Organic additive release from plastic to seawater is lower under deep-sea conditions

Global plastic production is vast and a fraction reaches the ocean each year, where macroplastics undergo slow photo-, bio-, and physical degradation leading to micro- and nanoplastics. Plastic additives (phthalates, organophosphate esters, bisphenols) represent a significant mass fraction of plastics and can leach into seawater, posing endocrine and toxic risks to marine life and potentially humans. Observations and models suggest most plastic mass resides below the surface, including deep waters and sediments. Yet, how environmental conditions across the water column—especially high hydrostatic pressure and differing prokaryotic assemblages—affect additive leaching is poorly constrained. This study tests the hypothesis that deep-sea conditions reduce additive release compared to the surface, and that prokaryotes enhance leaching, by comparing additive release from polyethylene (PE) and soft polyvinylchloride (PVC) under surface vs deep seawater chemistry, low (0.1 MPa) vs high (10 MPa) hydrostatic pressure, and abiotic vs biotic (natural prokaryotes) conditions.

• Plastic-derived dissolved organic carbon includes oligomers and additives; many detected additives in marine systems are endocrine disruptors. Release is governed largely by polymer–water partitioning and can be influenced by marine prokaryotes.
• Only ~1% of plastic waste is estimated to float; evidence points to substantial accumulation in the ocean interior and sediments, with elevated additive concentrations near the bottom in impacted seas (Yellow Sea, Mediterranean).
• Prior lab studies indicated salinity, UV, and turbulence affect additive leaching, and microbial presence can accelerate phthalate release and degradation; OPE and PAE half-lives vary with nutrient status. Data under deep-sea pressure conditions were previously lacking.

• Materials and sites: Recycled low-density PE pellets (PTX131; 3.7 mm average diameter; density 0.955 g mL−1; surface area 17 cm2 g−1) and pristine soft PVC pellets (PTX500; 3.6 mm; density >1.1 g mL−1; surface area 13 cm2 g−1) were exposed to filtered natural seawater.
• Seawater collection: Deep seawater (1000 m) from Cassidaigne Canyon (43.068223 N, 5.468057 E) collected 06/06/2018; surface seawater (0.5 m) from Marseille Bay (43.273624 N, 5.347348 E) collected 11/23/2018; both filtered (0.7 μm GF/F). Deep: salinity 38.5; DOC 1.3 mg C L−1. Surface: salinity 37.9; DOC 7.2 mg C L−1.
• Experimental design: 5.0 ± 0.1 g of PE or PVC pellets incubated in 130 mL filtered seawater, dark, 13 °C, for 30 days. Treatments varied by: (i) seawater source (surface vs deep), (ii) hydrostatic pressure (0.1 vs 10 MPa), (iii) biology (abiotic via HgCl2 10 mg L−1 vs biotic natural prokaryotes). Deep biotic at 0.1 MPa not run. Plastic-free controls matched each condition. Duplicate bottles per timepoint; two 130-mL bottles sacrificed per sampling.
• High-pressure setup: 500-mL stainless steel high-pressure bottles (HPBs) housed two sealed 130-mL glass bottles; pressure (10 MPa) transmitted via PTFE septa. Preliminary tests confirmed negligible sorption losses.
• Target analytes: 25 parent additives (9 PAEs: DMP, DEP, DiBP, DnBP, BzBP, DEHP, DnOP, DiNP, DiDP; 9 OPEs: TPP, TiBP, TnBP, TCEP, TCPP, TDCP, TPhP, EDHPP, TEHP; 7 BPs: BPA, BPAF, BPAP, BPF, BPP, BPS, BPZ) and 7 PAE monoester metabolites (MPAEs: including MMP, MEP, MiBP, MnBP, MBzP, MEHP, MnOP).
• Chemical analysis: Single extraction protocol; PAEs and OPEs analyzed by GC–MS/MS and LC–HRMS (MPAEs from second elution). DiNP identified via full-scan then quantified by GC–MS/MS. QA/QC included instrumental and procedural blanks (<LOQs; DEHP blank-corrected), spiked recoveries 63–107%, and control checks.
• Microbiology: Prokaryotic abundance by flow cytometry (FACS-Calibur; SYBR Green II staining) via PRECYM platform.
• DOC/DON/DOP: Filtration (0.7 μm), DOC on Shimadzu TOC-5000; DON/DOP by total dissolved minus inorganic N/P using wet oxidation and AutoAnalyzer.
• Kinetics/statistics: First-order release model X(t)=a(1−exp(−bt)); parameters estimated by nonlinear regression (Gauss–Newton). 95% confidence bands constructed via SSE-based confidence region sampling; dynamics compared by overlap of 95% CIs across treatments/compounds.

• Additives detected and general kinetics:
  • From PE, 5/25 additives leached: PAEs (DMP, DEP, DEHP) and OPEs (TnBP, TEHP). From PVC, DiNP and BPs (notably BPS) were detected.
  • Most additives released predominantly within the first week, then approached a plateau by 30 days. Exception: BPS from PVC showed linear, constant-flux release over 30 days under biotic conditions.
• Magnitudes of cumulative release (30 days):
  • PE: cumulative across detected additives 212.5–738.5 ng g−1 depending on treatment (lowest at deep 0.1 MPa abiotic; highest at surface 0.1 MPa biotic).
  • PVC: DiNP release 4.136–88.167 μg g−1; BPS in ng g−1 range. PVC released up to two orders of magnitude more additive mass than PE.
• Hydrostatic pressure effects (abiotic deep seawater 0.1 vs 10 MPa):
  • Low-MW, more hydrophilic additives (e.g., DMP, DEP) showed overlapping 95% CIs; no significant pressure effect.
  • Higher-MW, more hydrophobic additives (TEHP, DEHP, DiNP) exhibited significantly reduced leaching at 10 MPa after ~15 days (non-overlapping 95% CIs). Example: PE TEHP dropped to <1.6 ng g−1 at 10 MPa vs 7.8–8.3 ng g−1 at 0.1 MPa (abiotic); PVC DiNP 4.136 μg g−1 at 10 MPa vs ~17.6–18.2 μg g−1 at 0.1 MPa (abiotic).
• Prokaryote effects:
  • Presence of natural prokaryotes increased dissolved-phase additive levels across compounds and pressures (separated 95% CIs), except DEP in deep seawater where effect was not significant.
  • Biotic vs abiotic increases were observed both at surface (0.1 MPa) and deep (10 MPa) conditions.
• Transformation products (MPAEs):
  • MMP from PVC higher under biotic vs abiotic: surface 146 ± 12 vs 24 ± 5 ng g−1; deep 77 ± 17 vs 22 ± 3 ng g−1.
  • MEHP from PE detected under biotic only: surface 16 ± 4 ng g−1 (abiotic ND); deep 9.7 ± 2.2 ng g−1 (abiotic ND). Indicates microbial degradation of parent PAEs.
• Effects on prokaryotic abundance:
  • Initial free-living prokaryotes ~104 mL−1 rose to ~105 mL−1 in controls and PE incubations. PVC exposure reduced prokaryote abundance ~4-fold vs controls after 30 days (surface: 0.42 ± 0.09 vs 1.9 ± 0.06 ×105 mL−1; deep: 0.54 ± 0.12 vs 2.2 ± 0.22 ×105 mL−1).
  • Nutrients (C, N, P) were not depleted; DOC and DOP increased with PVC, implicating leachate toxicity (e.g., DiNP 527–3392 μg L−1; BPS ~230 ng L−1 under conditions) in growth limitation.
  • Under surface biotic conditions, DiNP carbon represented ~18% of total DOC released from PVC (65.8 μg C g−1 of PVC vs 368 μg C g−1 total DOC).
• Surface vs deep overall:
  • Considering pressure and biology, additive release during the first month was roughly two-fold lower under deep-sea conditions than at the surface (excluding UV effects).
• Global implications:
  • Extrapolated tentative first-week oceanic releases for target list: PVC 2.3–132 tons yr−1; PE 0.4–3.4 tons yr−1. Actual totals likely higher due to non-targeted compounds.

The study directly addresses how deep-sea conditions and microbial presence modulate additive leaching from common plastics. High hydrostatic pressure characteristic of 1000 m depth reduces leaching of higher molecular weight, hydrophobic additives (e.g., DEHP, TEHP, DiNP), while low-MW hydrophilic compounds are largely unaffected. Possible mechanisms include increased polymer–water partitioning favoring retention within the polymer and/or reduced water-accessible polymer layer and free volume limiting diffusion. Conversely, natural marine prokaryotes enhance additive release under both surface and deep conditions—likely via polymer surface oxidation that increases hydrophilicity and lowers Kpw, and by increasing accessible surface area via bio-degradation. Prokaryotes also transform parent PAEs into more hydrophilic monoesters (MPAEs), which have higher mobility in water and lower sediment affinity, potentially enhancing transport through the water column and sediments. The net result is more efficient additive release in surface waters compared to the deep ocean; however, deep-sea exposure may remain substantial due to larger plastic burdens and persistence, and potential microbial growth inhibition from leachates (observed with PVC). These findings refine understanding of plastic additive fate through the water column and suggest that combined surface (UV, microbial) and benthic (microbial) processes govern long-term additive release and transport.

This work demonstrates that (i) deep-sea hydrostatic pressure significantly inhibits leaching of heavier, more hydrophobic additives from PE and PVC, (ii) natural marine prokaryotes enhance additive release and generate more mobile transformation products (MPAEs), and (iii) cumulative additive release over the first month is higher at the surface than in deep seawater. PVC released substantially more additive mass (dominated by DiNP) than PE, while PE leached a more diverse additive suite (notably OPEs). These results imply greater persistence of both plastics and associated additives in the deep sea and highlight the importance of biological processes in controlling leaching. Future research should quantify UV effects in tandem with pressure and biology, characterize non-target additive and oligomer leachates, resolve mechanistic impacts of pressure on partitioning/diffusion in polymers, incorporate particle size/aging states and biofilms (including particle-attached communities), and extend investigations to sediments and longer timescales to assess chronic release and organism exposure.

• Test materials were specific PE and PVC examples and are not representative of all formulations or additive profiles.
• Only 25 target additives (7 detected) and 7 PAE monoesters were monitored; non-target additives and oligomers were not quantified.
• Experiments ran 30 days in the dark; UV-driven processes were not included.
• Deep biotic incubations at 0.1 MPa were not performed; deep biotic results pertain to 10 MPa only.
• Mechanistic attribution for pressure effects (Kpw changes vs polymer free-volume/pseudo-porosity changes) could not be distinguished.
• OPE biodegradation half-lives under the specific nutrient conditions are uncertain; apparent release under biotic conditions integrates leaching and biodegradation.
• Water column focus excludes sediment processes and resuspension dynamics.