Akbar SOLIEMANZADEHand Majid FEKRI

Department of Soil Science，College of Agriculture，Shahid Bahonar University of Kerman，Ker man 7718897111(Iran)

ABSTRACT Green iron nanoparticles(Fe NPs)can be a practical solution to combat iron(Fe)def iciency in calcareous agricultural soils.The main aim of the present work was to assess the effects of green Fe NPs on Fe availability in calcareous soils.For this purpose，green Fe NPs were synthesized using green tea(G-Fe NPs)，Shirazi thyme(T-Fe NPs)，walnut green hull(W-Fe NPs)，and pistachio green hull(P-Fe NPs)extracts and applied as a source of Fe fertilizer to sorghum(Sorghum bicolor L.Moench)plants.Results of X-ray diffraction(XRD)，scanning electron microscopy(SEM)，and dynamic light scattering(DLS)indicated that the green Fe NPs were amorphous in nature and the polyphenols obtained from plant-part extracts acted as both capping and reducing agents.Similar to the behavior of Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid(Fe-EDDHA)in calcareous soils，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs increased Fe release compared with the control and FeSO4 treatment.Cumulative Fe release data f itted well to the power function，intra-particle diffusion，and Elovich kinetic models.According to the pot experiment，the increment in soil Fe availability upon Fe-EDDHA and Fe NPs application led to an increase in Fe uptake，growth，and photosynthetic pigment contents of the sorghum plants.Although further research is needed to evaluate the residual effect and environmental impact of green Fe NPs，they may be an appropriate substitute for traditional Fe fertilizers in calcareous soils.

Key Words: chlorophyll content，Fe fertilizer，Fe release，green chemistry，natural polyphenols，sorghum

INTRODUCTION

Iron(Fe)is absolutely crucial for plant growth，owing to its involvement in vitally important processes such as electron transportation，synthesis of photosynthetic pigments，respiration，enzyme activation，and DNA production(Liet al.，2018).Although the solid phase of soils is enriched with Fe，only a negligible amount(＜1 mg L-1)of this element is available for plant uptake(Ruiet al.，2016).In calcareous soils with high pH and low organic matter content，the availability of Fe species is low，which leads to Fe def iciency chlorosis，a serious disorder that affects crop production(Soliemanzadeh and Mozafari，2014;Askaryet al.，2017).The yield and quality of agricultural products decrease as a result of Fe def iciency chlorosis，owing to the reduction in leaf chlorophyll content and net photosynthesis efficiency(Liet al.，2016).Various approaches，including protonation，chelation，and reduction have been frequently used to enhance Fe availability in soils.Among these techniques，the application of synthetic Fe chelates is effective in alleviating Fe chlorosis.However，chelating agents such as ethylenediaminetetraacetic acid(EDTA)，diethylenetriaminepentaacetic acid(DTPA)，and ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid(EDDHA)are often costly and pose a number of environmental risks.For example，the soil availability of heavy and radioactive metals can be increased by DTPA and EDTA application(Zhaoet al.，2011;Esringüand Turan，2012).

Recently，the application of nanomaterials as a source of Fe fertilizer has received special attention due to their unique characteristics，including size，surface area，and morphology(Raliyaet al.，2018).Wanget al.(2016)reported thatγ-Fe2O3nanoparticles(NPs)applied at an appropriate rate could effectively alleviate Fe def iciency chlorosis of watermelon plants and act as an Fe-fertilizer source in agriculture production.There have been further reports suggesting that nanomaterials such as zero-valent Fe，α-Fe2O3，γ-Fe2O3，and Fe3O4could replace traditional Fe fertilizers in agricultural activities(Almeelbi and Bezbaruah，2014;Liet al.，2018).

Green-synthesized zero-valent Fe NPs are one of the most highlighted nanomaterials of interest，with a wide versatility in environmental remediation due to their unique characteristics such as high surface area，eco-friendly and non-toxic nature，and cost effectiveness.Recently，renewable and sustainable polyphenols obtained from plant leaves，fruits，and stems have been applied to boost the production of green zero-valent Fe NPs(Soliemanzadehet al.，2016;Soliemanzadeh and Fekri，2017a).The polyphenols obtained from plant-part extracts can serve as both reducing and capping/stabilizing agents，which keep NPs dispersed in aquatic environments and prevent their further oxidation and agglomeration(Soliemanzadeh and Fekri，2017a).Several studies have recently demonstrated the use of green Fe NPs for removal of chromium(Soliemanzadeh and Fekri，2017a)，phosphorus(Soliemanzadehet al.，2016;Soliemanzadeh and Fekri，2017b)，arsenic(Poguberovićetal.，2016)，copper(Cu)(Wenget al.，2016)，and nitrate(Wanget al.，2014)from water media.Although much information is available on the behavior of green Fe NPs in aqueous media，few studies have investigated the fate of these nanomaterials in relatively complicated systems such as soils.Therefore，we used plant-part extracts，including green tea，Shirazi thyme，walnut green hull，and pistachio green hull，to develop green Fe NPs(G-Fe NPs，T-Fe NPs，W-Fe NPs，and PFe NPs，respectively)，which were subsequently used as a source of Fe fertilizer in calcareous soils.The objectives of the present study were to:i)prepare green Fe NPs using plant-part extracts and characterize them and ii)study Fe bioavailability in a calcareous soil with the green Fe NPs applied.

MATERIALS AND METHODS

Materials

All the chemicals used were of analytical grade(Merck Co.，Germany).To fabricate Fe NPs，the polyphenols extracted from green tea，Shirazi thyme，walnut green hull，and pistachio green hull were used as stabilizing and capping agents.

Preparation of green Fe NPs

The production of Fe NPs using plant-part extracts has already been reported previously(Soliemanzadehet al.，2016;Soliemanzadeh and Fekri，2017a，b).Brief ly，green tea，Shirazi thyme，walnut green hull，and pistachio green hull extracts were produced by heating the plant parts in distilled water(20 g L-1)at 80°C for 1 h，after which，the extracts were centrifuged and cooled down to 25—27°C.These extracts were added to 0.10 mol L-1FeSO4solution at a ratio of 1:1(volume/volume)，and G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs were generated.The black color of these Fe NPs indicated the reduction of Fe(II)to Fe(0).The pH of the f inal suspension was adjusted to 6 by adding 0.01 mol L-1NaOH.The analysis of the residual contents of Fe in G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs indicated that only about 35%，33%，33%，and 27%，respectively，of total Fe was reduced to the elemental state.The Fe NPs were synthesized at room temperature(25—27°C)and dried for composition and size analyses using X-ray diffraction(XRD)(Philips X’pert Pro MPD，Philips，Eindhoven Co.，The Netherlands)，scanning electron microscopy(SEM)(KYKY EM-3200，KYKY Technology Co.，China)，and dynamic light scattering(DLS)(ZEN3600，Malvern Co.，UK).

Experiment of Fe release kinetics

Surface soil samples(0—20 cm)were collected from the Mahan area，located 35 km southeast of Kerman，Iran.The soil samples were air-dried，crushed，and sieved(2 mm)，following which，their chemical and physical properties were determined according to standard methods(Sparks，1996).Soil pH was determined in a saturated paste using a pH meter(Jenway，UK).Cation exchange capacity was determinedviasubstitution of NaOAc by NH4OAc at pH 7.To quantify available Fe，20 g soil was extracted with 20 mL 0.005 mol L-1DTPA(containing 0.1 mol L-1triethanolamine and 0.01 mol L-1CaCl2)at pH 7.3 while shaking for 120 min(Sparks，1996).The concentration of Fe was then determined using a f lame atomic absorption spectrometer(AA240，Varian Co.，Australia).Soil organic matter was measured using the wet oxidation method.Selected soil characteristics are reported in Table I.

TABLE I Selected physicochemical characteristics of the soil collected from the Mahan area，located 35 km southeast of Kerman，Iran and used for the Fe release experiment

To investigate Fe release kinetics，FeSO4，Fe-EDDHA，G-Fe NPs，W-Fe NPs，P-Fe NPs，and T-Fe NPs were mixed with 200 g soil at 10 mg Fe kg-1soil and incubated at 23—27°C near f ield water capacity for 150 d.On days 1，2，4，8，30，60，and 150，soil available Fe was determined.At the end of incubation，the soils were air-dried and ground.One gram of the soils was weighed into 15-mL conical falcon tubes，and then 10 mL 0.005 mol L-1DTPA was added to each tube.Afterward，the mixtures were shaken at 250 r min-1for 30，60，120，240，480，960，1 440，2 880，and 4 320 min，followed by centrifugation at 4 000 r min-1for 15 min.The Fe concentration in the supernatant was measured by f lame atomic absorption spectrometry(AA240，Varian Co.，Australia).Finally，popular kinetic models such as power function(Eq.1)，pseudo f irst-order(Eq.2)，pseudo second-order(Eq.3)，Elovich(Eq.4)，and intra-particle diffusion(Eq.5)were f itted to the obtained data to describe the process of Fe release in the soil(Sparks，2003).

whereqtis the amount of Fe released at timetin mg Fe kg-1;ais initial Fe release rate constant in mg Fe kg-1min-1;andbis the Fe release rate constant in(mg Fe kg-1)-1.

whereqmaxis the maximum amount of Fe released at the timescale of the experiment in mg Fe kg-1andk1is the pseudo f irst-order constant in s-1.

wherek2is the pseudo second-order constant in(mg Fe kg-1)-1.

whereαis the initial Fe release rate constant in mg Fe kg-1s-1andβis the Fe release rate constant in(mg Fe kg-1)-1.

wherekpis the Fe release rate constant in(mg Fe kg-1)-0.5andCis the initial Fe release rate constant in mg Fe kg-1.

The goodness of f it of the kinetic models was estimated by coefficient of determination(R2)and standard error of estimate(SEE)calculated using the following equation:

whereqis the measured Fe release;q′is the predicted Fe release;andnis the number of measurements.

Pot experiment

Similar to the Fe release experiment，2.5 kg soil were mixed with FeSO4，Fe-EDDHA，G-Fe NPs，W-Fe NPs，PFe NPs，and T-Fe NPs at 10 mg Fe kg-1soil，placed in polyethylene pots，and incubated at 23—27°C near f ield water capacity for f ive months.At the end of the incubation，six seeds of sorghum(SorghumbicolorL.Moench)were sown in each pot and thinned to four plants 10 d after emergence.The seedlings were watered to maintain the soil moisture near f ield capacity.After 10 weeks，the seedlings were harvested and plant weights in each pot were recorded.For chlorophyll determination，0.2 g fresh leaves were mixed with 10 mL acetone(80%)and the mixture was centrifuged at 4 000 r min-1for 15 min.The absorbance(A)of supernatants was recorded spectrophotometrically at 470，647，and 663 nm(Mackinney，1941).Finally，the amounts of chlorophylla(Chla)，chlorophyllb(Chlb)，total chlorophylls(Chl t)，and carotenoids(Car)were calculated using the following equations(Lichtenthaler，1987):

whereA663andA647are the absorbance at 663 and 647 nm，respectively.

whereA470is absorbance at 470 nm.

To measure plant Fe content，root and shoot samples were dried at 75°C for 78 h，weighed，ground，dry-ashed at 550°C，and digested with 2.5 mol L-1HC1.The concentration of Fe in the extracts was determined using f lame atomic absorbance spectrometry(AA240，Varian Co.，Australia).The amount of plant Fe uptake was obtained as the product of Fe concentration and plant dry biomass.At the end of the growth period，soil Fe availability in each pot was determined using the DTPA method.

The pot experiment adopted a completely randomized design with three replicates.One-way analysis of variance(ANOVA)was performed using SAS 9.1 and the statistical differences between means were evaluated by Duncan’s multiple range test(P＜0.05).

RESULTS AND DISCUSSION

Characterization of the green Fe NPs

Based on the XRD patterns of G-Fe NPs，T-Fe NPs，and W-Fe NPs(Fig.1)，the intensity peaks at 23.85°，27.45°，44.9°，and 17.6°were assigned to goethite，magnetite，zerovalent Fe，and plant polyphenols，respectively.As the green zero-valent Fe NPs were poorly crystalline in nature，Fe oxides/hydroxides were also present(Soliemanzadehet al.，2016).However，P-Fe NPs were defective in the distinguished diffraction peaks(Fig.1)，indicating that the particles were amorphous in nature and not crystalline(Wanget al.，2014).Moreover，the sharp peak at 17.6°was associated with the polyphenols in plant-part extracts，known as capping and reducing factors in the green production of Fe NPs(Soliemanzadehet al.，2016).

Fig.1 X-ray diffraction(XRD)patterns of the green iron nanoparticles(Fe NPs)synthesized using green tea(G-Fe NPs)，walnut green hull(W-Fe NPs)，Shirazi thyme(T-Fe NPs)，and pistachio green hull(P-Fe NPs).

The SEM images of G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs are illustrated in Fig.2，showing that the green Fe NPs were universally spherical.Similar results have been reported by Soliemanzadehet al.(2016)and Wenget al(2016).Size distribution of the NPs was estimated by DLS and the size order was found to be:G-Fe NPs＜W-Fe NPs＜P-Fe NPs＜T-Fe NPs(Fig.3).

Fig.2 Scanning electron microscopy(SEM)images of the green iron nanoparticles(Fe NPs)synthesized using green tea(G-Fe NPs)(a)，walnut green hull(W-Fe NPs)(b)，Shirazi thyme(T-Fe NPs)(c)，and pistachio green hull(P-Fe NPs)(d).

Fig.3 Particle size distributions of the green iron nanoparticles(Fe NPs)synthesized using green tea(G-Fe NPs)，Shirazi thyme(T-Fe NPs)，walnut green hull(W-Fe NPs)，and pistachio green hull(P-Fe NPs).

Soil available Fe changes during incubation

As can be seen from Fig.4，soil available Fe in the treatments with Fe-EDDHA and the green Fe NPs was higher than that in the control and FeSO4treatment throughout the incubation period.Soil available Fe in the control and FeSO4treatment was 1.05—1.53 and 1.00—1.85 mg kg-1，respectively，indicating the inefficiency of FeSO4as an Fe source in calcareous soils(Garcia-Minaet al.，2003).Available Fe in the Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs treatments decreased with time from 12.53 to 8.85，12.62 to 7.73，12.47 to 5.58，12.21 to 6.13，and 11.99 to 4.52 mg kg-1，respectively，while that in the control and FeSO4treatment increased from 0.87 to 1.53 and 1.00 to 1.85 mg kg-1，respectively.Throughout the incubation period，the available Fe in the treatments with Fe-EDDHA and green Fe NPs were higher than 4.50 mg kg-1.The stable Fe-chelates，such as EDDHA and its isomers，are able to the increase Fe solubility in calcareous soils and transport Fe from the solid phase(poorly crystalline Fe mineral，ferrihydrite，and insoluble Fe-humic complexes)to the solution and plant root surfaces(Lucena，2006).Similar to Fe-EDDHA，the green Fe NPs showed considerable potential for maintaining Fe in the soil solution.According to the results obtained from SEM and XRD，the green zero-valent Fe NPs were amorphous in nature，and the plant-part polyphenols limited their crystal growth and oxidation.However，it seems that some characteristics of the natural polyphenols，including oxidation/reduction，acidif ication，and chelation reactions，might improve Fe mobility in calcareous soils(Colomboet al.，2014).The results of the present study were in agreement with the f indings of Schmidtet al.(2017)，who reported that the high mobility of Fe in the soils with the presence of natural polyphenols could be related to the reduction of less available Fe(III)to bioavailable Fe(II)and Fe-polyphenol complexes.

Fig.4 Available Fe changes in a calcareous soil with addition of various Fe sources(10 mg Fe kg-1 soil)during the 150-d incubation at 23—27°C near f ield water capacity.Control=no Fe source was added;FeSO4=FeSO4 was added;Fe-EDDHA=Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid was added;G-Fe NPs=iron nanoparticles(Fe NPs)synthesized using green tea was added;T-Fe NPs=Fe NPs synthesized using Shirazi thyme was added;W-Fe NPs=Fe NPs synthesized using walnut green hull was added;P-Fe NPs=Fe NPs synthesized using pistachio green hull was added.

Kinetics of Fe release

The Fe release patterns in the treatments are shown in Fig.5.The process of Fe release occurred fast initially and then proceeded slowly，similar to the results of Yuanet al.(2017)for Fe and Olamaet al.(2010)for Cu.

Fig.5 Iron release kinetics in a calcareous soil with addition of various Fe sources after 150 d of incubation at 23—27°C near f ield water capacity.Control=no Fe source was added;FeSO4=FeSO4 was added;Fe-EDDHA=Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid was added;G-Fe NPs=iron nanoparticles(Fe NPs)synthesized using green tea was added;T-Fe NPs=Fe NPs synthesized using Shirazi thyme was added;W-Fe NPs=Fe NPs synthesized using walnut green hull was added;P-Fe NPs=Fe NPs synthesized using pistachio green hull was added.

The successive phases of rapid and slow release of Fe could also be attributed to the exchangeable Fe and the sluggish process of dissolution of amorphous iron oxides，respectively(Yuanet al.，2017).The results showed that the rate of Fe release from the soils treated with Fe-EDDHA and the green Fe NPs was higher than that from the control，whereas the difference between the FeSO4treatment and the control was not signif icant(Fig.5).Over the timescale of this study，the cumulative Fe release from the control and the FeSO4，Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs treatments was 13.00，14.50，42.40，36.20，26.20，31.30，and 21.30 mg kg-1，respectively.Application of Fe-EDDHA to calcareous soils increases diffusion gradient，which promotes the transfer of Fe from the solid phase to the solution due to the high reaction affinity of EDDHA for Fe compared with other cations such as calcium，magnesium，and potassium(Lindsay，1995).Lucenaetal.(1987)reported that the addition of Fe-EDDHA to calcareous soils increased the total soluble Fe as compared with Fe-EDTA and Fe-DTPA after 43 d of incubation.This result was similar to that reported by Garcia-Minaet al.(2003)，who observed that the water-extractable Fe increased when Fe-EDDHA was applied，rather than FeSO4，Fe-EDTA，and Fe-DTPA.Although the highest amount of Fe release was observed in Fe-EDDHA，the rate of Fe release increased signif icantly upon the application of the green Fe NPs(especially GFe NPs and W-Fe NPs)compared with the control and FeSO4 treatment.As for the treatments with green Fe NPs，the slow release phase might be due to the dissolution of amorphous nano-oxides and oxidation-reduction processes，including electron transfer from plant polyphenols to Fe in the surface complexes，the oxidation of natural polyphenols，and the diffusion of soluble Fe into the soil solution.However，the exchangeable Fe and soluble Fe-polyphenol complexes apparently played a signif icant role in the rapid phase of Fe release.A similar interaction between Fe and natural polyphenols has been previously reported(Schmidtet al.，2017;Yuanet al.，2017).Based on Fig.5，the order of NPs in terms of Fe release was as follows:G-Fe NPs＞W-Fe NPs＞T-Fe NPs＞P-Fe NPs，which could be related to their sizes and the type of natural polyphenols extracted.

According to theR2and SE values presented in Table II，the cumulative Fe release data were better described by the power function，intra-particle diffusion，and Elovich equations，compared with the pseudo f irst-order and pseudo second-order kinetics.The current results are similar to those reported by Safarzadehetal.(2018)for Fe，Yuanetal.(2017)for Fe，aluminum，manganese，and silicon，and Olamaet al.(2010)and Sadeghet al.(2012)for Cu.The good f it of the power function model to describe the release kinetics of Fe from the control and treated soils may be due to the f lexibility of its empirical coefficients，includingaandb(Yu and Klarup，1994;Safarzadehet al.，2018).Based on the intra-particle diffusion model(Fig.6)，the relationship betweenqtandt0.5for all treatments was linear over the entire period，indicating that the diffusion of Fe ions，Fechelates，and Fe-polyphenol complexes played an important role in the release of Fe from the solid phase to solution.

Fig.6 Intra-particle diffusion plot for Fe release from a calcareous soil in treatments with addition of various Fe sources after 150 d of incubation at 23—27°C near f ield water capacity.Control=no Fe source was added;FeSO4=FeSO4 was added;Fe-EDDHA=Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid was added;G-Fe NPs=iron nanoparticles(Fe NPs)synthesized using green tea was added;T-Fe NPs=Fe NPs synthesized using Shirazi thyme was added;W-Fe NPs=Fe NPs synthesized using walnut green hull was added;P-Fe NPs=Fe NPs synthesized using pistachio green hull was added.

TABLE II Kinetic parameters of selected models for the release of Fe from a calcareous soil in treatments a)with addition of various Fe sources after 150 d of incubation at 23—27°C near f ield water capacity

In the Elovich equation，the values ofβfor the treated soils were higher than those for the control and FeSO4treatment，whereas application of Fe-EDDHA and Fe NPs led to a reduction inαvalue compared with the control and FeSO4treatment(Table II).Moreover，the reduction inαand increment inβwere indicative of an increase in the rate of Fe release from soil(Olamaet al.，2010;Islas-Espinozaet al.，2014).

Iron phytoavailability

The root and shoot dry weights of the sorghum seedlings increased signif icantly in the pots treated with Fe-EDDHA and the Fe NPs compared with the control and pots treated with FeSO4(Table III).The application of Fe NPs increased the shoot dry weight by an average of 18%compared with the FeSO4treatment.A similar result was reported in spinach，Catharanthus roseus，and wheat，where the weights of the plants increased signif icantly upon the application of chemical nanoscale zero-valent Fe，Fe2O3NPs，and biochar-Fe3O4NP complexes，respectively(Almeelbi and Bezbaruah，2014;Josephet al.，2015;Askaryet al.，2017).Ruiet al.(2016)reported that the application of Fe2O3NPs could effectively increase the dry biomass，height，number of branches，and root length of peanut seedlings.However，they suggested that Fe2O3NPs could be used as an Fe-fertilizer source in agricultural systems.Our f inding concerning the growth parameters of sorghum was similar to the results reported by Maet al.(2013)，who showed that theTyphagrowth parameters such as wet weight，dry weight，and shoot length were enhanced by the application of 25 and 50 mg L-1nanoscale zero-valent Fe.In the growth media of quartz sand，40 and 80μmol L-1of nanoscale zero-valent Fe particles，which were prepared based on the borohydride reduction method，improved the shoot and root growth of the peanut seedlings better than Fe-EDTA did(Liet al.，2015).

Compared with the control，10 mg kg-1Fe as FeSO4yielded no signif icant differences in plant Fe content，whereas the root Fe contents of the treatments with Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs were signif icantly higher by 79%，91%，81%，80%，and 56%，respectively，and the shoot Fe contents increased by 74%，78%，71%，52%，and 39%，respectively(Table III).Root Fe uptake in the G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs treatments increased by 77%，60%，68%，and 45%，respectively，relative to the FeSO4treatment，while the shoot Fe uptake increased by 127%，127%，97%，and 83%，respectively.Similarly，an increase in plant Fe content upon exposure to Fe NPs such as magnetite(Parionaet al.，2017;Sebastianet al.，2018)，carbon-bound Fe oxide compounds(Sebastianet al.，2017)，hydr(oxides)-humic substances(Kulikovaet al.，2017)，and Fe oxide-oxalate systems(Dasetal.，2016)has been reported previously.The increased uptake of Fe by sorghum seedlings in the presence of green Fe NPs might be due to the Fepolyphenol complexes and the reduction of Fe(III)to Fe(II)，which enhanced the plant available Fe in the soil.

The translocation factor(TF)of Fe，calculated as a ratio of Fe concentration in shoot to that in root，in the G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs treatments was signif icantly lower than that in the control，FeSO4，and Fe-EDDHA treatments，indicating that the precipitation of Fe oxides-hydroxides in the plant roots likely decreased the transfer of Fe ions from the root to the shoot.These f indings were in agreement with the observations of Liet al.(2018)，who studied the transfer ofα-Fe2O3，γ-Fe2O3，and Fe3O4NPs inCitrusmaximaseedlings and reported that a smaller amount of Fe in the form of NPs could penetrate into the shoot tissues in comparison with that in the form of FeCl3.

TABLE III Root and shoot weights and Fe content，uptake，and translocation factor(TF)of sorghum seedlings in the treatments with addition of various Fe sources at 10 mg Fe kg-1 soil in a pot experiment using a calcareous soil

The effects of the FeSO4，Fe-EDDHA，and green Fe NPs treatments on the contents of Chla，Chlb，Chl t，and Car are shown in Fig.7.The treatments with Fe-EDDHA and Fe NPs improved Chl t signif icantly(P＜0.05)，which ranged from 2.1 mg g-1fresh weight(FW)for the FeSO4treatment to 4.0，3.8，3.4，3.4，and 3.0 mg g-1FW for the treatments of Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs，respectively.However，the Car content was enhanced by 80%，110%，40%，40%，and 30%in the Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs treatments，respectively.Although the highest amount of Chl t was observed in the Fe-EDDHA treatment，there was no difference between the G-Fe NPs and Fe-EDDHA treatments.Thus，G-Fe NPs may be a suitable substitution for Fe-EDDHAin calcareous soils，as the evidence demonstrated that G-Fe NPs at the rate of 10 mg kg-1could increase the photosynthetic pigments in comparison with the control and FeSO4treatment.It is well known that Fe def iciency disrupts the synthesis of photosynthetic pigments such as Chlaand Chlb，which manifests as chlorotic symptoms in young plant leaves，severely decreasing plant yield(Ruiet al.，2016).Based on the results of the current experiment，it appears that an increase in soil Fe availability can justify the enhanced Fe content in plants，the increased chlorophyll content，and the improved growth of the treated plants.

Fig.7 Photosynthetic pigment contents of sorghum seedlings in the treatments of various Fe sources in a pot experiment using a calcareous soil.Different letters above the bars of a same pigment indicate signif icant differences between treatments based on Duncan’s multiple range test(P＜0.05).Chl a=chlorophyll a;Chl b=chlorophyll b;Chl t=total chlorophylls;Car=carotenoids.Control=no Fe source was added;FeSO4=FeSO4 was added;Fe-EDDHA=Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid was added;G-Fe NPs=iron nanoparticles(Fe NPs)synthesized using green tea was added;T-Fe NPs=Fe NPs synthesized using Shirazi thyme was added;W-Fe NPs=Fe NPs synthesized using walnut green hull was added;P-Fe NPs=Fe NPs synthesized using pistachio green hull was added.

Soil available Fe before and after planting in the treatments with FeSO4，Fe-EDDHA，and the green Fe NPs is shown in Fig.8.Soil available Fe in the soils treated with Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs was 5.2，5.0，3.0，3.3，and 2.4 times，respectively，higher than that in the FeSO4treatment before planting，and it was 4.7，4.6，3.3，3.4，and 2.3 times，respectively，higher after planting.Compared with the Fe-EDDHA treatment，the G-Fe NPs treatment did not have a signif icantly different Fe availability，whereas Fe availability upon the application of W-Fe NPs，PFe NPs，and T-Fe NPs decreased signif icantly by 27%，48%，and 32%，respectively.Before and after planting，the highest available Fe in the soils was observed in the Fe-EDDHA treatment，indicating that Fe-chelate complex was very stable and enhanced the solubility and diffusion of Fe species from the solid phase to the soil solution(Lindsay，1995).Based on the XRD patterns，the green Fe NPs were poorly crystalline in nature，and their potential for use as Fe fertilizers might be related to Fe-polyphenol complexes，electron transfer in the surface complexes，polyphenol oxidation，reduction of Fe(III)to Fe(II)，and diffusion of Fe ions in the soil solution.There were no signif icant differences in soil Fe availability before and after planting for any of the treatments(Fig.8)，ref lecting that the green Fe NPs(especially G-Fe NPs)could be an inexpensive，promising，and eco-friendly source of fertilizer to diminish Fe chlorosis in crops cultivated in calcareous soils suffering from Fe def iciency.

Fig.8 Soil available Fe before and after sorghum planting in the treatments of various Fe sources in a pot experiment using a calcareous soil.Different letters indicate signif icant differences between treatments for a same time period(before or after planting)based on Duncan’s multiple range test(P＜0.05).Control=no Fe source was added;FeSO4=FeSO4 was added;Fe-EDDHA=Fe-ethylenediamine-N，N′-bis(2-hydroxyphenyl)acetic acid was added;G-Fe NPs=iron nanoparticles(Fe NPs)synthesized using green tea was added;T-Fe NPs=Fe NPs synthesized using Shirazi thyme was added;W-Fe NPs=Fe NPs synthesized using walnut green hull was added;P-Fe NPs=Fe NPs synthesized using pistachio green hull was added.

Correlations

The correlation coefficients between the kinetic equation parameters and the results of the pot experiment are listed in Table IV.The amount of cumulative released Fe(qexp)was positively correlated with the shoot FW，plant Fe content，chlorophyll contents，and carotenoids content.However，thea，β，andkpparameters related to the Fe release rate indicated signif icant and positive relationships with plant yield，plant Fe content，chlorophyll contents，and carotenoids content，whereas thebandαparameters related to the initial Fe release rate showed negative relationships.These observations were in agreement with those made by Shariatmadariet al.(2006)and Brar and Vig(1988)，who reported a signif icant relationship between the cumulative desorbed phosphorus from calcareous soils and its uptake by wheat seedlings.They stated that desorption of phosphorus from the soils had a remarkable effect on plant yield.These f indings propose that the rate of Fe release restricted the amount of Fe uptake by the sorghum seedlings.In addition，our results support the view that the rate of replenishment of the soil solution with Fe is important in determining the Fe uptake of plants in comparison with the equilibrium Fe concentration in the solution.

TABLE IV Pearson correlation coefficients between kinetic equation parameters and pot experiment data a)

CONCLUSIONS

To our knowledge，our study is the f irst to investigate the application of green Fe NPs instead of traditional Fe fertilizers(FeSO4and Fe-EDDHA)to enhance Fe availability in calcareous soils.We also assessed the impacts on sorghum growth.The results indicated that the application of Fe-EDDHA，G-Fe NPs，T-Fe NPs，W-Fe NPs，and P-Fe NPs increased the release of Fe compared with the control and FeSO4treatment.In addition，cumulative Fe release data were better described by the power function，intra-particle diffusion，and Elovich equations than by the pseudo f irstorder and pseudo second-order kinetics.Compared with the control and FeSO4treatment，the application of Fe-EDDHA and the green Fe NPs signif icantly enhanced the growth，Fe uptake，and photosynthetic pigment contents of the sorghum seedlings.According to the XRD patterns and SEM images，the nanoscale zero-valent Fe synthesized using green tea，Shirazi thyme，walnut green hull，and pistachio green hull extracts was poorly crystalline in nature，and its potential for use as Fe fertilizer might be attributed to Fe-polyphenol complexes，electron transfer in the surface complexes，polyphenol oxidation，reduction of Fe(III)to Fe(II)，and diffusion of Fe ions in the soil solution.The green Fe NPs(especially G-Fe NPs and T-Fe NPs)might be a suitable substitute for the regular Fe-chelate fertilizers in calcareous soils，although further investigations are needed to evaluate their residual effects and environmental impacts.