Search

Additional file 1: Figure S1. Map of sampling sites in the Yangtze River covering the actual sinuous channel reach of length 4,300 km (equivalent to 2.05 times the 2,102 km straight line joining the start to end sampling sites). Figure S2. Venn diagrams showing unique and shared numbers of ARG types (a) and subtypes (b) among the water-spring (WS), water-autumn (WA), sediment-spring (SS), and sediment-autumn (SA) samples in the Yangtze River. Principal coordinates analysis (PCoA) diagrams showing the compositional dissimilarity of ARG type (c) and subtypes (d) in the Yangtze River. Analysis of similarity statistics (ANOSIM) displaying the dissimilarities of ARG subtypes in water (e) and sediment (f) between different seasons. Figure S3. Richness (a) and Shannon diversity (b) of ARG subtypes in each sample group of the Yangtze River. ARG types exhibiting significant seasonal differences in water (c) and sediment (d). Figure S4. (a) Principal coordinates analysis (PCoA) showing compositional dissimilarities (Bray-Curtis) of ARG hosts (at the species level) in the four sample groups of the Yangtze River. (b-d) ANOSIM statistics concerning differences in ARG hosts within and between sample groups. Figure S5. Richness (a) and Shannon diversity (b) of ARG hosts in each sample group of the Yangtze River. ARG hosts (at the species level) exhibiting significant seasonal differences in water (c) and sediment (d). Figure S6. (a) Distribution of MGEs co-existing with ARGs across four sampling groups. (b) Distribution of various types of MGEs (co-occurring with ARGs) across four sampling groups. Figure S7. Procrustes analysis depicting correlations between ARG subtypes and ARG hosts in water (a) and sediment (b). Solid and hollow circles represent ARGs and hosts, respectively. Figure S8. Relationships between ARG hosts at the phylum level (inner circle) and ARG types (outer circle) in water-spring (a), water-autumn (b), sediment-spring (c), and sediment-autumn samples (d). Figure S9. Networks displaying ARG hosts carrying multiple ARGs in water-spring (a), water-autumn (b), sediment-spring (c), and sediment-autumn samples (d) of the Yangtze River. The sizes of nodes correspond to the connection degree. Figure S10. Networks displaying HPB carrying multiple ARGs in water-spring (a), water-autumn (b), sediment-spring (c), and sediment-autumn samples (d) of the Yangtze River. The sizes of nodes correspond to the connection degree. Figure S11. Distance-decay relationships of Bray-Curtis similarity of sedimentary ARGs in spring (a) and autumn (b), hosts in spring (c) and autumn (d), and HPB with the geographical distance in spring (e) and autumn (f). Mantel-Spearman correlations (r) and probabilities (significance codes: ***≤ 0.001 **≤ 0.01 *≤ 0.05) are provided. Solid lines indicate the ordinary least squares linear regression across all samples. Slopes of regression lines are also provided. Figure S12. (a) Abundance (copy of ARG per copy of 16S-rRNA gene) and (b) richness of planktonic and sedimentary ARG subtypes in mainstream and seven tributaries (DTH: Dongtinghu, HBH: Huangbohe, HJ: Hanjiang, JLJ: Jialingjiang, MJ: Minjiang, PYH: Poyanghu, and WJ: Wujiang) along the Yangtze River. WS, WA, SS, and SA refer to water-spring, water-autumn, sediment-spring, and sediment-autumn samples, respectively. Figure S13. (a) Abundance (copy of ARG per copy of 16S-rRNA gene) and (b) richness of planktonic and sedimentary ARG types in mainstream and seven tributaries (DTH: Dongtinghu, HBH: Huangbohe, HJ: Hanjiang, JLJ: Jialingjiang, MJ: Minjiang, PYH: Poyanghu, and WJ: Wujiang) along the Yangtze River. WS, WA, SS, and SA refer to water-spring, water-autumn, sediment-spring, and sediment-autumn samples, respectively. Figure S14. (a) Abundance (coverage/Gb) and (b) richness of planktonic and sedimentary ARG hosts in mainstream and seven tributaries (DTH: Dongtinghu, HBH: Huangbohe, HJ: Hanjiang, JLJ: Jialingjiang, MJ: Minjiang, PYH: Poyanghu, and WJ: Wujiang) along the Yangtze River. WS, WA, SS, and SA refer to water-spring, water-autumn, sediment-spring, and sediment-autumn samples, respectively. Figure S15. (a) Abundance (coverage/Gb) and (b) richness of planktonic and sedimentary HPB in mainstream and seven tributaries (DTH: Dongtinghu, HBH: Huangbohe, HJ: Hanjiang, JLJ: Jialingjiang, MJ: Minjiang, PYH: Poyanghu, and WJ: Wujiang) along the Yangtze River. (c) Abundance (coverage/Gb) of planktonic and sedimentary supercarriers in mainstream and seven tributaries. WS, WA, SS, and SA refer to water-spring, water-autumn, sediment-spring, and sediment-autumn samples, respectively. Figure S16. LEfSe cladogram depicting the taxonomic differences of ARG hosts in water (a) and sediment (b) in spring for four landform types. Differentially abundant taxa (biomarkers) are colored according to their most abundant landform habitats. Figure S17. ARG types (water-spring group) exhibiting significant differences among four landform types along the Yangtze River. Figure S18. Occurrence frequency of sedimentary ARG subtypes in spring (a) and autumn (b) as well as hosts in spring (c) and autumn (d) fitted to mean relative abundance using Sloan et al.’s neutral community model. Inserts in (a-b) and (c-d) show the neutral community model fits to ARG type and HPB. Purple and green dots indicate ARGs/hosts that occur more (‘Above’) and less (‘Below’) frequently than given by the neutral model (gray dots, ‘Neutral’). R2 indicates the fit to the neutral community model, and m indicates the immigration rate. Dashed lines represent 95% confidence intervals about the model prediction. Figure S19. Distribution of occurrence frequency (a) and mean relative abundance (b) of ARGs in different sampling groups. ‘Above’ and ‘Below’ indicate ARGs that occur more and less frequently than given by the neutral model (‘Neutral’). Figure S20. The partial least squares path models showing the effects of spatial variables, anthropogenic variables, physicochemical variables, antibiotics, MGEs, and ARG hosts on ARG compositions in water (a) and sediment (b) of the Yangtze River. Solid and dashed lines indicate positive and negative effects, respectively. Numbers adjacent to each arrow denote partial correlation coefficients (significance codes: ***≤ 0.001 **≤ 0.01 *≤ 0.05). R2 values display the proportion of variance explained for each factor. The bar-chart showing the standardized total effect of each factor on the ARG composition in water (c) and sediment (d). Figure S21. Spearman’s correlations (R) between the richness of ARG host and ARGs in water (a) and sediment (b), and between the richness of HPB and the richness of ARGs in water (c) and sediment (d). Ordinary least square linear regressions and 95% confidence intervals are also displayed.

The ecofriendly waterborne alkyd coatings generally compromise the coating performance. Incorporation of polyaniline into alkyd to form nanocomposites can improve the corrosion resistance; however, the uniform distribution of polyaniline (PANI) within the alkyd resin remains a challenge. In this work, PANI is grafted onto epoxy functionalized phosphorylated poly (vinyl alcohol) (PPVA) with different side-chain configurations in order to improve compatibility and dispersion stability. These functionalized PPVA are synthesized using epichlorohydrin (ECIP) or γ-glycidyl methacrylate (GMA) as modifying agents, and respectively labeled as E-PPVA and G-PPVA. G-PPVA/PANI dispersion is stable even at a high PANI loading of 50%, while the maximum PANI loading for E-PPVA/PANI is only 30%. With the incorporation of G-PPVA/PANI into waterborne alkyd resins (AR), the coating resistance is increased by 7 orders of magnitude, and the corrosion inhibition efficiency is up to 99.9%. The coating with the incorporation of G-PPVA/PANI also demonstrates superior long-term corrosion resistance in comparison with that AR/E-PPVA/PANI nanocomposite coating.

Graphitic carbon nitride (g-C3N4) has been considered as a candidate for anti-corrosion material owing to its barrier property and photoelectrochemical cathodic protection. However, the related mechanisms remain obscure and are not fundamentally investigated. Herein, g-C3N4 coating on Q235 CS plate matrix can be simply made via conductive adhesive, which helps to achieve scalable production and high cost-effectiveness. It can also ensure the unidirectional transfer of excited electrons from g-C3N4 to metal to prevent the consumption of free electrons on metal and reduce Fe2O3 onto Fe. During the photocatalytic anti-corrosion process, the electron holes were simultaneously consumed via oxidizing water to achieve the balance between electron and hole transport. By combining theoretical calculation and experimental analysis, the anti-corrosion mechanisms of g-C3N4 were systematically investigated. g-C3N4 displayed superior long-term and remote anti-corrosion properties, especially for remote and difficult-to-reach areas, which will make it become a new generation of anti-corrosion 2D material with promising performance.

Graphene oxide/polyaniline (GO/PANI) has been demonstrated to be an effective anti-corrosive nanofiller in polymer matrix, however, its hydrophobility restricts its application in waterborne polymer coatings. Here, cetyltrimethylammonium bromide (CTAB) and cationic gemini surfactant (GS) with the same constitutional unit as CTAB were adopted to fabricate CTAB modified GO/PANI (CGO/PANI) and GS modified GO/PANI (GGO/PANI). The incorporation of GS not only increased the intercalation degree of GO nanosheets but also functionalized as dopant to improve the electrochemical properties of PANI. The dispersibility and compatibility of GGO/PANI in waterborne alkyd resin (WAR) were also higher than that of CGO/PANI, resulting in the formation of more uniform nanocomposite coating. Compared with pure WAR, the coating resistance of GGO/PANI/WAR increased by 4 orders of magnitude, the corrosion density decreased by 3 orders of magnitude and the corrosion potential shifted from −0.72 V to −0.28 V. GGO/PANI/WAR also displayed enhanced long-term anti-corrosive properties in comparison with CGO/PANI/WAR.

Simultaneous enhancement of conductivity and mechanical properties for polyaniline/polymer nanocomposite still remains a big challenge. Here, a reverse approach via in situ polymerization (RIP) of vinyl monomers in waterborne polyaniline dispersion was raised to prepare conductive polyaniline (GPANI)/polyacrylate (PMB) interpenetrating polymer (GPANI-PMB) nanocomposite. GPANI/PMB physical blend was simultaneously prepared as reference. The conductive GPANI-PMB nanocomposite film with compact pomegranate-shape morphology is homogeneous, ultraflexible and mechanically strong. With incorporating a considerable amount of PMB into GPANI via the RIP method, only a slight decrease from 3.21 to 2.80 S/cm was detected for the conductivity of GPANI-PMB, while the tensile strength significantly increased from 25 to 43.5 MPa, and the elongation at break increased from 40% to 234%. The water absorption of GPANI-PMB3 after 72 h immersion decreased from 24.68% to 10.35% in comparison with GPANI, which is also higher than that of GPANI/PMB. The conductivity and tensile strength of GPANI-PMB were also much higher than that of GPANI/PMB (0.006 S/cm vs. 5.59 MPa). Moreover, the conductivity of GPANI-PMB remained almost invariable after folding 200 times, while that of GPANI/PMB decreased by almost half. This RIP approach should be applicable for preparing conventional conductive polymer nanocomposite with high conductivity, high strength and high flexibility.

Diisocyanates functionalized graphene nanosheets with high level exfoliation have been demonstrated as effective anticorrosion fillers in waterborne polymer coatings. However, diisocyanates can cause potential environmental and health hazard. Here, we propose to use polyiscocyanate (Tri-HDI) with an isocyanurate-ring structure yet low volatility and toxicity, to functionalize reduced graphene (RGO). Compared with hexamethylene diisocyanate (HDI), the incorporation of Tri-HDI simultaneously increased the exfoliation degree and the dispersibility of RGO. The reactivity between Tri-HDI and GO was also enhanced, resulting in the increase of graft ratio. The addition of Tri-HDI-RGO (Tri-HDIG) was able to improve the colloidal stability of waterborne epoxy resin (ER), as well as the interaction and compatibility between Tri-HDIG and ER. The tensile strength of ER/Tri-HDIG thereby increased from 50.1 MPa to 81.8 MPa, and the water absorption decreased from 18.0 % to 5.90 %. The long-term corrosion resistance of ER/Tri-HDIG also got significant improvement, its impedance modulus increased by 5 orders of magnitude compared to pure ER coating. The broad applicability of Tri-HDIG as functional nanofiller in polymer matrixes has been confirmed.

Hydrophilic functionalization of reduced graphene (RGOs) is able to improve its dispersibility in aqueous polymer. However, the influence of hydrophilic segments on the chemical resistance of RGOs/polymer nanocomposites keeps indistinct. Here, we prepared RGOs functionalized with hydrophilic chelated titanate (TA) and non-chelated titanate (T), named as TARGOs and TRGOs, respectively. The intercalation reaction mechanism between titanate and GO was elucidated. The presence of hydrophilic triethanolamine in titanate had little impact on the structural disorder of RGOs, but can effectively improve the exfoliation degree and the dispersibility of functionalized RGOs in waterborne alkyd resin (WAAR). The water resistance and thermal stability of TARGOs/WAAR were inferior to that of TRGOs/WAAR. The initial impedance modulus and coating resistance of TARGOs/WAAR were higher than TRGOs/WAAR at the same RGOs content when the RGOs content is 0.5% owing to the more homogeneous distribution of TARGOs in WAAR. However, the corrosion resistance of TARGOs/WAAR significantly decreased when the TARGOs content increased to 0.7%, which can be attributed to the presence of hydrophilic regions. In contrast, TRGOs/WAAR displayed the optimum long-term corrosion resistance when the TRGOs content was 0.7%, even though the dispersion of TRGOs in WAAR cannot be comparable to TARGOs in WAAR.

The compatibility between polypyrrole (PPy) and waterborne epoxy coating (EP) remains the main challenge. Here, phosphorylated poly (vinyl alcohol) (PPVA) and p-toluene sulfonic acid sodium (p-TSA) were simultaneously utilized as dopants to modify PPy and prepare water dispersible PPVA/p-TSA/PPy (PPVAT/PPy) solution. Additionally, poly (vinyl alcohol) (PVA) and TSA were also adopted to prepare PVA/p-TSA/PPy (PVAT/PPy) for comparison. Denser film with granular morphology was observed for PPVAT/PPy nanocomposite film. Compared with PVAT/PPy-40 %, the tensile strength of PPVAT-8.0 %/PPy-40 % increased from 248.8–283.9 MPa. PPVAT/PPy displayed higher compatibility and interactions with EP compared to PPy-TSA and PVAT/PPy. Compared with EP, the impedance modulus of PVAT/PPy+EP and PPVAT/PPy+EP increased from 3.17×105 to 1.07×108 and 1.62×109 Ω·cm2; corrosion current density Icorr decreased from 1.49×10−6 to 7.72×10-8 and 4.01×10-9 A·cm-2; corrosion potential Ecorr displayed a positive shift of 230 mV and 269 mV, respectively. The corrosion inhibition efficiency of PPVAT/PPy+EP reached 99.83 %, which is much higher than that of PVAT/PPy+EP (93.97 %). PPVAT/PPy+EP nanocomposite coating has been proved to have preferable long-term corrosion resistance based on electrochemical impedance spectra (EIS), polarization curves and salt water spray test. The simultaneous incorporation of phosphate anions, sulfate ions and PPy into EP was beneficial to fabricate high-performance anticorrosion coating.

Phosphorylated poly(vinyl alcohol) (PPVA) was demonstrated to be more effective to improve the dispersion stability of polyaniline (PANI). Phosphate groups also can improve corrosion resistance, but they may spatially interfere with graft reaction of PANI onto PPVA. Therefore, it is of great significance to explore the balance between phosphorylation degree, graft reaction and composite performance. Here, EPPVA-grafted-PANI (EPPVA-g-PANI) dispersions were prepared by chemical grafting polymerization of aniline on the epichlorohydrin modified PPVA (EPPVA) with different degree of phosphorylation (DP). It was found that DP has a considerable effect on microstructure, rheological behaviors, as well as the compatibility between EPPVA and PANI. Enhanced interactions and compatibility were demonstrated for EPPVA-g-PANI-8.00 %, resulting in the increase of tensile strength from 26.84 to 38.62 MPa. As-prepared EPPVA-g-PANI dispersions can be directly blended with waterborne alkyd resins (WAR) without complex post-process to fabricate high-performance WAR/EPPVA-g-PANI nanocomposites. Compared with WAR, the impedance modulus of WAR/EPPVA-g-PANI increased from 8.81×104 to 3.03×1010 Ω·cm2, corrosion current density decreased to 3.02×10−8 from 1.78×10−6 A·cm−2. Additionally, only one time constant was detected after 480 h immersion, further demonstrating excellent long-term corrosion resistance for WAR/EPPVA-g-PANI.

Waterborne polyaniline (PANI) dispersion has got extensive attention due to its environmental friendliness and good processability, whereas the storage stability and mechanical property have been the challenge for the waterborne PANI composites. Here we prepare for waterborne PANI dispersion through the chemical graft polymerisation of PANI into epichlorohydrin modified poly (vinyl alcohol) (EPVA). In comparison with waterborne PANI dispersion prepared through physical blend and in situ polymerisation, the storage stability of PANI-g-EPVA dispersion is greatly improved and the dispersion keeps stable for one year. In addition, the as-prepared PANI-g-EPVA film displays more uniform and smooth morphology, as well as enhanced phase compatibility. PANI is homogeneously distributed in the EPVA matrix on the nanoscale. PANI-g-EPVA displays different morphology at different aniline content. The electrical conductivity corresponds to 7.3 S/cm when only 30% PANI is incorporated into the composites, and then increases up to 20.83 S/cm with further increase in the aniline content. Simultaneously, the tensile strength increases from 35 MPa to 64 MPa. The as-prepared PANI-g-EPVA dispersion can be directly used as the conductive ink or coatings for cellulose fibre paper to prepare flexible conductive paper with high conductivity and mechanical property, which is also suitable for large scalable production.