| 16soil samples from agricultural field in China;soil, plant, water and human hairsamples from Ziyang seleniferous region, soil and plant samples from Ping’an-Ledu adequateselenium(Se) region in Qinghai were collected in this study. Total Se content were analyzedin all the samples. A five-step sequential extraction procedure was used for Se fractionation,including soluble Se, exchangeable Se and carbonate-bound Se, iron and manganeseoxides-bound Se, organic matter-bound Se, and the residual Se fraction. Species of soluble Sein upland soils included Se2-, Se4+, and Se6+. The relationship between Se fractions andspeciations, and soil properties and Se bioavailability were analyzed. The transformation ofSe fractions in soil from Ziyang seleniferous region and calcareous soil treated with additionalselenite and selenate was also investigated. The results were shown as below:1) Total Se content in16agricultural soils averaged at292μg/kg,60%of soil samplesfrom Ping’an-Ledu region had adequate Se content, and soils from Ping’an county was themost abundant in Se; Wheat and corn samples had more Se than the other crops, and garlicand carrot had more Se than the other vegetables. Total soil Se content in Ziyang seleniferousregion was generally more than5mg/kg with the maximum33.4mg/kg, corn and humanhairs Se content averaged at1.89mg/kg and11.6mg/kg, respectivley. Se content in mostwater samples was less than10μg/L, while in some samples it reached up to40μg/L,irrigation water had generally high Se content (no less than5μg/L). The bioconcentrationfactor (BCF=Seplant/Sesoil) was0.08in Ziyang corn samples, and was0.20in wheat,0.15incorn carrot and garlic,0.05in horse bean and benne, respectively, which incidated the low Sebioavalability in these soils2) Organic matter-bound Se and residual Se were the dominant Se fractions in16agricultural soils and soils from Pingan-Ledu region, exchangeable Se and carbonate-boundSe content was genrally higher than iron and manganese oxides-bound Se, Soluble Se contentwas the lowest. Se2-content was the highest both in soluble, exchangeable Se andcarbonate-bound Se fractions in soil from Qinghai, while Se6+was the lowest. In soil formZiyang, Soluble Se, exchangeable Se and carbonate-bound Se fractions(3-fold higher thansoluble Se) accounted for less than1%of the total Se in the upland soil, but approximately16.1%in the paddy soil; Concentrations of residual Se were lower than those of iron and manganese oxides-bound Se and organic matter-bound Se in both upland and paddy soils;Iron and manganese oxides-bound Se was the dominant fractions in upland soil, whereasorganic matter-bound Se abounded in paddy soil. Se2-and Se6+were the main soluble Sespecies in upland soil from Ziyang; Se4+was the only Se species in exchangeable Se andcarbonate-bound Se fraction in most samples.3) Organic matter, Clay, pH and amorphous iron had significant correlation with soluble,while most soil physico-chemical properties had no significant correlation with other four Sefractions. Amorphous iron had significant correlation with all Se fractions, while organicmatter had no significant correlation with organic matter-bound Se. Path analysis showed thatorganic matter negatively affected iron and manganese oxides-bound Se, while positivelyaffected other species. Amorphous iron showed no effect on soluble Se, but it did strongereffect on the other four Se species than organic matter did. Except for soluble Se, organicmatter and amorphous iron had the strongest impact on organic matter-bound Se and residualSe, respectively. pH and clay influenced Se speciation via their negative or positive effect onorganic matter and amorphous iron, respectively.4) Se concentration in corn tissues was significantly correlated with the Se6+concentration in soil soluble, exchangeable and carbonate-bound, and organic matter-boundfractions, but not with other Se fractions. Path analysis indicated that soluble Se6+significantly affected Se accumulation in corn tissues directly, whereas the organicmatter-bound Se had a significant indirect effect. Soluble Se6+had greatest effect on Seaccumulation in corn leaf, and smallest in corn root. While exchangeable andcarbonate-bound Se4+had greatest effect on Se accumulation in corn root and seed, andsmallest in corn leaf. Monod model fitting results showed that estimated maximal seleniumcontents of pakchoi shoots and roots in selenate-treated soils were5times and2times largerthan in selenite-treated soils. In selenite-treated soils, estimated maximal selenium content ofbokchoy roots was2.3times larger than that in shoots. While in selenate-treated soils,estimated maximal selenium contents of bokchoy shoots and roots were approximately thesame.5) After removal of soluble, excheangeable and carbonate-bound Se by phosphatesolution in soil from Ziyang, Se in other fractions would transform to the two fractions. Thesum of these two fractions were approximately64%of the content before removal (content ofsoluble Se was higher than before while excheangeable and carbonate-bound Se was lower),and had significant correlation with preceding organic matter-bound Se. Among the Sespecies that could be detected, amount of soluble Se4+was the highest (4.4times higher thanbefore), followed by exchangeable and carbonate-bound Se4+(23%of the content before), with the lowest soluble Se6+(the same as before). In selenite and selenate treated soil, soil Utsvalue increased while IRvalue decreased, and this phenomenon was the most obvious inselenate treated soil. Proportion of iron and manganese oxides-bound Se remained the samewith time increase in2.5-20mg/kg selenite treated soil. While soluble, exchangeable andcarbonate-bound Se decreased and organic matter-bound and residual Se increased. In1-5mg/kg selenate treated soil, Proportion of soluble Se decreased dramatically at the beginning66days while then had no significant change. By contrast, the other four Se fractionsincreased dramatically at the beginning66days while then had no significant change. In40-60mg/kg selenite and10-40mg/kg selenate treated soil, transformation mainly occurredbetween soluble Se and exchangeable and carbonate-bound Se.2.5and5.0mg/kg selenatetreated reached to the equilibrium state more quickly than selenite treated soil. With the timeincrease, soil Uts values decreased more obviously in selenate treated soil than selenite treatedsoil, while it might increase in40-60mg/kg selenite treated soil. Soil IRvalues increased moreobviously in selenate treated soil than selenite treated soil, while it remained the same whenhigh content selenate added to the soil. |