Degradation of Bio-Based and Biodegradable Plastic and Its Contribution to Soil Organic Carbon Stock

Expanding the use of environmentally friendly materials to protect the environment is one of the key factors in maintaining a sustainable ecological balance. Poly(butylene succinate-co-adipate) (PBSA) is considered among the most promising bio-based and biodegradable plastics for the future with a high number of applications in soil and agriculture. Therefore, the decomposition process of PBSA and its consequences for the carbon stored in soil require careful monitoring. For the first time, the stable isotope technique was applied in the current study to partitioning plastic- and soil-originated C in the CO2 released during 80 days of PBSA decomposition in a Haplic Chernozem soil as dependent on nitrogen availability. The decomposition of the plastic was accompanied by the C loss from soil organic matter (SOM) through priming, which in turn was dependent on added N. Nitrogen facilitated PBSA decomposition and reduced the priming effect during the first 6 weeks of the experiment. During the 80 days of plastic decomposition, 30% and 49% of the released CO2 were PBSA-derived, while the amount of SOM-derived CO2 exceeded the corresponding controls by 100.2 and 132.3% in PBSA-amended soil without and with N fertilization, respectively. Finally, only 4.1% and 5.4% of the PBSA added into the soil was mineralized to CO2, in the treatments without and with N amendment, respectively.


Introduction
A large amount of plastic is applied to the packaging of commercial products such as food, pharmaceuticals, cosmetics and chemical products and in modern agriculture (e.g., mulching plastic films). The incineration of this traditional plastic waste releases a large amount of carbon dioxide, which contributes to global warming [1,2]. Plastic debris and the toxic gases produced as intermediate products of its degradation result in environmental pollution, thereby increasing the demand for consumers to use more natural and recyclable packaging materials [1,3]. Therefore, it is preferable and efficient to replace conventional plastics with a sustainable alternative, such as bioplastics, which are produced from renewable biomass sources. Expanding the use of environmentally decomposition [36]. Therefore, it is necessary to consider PE and the transformation of the SOM in the studies on BBP decomposition in soil.
Among several techniques to track the decomposition of plastics [1,37], stable isotope analysis is very promising to distinguish the origin of the intermediate products of plastic decomposition in the case of difference in isotopic signature between plastics and SOM [38][39][40]. The use of naturally occurring 13 C tracers in soil studies is based on the different photosynthetic pathways between C3 and C4 plants [41]. The δ 13 C values of C4 plants, such as maize or sugarcane, range from −12‰ to −10‰ [42]. C4 plant-derived products demonstrate higher 13 C content than C3 plants. Thus, an application of C4-bio-based plastics to a C3 soil enables the identification of the origin of the end products. As plastic decomposition is monitored indirectly by the CO 2 emission, the possible shift in the 13 C signature of the SOM and its mineralization products due to: (i) 13 C isotopic fractionation, i.e., the difference between heavy and light coexisting isotopes of an element; or (ii) 13 C isotopic discrimination (see below) needs to be considered in the calculations [40]. Such a shift is caused either by: (i) the preferential use of substances with light 12 C versus heavy 13 C [43]; or (ii) by the preferential utilization of heavy 13 C-substrate of high availability (sugars, cellulose) versus the light 12 C of low availability (lignin, lipids) [44]. The discrimination between SOM-C and SOM-respired CO 2 can be 13 C-enriched, near-zero difference and 13 C-depleted [45][46][47][48]. Since the N is the driver for the microbial utilization, its availability might strongly affect the carbon isotope fractionation in microbial processes [49,50].
Our study attempted to distinguish plastics and soil C sources in the course of the decomposition of bio-based and bio-degradable plastics using stable isotope techniques. For this purpose, we used the isotope partitioning approach to differentiate the plasticderived CO 2 from the soil-derived CO 2 . Our aim was to determine: (i) the degradation rate of BBPs by the contribution of plastic-derived C to the CO 2 emission; (ii) the effect of N fertilization on the plastic degradation rate: and (iii) the priming effect of soil organic matter induced by plastic addition, using δ 13 CO 2 source partitioning. We hypothesize that: (i) adding N (as a limiting nutrient in the BBPs) significantly increases plastic-derived CO 2 emission as the biodegradation of BBPs increases; (ii) BBPs are a potential C source that can cause PE.

Material
We tested the renewable-based PBSA (poly(butylene succinate-co-adipate)) as one of the most promising and new biopolymers. The PBSA is the combination of succinic acid and adipate acid with 1-4-butanediol. The PBSA plastic films were purchased from PTT MCC Biochem Company Limited (Thailand) with the BioPBS FD92 trade name [8,11]. The 50 µm thick, double-layered PBSA film contains 55% C and consists of 35% bio-based carbon. Before being applied to soil, the PBSA was sterilized with 70% ethanol and cut into small pieces (2-5 mm × 2-5 mm). The δ 13 C of PBSA was determined by elemental analysis coupled to an isotope ratio mass spectrometer (Delta Plus XL IRMS, Thermo Finnigan MAT, Bremen, Germany).

Design of Experiment
The laboratory experiment was conducted on soil sampled from a conventional farming plot of the Global Change Experimental Facility (GCEF), Bad Lauchstädt, Central Germany (51 • 22 60 N, 11 • 50 60 E, 118 m a.s.l.). The soil was classified as a Haplic Chernozem with water-holding capacity = 35%, total organic C = 2%, C: N ratio = 10, pH = 7. 5 and silt-loam soil texture with a 70% silt, 20% clay and 10% sand combination [51,52]. Considering the δ 13 C of PBSA used, the conventional farming plot was chosen with most distinct difference in δ 13 C between SOM and PBSA. The plot is under a typical regional crop rotation consisting of winter rape, winter wheat and winter barley [53]. The soil was sampled from winter barley, homogenized passing through a 2 mm sieve and divided into the following treatments: (1) control soil without PBSA (S); (2) control soil without PBSA and with (NH 4 ) 2 SO 4 addition (SN); (3) PBSA-soil-soil with PBSA addition (PS); (4) PBSA-soil-N-soil with PBSA and (NH 4 ) 2 SO 4 addition (PSN). The PBSA treatment contained 28.5 g of soil (equivalent to 23.5 g dry soil) and 1.5 g (6% PBSA, w/w) of plastic material [11,53,54]. The five replicates of each treatment were put into a glass container and placed into plastic 250 mL jars in the respirometer for CO 2 emission measurement. In the PSN and control SN treatments, 0.0825 g N was added in soil as 2.1 mL of 1.42 M (NH 4 ) 2 SO 4 to maintain C: N ratio (10:1) and simulate fertilization in agricultural system. For PS and control treatments (without N), 2.1 mL sterile Milli-Q water was added to equalize water content as in the N added treatments (17.5%, accounting for 50% of the water-holding capacity). All treatments were incubated at a constant soil moisture and air temperature (22 • C) during 80 days of the experiment.

Soil Respiration
Soil respiration was determined using an automated respirometer Respicond V (Nordgren Innovations, Sweden) during 80 days at 22 • C. The system enabled a continuous measurement of CO 2 trapped by 0.6 M KOH by a decrease in electrical conductivity of KOH solution [55]. During the experiment, the KOH with the trapped CO 2 was collected to measure δ 13 C in dynamics. Thereafter, a new KOH was added to the system at each sampling time.

Source-Partitioning of CO 2 Emitted from Soil and from PBSA
Samples for determination of δ 13 C signature in CO 2 were obtained by C precipitation [40]. At first, the CO 2 trapped as K 2 CO 3 in 10 mL KOH was precipitated with 15 mL of 0.5 M BaCl 2 aqueous solution. Then, the KOH solutions containing the BaCO 3 precipitate were centrifuged three times at 3000× g for 5 min and washed in between with deionized and degassed water to remove an excess of KOH and to reach a pH of 7. After washing, the supernatant water was removed from the vials and the BaCO 3 was dried at 105 • C for 24 h. The solid samples of BaCO 3 were analyzed for δ 13 C values by elemental analysis coupled to an isotope ratio mass spectrometer (Delta Plus XL IRMS, Thermo Finnigan MAT, Bremen, Germany).
Isotope fractionation between soil-C and CO 2 -C was estimated as difference between the δ 13 C value of soil (−26‰) and the corresponding δ 13 C value of CO 2 -C released from this control soil (−23‰). 13 C enrichment in respired soil CO 2 is a common phenomenon and can be explained by preferential use of 13 C-enriched SOM compounds such as sugars [41]. The N effect on 13 C fractionation was considered as a difference between δ 13 C of CO 2 -C from control soil with and without added N ( Figure 1). The isotope fractionation accounted for 3‰ (S) and 5‰ (SN) as compared to the δ 13 C of corresponding SOM. Thus, N addition further enriched δ 13 C of CO 2 -C by 2‰ compared to SOM mineralization without additional N. The further δ 13 C source partitioning was based on the assumption of similar isotope fractionation of CO 2 -C originated from both soil and plastics [42,56].
The δ 13 C notation expresses carbon isotope ratios of sample and standard [38,57]: where δ 13 C is the parts per thousand difference between the 13 C content of the sample and the standard. The δ 13 C values are expressed relative to the reference standard Vienna-PeeDee Belemnite (VPDB), being equal to 0.00112372. The fraction of plastics-derived C in total C in CO2 was estimated according to Ame lung et al. [38]: where δ 13 Ct is the δ 13 C value of the CO2-C released from the PBSA-amended soil; δ 13 Ccs i the δ 13 C value of coinciding CO2-C pool, which is referred to control soil; δ 13 CPBSA is th δ 13 C value of PBSA. Before the CO2 source partitioning, we corrected the δ 13 Ccs by the CO2 possibly en trapped from the air: where the Ccs and Ca are the amount of C derived from soil-respired CO2 and atmospheri CO2, respectively; the Ccs+a is value of the mixture of C derived from soil-respired CO2 and from atmospheric CO2 in the incubation jar; δ 13 Ca is the δ 13 C value of the air CO2; th δ 13 Ccs+a is the δ 13 C value of the mixture of CO2 originated from soil and from air in th incubation jar. The amount of plastics mineralized to CO2 was calculated according to the followin equation: where Ct is the total amount of CO2-C released from PBSA-amended soil.
To estimate priming effect (PE), we used the mass balance equation [40,57]: where the CO2-C is an amount of C in the corresponding pool, i.e., "PBSA" indicates C derived from PBSA, "soil"-C derived from soil, "PE" is the priming effect and "total" i the sum of all pools.

Statistical Analysis
The open-source R software version 3.6.1 was used to perform all statistical analysis The effects of soil treatments on CO2 emission and fractions were analyzed using analysi Figure 1. Isotope signature of 13 C sources: Soil-δ 13 C in soil, S-CO 2 -δ 13 C in CO 2 respired from soil without N addition, SN-CO 2 -δ 13 C in CO 2 respired from soil with N addition, PBSA-δ 13 C of PBSA, PS-CO 2 -an estimated shift in δ 13 C of PBSA in CO 2 respired from soil in the PS treatment; PSN-CO 2 -an estimated shift in δ 13 C of PBSA in CO 2 respired from soil in PSN treatment.
The fraction of plastics-derived C in total C in CO 2 was estimated according to Amelung et al. [38]: where δ 13 C t is the δ 13 C value of the CO 2 -C released from the PBSA-amended soil; δ 13 C cs is the δ 13 C value of coinciding CO 2 -C pool, which is referred to control soil; δ 13 C PBSA is the δ 13 C value of PBSA. Before the CO 2 source partitioning, we corrected the δ 13 C cs by the CO 2 possibly entrapped from the air: where the C cs and C a are the amount of C derived from soil-respired CO 2 and atmospheric CO 2 , respectively; the C cs+a is value of the mixture of C derived from soil-respired CO 2 and from atmospheric CO 2 in the incubation jar; δ 13 C a is the δ 13 C value of the air CO 2 ; the δ 13 C cs+a is the δ 13 C value of the mixture of CO 2 originated from soil and from air in the incubation jar. The amount of plastics mineralized to CO 2 was calculated according to the following equation: where C t is the total amount of CO 2 -C released from PBSA-amended soil.
To estimate priming effect (PE), we used the mass balance equation [40,57]: where the CO 2 -C is an amount of C in the corresponding pool, i.e., "PBSA" indicates C derived from PBSA, "soil"-C derived from soil, "PE" is the priming effect and "total" is the sum of all pools.

Statistical Analysis
The open-source R software version 3.6.1 was used to perform all statistical analysis. The effects of soil treatments on CO 2 emission and fractions were analyzed using analysis of variance (ANOVA), incorporating the Shapiro-Wilk test for normality and Levene's test to assess the equality of group variances. If variance homogeneity and normality were given, an ANOVA was conducted by following the Tukey post hoc test to compare the means with the p < 0.05 level of significance. Otherwise, the Kruskal-Wallis rank sum test served as a non-parametric alternative by following Dunn's test as a post hoc test to compare the treatments for significant differences of multiple comparisons. Barplots were used to illustrate the data to display the distribution and changes in the variables. Since the values were used from different devices, the propagation of error was calculated to perform variation between replicates.

CO 2 -C Respired
The N addition to the control soil reduced the CO 2 emission by 13% at the end of the experiment. From the beginning of the experiment, PBSA-amended soils respired greater amounts of CO 2 as compared with the corresponding non-amended controls ( Figure 2; Table S1). The respiration of the PBSA-amended soil without N gradually increased. After a week of the experiment, the N raised the CO 2 efflux by 3.5 times in PSN as compared to PS treatment (Figure 2 insert). Larger amounts of CO 2 were respired from the PSN versus the PS treatment until 28 days and thereafter the N effect was insignificant. In the first two weeks of PBSA decomposition, the CO 2 released from the PS and PSN treatments exceeded that from control by the factors of 2 and 6, respectively.
Polymers 2023, 15, x FOR PEER REVIEW 6 o test to assess the equality of group variances. If variance homogeneity and normality w given, an ANOVA was conducted by following the Tukey post hoc test to compare means with the p < 0.05 level of significance. Otherwise, the Kruskal-Wallis rank sum served as a non-parametric alternative by following Dunn's test as a post hoc test to c pare the treatments for significant differences of multiple comparisons. Barplots w used to illustrate the data to display the distribution and changes in the variables. S the values were used from different devices, the propagation of error was calculate perform variation between replicates.

CO2-C Respired
The N addition to the control soil reduced the CO2 emission by 13% at the end of experiment. From the beginning of the experiment, PBSA-amended soils respired gre amounts of CO2 as compared with the corresponding non-amended controls (Figur Table S1). The respiration of the PBSA-amended soil without N gradually increased. A a week of the experiment, the N raised the CO2 efflux by 3.5 times in PSN as compare PS treatment (Figure 2 insert). Larger amounts of CO2 were respired from the PSN ver the PS treatment until 28 days and thereafter the N effect was insignificant. In the first weeks of PBSA decomposition, the CO2 released from the PS and PSN treatments ceeded that from control by the factors of 2 and 6, respectively.

13 C Signature of Respired CO2-C
The δ 13 CO2-C values of the control soils (S and SN) were relatively constant du the experiment (Figure 3). A significant N fertilization effect resulting in a 2‰ greater enrichment of the CO2 respired from the SN versus S treatment was stable throughout experiment ( Figure 3). The 13 C signature in the PS treatment gradually shifted towards δ 13 C signature of the PBSA during the experiment (Figure 3). The δ 13 CO2-C in the P treatment was 3‰ more 13 C enriched as compared to soil without N addition an reached its highest level at 6 weeks of the incubation. After 6 weeks, the δ 13 C signatur

13 C Signature of Respired CO 2 -C
The δ 13 CO 2 -C values of the control soils (S and SN) were relatively constant during the experiment (Figure 3). A significant N fertilization effect resulting in a 2‰ greater 13 C enrichment of the CO 2 respired from the SN versus S treatment was stable throughout the experiment (Figure 3). The 13 C signature in the PS treatment gradually shifted towards the δ 13 C signature of the PBSA during the experiment (Figure 3). The δ 13 CO 2 -C in the PSN treatment was 3‰ more 13 C enriched as compared to soil without N addition and it reached its highest level at 6 weeks of the incubation. After 6 weeks, the δ 13 C signature in the PSN treatment returned to the original level and decreased slightly for the rest of the experiment (Figure 3).
Polymers 2023, 15, x FOR PEER REVIEW 7 Figure 3. Dynamics of δ 13 CO2-C during experiment: S indicates control soil without N addi SN⎯control soil with N addition, PS⎯PBSA-amended soil without N addition and PSN⎯PB amended soil with N addition. The letters indicate significant differences between treatments a corresponding time point.

Source Partitioning of the Respired CO2-C
Despite the fact that PBSA was degraded both with and without N addition, the of decomposition was not constant. Without N, PBSA was decomposed relatively slo during the first month of the experiment, whereas N addition accelerated the decomp tion rate in the first 4 weeks (Figure 2). The decomposition rate of PBSA was 57 ti higher in the PSN versus the PS treatment after 14 days of incubation (Figure 4a; T S2). The 28-46-day period of the experiment was a tipping point, after which the dec position rate of PBSA in N-amended soil decreased dramatically. In terms of absolute ues, PBSA decomposition gradually increased during the experiment when no N added (Figure 4a). The decomposition of plastic in the PSN treatment was most inten in the first 2 weeks of the experiment and strongly decreased after 60 days of incuba ( Figure 4).

Source Partitioning of the Respired CO 2 -C
Despite the fact that PBSA was degraded both with and without N addition, the rate of decomposition was not constant. Without N, PBSA was decomposed relatively slowly during the first month of the experiment, whereas N addition accelerated the decomposition rate in the first 4 weeks (Figure 2). The decomposition rate of PBSA was 57 times higher in the PSN versus the PS treatment after 14 days of incubation ( Figure 4a; Table S2). The 28-46-day period of the experiment was a tipping point, after which the decomposition rate of PBSA in N-amended soil decreased dramatically. In terms of absolute values, PBSA decomposition gradually increased during the experiment when no N was added (Figure 4a). The decomposition of plastic in the PSN treatment was most intensive in the first 2 weeks of the experiment and strongly decreased after 60 days of incubation (Figure 4).
In terms of relative values, the proportion of CO 2 factions originated from soil and PBSA were strongly affected by the treatment and incubation time (Figure 4b; Table S3). In the PS treatment, the PBSA fraction in the respired CO 2 generally increased over time (Figure 4b). The N amendment increased the PBSA fraction in CO 2 by almost 21 times (Figure 4b). The two decomposition phases of PBSA, intensive (up to 6 weeks) and relatively slow, were distinguished in the PSN treatment. The largest contribution of PBSA-originated C to the CO 2 respired was observed between 4 and 6 weeks of PBSA decomposition when N was added. After 6 weeks, however, the PBSA fraction in the CO 2 emission decreased 4.7 times compared to the previous period. In terms of relative values, the proportion of CO2 factions originated from soil and PBSA were strongly affected by the treatment and incubation time (Figure 4b; Table S3). In the PS treatment, the PBSA fraction in the respired CO2 generally increased over time (Figure 4b). The N amendment increased the PBSA fraction in CO2 by almost 21 times (Figure 4b). The two decomposition phases of PBSA, intensive (up to 6 weeks) and relatively slow, were distinguished in the PSN treatment. The largest contribution of PBSAoriginated C to the CO2 respired was observed between 4 and 6 weeks of PBSA decomposition when N was added. After 6 weeks, however, the PBSA fraction in the CO2 emission decreased 4.7 times compared to the previous period.
Despite insignificant differences in the total respired CO2-C between PSN and PS (Figure 2), the contribution of PBSA was larger by 22 % in the PSN versus the PS treatment (Figure 4c). At the end of experiment, only about 4.1% and 5.4% of the PBSA-C added in the PS and PSN treatments, respectively, were mineralized to CO2 ( Figure 5). Despite insignificant differences in the total respired CO 2 -C between PSN and PS (Figure 2), the contribution of PBSA was larger by 22% in the PSN versus the PS treatment (Figure 4c). At the end of experiment, only about 4.1% and 5.4% of the PBSA-C added in the PS and PSN treatments, respectively, were mineralized to CO 2 ( Figure 5).

Priming Effect
Essentially, a large fraction of primed soil C was detected in both treatments. The largest fraction of primed C was detected after 60 days in the PSN while this was not a

Priming Effect
Essentially, a large fraction of primed soil C was detected in both treatments. The largest fraction of primed C was detected after 60 days in the PSN while this was not a case in the PS treatment (Figure 4b). The highest intensity of PE in the PS and PSN treatments was detected at 3 and 2 weeks after PBSA application to the soil, respectively. After reaching the highest SOM decomposition rate by PE in the PS treatment, it decreased towards the end of the incubation time. The intensity of the priming effect in the PS treatment decreased by 5 times between 3 and 11 weeks of plastic decomposition (Figure 4a). In contrast, the fraction of PE in the PSN treatment increased during the slow (after 46 days) as compared to the fast decomposition phase of plastic (Figure 4b), whereas the highest SOM decomposition rate was observed by primed C after 2 weeks. At 46 days of incubation, the decomposition of PBSA in the PSN treatment resulted in negative PE, which indicated 17% retarded SOM decomposition. Due to the consistently high fraction of PE in the PS treatment in the course of PBSA decomposition, the cumulative PE during 80 days of PBSA decomposition was 25% greater in the PS vs. the PSN treatment (Figure 4c). Remarkably, the C losses from soil by PE were comparable (PSN) or even exceeded by the factor of 1.3 (PS) the losses by basal respiration of the control soils.

Discussion
The 3‰ and 5‰ 13 C-enriched CO 2 from control soil without and with N indicated isotope fractionation (Figure 1). Isotope fractionation between soil and soil-respired CO 2 is the result of two fractionation steps: the first isotope discrimination occurs during the metabolic processes, which leads to the preferential use of light 12 C for the synthesized compounds and results in the 13 C enrichment of the residual compounds, and the second fractionation is related to specific compound utilization by microorganisms [58][59][60]. The 13 CO 2 -C enrichment during the utilization of complex soil organic matter can be explained by the preferential utilization of specific compounds strongly varying in 13 C values [61,62]. The selective utilization of 13 C-rich organic compounds such as sugars can cause an enrichment of 13 C in soil-respired CO 2 during heterotrophic respiration [42,45]. As living plants were removed from the soil, we assume that only microbial (heterotrophic) respiration was the source of soil-originated CO 2 . The quality of carbon that cycled between organic debris and biomass in the course of the microbial re-utilization of C differs from the originally assimilated carbon [63,64]. We demonstrated that N availability affected the microbial metabolic pathways of isotope discrimination and fractionation. At a low C: N ratio, soil microorganisms preferentially utilize easily decomposable SOM with 13 C-enriched organic compounds such as sugars, starch, cellulose, proteins, or organic acids. Therefore, the difference between soil-C and CO 2 -C was much lower in non-amended versus N-amended soil (3‰ versus 5‰).
The pattern of shifts in 13 C values towards PBSA in PBSA-added soils and without PBSA-added soils are confirmed by other studies [37,65]. However, compared to other BBP investigations, the 13 C values of PBSA-added soil shifted more quickly toward PBSA. In our study, the shifts in the PBSA-added soil with and without the addition of N were 3.7‰ and 2.0‰ after 6 weeks, respectively. The differences in the decomposition of PBSA between N addition in both scenarios (with and without N amendment) indicated a nutrient limitation in the soil. The nitrogen amendment stimulated the degradation of easily decomposable organic material, which contained 13 C-enriched C compounds. The differences between the 13 C signature of the PBSA-amended soil without and with N fertilization indicated that N availability changes the pathway of the decomposition of PBSA. The slowly increasing PBSA degradation when no N was added can be explained by the lack of N. Since PBSA does not contain nitrogen, the PBSA addition to soil increases the C:N ratio and the N deficiency in the PBSA-soil system. Microorganisms are able to colonize the plastic films, but because of the N requirement they are not efficient in PBSA degradation [8]. As the microorganisms that colonized the polymer surfaces [12] require N to grow [66], we assume that the microorganisms in the PS treatment without the N addition invested an energy from the carbon source to mine for N from SOM. Therefore, we assume that in the case of pure PBSA addition, the decomposition of PBSA was slowly fueled by N obtained through priming. Priming in such a case could serve as an indication of microbial growth. An alternative N source could be gained through N-fixing bacteria, which contributed to N acquisition in the PS treatment [8]. In the N-amended PBSA treatment, the PBSA decomposition rate was faster from the beginning of the experiment until the N requirements were met. However, a reduced CO 2 emission rate and increased PE after 46 days in PSN treatment could be an indication of N limitation in this treatment in the last 5 weeks of plastic decomposition [8,11].
The input of PBSA served as a C source in both treatments [67,68] and altered the CO 2 release from the soil causing a priming effect. Indeed, the PE, in other words the stimulation of SOM mineralization, was stronger in the N-poor PS treatment during the first 4 weeks as compared to the remainder of the experiment. Thus, the N requirements of microorganisms involved in PBSA degradation were covered by intensive SOM mineralization, which proves the "N-mining" mechanisms of PE. The PE during the 2 week period of the experiment in the PSN treatment can be explained by "stoichiometric decomposition" mechanisms, which are related to the increasing biomass and enzymatic activity at high C and N availability. Our assumption in the case of N fertilization is that an input of C and N matches microbial demands at the early stage of PBSA decomposition. In this case, microbial decomposers release extracellular enzymes, which are able to break down SOM along with PBSA. Thus, microbial activity increased the losses of the soil-originated CO 2 -C. The strong decrease in PE by a factor of 16 after two weeks and even the negative PE observed during days 28-46 of the experiment (Figure 4b) compared with the first two weeks of the PSN treatment indicated the microbial switch from SOM decomposition to PBSA utilization, which can be explained by shifting from "stoichiometric decomposition" to "preferential utilization" mechanisms. At the later decomposition stage, however, when added N was utilized, microorganisms growing on PBSA faced the lack of N and possibly of other nutrients. We assume, therefore, that the "tipping point" of 46 days can be explained by shifting from "preferential utilization" to a "N-mining" mechanism due to the depletion of the added N.
The contrasting dynamics of PE in the treatments with and without N could also indicate the successional changes in the community of PBSA decomposers as well as corresponding changes in dominating microbial taxa. This was supported by the observed assembly of archaeal (Nitrososphaeraceae), bacterial (e.g., N fixing bacteria Bradyrhizobium spp.) and fungal (Sordariomycetes, Eurotiomycetes, Leotiomycetes) taxa dominating plastic decomposition under the lack of N [53]. In contrast, N-fixing bacteria disappeared from the N-amended PBSA decomposition and the community domination shifted towards nitrophilic microorganisms such as ammonia-oxidizing archaea (Nitrocosmicus and Nitrososphaeraceae) and denitrifying bacteria (Achromobacter insolitus, Achromobacter denitrificans) [8].
In total, 4.1% and 5.4% of the added PBSA was respired as CO 2 without and with N-amended soils, respectively, during 80 days, which is consistent with other PBSA decomposition studies [11,27]. However, the mineralization of PBSA to CO 2 is much lower as compared to the mineralization of natural substrates in soil, whereas PBSA-induced primed C is near the highest border of natural source-induced primed C ( Table 1). The degradation of BBPs has been demonstrated by the weight loss approach with other studies, whereas the fate of PBSA is unknown: whether it degraded into micro-and/or nano-plastics or became part of the biomass and mineralized into CO 2 . In our study, we used the "source partitioning of the CO 2 -C respired" approach to trace the decomposition of BBPs, which is more relevant to identifying the fate of plastics. Assuming the carbon use efficiency ≈ 50-60% [69], we consider that roughly 9-10% of the BBPs were degraded during 80 days. That means about 75 mg g −1 of added plastic was either incorporated in the microbial biomass or transformed into secondary metabolites and micro-size particles, while 1350 mg g −1 of plastic remained unaffected. N accelerated the degradation of plastic at least during the period of 80 days, as shown by the 31.6% greater PBSA decomposition in the PSN as compared to the PS treatment, but this tendency retarded with time due to nutrient limitation.

Conclusions
We demonstrated the possibility of estimating the contribution of PBSA-derived C in CO 2 by a stable isotope partitioning approach. The PBSA increased both the CO 2 release and SOM mineralization, which may have been caused by increased microbial activity. In both cases-with and without N fertilization-the PBSA decomposition raised the overall CO 2 output by a factor of around 2.5. However, the decomposition rate of PBSA depended on N availability. The N fertilization increased PBSA mineralization to CO 2 by 31.6% and decreased the priming effect by 36% compared with the non-N treatment. Therefore, N fertilization coupled with optimal watering (to ensure appropriate redox conditions) could be recommended as an environmentally friendly strategy to facilitate plastic decomposition and reduce soil C and N 2 O losses under field conditions. Since the biodegradability of plastics is dependent on their quality, soil type and environmental conditions, the strength and duration of PE needs to be investigated in the future with a combination of various soils and plastics to reveal the mechanisms of C sequestration in soil.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/polym15030660/s1, Table S1. The statistic parameters of analysis of variance (ANOVA) of the CO 2 emission of treatments during the experiment (correspond to Figure 2); Table S2. The statistic parameters of analysis of variance (ANOVA) of the contribution of three fractions in the decomposition rate (plastic, soil and priming) during the incubation experiment (correspond to Figure 3a); Table S3. The statistic parameters of analysis of variance (ANOVA) of the proportion of three fractions (plastic, soil and priming) during the incubation experiment (correspond to the Figure 4b).