Catalytic degradation of microplastics

The environmental problem associated with the spread of microplastics took shape only in the last five years, as a result of numerous thorough studies on the determination of these micro- and nanoparticles in various environmental objects.1 This became possible owing to advancement of the methods for chemical analysis of micro- and even nanoplastics,2 which revealed the presence of these pollutants in various environments, including soils3 and atmosphere.4 In any case, the provided data require careful interpretation in order to avoid false-positive results of chemical analysis.5

There is still no generally accepted term defining what is meant by microplastic: most often, this term refers to any plastic particles of <5 mm size, i.e., particles that are not retained on a metal grid with a step size of 4.76 mm (mesh 4) and a single cell diagonal of 6.7 mm. A more detailed and, hence, less contradictory classification of microplastics was proposed in the ISO/TR 21960:2020 standard and in the currently developed ISO/DIS 24187 standard, distinguishing nanoplastics (<1 μm), microplastics (1--1000 μm) and large microplastics (1--5 mm). In this review, plastic particles of up to 6 mm in size capable of passing through a metal grid with the above-indicated size are considered to be microplastics.

The formation of microplastic pollutants from items used in the household and in the everyday life has been noted throughout the world.6 The chemical analysis of samples taken from different water bodies demonstrated a broad variability of structural characteristics and chemical composition of microplastic particles.7 In addition, it is evident that the composition of pollutants can rapidly change under environmental conditions;8 therefore, it is reasonable to consider polymer products in the overall microplastic pollution. It is obvious that the component composition of microplastics generally coincides with the composition of abundant polymer sources, that is, natural, semisynthetic and synthetic polymers. The main domestic sources of microplastics are various fabrics, which may release particles during washing with detergents at elevated temperatures.9 Most part of fabric fibres are cellulose- (cotton, linen, rayon) and protein-containing (silk, wool) natural materials with some fraction of synthetic materials (elastane, polyamides, polyethylene terephthalates), which are not fully decomposed by waste water treatment.10

The main hazard of microplastics (unlike, for example, solid municipal waste) is that they are able to form relatively stable disperse systems in water. As a result, microplastic particles get into rivers and then to seas and oceans together with waste water; they are carried by air mass over large distances, fall on the ground with rain drops and are involved in trophic chains at different levels.11--13 As a result, microplastics are gradually accumulated in living organisms.14 Currently, microplastics have been found in many food products and in potable water. It was found that, on average, people throughout the world can intake up to 0.1--5 g of microplastics per week.15

The ecotoxicological effects of microplastic particles are mainly manifested as damage to internal organs and tissues (liver, intestines), adverse effects on fetal development and reproductive function, disorders of intestinal microbiome and metabolism, immune dysfunction and oxidative stress.16,17 Unfortunately, modern methods for waste water treatment (and even potable water treatment) are not meant for the removal of this type of pollutants; hence, the development of such approaches is highly relevant. Furthermore, chemisorption of other organic and inorganic pollutants on these microparticles18 leads to an adverse cumulative effect. Meanwhile, the extraction of microplastics from, for example, waste water, would mean solution to only a smaller part of the existing problem. After that, it would be necessary to process the microplastics. Combustion or landfilling of microplastics seems to be the simplest solution. However, after conventional combustion, most of microplastics are either discharged to the atmosphere together with effluent gases or retained in the residue, while in the case of landfilling, they may remain non-degraded in soil for long periods of time, despite the presence of various microbial associations.19 Moreover, there are studies in which the agricultural lands were irrigated with water containing particles of microplastics, hoping that they will be subsequently biodegraded, but this resulted only in soil accumulation of the microplastics,20,Ђ21 the particles of which were taken up by plants through the roots.22 Wild type (wt) microorganisms did not decompose these particles, but actually, like plants, accumulated them in their cells and on the surface in the attached state. It was noted that the accumulated particles have an adverse effect on the microorganisms (bacteria and fungi);23 the toxic effect on the biota is enhanced with decreasing particle size.

It was found that microplastics can change the biophysical properties and the bulk density of soil and the water retention behaviour of soils and can affect the formation of soil aggregates. It was shown24 that plastic particles of 1 to 5 mm in size brought into soil in an amount of 0.5 mass% create channels for water flow, resulting in enhanced water evaporation and thus lead to soil drying; this is unfavourable for crop yield.25 Changes in the soil structure have also other side effects, including modification of the composition of the microbial community (mycorrhiza and nitrogen fixers), which further affect plant growth and development. In addition, microplastics may contain substances toxic to plants. Studies of the mechanisms that underlie the plant uptake of microplastic particles using various analytical methods showed that these particles can penetrate into plant roots by changing (disturbing) the cell membrane structure, damaging the intracellular molecules and generating oxidative stress because of involvement of transport membrane proteins.22,23 For example microplastic particles considerably decreased the sprout (by 16--40%) and root (by 20--50%) biomass for corn, wheat, rice, onions, beans and many other plants.22,26

In this connection, it appears rational to perform the catalytic conversion of microplastics into products that would be, at least, not hazardous and, at best, useful, because microplastics are rather abundant organic resource, in some cases containing also nitrogen (polyurethanes).27--30

The catalytic conversion of macroplastics is addressed in numerous reviews (see, for example, Refs 31, 32); however, analysis of the influence of the micro- and nano-size of polymer particles on the efficiency of their catalytic degradation also seems to be relevant. To date, data representing state-of-the-art research in the (bio)catalytic transformation of microplastics have been analyzed in a number of publications,27--30 which, however, almost neglect the chemical aspects of the microplastic degradation. In order to fill this gap, here we analyze publications of the period from 2012 to 2022 considering polymer particles of up to 6 mm size converted to various products using (bio)catalysts; the attention is focused on the chemical aspects of the reactions. For a correct practical comparison of different catalysts in the presence of a sufficient amount of data, a common characteristic was calculated, that is, the performance (P) for each process in mg of products (or litres of gases) formed in 1 h under the action of 1 mg of the catalyst.

Microplastics are chemically indistinguishable from macroplastics; therefore, they can be degraded using catalysts with various chemical structure, cost, performance and stability, including catalysts developed for the destruction of plastics and waste from their production and use. In this review, the data on the catalysts are classified, according to the type of catalyzed reactions (Fig. 1, Table 1, Table 2, Table 3, Table 4, Table 5, Table 6), into thermal conversion, hydrogenolysis, silylation, electro-oxidation, photolysis and solvolysis (including alcoholysis, aminolysis, and hydrolysis).

Fig. 1

Examples of some catalytic processes used for degradation of microplastics. (a) thermal conversion, (b) hydrogenolysis, (c) silylation, (d) oxidation, (e) solvolysis. The thermal conversion usually gives a variety of products, which is shown as a hypothetical diagram. Percentage in the products. Number of carbon atoms

Table 1 part 1

\[ \]

Chemical catalysts for the thermocatalytic reactions of microplastics. All abbreviations and symbols are indicated in the Notes to the Table

(1)

Table 1 part 2

\[ \]

(7)

Table 1 part 3

\[ \]

(8)

Table 1 part 4

\[ \]

(9)

Table 1 part 5

\[ \]

(10)

Table 2

\[ \]

Chemical catalysts for the hydrogenolysis of microplastics

(2)

Table 2 part 2

\[ \]

(11)

Table 3

\[ \]

Chemical catalysts for the silylation of microplastics

(3)

Table 4

\[ \]

Chemical catalysts for electro- and photocatalytic reactions of microplastics

(4)

Table 4 part 2

\[ \]

(12)

Table 5

\[ \]

Chemical catalysts for the oxidative reactions of microplastics

(5)

Table 6

\[ \]

Chemical catalysts for the solvolysis of microplastics

(6)

Table 6 part 2

\[ \]

(13)

Table 6 part 3

\[ \]

(14)

Table 6 part 4

\[ \]

(15)

Table 6 part 5

\[ \]

(16)

Table 6 part 6

\[ \]

(17)

Pyrolytic processes are carried out in the presence of various zeolites — the most accessible and abundant catalysts used, in particular, to upgrade oil residues.150 The conversion can be carried out not only for single polymers, but also for polymer blends with accompanying biomass, which, in principle, can be isolated, for example, from wastewater, together with microplastics. The combined pyrolysis of polymers and biomass results, as expected, in a high content of organic acids in the products,34,35 which is undesirable if they are meant for the use as fuels. However, the relative content of oxygenates and nitrogen-containing compounds in the products of the catalytic co-pyrolysis can be significantly reduced compared to those in the product of the non-catalyzed process.35 The halogenated polymers present in the blend would be converted into halo derivatives, which may, in turn, partly bind to bases that are additionally introduced into the catalyst [30% binding for Ca(OH)₂ (Ref. 36) and 20--25% for Al₂O₃ (Ref. 40)].

Regarding the composition of the product, important characteristics are the Si:Al ratio,39 zeolite content in the catalyst,47 zeolite type 40,41 and porosity 41 and doping with metals.37,45 It is to note important that repeated thermal conversion using metal-doped catalysts may lead to additional deposition of some metals on the catalyst,45 resulting in variation of the product composition from one cycle to another.

Comparison of the catalyzed and non-catalyzed pyrolysis of microplastics gives ambiguous results. A number of publications attest to a decrease in the pyrolysis duration or temperature and/or a change in the product composition upon addition of a catalyst.36--38,41--43,45,47--51,53,54,56,59 The authors of other studies either did not observe such differences or noted some deterioration of the process characteristics.40,44

It is of interest that upon zeolite conversion into a nano-sized form, which is accompanied by a pronounced increase in the contact surface area, the product yield increases 3.2-fold and the required process temperature decreases by 40 °C (i.e., by ≈10%).53

An alternative way for supplying the thermal energy to a microplastic subjected to pyrolysis, that is, microwave (MW) heating of the reaction medium should be mentioned.54,59--61 Apart from pyrolytic processes, this design was implemented in other catalytic processes (see below).

The composition of not only zeolites, but also other mixed (composite) catalysts markedly affects the process performance and the yield of the final products: even a minor deviation leads to a decrease in both parameters.59 Moreover, impurities of other polymers in the major polymer (e.g., PS impurity in PE or PP)59 can also have adverse effect on these characteristics of the process.

In most studies, the stages of polymer pyrolysis and subsequent catalytic conversion of the formed intermediate products are separated in time and space. Such studies are not addressed in this review, although they have their own benefits, but also drawbacks. A comparison of the two-stage process with one-stage catalytic pyrolysis showed an advantage of the latter.34 In the development of the concept of two-stage process, it was proposed to use ­functionally different catalysts in the two stages: Ni/Al–SBA-15 for the primary gasification and generation of carbon nanotubes (CNTs) and Ni-Cu/CaO–SiO₂ for the conversion of effluent gases to H2.58

A study by Jung et al.,57 in which microplastic samples were isolated directly from soil, deserves special attention. For this purpose, polymer particles were salted-out with a concentrated NaCl brine, which was followed by mineralization of labile (bio)organic compounds using 30% H₂O₂. According to an original publication,151 the recovery of microplastics was, on average, 30--60% and was most efficient for HDPE, LDPE, PP and PS (without changes in the chemical composition). Despite the fact that the performance of the used catalyst was rather low, transition from the model microplastics to real samples should be highly encouraged and implemented in other works.

Upon pyrolysis, polymers such as PE, PP, PS, PVC, PET, PLA are converted to hydrocarbons and CO₂/CO or, more rarely, to H₂ and CNTs. Cai et al.62 combined the PP decomposition with the simultaneous preparation of the catalyst for the subsequent electrolysis of water.

The steam hydrolysis of PBAC to give bisphenol A and other phenolic compounds can be considered separately.63,64 The use of ionic liquid64 made it possible, on the one hand, to considerably decrease the reaction temperature and, on the other hand, to increase the product yield as a result of dissolution of the polymer and, as a consequence, homogeneous catalysis. High temperatures63 caused partial decomposition of bisphenol A to phenol and p-isopropenylphenol. During pyrolysis of PBAC, destruction gave only phenol and polyphenol compounds, which formed a solid residue.

Nabgan et al.65,66 carried out steam reforming of microplastics together with phenol. With this experimental design, it was impossible to calculate the yields of H₂ separately from phenol and from microplastics.

To summarize the discussion of thermocatalytic reactions for the degradation of microplastics, it should be noted that this process consumes a lot of energy and requires heating, on average, up to 500 °C and higher (Fig. 2). Most often, researchers try to minimize the energy consumption by reducing the process time and/or temperature. With few exceptions, for example, when specific polymers are used,64 the temperature cannot be markedly reduced, because this would decrease the performance and the yield of the reaction products.55 This results in the loss of the main benefit of the pyrolytic conversion, that is, versatile degradability irrespective of the composition of applicable microplastics. A possible solution to reduce the energy consumption is to combine catalysts with different types of action, e.g., ionic liquids and nano-sized zeolites, in the same catalytic system.

Fig. 2

Statistical treatment of the results of application of various catalytic systems. (a) yield of products, (b) performance, (c) temperature. The interquartile ranges (25% - 75%) are enclosed by rectangles, which are divided by a line corresponding to the median value. The averages over all values and particular values are marked by square and round dots, respectively; the statistical outlier areas are indicated (Tukey's value is 1.5). The Figure was created by the authors using published data presented in Tables 1-6.

Unlike pyrolysis, hydrogenolysis of microplastics is more often catalyzed by organic metal complexes structurally similar to the complexes used in the polymerization152 and by platinum group metals. Zeolites, including those doped with noble metals (Pt, Pd, Ru), can also be successfully used, but at higher temperatures (they demonstrate higher performance than other types of catalysts, see Table 2).

In the case of nano-sized catalysts, the metal nature is of primary importance; most often, the Ru-based catalysts are more active than samples based on Ni, Co, Pt,71 Cu, Fe, Ni, Pt, Pd, Rh,72 Ir, Rh, Pt, Pd, Cu, Co and Ni.74 One more important aspect in the hydrogenolysis of microplastics is the type of the support; the lowest activity was observed for Al₂O₃, TiO₂, MgO, ZrO₂ and SiO₂.71,74 It is of interest that in the reaction medium without a solvent, a CeO₂-supported catalyst was more active than a carbon-supported one.74

In the studies of organic metal complexes as catalysts, Ru complexes were used most often,76--78 and the results were comparable with those obtained for Ir and Mn complexes.75 However, the use of ruthenium compounds required lower temperatures (140--180 °C) for reactions to proceed. The structure of the organic ligand, the presence of dopants and the solvent are important not only for maximizing the activity, but also for the mere possibility of the catalyzed reaction.77

The yield of the target products and the performance of processes substantially depended on the polymer type,69,74,76--79 molecular weight (MW),70,74 the presence of additional cross-links75 and impurities of other polymers.73,81 The adverse effect of some factors can be partially counterbalanced by increasing the reaction time and/or temperature (for example, the effect of polymer impurities have been avoided78,80).

Interestingly, the Cu₄Fe₁Cr₁ catalyst can successfully perform methanolysis when the gas mixture (CO₂–H₂–Ar) is replaced by methanol.79 This catalyst can be considered as versatile, although its performance is still lower compared to more specific catalysts.

Wu et al.81 are among the few researchers who deve­loped a catalyst based on the UiO-66 metal-organic framework (MOF) containing zirconium. Currently, these materials attract increasing attention for a number of reasons.153 Anticipating a little bit, we would like to note that this catalyst is not the only one MOF used for the degradation of microplastics; some other compounds of this type are considered below. During the microplastic degradation, UiO-66 undergoes an intriguing transformation into MIL-140A, which is also a Zr-containing MOF with a somewhat lower activity compared to that of UiO-66. Furthermore, repeated use of the same catalyst resulted in lower yields of the products. It is noteworthy that Zr or ZrO₂ are not catalytically active, and the reaction in the presence of these compounds does not differ from the reaction in the absence of a catalyst. The catalytic activity is determined particularly by the supramolecular structure of MOF, which facilitates β-scission of the C–C bond in the ethylenediol substituent.81 The MOF potential for the degradation of microplastics is very high, since MOFs can be designed using computer modelling methods before their actual synthesis.

Unlike thermal conversion catalysts, computer modelling was applied to some catalysts for hydrogenolysis. Most often, quantum chemical calculations of the surface potentials for an infinite plane have been carried out.69,70 More rarely, the molecular dynamics was studied, in particular the conformational changes in the C₂₀ polymer chain in various solvents.72 Density functional theory (DFT) calculations for the interaction of the catalyst with the low-molecular-weight analogues of substrates80 or with solvents77 provide much more information, although they are less relevant to degradation of polymer substrates.

Usually silylation of various organic and inorganic compounds is reduced to modification of some chemical groups with silicon-containing substituents. However, appropriate choice of catalysts and reaction conditions may result in the destruction of polymers that form macro- and microplastics (see Table 3). Strong Lewis acids serve as catalysts; as the acidity increases, the reaction temperature decreases (down to room temperature, see Fig. 2). It was shown that the silylating agent affects the nature of the final product: reactions with Et₃SiH, (Me₂SiH)₂NH and PhSiH₃ give silylation adducts,82--84 which can be converted to alcohols by alkaline hydrolysis82,83 or even to hydrocarbons (by increasing the reaction time and/or temperature).82 When (Me₂SiH)₂O is used, such hydrocarbons can be obtained in one step.84

It is important to note that neither the catalyst nor the silylating agent alone is able to degrade a polymer under conditions of this type.83 Currently, this accounts for some difficulties in the selection of components for this catalytic system.

Perhaps the only disadvantage of silylation-induced degradation of microplastics is the limited range of polymers that can be successfully degraded in this way; moreover, many of such polymers are polyesters, which are often considered to be biodegradable. Much more common aliphatic polymers such as PE, PP, PVC and, for example, PS cannot bet degraded by this method.

All catalysts of the destruction of microplastics exploit a common mechanism consisting in the artificial ageing of polymers induced by reactive oxygen species. These species can originate either from O₂ or water subjected to electrolysis or photolysis, or from chemically decomposed H₂O₂, or from various combinations of the above sources.

The degradation of microplastics is often impossible without preliminary modification (e.g., by hydrolysis or sulfonation) of the polymer.86,92--95 In most of other cases, the yield of products and the process performances are very low, being markedly inferior to analogous parameters of the processes discussed above. The best characteristics, comparable with the average characteristics obtained using thermal conversion, hydrogenolysis and silylation, were attained91,97 using strong acid catalysts. In both of these studies, various acids were tested and, along with the reaction conditions, the nature of the acid affected the composition of the reaction products.

Studies of so-called nanoplastics, that is, submicrometre-size microplastics, should also be noted.89,90 Despite some drawbacks (in particular, the sorption of nanoplastics on porous photocatalysts was not studied), these works deserve attention because they clearly demonstrate that turbidimetry is inapplicable for determining the catalyst activity, unlike, for example, the reliable method based on determination of the content of elemental carbon. Moreover, chromatography, NMR and other quantitative techniques are even more trustworthy for identification of the products of degradation of microplastics. Therefore, in this review, the priority is given to the studies that use particularly these methods for analysis of the results; otherwise, the method of determination of the products is specially noted. Meanwhile, the frequently used determination of the loss of mass of the initial polymer should be mentioned: according to Ouyang et al.96 and some other authors, treatment of microplastics may lead to reduction of the particle size without the formation of degradation products. Therefore, the results of determining the process efficiency obtained by this method were not considered in this review.

In some studies dealing with photocatalysts, quantum chemical calculations are performed to examine the interaction of the hypothesized intermediates ([COOH], [CO], [H]) with the catalyst surface during the electroreduction of CO₂.87,88 However, the authors did not calculate any physicochemical parameters that could be compared with the characteristics of other catalysts.

Huang et al.91 carried out DFT calculations for 1,3-­Ph₂Bu used as a model of PS. The energy barriers for pathways of the reactions of 1,3-Ph₂Bu with various reactive oxygen species were determined, but in the absence of catalysts; therefore, the results obtained in this work are of limited utility for our subject matter.

Solvolysis results in decomposition of mainly polyesters and, much more rarely, polycarbonates. However, the number of catalysts developed for this process markedly exceeds the number of thermal conversion catalysts (see Table 1 and Table 6). The types of this method can be classified into methanolysis, ethanolysis, glycolysis (i.e., reactions with ethylene glycol and other diols), aminolysis, and hydrolysis.

A comparison of catalysts containing different metals showed the highest efficiency for Zn²⁺-containing catalysts (see101,110,113,115,118,122,127). In rare cases, the activity of Co²⁺ catalysts was comparable with that of Zn²⁺-based samples.117 When Mn²⁺, Cu²⁺, Ni²⁺, Fe³⁺ or Cr³⁺ was used, either the yields of products considerably decreased or no catalytic activity was detected. The anion present in the catalyst also played a role; for example, AcO–containing catalysts were more active than the catalysts containing Cl^–, Br^–, CF₃SO^3-, CH₂=C(Me)COO^– or NO^3- anions.99,110,148 Similarly, in the case of ionic liquids, the catalysts containing acetate anions showed the highest activity100,121 {in combination with the [1-Bu-3-MeIm]+ cation, they provided yields comparable with those obtained with Zn(OAc)₂}.100 Some authors made attempts to combine different catalysts, e.g., ionic liquids with ZnCl₂. However, it must be admitted that the performance of such processes remained moderate or even decreased.122 A better performance can be attained by alternative methods, particularly, by increasing the polymer solubility by adding one more solvent103,110 or other additives,144 by increasing the temperature,102,104 by conducting the reaction in a supercritical fluid 147 and using ultrasonic treatment.144

It is noteworthy that various types of reactions can be carried out with the same catalyst, e.g., methanolysis,100 glycolysis110 and hydrolysis.148 Of course, the reaction conditions should be optimized for each particular case to maximize the product yields. However, the reactivity of glycols significantly decreases with increasing molecular weight,112 while some alcohols, in particular BuiOH, cannot react;104 therefore, the set of solvents that are able to provide satisfactory yields of products upon solvolysis is limited.

Organic zinc complexes can be used to carry out destruction of microplastics at the lowest temperatures (40--50 °C) with retention of high yields and a satisfactory performance of the process;102,103 however, the presence of an additional solvent is necessary.

In the case of nanocatalysts, the highest activity is observed for other metals, for example, Co or Fe.105,129 The catalyst activity increased by a large factor105 or even appeared particularly owing to the use of nano-sized catalyst forms.128 A decrease in the nanoparticle size128 or doping with reactive metals131,134 resulted in higher pro­duct yields. In view of the above, primary attention should be paid to trace impurities present in the catalyst, e.g., noble metals, the presence of which was neglected in some studies131,132 addressing the degradation of PET.

Combination of various nanocatalysts is a more efficient approach for improving the performance of the overall process than the use of composites with ionic liquids.125,130,133 Special mention should be made of the composite catalyst containing Zn-based MOF,135 which markedly improved the characteristics of catalysis. Among the studied MOFs, the highest surface area was inherent in ZIF-8; the same sample was most efficient in comparison with ZIF-67 and MOF-5.136 A similar beneficial effect of increasing surface area was also found for other catalysts.125,126 Yang et al.137 found a higher activity for MAF-6 over MAF-5 or MAF-32 and also over Zn(OAc)₂ in the glycolysis of PET.

It is of interest that the particle size of microplastic may have some influence on the product yields. According to various studies, as the size increases, the yield increases,113,114,117,143 decreases111,137 or remains virtually unchanged.147 A decrease in the yield with increasing microplastic particle size was also detected in the electrophotocatalytic process.90

A combination of solvolysis with photothermolysis138 in which CNTs present in the composite catalyst converted the incident light energy into heat is worth noting; hence, no additional heating of the reaction mixture is required. When the reaction was conducted in this regime, the product yield increased 1.5-fold in comparison with the conventional solvolysis.

Density functional theory calculations for the interaction of 3-tropanol (one of the catalyst components) with ethylene glycol were carried out;113 however, these results do not describe the catalytic glycolysis of PET as a whole. According to DFT calculations for the interactions of the product with various solvents,110 the complex with DMSO had the highest energy, which was partially correlated with experimental data for PET solubility. The DFT calculations for the complex of the ionic liquid cation with the product100 are also of limited interest.

The energy barrier (the activation energy E_a) for dissociation of the components of the TBD:MSA composite catalyst was calculated by quantum chemistry methods; the result of 156.5 kJ mol^–1 served as the basis for theoretical substantiation of the catalyst thermal stability.106 More recently,98 the dissociation energy of this complex in EG was determined in the same way; it was found to be 3.3 times lower than that calculated earlier.106 The ionization energy in the gas phase (413 kJ mol^–1) was several-fold higher than the initially calculated value. Thus, the original hypothesis about the cause for the catalyst thermal stability was, if not completely rejected, then at least, found to be of low significance for real conditions. Unfortunately, the subsequent calculations of the energy barriers along the reaction pathway were performed for low-molecular-weight models of substrates: BzMe for PET and (PhO)₂CO for PBAC. As applied to microplastics, these results provide only a qualitative estimate (2 and 4 barriers for PET and PBAC, respectively, without considering the initial ionization and the interaction of the anion with EG). A similar situation exists for Me₂TPA140 and BzMe.141

According to DFT calculations,108 the formation energy of [Me₃N(CH₂)₂OH]⁺ complexes with two EG–TPA monomer units varied in the 400--450 kJ mol^–1 range depending on the used anion. However, the yield of the reaction products did not correlate with the calculated values for various anions. Nevertheless, all of the models were found to have a hydrogen bond between the OH group of the cation and the oxygen atom of the TPA carboxyl group.

Relying on DFT calculations for the complex of a variable phenol-containing moiety of the DBN:[4-MePh] catalyst with EG, Wang et al.107 noted that the calculated energy of the O-H bond within EG was somewhat correlated with the product yield (the lower the energy, the higher the yield). However, these results should be considered to be tentative, since they are clearly incomplete, as there are also other energy barriers along the reaction pathway.

Looking again at the issue of combining different methods for the degradation of microplastics, it should be emphasized that the process can be conducted in two stages, e.g., using solvolysis and the subsequent electrochemical conversion of the obtained products.145,146

The results of two successful studies,148,149 in which solvolysis (in the first stage) was combined with enzymatic hydrolysis (in the second stage) are also noteworthy. Since the BsEst-^N,CHis₆ enzyme is able to directly hydrolyze PET and HET, the yield of TPA exceeded 100% (based on BHET determined after the first stage) as a result of hydrolysis of the residual amount of the initial polymer, oligomers and HET. In addition, the enzyme was not inhibited by the final product (up to 20 mmol L^–1) but was inhibited by intermediates (HET and BHET), especially when the enzymatic reaction was carried out in the TRIS buffer at the same pH.

The possibility of using biocatalysts (Fig. 3) in the biodegradation processes of macro-154,155 and microplastics has long been discussed, and considerable advances have been made in this area. Meanwhile, biocatalysts are still much inferior in the performance to high-temperature chemical catalysts, but they are comparable or even superior to the low-temperature catalysts (Table 18). It is noteworthy that not all polymers that are claimed to be biodegradable are susceptible to microbial degradation under environmental conditions.241 This may be caused by a variety of factors starting with the reduced bioavailability of the substrate for degradation and ending with the presence of various chemical agents (modifiers, dyes) that are additionally introduced into the polymer and are toxic particularly to biological objects.242

Fig. 3

Structures of enzymes: IsPETase (a), TfCut (b), ThcCut2 (c), LCC-ICCG (d), RgPETase (e), HiCut (f), PHL7 (g), CaLipB (h) and AoCut (i); the crystallographic data were taken from the RSCB PDB: 6EQD, 5ZOA, 5LUJ, 6THT, 7DZT, 4OYY, 7NEI, 6TP8 and 3GBS, respectively, and illustrated using PyMOL (version 1.7.6, Schrodinger, LLC). The catalytic triads (Ser His Asp) in the active sites are highlighted with a colour and marked by an arrow.

Table 7

\[ \]

Enzyme biocatalysts for the degradation of microplastics

(18)

Table 7 part 2

\[ \]

(19)

Table 7 part 3

\[ \]

(20)

Table 7 part 4

\[ \]

(21)

Table 7 part 5

\[ \]

(22)

Table 7 part 6

\[ \]

(23)

Table 7 part 7

\[ \]

(24)

Table 7 part 8

\[ \]

(25)

Table 7 part 9

\[ \]

(26)

Table 7 part 10

\[ \]

(27)

Table 7 part 11

\[ \]

(28)

Table 7 part 12

\[ \]

(29)

Table 7 part 13

\[ \]

(30)

The same inactivation problems may arise in the enzymatic catalysis; therefore, a number of methodological solutions that can somewhat level off the possible deterioration of the process efficiency have been proposed.

First, the bioavailability of the polymer substrate for the biocatalytic transformation can be improved by adding organic solvents,156,162,166,173,226,230 detergent emulsifierss158,183,187,194,195,221,234,238 or hydrophobic binding proteins (i.e., proteins that have high affinity to hydrophobic surfaces)178--180 into the reaction medium or by incorporating the enzyme into the polymer matrix directly during its formation.218 As a result, the reaction rate may increase up to 129-fold158 (however, in practice, the typical improvement of biocatalysis is much more modest). Also, the additional component introduced into the reaction medium can, in some cases, lead to inactivation of the enzyme;83,187,234 therefore, a trade-off adjustment of suitable conditions is necessary.

Second, the biocatalytic reaction can be combined, for example, with ultrasonic treatment,204 or the starting polymer substrate can be additionally pretreated before reaction with microwave163 or conventional heating.215 However, the enzyme efficiency is not always improved upon this type of pretreatment.172 For example, Kaabel et al.208 combined enzymatic hydrolysis with simultaneous mechanochemical treatment of microplastics in a ball mill. However, to attain 49% yield of products, repeated addition of fresh portions of the enzyme was required, since the enzyme was inactivated in the reaction medium.

Third, it is possible to improve binding of the biocatalyst to the polymer substrate by modification of the enzyme itself, e.g., by conjugation176 or insertion of an additional high-affinity amino acid sequence,175,177,193,203 which promotes better enzyme--substrate binding.

Enzyme binding to the substrate requires special attention. Unlike the chemical catalysts considered above, a soluble biocatalyst should first interact with the insoluble substrate for the subsequent efficient biocatalysis. Therefore, it is impossible to transfer the biocatalytic process into a single phase, thus switching to homogeneous catalysis. The adsorption of various enzymes on diverse polymer substrates follows different kinetics; for example, the adsorption of IsPETase enzyme on PET takes 2 h.158 In some studies, adsorption isotherms for various enzymes on PET were measured: the maximum PET capacity for HiCut enzyme, which has the minimum size (1.5--1.6 times smaller than IsPETase or TfCut), was 1.7 times higher, but the calculated dissociation constant (Kd) was worst (3--5 times higher).168 The PET binding capacity for TfCut and LCC enzymes with approximately equal weights differed by a factor of two (0.5 and >1 mg g–1, respectively) at 40 °C, and TfCut was adsorbed on PET faster than LCC.187 ThcCut1 and ThcCut2 enzymes, which are even more similar in the structure and properties, differed fundamentally in Kd (by a factor of 4 at 60 °C), and differed in the binding capacity to PET (0.25 and 0.28 μg g^–1, respectively), but the differences were not statistically significant.200 The increase in both characteristics with decreasing temperature193,200 attests to a complex character of the enzyme interaction with the solid surface of a substrate, comprising contributions of both electrostatic (including hydrogen bonding) and hydrophobic interactions.

Of certain interest are studies of the adsorption of biocatalysts with fluorescence-labelled enzymes on a polymer substrate219,228 and their direct determination using a quartz microbalance.202,203 By introducing point mutations,219 it was possible to attain faster binding of lipase to PET surface and slower dissociation of the enzyme--substrate complex. The bimodal dissociation attested to the following processes taking place simultaneously:

- true dissociation followed by the enzyme migration to the bulk of the solvent;

- change in the position and/or conformation of the enzyme on the surface of substrate particles.

From this standpoint, the insertion of a high-affinity sequence into the protein molecule not only accelerates the formation of the enzyme--substrate complex, but also decreases the enzyme desorption from the substrate surface, which was established by studying competitive binding of native and modified cutinase ThcCut1.203 Upon enzyme modification by adding polyhydroxybutyrate-binding module (PBM) as a fusion partner, the adsorption capacity of the polymer increased by only 15%;203 however, this value can be increased by changing the high-affnity module, which is genetically introduced into the enzyme molecule; for example, the use of the amino acid sequence of the chitin binding domain (ChBD) led to a threefold increase in the adsorption capacity.193 It should be noted that the effect of introducing a particular high-affinity sequence into the molecules of protein catalysts that degrade microplastics is difficult to predict. The introduction of hydrophobin HFB4 or PBM not only did not give useful result, but, conversely, it rather impaired the parameter in question, especially at elevated temperatures, whereas the cellulose-binding domain (CBM) attached to the N-terminus of IsPETase did not give any result.175

A large number of studies have been devoted to the genetic modification of enzymes with the goal to improve their catalytic activity in the microplastic transformation reactions158,170,183,186,205,225,226,237 and/or their stability,166,167,169,173,174,177,191,194,195,219,221,230 to increase the protein yields in the biosynthesis160,161,179,227,236--238 and/or to facilitate the isolation and purification of proteins for obtaining the most efficient biocatalysts. Currently, modifications facilitating the isolation and purification of the target enzymes are applied almost in all studies.

Usually, the defined goals can be achieved by genetic modification, although there are some exceptions.221,237 In any case, discussion of the effect of particular mutations is beyond the scope of this review, and to get acquainted with this topic, we recommend specialized reviews (e.g., Refs 243, 244).

The enhancement of stability, in particular thermal stability, of the biocatalysts is aimed at increasing the process performance, which is easily achieved by increasing the reaction temperature, i.e., enzymatic biocatalysts become comparable in temperature characteristics (Fig. 4) with the low-temperature chemical catalysts considered above (see Fig. 2). However, an advantage of enzymes, that is, moderate temperatures of biocatalysis, is thus lost.

Fig. 4

Statistical treatment of the results of application of various biocatalytic systems for the degradation of microplastics. (a) performance, (b) temperature, (c) pH of the reaction. The interquartile ranges (25% - 75%) are enclosed by rectangles, which are divided by a line corresponding to the median value. The averages over all values and particular values are marked by square and round dots, respectively; the statistical outlier areas are indicated (Tukey's value is 1.5). The Figure was created by the authors using published data presented in Table 7.

Since a number of enzymes, for example, TfCut2, SvCut190 and LCC, were identified and isolated from thermophilic microorganisms, they have high thermal stability. The information on the structure of these proteins served as the basis for comparative analysis and introduction of necessary mutations into the structures of similar mesophilic analogues possessing no thermal stability. In addition, some enzymes, e.g., SvCut190,194 MtCut 230 and PpEst,232 can be additionally stabilized by the Ca2+ ions present in the system. However, the metal binding site is usually modified by introducing instead a disulfide bridge, which increases the thermal stability of enzymes to even a higher extent. The following important point should be mentioned: the biosynthesis of target enzymes in thermotolerant yeast is often accompanied by their glycosylation at the Asn residues. This may not only result in thermal stabilization,220 but also increase the aggregative stability and resistance to chaotropic agents;188 in some cases, this markedly decreases the catalytic activity.202,225

It is often difficult to make a direct comparison of different enzymes and draw objective conclusions within one study, because enzymes are characterized by different optimal conditions of biocatalysis, and the use of some conditions is a priori unfavourable for some enzymes.158,165,222,227,231,235,239 Comparison of the results obtained in different works is even more difficult due to variability of conditions and, first of all, the used substrates (see Fig. 4). First, like in the case of chemical catalysts, upon a decrease in the microplastic particle size, the enzymatic activity may either increase190,192,201,209--211,214 or decrease.207,212 Second, the substrate bioavailability starts to play an enormous role; as the crystallinity of the substrate increases, the enzyme activity decreases several-fold159,176,177,189,191,193,196,201,204,208,210,219,224,236 or even several hundred-fold,172,183,214 down to complete loss of the catalytic activity.226 Apparently, in the latter case, an important role may be played by the form of the polymer material; for example, micrometre-thick films are often hydrolyzed less efficiently than separate microparticles.230,235

In this aspect, it is worth noting the study by Brizendine et al.,189 in which, in the presence of 0.3% LCC-ICCG enzyme, the microplastic particle size did not affect the reaction rate, but the crystallinity of PET had a pronounced effect (the yield was 3--4 times lower when the degree of PET crystallinity increased from 8--11 to 33--36%). However, as the amount of the enzyme decreased 33-fold, the particle size of microplastics in a similar reaction started to be significant, so that smaller particles were hydrolyzed faster. This was accompanied by a change in the composition of products towards the formation of HET. In the study of Eugenio et al.,211 the Michaelis constants (KM) for different-size PET particles were comparable, while the catalytic constant (k_cat) was 2.4 times higher for smaller particles.

An interesting observation was made by Chang et al.,190 who found a direct correlation between the lag period preceding the hydrolysis of the substrate and the particle size of the microplastic. As a result, the performance was 2.8 lower in the case of larger particles. However, grinding was accompanied by a twofold increase in the crystallinity (from 6.5 to 12.6%); therefore, as previously, hydrolysis of large amorphous particles proceeded to a larger extent.

Due to the limited information on the size and crystallinity of the used microplastics reported in different studies, the contradictory data on the influence of the glass transition temperature (T_g), melting point (T_m) and the number-average (M_n) and weight-average (M_w) molecular weights of the polymer on the enzymatic activity are also difficult to interpret: in some cases, the biocatalyst activity was positively correlated with one or several parameters,197 while in other cases, there was no such correlation;198 in some studies, the correlation was observed only over a narrow range of parameters, but it was absent in a broad range.199 The chemical structure of the polymer has a much greater effect on the enzyme activity than the variation of the physicochemical characteristics of the polymer material.198,199 It is noteworthy that ageing of polymer materials under environmental conditions can also result in a considerable (1.6--2.6-fold)207 decrease in the efficiency of enzyme action. Apparently, this is due to structural changes in the polymer substrate caused by its modification with functional groups that prevent normal catalysis, adsorption of compounds that inhibit the enzyme and changes in the rheological properties, including (bio)degradation of the bioavailable parts of microplastics.

A number of studies present quite successful attempts to improve the enzymatic activity of biocatalysts towards crystalline polymer substrates.193,226 The number of such studies is small, since the main efforts were directed to other goals, for example, to increasing the thermal stability of enzymes. Indeed, considering the catalytic characteristics of IsPETase, the KM value proved to be much lower (that is, better) in the case of crystalline than amorphous PET for temperatures of 30 and 40 °C.166 In addition, the constant kcat depended on both the temperature and crystallinity of the polymer, with the highest kcat value being observed at elevated temperature with crystalline PET. The highest efficiency of enzyme action (kcat/KM) was observed under the same conditions. After two amino acid mutations, W159H/S238F, in the protein the thermal stability of the enzyme increased, but this had an adverse effect on both catalytic characteristics of the enzyme for both amorphous and crystalline PET at the two studied temperatures.

The poor applicability of optical, in particular, turbidimetric methods of product determination in the studies of biocatalytic decomposition of microplastics should be noted again. These methods give unreliable results for mixtures of products.173 Therefore, the data obtained by high-performance liquid chromatography (HPLC)216 and turbidimetry217 for one and the same enzyme under the same conditions may differ by large factors, being 10--20 time overestimated for turbidimetry. Similar discrepancies are also observed when one compares the results of gravimetric analysis of the initial substrate (or determination of polymer Mn) with the product yields measured by HPLC.212 There are strong reasons to believe that determination by titration also gives overestimated results, since oligomeric products are formed.

An important problem is the inhibition of enzymes by inorganic and organic compounds present in the reaction medium with microplastics. Most of the enzymes that catalyze the destruction of microplastics are hydrolases containing the serine amino acid residue in the active site and are usually not inhibited by chelating agents.194,222 However, these enzymes can contain additional structuring metal ions, which can be bound to chelating agents or replaced by other metals, thus leading to detectable decrease in the biocatalytic activity.206,230,236 The intermediates of PET hydrolysis (HET and BHET), which can be formed during the enzymatic reaction, are often also substrates of these enzymes; hence, they would function as competitive inhibitors, e.g., for TfCut2,181,182 MbPles629227 and PpEst.232 However, some enzymes such as SvCut190 and Novozym® 51032 are quite tolerant to the presence of both intermediates over a broad range of their concentrations;194,211 however, this feature may be no longer relevant at high degrees of substrate conversion.208 A genetic modification of these enzymes directed towards the change in their catalytic characteristics with the intermediate compounds may decrease the inhibitory activity of these compounds towards PET hydrolysis.182 An alternative method is to introduce combinations of several enzymes into the biocatalytic process162,210,213,214,240 and/or to perform a two-stage treatment of the polymer substrate with different enzymes.170 In the former case, there is requirement that all of the enzymes function under the same reaction conditions. In the latter case, identical conditions are not required, although finally this design would be inferior in the performance due to increase in the process duration.

Computer simulation methods are widely applied to enzyme reactions. First, the appearance and upgrading of extensive databases of nucleotide and amino acid sequences for a variety of (micro)organisms fundamentally changed the strategy of the search for new enzyme biocatalysts for the degradation of microplastics. Whereas earlier it was necessary to search for the potential dectructors of microplastics and isolate them from the environmental objects, today the bioinformatic screening is performed in the automatic mode: it is sufficient to know the amino acid sequence of another enzyme with similar function (see Fig. 3). Moreover, some researchers have already prepared specialized databases on esterases that hydrolyze PET and polyurethanes.245 Thus, before direct biocatalytic experiments with microplastics, it is possible to select, for example, enzymes characterized by optimal operation at low temperature,222 or, conversely, thermally stable enzymes,223 or those carrying a definite set of mutations.225 Therefore, diverse sources in the form of non-culturable microorganisms, which are virtually impossible to isolate as pure cultures, do not even need to be isolated, but they remain at the level of identified metagenomes.219,223,224,228,231

Second, if the amino acid sequence is available, it is possible to predict the structure and even the properties of enzymes as biocatalysts for the degradation of polymers.180,182,219,220,223,224,228,230,232 Third, using the predicted structure, at this stage, it is already possible to plan, for example, the introduction of point mutations to modulate the catalytic activity of enzymes185,219 or, what is more important, to model the interaction with the substrate, in particular, by molecular docking techniques. Docking can be carried out using conformationally rigid171,184 or flexible163,165,167 enzyme molecule; usually the procedure involves up to 4--5 monomer units of PET (or up to 7 monomers, like in the case of PLA205). Too short HET and BHET molecules may be considered to be inapplicable to reactions with microplastics. Docking offers more opportunities for the introduction of genetic modifications into the enzymes. Fourth, the resulting enzyme--substrate binding models can be used to study the mechanism of enzyme action by molecular dynamics.156,163,171,177,191,230 In exceptional cases, studies achieve the state of combined QM/­MM calculations.162,184 The results of determining the energy barriers for some enzymes are summarized in Table 31.

Table 8

\[ \]

Activation energies (E_a) of various reactions in the presence of (bio)catalysts

(31)

Table 8 part 2

\[ \]

(32)

Certainly, there are much more papers on the computer simulation of enzymes capable of degrading polymers than have been mentioned, but not all of them refer directly to the decomposition of microplastics and, therefore, they are not discussed here. In any case, the efforts of many researchers produced theoretical and experimental data that served as the basis for the development of a variety of biocatalysts degrading various microplastics. The range of microplastics that can be decomposed by enzymes is still limited to polyesters and polyamides and is comparable only with the range of microplastics that are decomposed by solvolysis. Nevertheless, there are other known enzymes that catalyze, for example, redox processes involving polymers, but they have not yet been investigated in reactions with microplastics, although they have practical potential in this field. The specific features of enzymes used to decompose microplastics include the following: moderate temperatures and mild conditions of catalytic reactions (in particular, the absence of organic solvents); the lack of toxicity and biodegradability of the proper biocatalysts, and, in the absence of additional stimulating additives, the lack of toxicity of reaction media; specificity of action; modular structure of biocatalysts (i.e., different catalytically active and/or auxiliary modules can be combined in different ways or be replaced with one another in the same catalytic system).

Special discussion is required for the small group of publications addressing immobilized enzyme catalysts (Table 33). It is known 255 that, on the one hand, immobilized enzymes are stabilized against inactivating factors and, on the other hand, they can be used in the catalytic process many times. Both these factors were successfully demonstrated in experimental works with microplastics.249,251--­254 Certainly immobilization of enzymes by treatment with cross-linking agents is likely to decrease the catalytic activity by a large factor.251 However, this loss can be minimized by immobilization (Fig. 5),250,252 which provides impressive binding capacity of the carriers with these enzymes (up to 0.47 g g^–1).

Table 9

\[ \]

Immobilized enzyme biocatalysts

(33)

Fig. 5

Basic schemes of various types of immobilization of enzymes. (a) embedding and/or absorption (adsorption) in/on a carrier; (b) covalent binding to the carrier; (c) high affinity interaction. The structure of IsPETase enzyme (PDB 6EQD) was illustrated using PyMOL (version 1.7.6, Schrodinger, LLC).

Meanwhile, immobilization opens up new prospects. For example, the use of magnetic nanoparticles as carriers allows easy separation of the biocatalyst from the reaction media or additional stimulation of the catalytic activity by using MW field and/or light.250,251 A study by Li et al.,253 who incorporated simultaneously two enzymes into MOF is quite promising. Despite the fact that the activity of enzymes decreased 1.5--­2-fold, an approach of this type could be used in the future for combining various enzymes with chemical catalysts degrading microplastics (Fig. 6). In turn, enzymes can also influence the partner introduced into the process, resulting, for example, in an increase in the size of metal nanoparticles by dozens of times.250 Thus, careful selection of the chemical and biocatalytic components is required, which can be greatly facilitated by using computer simulation techniques to calculate the interaction of enzymes with MOFs256 and with nanoparticles.257

Fig. 6

Some examples of (bio)catalysts for the degradation of microplastics. The images of a typical zeolite,53 nanomaterial,70 organic metal complex,77 MOF135 and their combination (exemplified by the enzyme and MOF)256 were adapted from original sources with permission from Elsevier and the American Chemical Society. The structure of the IsPETase enzyme (PDB 6EQD) was illustrated using PyMOL the (version 1.7.6, Schrödinger, LLC).

A separate option is to use living cells of microorganisms that synthesize the target enzymes to degrade microplastics directly during their culturing.258--260 The performance of these catalyst systems is comparable to that of immobilized enzymes. In principle, several enzymes can be simultaneously expressed in a cell, thus performing multistage conversion of microplastics.

Returning to the issue of the energy barriers of the reactions, it should be emphasized that in some studies, the activation energies were determined experimentally and/or calculated for a number of (bio)catalysts (see Table 31). As expected, catalysts decrease the energy barriers by large factors. The energy barrier depends on not only the composition and amount of the catalyst,39,102,119 but also on additionally introduced solvents110 and the operating pressure.147 It is quite possible that the activation energy of the reaction also decreases under MW irradiation111 as compared to the activation energy of a similar process without it.110 It is noteworthy that almost in all studies, the activation energy in the presence of (bio)catalysts was determined via the degree of conversion of the substrate, which does not coincide with the product yield. Therefore, for correct comparison of different (bio)catalysts, it would be reasonable to revise or even redefine the activation energies determined previously, in order to attain certain common conditions for comparison.

With rare exceptions, the energy barriers for enzyme-catalyzed reactions are usually determined by computer simulation. According to a typical mechanism of action of these catalysts, the first step is the acylation of the Ser residue (see Fig. 3), which is deacylated in the second step. This reaction pathway involves two energy barriers for both monomeric162 and dimeric substrates246--­248 and higher-molecular-weight models. However, four or more energy barriers were identified in many studies;184,261,262 most likely, these barriers are artifacts and/or are due to the use of erroneous initial state(s) or interpretations.

Among other drawbacks, the use of truncated models for polymer substrates is noteworthy. Of course, computer modelling is too resource-demanding, but it is already evident that reduced (shortened) models give distorted results. For example, successive increase in the model length gives different values for energy barriers and different sets of amino acid residues that interact with polymers.262 This set of definite amino acid residues of biocatalysts can serve for rational choice of the targets246--­248 for their subsequent modification by rational design methods. Furthermore, the roles of the cap domain of the enzyme active site162,235 and metal ions (able to induce conformational changes in the protein structure),263 which can affect the enzymatic activity, are often neglected.

As opposed to macroplastics visible by naked eye, the (bio)degradation of microplastics isolated from environmental objects is addressed in rare instances.57 Most often, these studies use model polymer substrates or (more rarely) their mixtures. Certainly, this makes it possible to precisely control the conditions for laboratory processes, but does not appear adequate for real-life implementation.

It is also noteworthy that owing to the deficiency of studies of real samples of microplastics, the issue of how microplastics can be extracted from environmental specimens has not yet been clarified. As initial conditions, it is rational to resort to the existing practices used for analysis of microplastics in water, soil and atmosphere.1--­4 It is evident that in the case of catalytic processes, it is likely that some procedures will be eliminated or, conversely, additional isolation stages would be required. For instance, it is known that the concentrations of microplastics in water specimens vary over a broad range: from a few to tens of thousands of particles in cubic metre of water;264,265 therefore, for more efficient catalysis, it is necessary, at least, to separate the microplastics from the embedding matrix and, at most, to concentrate it to an appropriate level. For this purpose, it is expedient to use, for example, membrane technologies, which have been rather advanced to date.266 Here, the problem of nanoplastics deserves mention: among the publications discussed in this review, only a few papers89,90,174,227 address model polymer nanoparticles with reliably determined size. The small number of papers allows one to hope that studies in this area, especially as applied to real samples of nanoplastics, would be intensified in the near feature. This is a nontrivial and highly important task, which has a high scientific and practical potential.

Analysis of published works makes it possible to formulate several criteria that are necessary for studying the degradation of microplastics:

- identification and quantitative determination of specific decomposition products by relevant methods; semiquantitative and qualitative determinations are acceptable only in the early and preliminary stage of investigation;

- publication of complete exhaustive procedures for conduction of the reactions; regarding the currently reported procedures, the results of many published studies cannot, in principle, be reproduced; hence, they objectively cannot be included into the list of procedures available for the review analysis;

- as full as possible characterization of the polymer substrate and the catalyst; as shown in this review, quite a number of parameters affect the process performance and the yields of products; characteristics of commercial products may vary over a wide range without notice from the manufacturer;

- determination of the toxicity of chemical catalysts and reaction products formed with these catalysts; currently this issue is ignored by authors who develop catalytic processes; however, the toxicity (gene, eco-, cyto- and immunotoxicity and other types of toxicity) should be determined, if possible, for different biological objects; this problem is especially acute when the research is switched from model to real microplastics;

- ruling out contamination of the polymer substrate by other polymers during the process; this problem is neglected in the studies discussed here, although it has been known for a long time5 and is discussed in the analysis of microplastics; the contaminants may be not converted to reaction products themselves, but affect the efficiency of catalysis.

The compliance with these criteria would markedly simplify both selection of the catalysts for practical application and development of new catalyst samples.

Currently, computer simulation methods are mainly used in the development of enzyme biocatalysts and are rarely used for other purposes. This obvious imbalance already affects the progress in the development of chemical catalysts; in the future, this effect would increase unless decisive measures are taken to improve the situation. One more aspect is that computer simulation is still used only to study the mechanisms of interaction of enzymes with low-molecular weight substrates. Meanwhile, these methods are suitable for much more complex systems, e.g., proteins interacting with nanomaterials,267,268 MOFs269 or polymers.270 The latter is especially important, because, as has been noted above, the artificial shortening of the model for the interacting substrate gives unpredictable results unrelated to the actual experiment. The use of micro- or nanoparticles composed of a definite polymer would be an ideal model.271 In any case, irrespective of the chosen option (nanomaterial, MOF, polymer), this approach enables:

- design of (bio)catalysts of a new type, for example, those capable of targeted delivery of the (bio)catalyst towards a microplastic particle via its conjugation or merging with a partner that has a high affinity for the substrate;272

- fabrication of (nano)biocatalysts with enhanced stability and without an adverse influence of the carrier and/or the preparation itself on the enzymatic activity;

- combining of (bio)catalytic components with (bio)sensors and/or coagulating agents;

- combining of different type (bio)catalytic components that are allowed to react with microplastics.

In principle, this combination has already been implemented in a number of catalytic processes, and the advantages of this approach have been demonstrated. However, the combinations of catalysts are now selected by the trial and error method. Meanwhile, simulation will elevate the efficiency of such system to a new research and practice level. Indeed, using various starting enzymes, it is now possible to modify their structures and, hence, to improve the catalytic performance on the basis of a reference sample with the highest biocatalytic activity.225

A few words should be said about enzyme kinetics in heterogeneous systems. The authors of many studies actively promote the so-called inverse Michaelis--Menten kinetics for the explanation of the observed atypical decrease in the enzyme activity with increasing enzyme concentration. Moreover, some authors go even further and take the calculated values as the classical KM and kcat constants, although Scandola et al.,273 who were the first to propose this model, interpret the results more properly. Further studies and development of an adequate kinetic model are required to explain this unusual behaviour of enzymes. As a rule, a few variants are generated, and the one describing most closely the experimental data is chosen.209 That is not to say that the model and preconditions used by de Queiros Eugenio et al.209 are exhaustive (especially considering the specific adsorption and desorption processes established in other studies), but these examples demonstrate that an appropriate kinetic model for heterogeneous enzymatic reactions, including conversion of microplastics, could be developed. Adequate interpretation of the data derived from catalytic degradation of microplastics is an important part of the study of the conditions of these processes, their catalysts and products.

This review was written within the framework of the State task of the Lomonosov Moscow State University (121041500039-8) and with the support of the Interdisciplinary Scientific and Educational School of ­Moscow University “The Future of the Planet and Global Environmental Change”.

AA — adipic acid;

AE — 2-aminoethanol;

Amim — 1-allyl-3-methylimidazolium;

BHBA — bis(4-hydroxybutyl) adipate;

BHBS — bis(4-hydroxybutyl) succinate;

BHBT — bis(4-hydroxybutyl) terephthalate;

BHDET — bis(hydroxydiethylene) terephthalate;

BHEF — bis(2-hydroxyethyl) 2,5-furandicarboxylate;

BHET — bis(2-hydroxyethyl) terephthalate;

Bmim — 1-butyl-3-methylimidazolium;

[Ch]3PO4 — choline phosphate;

CNTs — carbon nanotubes;

DBN — 1,5-diazabicyclo[4.3.0]non-5-ene;

DEG — diethylene glycol;

DFT — density functional theory;

DPG — dipropylene glycol;

DTMAC — dodecyltrimethylammonium chloride;

EDC — 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide;

EG — ethylene glycol;

FCC — fluid cracking catalyst;

FDA — 2,5-furandicarboxylic acid;

HBA — 4-hydroxybutyl adipate;

HBS — 4-hydroxybutyl succinate;

HBT — 4-hydroxybutyl terephthalate;

HDPE — high-density polyethylene;

HDPS — high-density polystyrene;

HEF — 2-hydroxyethyl 2,5-furandicarboxylate;

HET — 2-hydroxyethyl terephthalate;

Hmim — 1-hexyl-3-methylimidazolium;

LDPE — low-density polyethylene;

MAA — methacrylic acid;

MCC — microcrystalline cellulose;

Me₂TPA — dimethyl terephthalate;

MeTPA — monomethyl terephthalate

Mmim — 1-methyl-3-methylimidazolium;

[Mmim]⁺-2-COO– — 1-methyl-3-methylimidazolium 2-carboxylate;

MOF — metal-organic framework;

MSA — methanesulfonic acid;

MW — molecular weight;

N-Melm — N-methylimidazole;

NP — nanoparticle;

NPG — neopentyl glycol;

P3HB — poly(3-hydroxybutyrate);

P3HP — poly(3-hydroxypropionate);

PBB — poly(1,4-butylene butanedioate);

PBBH — poly(1,4-butylene butanedioate-co-hexanedioate);

PBBC — poly(1,3-bis(aminomethyl)benzenecapramide);

PBBO — poly(1,3-bis(aminomethyl)benzyloxalamide);

PBBT — poly(1,3-bis(aminomethyl)benzylterephthalamide);

PBVI — poly(1-butyl-3-vinylimidazolium bis(trifluoromethane)sulfonimide);

PBH — poly(1,4-butylene hexanedioate);

PBHT — poly(1,4-butylene hexanedioate-co-terephthalate);

PBDT — poly(1,4-butylene decanedioate-co-terephthalate);

PBT — poly(1,4-butylene terephthalate);

PBAC — poly(bisphenol-A-carbonate);

PVP — polyvinylpyrrolidone;

PVC — polyvinyl chloride;

PHBHP — poly(3-hydroxybutyrate-co-3-hydroxypentanoate);

PHT — poly(hexamethylene terephthalate);

PHMA — poly(hexamethylene adipate);

PDHT — poly(1,6-diaminohexaneterephthalamide);

PDMAEMA — poly(dimethylaminoethyl methacrylate);

PDO — polydioxanone;

PDCT — poly(1,4-dihydroxymethylenecyclohexane ­terephthalate);

PCL — poly(_e-caprolactone);

PLGC — poly(lactide-co-glycolide-co-e-caprolactone);

PMMA — poly(methyl methacrylate);

PMBS — poly(4,4ʹ-methylene-bis(cyclohexanamide)succinamide);

PLA — polylactic acid;

PP — polypropylene;

PPC — poly(propane-1,2-diol carbonate);

PS — polystyrene;

PTMT — poly(trimethylene terephthalate);

PU — polyurethane;

PPS — poly(phenylene sulfide);

PEB — poly(ethylene butanedioate);

PEVIA — poly(1-ethyl-3-vinylimidazolium acetate-co-acrylate);

PE — polyethylene;

PEG — polyethylene glycol;

PEC — poly(ethylene carbonate);

PET — poly(ethylene terephthalate);

PEF — poly(ethylene 2,5-furandicarboxylate);

SA — succinic acid;

TBD — 1,5,7-triazabicyclo[4.4.0]dec-5-ene;

TMAD — trimethylolpropane allyl ether;

TMADC — trimethylolpropane allyl ether carbonate;

TMGasme — 2-(tetramethylguanidinium)-5-R¹-benzoic acid methyl ester, where R¹=Cl, H or NMe₂;

TOC — total organic carbon;

TPA — terephthalic acid;

THF — tetrahydrofuran.