Agglomeration Determines Effects of Carbonaceous Nanomaterials on Soybean Nodulation, Dinitrogen Fixation Potential, and Growth in Soil

Abstract

The potential effects of carbonaceous nanomaterials (CNMs) on agricultural plants are of concern. However, little research has been performed using plants cultivated to maturity in soils contaminated with various CNMs at different concentrations. Here, we grew soybean for 39 days to seed production in soil amended with 0.1, 100, or 1000 mg kg^-1 of either multiwalled carbon nanotubes (MWCNTs), graphene nanoplatelets (GNPs), or carbon black (CB) and studied plant growth, nodulation, and dinitrogen (N₂) fixation potential. Plants in all CNM treatments flowered earlier (producing 60% to 372% more flowers when reproduction started) than the unamended controls. The low MWCNT-treated plants were shorter (by 15%) with slower leaf cover expansion (by 26%) and less final leaf area (by 24%) than the controls. Nodulation and N₂ fixation potential appeared negatively impacted by CNMs, with stronger effects at lower CNM concentrations. All CNM treatments reduced the whole-plant N₂ fixation potential, with the highest reductions (by over 91%) in the low and medium CB and the low MWCNT treatments. CB and GNPs appeared to accumulate inside nodules as observed by transmission electron microscopy. CNM dispersal in aqueous soil extracts was studied to explain the inverse dose-response relationships, showing that CNMs at higher concentrations were more agglomerated (over 90% CNMs settled as agglomerates >3 μm after 12 h) and therefore proportionally less bioavailable. Overall, our findings suggest that lower concentrations of CNMs in soils could be more impactful to leguminous N₂ fixation, owing to greater CNM dispersal and therefore increased bioavailability at lower concentrations.

Keywords: Carbonaceous nanomaterials; agglomeration; bioavailability; carbon nanotubes; dinitrogen fixation; graphene; soybean.

Figure 1

Transmission electron microscopy (TEM) images of (A) carbon black (CB), (B) multiwalled carbon nanotubes (MWCNTs), and (C) graphene nanoplatelets (GNPs). White arrows indicate the representative features of corresponding nanomaterials. Scale bars are indicated in each image.

Figure 2

Time course of soybean plant vegetative development post-transplantation according to either (A–C) stem length or (D–F) leaf cover (% coverage projected onto the pot soil surface). Ctrl = control without nanomaterial amendment, CB = carbon black, MWCNT = multiwalled carbon nanotubes, and GNP = graphene nanoplatelets. Low, med (medium), and high concentrations correspond to 0.1, 100, and 1000 mg kg⁻¹ nanomaterial on a dry soil basis. Gray arrows indicate when the intermediate harvest took place. Error bars are±SE (n = 5 plants, except n = 4 for the Ctrl, CB_High, and MWCNT_Low treatments).

Figure 3

Soybean root nodule production and N2 fixation potential according to treatment at the intermediate and final harvests, as measured by (A) nodule count per plant, (B) total nodule dry biomass per plant, (C) dry biomass per nodule, and (D) N2 fixation potential (normalized to dry nodule biomass). Ctrl = control without nanomaterial amendment, CB = carbon black, MWCNT = multiwalled carbon nanotubes, and GNP = graphene nanoplatelets. Low, med (medium), and high concentrations correspond to 0.1, 100, and 1000 mg kg⁻¹ nanomaterial on a dry soil basis. Error bars are ±SE (n = 3 plants at the intermediate harvest; n = 5 plants, except n = 4 for the Ctrl, CB_High, and MWCNT_Low treatments at the final harvest). *P < 0.05, **P < 0.01, and ***P < 0.001, as compared to the control (blue *, intermediate; red *, final).

Figure 4

Transmission electron microscopy (TEM) images of soybean root nodules at the final harvest from either (A, B) the controls (Ctrl), or the low exposures (0.1 mg kg⁻¹) of either (C, D) carbon black (CB), (E, F) multiwalled carbon nanotubes (MWCNTs), or (G, H) graphene nanoplatelets (GNPs). Scale bars are indicated in each image. Dense accumulations of bacteroids are evident in the Ctrl images at (A) high and (B) low magnifications. Within some bacteroids (e.g., Ctrl in part A), electron transparent (white-appearing) features (indicated by white arrows) are inferred to be poly(β-hydroxybutyrate) (PHB) inclusions. White double arrows point to apparent nanomaterials inside root nodule cells in the images for the CB (C) and GNP (G, H) treatments, only. The center double arrow in the CB image C indicates where nanomaterials appear associated with a bacteroid. Densely packed bacteroids are evident in one image (H) for the GNP treatment. Accumulations of putative starch granules (indicated by white arrowheads, external to bacteroids) are observed in the images for the CB (D), MWCNT (F), and GNP (G) treatments.

Figure 5

Stability of either 10 or 300 mg L⁻¹ carbon black (CB), multiwalled carbon nanotubes (MWCNTs), or graphene nanoplatelets (GNPs) in the filtered soil extract as evidenced by (A) the early (first 2 h) time course of nanomaterial hydrodynamic diameter as measured by dynamic light scattering (DLS), and (B) the time course to 7 d of normalized nanomaterial suspension absorbance at 600 nm (A/A₀, where A₀ was at time 0 of the experiment; the suspension absorbance at 600 nm used as a proxy for suspended nanomaterial concentration). Error bars are ± SE (n = 3). Note that closed symbols in panel A are clustered along the ordinate for all CNMs at 10 mg L⁻¹, indicating that over the first 2 h of the stability assessment the hydrodynamic diameters of CNMs at 10 mg L⁻¹ were relatively low and constant.

Figure 6

Environmental scanning electron microscopy (ESEM) images of either 10 or 300 mg L⁻¹ (A, B) carbon black (CB), (C, D) multiwalled carbon nanotubes (MWCNTs), or (E, F) graphene nanoplatelets (GNPs) dispersed into the filtered soil extract then deposited onto clean quartz sand. Scale bars are either 5 (A, B, D) or 2 μm(C, E, F). White arrows indicate either dispersed nanomaterials at the lower concentration (10 mgL⁻¹; A, C, E) or larger nanomaterial agglomerates at the higher concentration (300 mg L⁻¹; B, D, F).

Figure 7

Conceptual inverse dose–response relationship for an agglomerating, nondissolving nanomaterial that negatively affects soybean N2 fixation potential on a whole plant basis in moist soil. Results herein for carbon black (CB) exemplify observed data (actual, red squares connected by red solid lines, relative to red axes on the bottom and left; Table S11) for the tested CB doses of 0.1, 100, and 1000 mg kg⁻¹ dry soil at the final harvest. The black filled circle represents the observed control value (Table S11). Note the logarithmic scale of the x-axis for CB concentration. The whole-plant N2 fixation potential was reduced in all three CB treatments compared to the control but increased with CB concentration (r = 0.90, P < 0.001; Figure S6F). The gray solid line represents a sigmoidal dose–response curve for a toxicant that dissolves into the soil solution (with gray axes labeling on the top and right). This curve is based on data (gray triangles) generated by Vesper and CraigWeidensaul, showing the effect of dissolved metal toxicant, copper, on soybean whole-plant N2 fixation potential measured by the acetylene reduction assay. Two alternative very low dose–response regimes are hypothesized. Scenario 1 (blue long dashed line) represents a continuum of responses between the control value and the lowest CB dose (0.1 mg kg⁻¹) tested herein. Scenario 2 (green short dashed line) represents a threshold dose, possibly close to the lowest dose (0.1 mg kg⁻¹) tested herein, below which CB inhibition of N2 fixation potential is not observed but above which it is. Although there is uncertainty in the untested low concentration regimes, this study supports that negative impacts are partly mitigated by CB agglomeration at the highest dose (1000 mg kg⁻¹).