Cadmium accumulation and tolerance of two castor cultivars in relation to antioxidant systems

To investigate the effects of Cd on tolerance and antioxidant activities of castor, two different castor（Ricinus communis L.） cultivars（Zibo No. 5 and Zibo No. 8） were used for a hydroponic experiment（0, 1 and 2 mg/L Cd） and a pot experiment using Cd-contaminated soil（34 mg/kg） with the addition of ethylenedinitrilotetraacetic acid（EDTA）. The results indicated that there were significant differences between the two cultivars with respect to Cd uptake in shoots（113–248 mg/kg for Zibo No. 5 and 130–288 mg/kg Zibo No. 8）, biomass tolerance indexes（64.9%–74.6% for Zibo No. 5 and 80.1%–90.9% for Zibo No. 8） in the hydroponic experiment and survival rates（0% for Zibo No. 5 and 100% for Zibo No. 8）determined by the addition of EDTA in the pot experiment, suggesting that Zibo No. 8 has higher tolerance than Zibo No. 5. Moreover, the castor cultivars have low bioconcentration factors（4.80% for Zibo No. 5 and 5.43% for Zibo No. 8） and low translocation factors（〈1%）.Consequently, Zibo No. 8 can participate in Cd phytostabilization in highly Cd-polluted areas. The results indicated that glutathione（GSH） as a non-enzymatic antioxidant, and antioxidant enzymes including superoxide dismutase（SOD）, catalase（CAT） and guaiacol peroxidase（GPX）, were cultivar- and dose-dependent. The higher tolerance of Zibo No. 8compared with Zibo No. 5 can be attributed to the higher GSH levels in the root and higher GPX activity in the leaf.

Washing out heavy metals from contaminated soils from an iron and steel smelting site

Washing is a promising method for separating contaminants bound to the particles of soil ex-situ by chemical mobilization. Laboratory batch washing experi- ments were conducted using deionized water and varying concentrations of oxalic acid, citric acid, tartaric acid, acetic acid, hydrochloric acid and ethylenediaminetetra acetic acid （EDTA） to assess the efficiency of using these chemicals as washing agents and to clean up heavy metals from two heavily polluted soils from an iron and streel smelting site. The toxicity reduction index and remediation costs were analyzed, and the results showed that the soils were polluted with Cd, Pb and Zn. Hydrochloric acid and EDTA were more efficient than the other washing agents in the remediation of the test soils. The maximum total toxicity reduction index showed that 0.5 mol·L^-1 hydro- chloric acid could achieve the remediation with the lowest costs.

Training time minimization for federated edge learning with optimized gradient quantization and bandwidth allocation

Training a machine learning model with federated edge learning(FEEL)is typically time consuming due to the constrained computation power of edge devices and the limited wireless resources in edge networks.In this study,the training time minimization problem is investigated in a quantized FEEL system,where heterogeneous edge devices send quantized gradients to the edge server via orthogonal channels.In particular,a stochastic quantization scheme is adopted for compression of uploaded gradients,which can reduce the burden of per-round communication but may come at the cost of increasing the number of communication rounds.The training time is modeled by taking into account the communication time,computation time,and the number of communication rounds.Based on the proposed training time model,the intrinsic trade-off between the number of communication rounds and per-round latency is characterized.Specifically,we analyze the convergence behavior of the quantized FEEL in terms of the optimality gap.Furthermore,a joint data-and-model-driven fitting method is proposed to obtain the exact optimality gap,based on which the closed-form expressions for the number of communication rounds and the total training time are obtained.Constrained by the total bandwidth,the training time minimization problem is formulated as a joint quantization level and bandwidth allocation optimization problem.To this end,an algorithm based on alternating optimization is proposed,which alternatively solves the subproblem of quantization optimization through successive convex approximation and the subproblem of bandwidth allocation by bisection search.With different learning tasks and models,the validation of our analysis and the near-optimal performance of the proposed optimization algorithm are demonstrated by the simulation results.