in the mixture (mg/L) and Ct is the concentration of component at a measurement time t (mg/L). The mixture volume is designated by V and mass of adsorbate m.

The reaction performed in the dark phase of experiment is considered to be entirely based on the adsorption mechanism because of the UV absence. For this case, we studied both first and the second-order pseudo-kinetics for comparison of the cases analyzed. The Langmuir, Freundlich, Temkin and Dubinin-Radushkevich isotherms (Wołowicz et al. 2022; Doan et al. 2021; Kalam et al. 2021) were tested.

The Langmuir isotherm formulation (8) used in this work reads:

$$ q_{e} = \frac{{k_{L} Q_{0} C_{e} }}{{1 + C_{e} k_{L} }} $$
(8)

In the above, the Q0 is the monolayer adsorption capacity (mg/g) and kL is the Langmuir constant (related to the free energy of adsorption) (L/mg).

The Freundlich isotherm Eq. (9) is formulated as follows:

$$ q_{e} = k_{F} C_{e}^{1/n} $$
(9)

The two variables kF (mg1−1/n L1/ng−1) and n are the constants of Freundlich isotherm related to adsorption capacity of adsorbent and the surface heterogeneity, respectively.

The Temkin isotherm relationship is presented below (10):

$$ q_{e} = \frac{RT}{{b_{T} }}lnAC_{e} $$
(10)

The parameters A (L/g) and bT (J/mol)—the Temkin constants, R—universal gas constant (8.314 J/mol K), T—the temperature (K).

The Dubinin-Radushkevich isotherm is mathematically the most complex mathematical expression of those tested in this work and is presented below (11):

$$ q_{e} = q_{m} e^{{ - k_{DR} \varepsilon^{2} }} $$
(11)

In the above, the term ε is the adsorption potential (J/mol), given by (12):

$$ \varepsilon = RTln\left[ {1 + \frac{1}{{C_{e} }}} \right] $$
(12)

kDR (mol2 J2) is the parameter associated with the energy of adsorption, and qm (mg/g) is the maximum adsorption capacity. The Dubinin-Radushkevich isotherm allows furthermore to calculate the mean free energy for departure of molecules from its adsorption site to the infinity, E (J/mol) (13):

$$ E = \frac{1}{{\sqrt {2k_{DR} } }} $$
(13)

The results for each case are presented in Tables 2 and 3.

Table 2 Dark stage of reactions results for both first kq1 and second kq2 order kinetics (2) and (3)
Full size table
Table 3 Equilibrium results for the dark stage (adsorption) experimental setup
Full size table

The results in Table 3 show that for reactions carried out without UV, the rate decreases in the order of anionic, non-ionic, and slowest cationic substances. The results of these experiments show that the type of surfactant has a big influence on the rate of the reaction in the dark stage.

Equilibrium isotherm studies (Table 3) indicate that for the case without UV excitation (adsorption stage), the most suitable models from statistical point of view are Freundlich, Langmuir or Dubinin-Radushkevich. The lowest R2 values in case of the Temkin isotherm indicate that it is the weakest to explain the results of the experiment indicating that it poorly explains the experimental data from the adsorption process without UV excitation. This can be attributed to the assumptions made by the Temkin model, which assumes a linear decrease in the heat of adsorption with coverage due to adsorbate-adsorbent interactions. However, it appears that this assumption does not hold for the adsorption systems under our studies. The physical interpretation of the Temkin isotherm conflicts with the observed adsorption behavior, indicating that the strength of the adsorbate-adsorbent interactions does not decrease linearly with increasing coverage. Consequently, the Temkin model fails to capture the underlying mechanisms governing the adsorption process in this particular system. In contrast, other models we tested such as Freundlich, Langmuir or Dubinin-Radushkevich isotherms, provide better fits to the experimental data and offer more appropriate physical interpretations for the adsorption process without UV excitation. The Freundlich isotherm assumes adsorbate-adsorbent interactions for heterogenous surfaces, the Langmuir isotherm assumes independent adsorption sites, and the Dubinin-Radushkevich isotherm provides information on the energetics of the adsorption process. These interpretations are more consistent with the observed adsorption behavior and are therefore more suitable for describing and understanding the specific adsorption system under investigation.

The purpose of the calculations which the results are presented in Table 4 was to compare the kinetic parameters for the both first- and second-order models for different types of substances. In particular, the focus here was on the reactivity of cationic, non-ionic, and anionic compounds in the presence of ZnO and UV light. The results showed that cationic substances were the most reactive, followed by non-ionic substances. Anionic compounds were the least reactive of the three groups. The photodegradation results showed a distinct reactivity among the three types of surfactants tested. Cationic surfactants exhibited the highest reactivity, likely due to their positively charged head groups, which allow for strong electrostatic interactions with target compounds and catalysts. Non-ionic surfactants exhibited moderate reactivity, relying on weaker interactions such as hydrogen bonding and van der Waals forces. In comparison, anionic compounds, characterized by negatively charged head groups, exhibited the lowest reactivity, as the repulsive electrostatic interactions hindered their adsorption and reactivity with the reactive species and catalyst surfaces. These results highlight the significant role of surfactant type in determining photodegradation reactivity, with cationic surfactants being the most reactive, followed by non-ionic surfactants, and anionic surfactants exhibiting the least reactivity.

Table 4 ZnO + UV (photocatalytic degradation) stage of reactions results for both first kq1 and second kq2 order kinetics (2) and (3)
Full size table

For the case in which the reactions were carried out without a catalyst, using UV only, the kinetic constants in Table 5 indicate that the highest reaction rates are achieved for anionic compounds, followed by non-ionic and finally cationic compounds. The percentage removal of anionic, non-ionic and cationic compounds under UV-only conditions varies due to their different chemical properties and reactivity. Anionic compounds are more easily removed due to their increased interaction with UV radiation. This results in direct photochemical reactions and degradation. Non-ionic compounds have intermediate removal rates because their direct interaction with UV radiation is limited. However, they can still be degraded by indirect processes initiated by UV-generated transient species or free radicals. Cationic compounds have lower removal rates than anionic and non-ionic compounds. The positive charge hinders their direct interaction with UV radiation, thereby reducing their reactivity. The absence of sensitive functional groups or bonds also contributes to their lower removal rates. The differences in percentage removal among these compounds can be attributed to their interaction with UV radiation, susceptibility to direct and indirect photodegradation, and the absence of receptive functional groups or bonds.

Table 5 UV only—reactions kinetics without catalyst, Eq. (1)
Full size table

The three groups react differently under UV-only conditions due to their distinct chemical structures and reactivity. Anionic compounds have higher reaction rates, possibly due to direct photodegradation or sensitization processes. Non-ionic compounds undergo indirect photodegradation through the generation of reactive species like hydroxyl radicals. Cationic compounds, hindered by their positive charge, have reduced reactivity, resulting in slower degradation compared to anionic and non-ionic compounds.

Identification of degradation products

The LC–MS/MS analysis was carried out to obtain information regarding compounds formed during the degradation of tested surfactants. Three sets of samples for each surfactant were investigated using the LC–MS/MS system: at the start of testing, as well as after 30 and 60 min of degradation.

For SDS analyzed in the negative ion mode, a peak at m/z = 265 was found corresponding to the desodiated molecule. During the degradation process, several compounds were formed and the changes in the structure include two processes—addition of the OH group (mass shift of 16 Da) and addition of the OH group with simultaneous elimination of hydrogen (mass shift of 14 Da). As a result, polyhydric alcohols and ketones were detected in the degradation mixture. Characteristic ions for these compounds were detected at m/z = 279, 297, 311, and 329 as well as the ions of lower intensity at m/z = 281 and 295 representing a series of compounds with a random number of hydroxy and keto groups (Fig. 1A).

Fig. 1
figure 1

Mass spectra of A degradation products of SDS, B SDBS and its degradation products

Full size image

SDBS was also analyzed in the negative ion mode. Since SDBS is a mixture of compounds, five ions of their desodiated molecules were found at m/z = 283, 297, 311, 325, and 339. During the degradation process the alkyl chain was oxidized (with simultaneous chain shortening) forming a carboxylic acid and a hydroxyl group was attached to the aromatic ring, thus resulting in the mass shift of 18 Da as described for photooxidative degradation of SDBS with ozone (Li et al. 2023). Characteristic ions were observed at m/z = 315, 329, 343, and 357 (Fig. 1B).

Manoxol has a molecular weight of 444 Da. Therefore, the loss of the sodium ion should result in the formation of an ion at m/z = 421. As in the case of SDS, ions corresponding to the addition of the OH group (mass shift of 16 Da) or addition of the OH group with simultaneous elimination of hydrogen (mass shift of 14 Da) were observed. They were represented by the ions at m/z = 435, 437, 453, and 467 (Fig. 2A). Furthermore, another group of degradation products was also found which resulted from the cleavage of the Manoxol molecule. These degradation products were eluted approx. two times earlier than Manoxol and their ions were found at m/z = 291, 293, 309, 311, 323, 325, 339, 341, 355, and 357 (Fig. 2B). Similarly, as described above, these ions were also formed by the addition of the OH group with or without the elimination of hydrogen.

Fig. 2
figure 2

Mass spectra of A Manoxol and its oxidative degradation products, B cleavage oxidative degradation products of Manoxol

Full size image

Triton X-100 belongs to ethoxylated surfactants. Its characteristic feature is a long chain containing repeated ethoxy groups with a mass of 44 Da, which allow the compound to be soluble in water. The hydrophobic part of the molecule contains the benzene ring and a branched alkyl chain in the para position to the ethoxy chain. Triton X-100 was analyzed in the positive ionization mode in the form of ammonium adducts. At the beginning of the process, the molecules contained from 4 to 22 ethoxy groups and the corresponding ions were found at m/z = from 356 to 1148 (Fig. 3A). The degradation process resulted in the formation of two types of compounds resulting from oxidation of the parent molecules as well as those formed during the molecule scission with subsequent formation of polyethylene glycols (PEGs). The first group included aldehydes, which formed during oxidation of the ethoxy chain (at m/z ranging from 354 to 882, i.e., 4–16 ethoxy units), and alcohols formed by oxidation of the aliphatic chain (m/z 416–900 corresponding to 5–16 ethoxy units) (Fig. 3B). Among the compounds resulting from the molecular scission, two groups were detected–PEGs (at m/z = 256 to 872, i.e., these containing 5 to 19 ethoxy units) and dicarboxylated PEGs with two carboxyl groups in a molecule. These were found in the m/z range from 152 to 636 corresponding to 2 or 13 ethoxy groups (Fig. 3C).

Fig. 3
figure 3

Mass spectra of A Triton X-100, B aldehyde and alcohol degradation product of Triton X-100, C) PEG degradation products of Triton X-100

Full size image

Dodecyl alcohol ethoxylate contains numerous compounds as the molecule may contain not only different numbers of ethoxy units but also its hydrophobic chain may slightly vary in terms of length. The results obtained using LC–MS/MS showed that the tested surfactant is a mixture of ethoxylated dodecanol, tetradecanol, and hexadecanol containing from 3 to 22 ethoxy units, i.e., for dodecyl alcohol ethoxylates the ammonium adduct ions at m/z from 336 to 1172 were observed, for tetradecyl alcohol ethoxylates at m/z = 364 to 1068, and for hexadecyl alcohol ethoxylates at m/z = 480 to 1096 (Fig. 4A). During the degradation process, two pathways were observed in analogy to the test performed for Triton X-100, i.e., oxidation of the molecules and their scission. The oxidation resulted in the formation of carboxylic acids, e.g., for carboxylated dodecyl alcohol ethoxylates the ions at m/z from 350 to 834 were found which contained from 3 to 14 ethoxy units (Fig. 4B). Scission of the molecule resulted in the formation of PEGs at m/z ranging from 256 to 608 corresponding from 5 to 13 ethoxy units (Fig. 4C).

Fig. 4
figure 4

Mass spectra of A C12E10 (ethoxylated dodecanol, tetradecanol, and hexadecanol), B carboxylated degradation products of C12E10 (carboxylated dodecyl alcohol ethoxylates), C PEG degradation products of C12E10

Full size image

CTAB is visible in the LC–MS/MS analysis as the positive ions of m/z = 284 due to the loss of the bromide ion from its structure. Its degradation products were formed in the oxidation process of its one or two alkyl chains. A series of observed ions indicated the oxidation of one alkyl chain and formation of alcohol (m/z = 300), aldehyde (m/z = 298), and carboxylic acid (m/z = 314). It is followed by oxidation of a second alkyl chain to alcohol (m/z = 330), aldehyde (m/z = 328), and acid (m/z = 344). Due to the increasing polarity of the degradation products, a considerable shift in their retention time was also observed from 5.1 min (for CTAB) down to 1.1 min (for dicarboxylated CTAB).

DDAC was found at m/z = 362. Similarly, to CTAB, shifts in m/z values were observed which were characteristic for oxidation and formation of alcohols (mass shift of 16 Da), aldehydes (mass shift of 14 Da), and carboxylic acids (mass shift of 30 Da). Finally, ions of a dicarboxylic acid were detected at m/z = 386.

Benzalkonium chloride is a mixture of compounds differing in terms of the alkyl chain length. Two of its most abundant ions were found at m/z = 304 and 332. During the degradation of benzalkonium chloride, no mass shift to greater m/z values was found, which is characteristic for oxidation as described for other tested compounds. Instead, ions of lower mass to charge ratio were found at m/z = 200, 214 and 274 resulting from the decomposition of the original structure. However, compounds related to these ions were not identified.

Evaluation of phytotoxicity after photodegradation

This part of the experiments was conducted in order to investigate the actual impact of the photodegradation on the toxicity of the treated solution toward plants. The goal of the treatment is to decompose the compounds in a manner which would warrant the lowest possible impact on the environment; however, since the mechanism of photodegradation often involves radicals, it is difficult to predict the exact outcome. This approach is further justified by the limited amount of data in this field and the fact that purified graywater (domestic wastewater which usually includes surfactants) is often used for irrigation of crops. The results of phytotoxicity evaluation are presented in Fig. 5.

Fig. 5
figure 5

Germination index of sorghum seeds after 7 days of exposure to non-treated surfactant solutions (black bars) and solutions subjected to photodegradation for 60 min (white bars)

Full size image

The impact of the studied surfactants on sorghum seeds varied from moderate inhibition, through no effect, to stimulation. Overall, the ranges of germination index (GI) values were equal to 85–125% for anionic surfactants, 87–121% for non-ionic compounds, and 75–113% for the cationic group, respectively. There was no clear correlation in terms of the effect of photodegradation on the level of surfactant toxicity, nevertheless the treatment resulted in acceptable results as the value of the indices did not decrease below the threshold of 75%. An increase of the GI (which indicates a reduction of the toxic effect) was observed in 4 cases (SDS, Triton X-100, C12E10 and DDAC), an inhibition of seed germination was observed in 4 cases (SDBS, Manoxol, CTAB and benzalkonium chloride), while no change was observed for Rokopol. The last result corresponds well with the previous observation regarding the low efficiency of photodegradation in case of Rokopol. The treatment was the most beneficial in case of two non-ionic surfactants (Triton X-100, C12E10) and a single anionic compound (SDS) as the GI increased by approx. Thirty percentage points (the increase in case of DDAC) was insignificant. In contrast, it seems that the negative influence of photodegradation was more profound in case of cationic surfactants (CTAB and benzalkonium chloride) as the GI was decreased by approx. 20 percentage points (in case of SDBS and Manoxol the difference) was equal to 10 percentage points and was within the error margin of the method. It can be assumed that the enhancement of phytotoxic effects in case of CTAB and benzalkonium chloride was associated with the formation of more toxic degradation products, since the HPLC–MS analysis confirmed that both surfactants were transformed into different compounds during treatment. Identification of these structures and modification of the photodegradation process in order to ensure their elimination are a topic worthy of future consideration.

High concentrations of surfactants in wastewater subjected to WWTP may result in low treatment efficiency, which results in a notable amount of surfactant residues present at the end of the process. Interestingly, although the microorganisms present in the activated sludge will be exposed to toxic effects of surfactants during the treatment process, they are usually able to cope with such stress factors, due to constant contact with such contaminants, which allows them to tolerate relatively high doses. In extreme cases, the microbes can utilize several survival strategies in order to negate their impact. This includes, e.g., the formation of biofilms or endospores, which exhibit higher mechanical, chemical, and biological resistance. In contrast, if the surfactant-rich effluent is used to water agricultural fields, the plants will likely be more vulnerable since they are not usually exposed to such compounds on a daily basis.

Conclusions

The experimental data presented in this study confirm that the heterogenous photocatalysis can be effectively applied for the neutralization of all types of surfactants from the aqueous environment at low concentrations (up to 10 mg/L). However, in the framework of this study, the catalytic photodegradation of highly concentrated surfactant solutions has been performed for the first time. The obtained results confirm that ZnO-catalyzed photocatalytic degradation can be effectively applied for removal of 100 mg/L benzalkonium chloride with efficiency up to 94%.

Nevertheless, the obtained results confirm that the proposed described photocatalytic degradation demonstrates potential to be applied as a pretreatment procedure for all types of surfactant solutions at concentrations between 100 mg/L and 1000 mg/L prior to further wastewater treatment methods (such as biodegradation).

Among all tested surfactants similar observation was done, i.e., the degradation compounds formed in the photocatalytic oxidation included hydroxylated products. For some of them, this was only the first step of oxidation leading further to carboxy derivatives (e.g., for CTAB, DDAC, Triton X-100, and C12E10). Also, cleavage of surfactant molecules was observed for Triton X-100, C12E10, and Manoxol. In the case of Triton X-100, C12E10 this led to the formation of PEGs.

The impact of photodegradation on the phytotoxicity of the studied surfactants varied between moderate inhibition and stimulation of seed germination. In general, the best results in terms of reducing the toxicity were observed for non-ionic surfactants, followed by anionic compounds, whereas the negative impact of cationic surfactants was slightly amplified. This effect seems to be structure-related and associated with the formation of photodegradation products characterized by higher toxicity compared to the parent molecule.