Three types of plastics were evaluated: low-density polyethylene with and without pro-oxidant additive (OXOLDPE and LDPE, respectively), and the compostable plastic Ecovio^®. Both polyolefins were supplied by Artes Gráficas Unidas. S. A. de C. V. AGUSA (Mexico), who used the d2w^® pro-oxidant additive from Plásticos Degradables S.A. de C.V. (México), subsidiary of Symphony Environmental. The compostable plastic was provided by BASF-México. It is composed of an aliphatic-aromatic mixture of the copolyester Ecoflex^® (also produced by BASF) and polylactic acid (PLA). All the plastics were cut into 150 × 10 mm probes, and half of each type of plastic was abiotically oxidized in a previously described weathering chamber built at the university [15] , in order to simulate the degradation produced by use, UV radiation and temperature. In the chamber, the materials were exposed to 50˚C, 80% relative humidity and a radiation interval 300 - 460 nm during 216 hours, the time needed for the OXOLDPE to decrease its elongation at break to values near to 100%, because of the abiotic degradation process. First, confirm that you have the correct template for your paper size. This template has been tailored for output on the custom paper size (21 cm × 28.5 cm).

The seawater used in this study was obtained from the Barra Norte beach in Tuxpan (Veracruz, Mexico, Figure 1). The seawater was transported in sterilized containers to the laboratory, where it arrived 4 hours after sampling. It had a pH of 7.85 ± 0.07 and salinity of 50.5 ± 1.2 mS/cm. The reactors used to assess degradation had a volume 2 L, and were filled with 1.2 L of seawater amended with inorganic nutrients (0.5 g/L of NH₄Cl and, 0.1 g/L of KH₂(PO)₄) and was incubated at 30˚C ± 1˚C for 48 days. Air was continuously flown into the reactors during the respirometric test.

The corresponding plastics were introduced into the reactors containing amended seawater. The mass of each plastic is shown in Table 1. Cellulose was used as a positive control, and reactors containing only amended seawater were used as blanks. Three replicates were setup for each plastic and control (Table 2). They were subsequently connected to a respirometric system (Figure 2), which included a continuous flow or CO₂-free, saturated air to guarantee aerobic conditions and a series of traps to capture the CO₂ produced by the plastic degradation.

Solutions of NaOH (0.25 N) were used as traps for CO₂. They were subsequently titrated with 0.5 N HCl solution twice a week. Biodegradation of samples was determined in terms of mineralization (%), with Equation (1):

where C_test is the mean of the amount of CO₂ produced in the replicates for each plastic (mg), C_blank is the mean CO₂ production in the blanks (inoculum only; mg) and C_i was the total amount of polymer-C added to the test reactors (mg).

The tensile elongation at break was assessed as an indicator of degradation (abiotic + biotic), as done previously

[16] [17] . It was measured based on the standard ASTM D882 [18] . The machine used was a Lloyd LFPlus (AMETEK Lloyd Instrument Ltd. UK). The testing speed was maintained at 300 mm/min and the grip-to-grip separation was 30 mm. Conditioning and testing were performed at standard atmospheric conditions (24˚C ± 1˚C and 50% ± 10% RH). The residual elongation at break (%REB) was determined using Equation (2).

The data obtained from tensile tests underwent an analysis of variance (ANOVA) and a Tukey’s multiple range test (MRT), both with 95% significance. The biodegradation stage was not included in the statistical analysis because it did not have a normal distribution.

After 48 days of experimentation, the cellulose reactors showed 68.3% of mineralization. This value is consistent with others reported before in composting conditions: 80% in 90 days and 100% in 317 days [19] [20] , and it can be considered as a measurement of the good performance of the respirometric system. The two polyolefins, with and without pro-oxidant additive or previous degradation process, did not show a significant difference in mineralization, which ranged from 2.06% to 2.78%. Ecovio^®, on the other hand, achieved 10.11% and 10.38% for oxidized and no-oxidized samples, respectively (Figure 3). There are no similar researches, which allow comparing of these results; however, biodegradation of these plastics in other environments has been reported previously. Mineralization values from 2% to 2.5% and 7% to 10% were reported for LDPE and OXOLDPE, after incubation in soil at 45˚C, during 90 days [21] . In a composting process inoculated with Rhodococcus rhodochrous LDPE reached 9% of mineralization, while OXOLDPE achieved 16%, after 317 and 352 days of degradation [20] . As it can be observed, the low degree of mineralization achieved for LDPE and OXOLDPE in our experiment is consistent with reported values.

Statistical analysis showed significant differences (p < 0.05) between the different plastics. The MRT analysis yields five homogeneity groups, where LDPE and previously oxidized LDPE (LDPE-O) had the lower %REB, with 9.3% and 17.1%, respectively. OXO-LDPE achieved a 29.11%, while Ecovio^®, with and without oxidation, showed a similar behavior. The highest loss of elongation was found for previously oxidized OXO-LDPE (OXO-LDPE-O), which decreased its mechanical resistance in 67.85% during the test (Figure 4).

Biodegradation rates measured by respirometric tests cannot be directly compared to the loss of elongation at break, which is a result of both, biotic and abiotic factors. In this research, Ecovio^® achieved the highest degree of biodegradation, while previously oxidized OXOLDPE had the higher loss of elongation at break. Degradation on natural marine environments is a complex multifactorial process that could be affected by the types and amount of living organisms, presence of soluble salts, temperature, dissolved oxygen, UV radiation and by the movement of the water. There is not available information about the degradation of plastics by sea salts, although it is known that soluble marine salts cause alterations on physical and chemical properties of anticorrosive paints and concrete [22] [23] . It also has been reported that microorganisms such as bacteria and fungi can form a biofilm on plastics, producing enzymes that promote the change of physicochemical properties of the plastic [24] [25] .

Previous oxidation by UV and temperature clearly promoted further degradation in the polyolefins, which consistently showed higher values of %REB, compared to those that had no treatment. It is known that for this plastics once light begins degradation, the process tend to continue [26] , favoring subsequent biotic and abiotic processes. While is common for polyolefins to begin their degradation by this kind of chemical interaction, Ecovio^® contains PLA, which begins its degradation by non enzymatic hydrolysis [27] . Previous exposition to UV light did not affect its degradation. In terms of percent of mineralization, compostable plastic presented faster than both polyolefins mineralization. The values obtained for all the plastics are similar to those reported before for direct exposure on marine environments [28] ; this shows that the respirometric system used in this experiment is a useful tool to an easier and cheaper screening of the performance of this plastics.

It is relevant to consider that loss of physical integrity, which is related to the loss in elongation at break, cannot be considered inherently beneficial. If the fragmentation process occurs at higher rates than the biodegradation process, microplastics could be produced. The presence of these small particles of plastics in the sea has been widely reported in literature, as well as their transport through the food chain [29] [30] .