Efficiency of soil-applied 67Zn-enriched fertiliser across three consecutive cro

Edson M.MATTIELLO,Eduardo L.CANCELLIER,Rodrigo C.DA SILVA,Fien DEGRYSE,Roslyn BAIRD and Mike J.MCLAUGHLIN

2Department of Soil Science,Universidade Federal de Lavras,Minas Gerais(Brazil)

3The University of Adelaide,School of Agriculture,Food and Wine,Fertiliser Technology Research Centre,PMB 1,Waite Campus,Glen Osmond SA 5064(Australia)

ABSTRACT A very small amount of applied zinc(Zn)is taken up by crops,resulting in low recovery by plants.Adding elemental sulphur to zinc oxide(ZnO)fertiliser could improve Zn solubilisation and exert a higher residual effect on crops than soluble Zn sources.We produced an isotopically labelled Zn-elemental sulphur fertiliser and evaluated its performance in comparison to traditional Zn sources during sequential crop cultivation.Three 67Zn-labelled fertilisers,ZnO,zinc sulphate(ZnSO4),and ZnO co-granulated with elemental sulphur(ZnOS0),were soil applied,and their contributions to the uptake of Zn by three consecutive crops,wheat,ryegrass,and corn,were assessed in a 294-d pot experiment.The contributions of Zn fertilisers followed the order:ZnSO4＞ZnO=ZnOS0.The relative contributions of Zn fertilisers were lower in the first crop than in the subsequent crops.The overall recovery of applied Zn by the three crops was higher for ZnSO4 than for ZnO and ZnOS0,reaching 1.56%,0.45%,and 0.33%of the applied Zn,respectively.Zinc recovery by plants was very low,regardless of the source of Zn.Adding elemental sulphur to ZnO did not increase its effectiveness up to 294 d after application.Fertiliser contribution was higher for the subsequent crops than for the initial crop,indicating the importance of assessing the residual effects of Zn fertilisers.

Key Words:agronomic biofortification,elemental sulphur,fertilizer residual effect,plant uptake,Zn bioavailability,Zn fertilization,Zn source

INTRODUCTION

Approximately half of the agricultural soils destined for cereal cultivation are zinc(Zn)deficient(Alloway,2008).Under such conditions,plant growth and productivity are negatively affected and the low Zn concentration in grains reduces the nutritional value of foods.Applying Zn fertilisers to soil and using foliar sprays have been common practices for correcting plant Zn deficiency,thereby increasing productivity and/or food quality(Liuet al.,2017;Cakmak and Kutman,2018).

Zinc applications vary in their efficiency,depending on the source,soil condition,application method,and crop system.Although many Zn compounds are used as fertilisers,zinc sulphate(ZnSO4)and zinc oxide(ZnO)are the most common sources applied.Soluble Zn fertilisers increase the concentration of Zn in soil solution rapidly and are useful for correcting plant Zn deficiency.However,several studies have shown a decrease in the availability of applied Zn(as ZnSO4)over time(Brennan,1990;Brennan and Gartrell,1990).Its effectiveness may be reduced by its adsorption to,and precipitation in,soil.Adsorption of Zn to soil occurs mainly on aluminium-iron oxides(Casagrandeet al.,2008),while both adsorption and precipitation occur in neutral and alkaline soils(Mesquita and Vieira e Silva,1996;Karimian and Moafpouryan,1999).Moreover,organic matter(OM)is an important adsorbent for Zn,especially in soil at pH below 7,and determines Zn bioavailability in soil(Fanet al.,2016).On the other hand,ZnO is a water-insoluble source of Zn,and its solubilisation and plant uptake depend on soil pH.When ZnO is mixed with soil,it tends to be quickly transformed into its soluble forms,especially at a soil pH below 7,increasing the Zn concentration in soil solution and behaving in a similar manner to water-soluble Zn fertilisers(McBeath and McLaughlin,2014).In contrast,ZnO is usually less efficient than ZnSO4when applied as a band application to high-pH soils,because of its slower solubilisation(Mortvedt,1992;McBeath and McLaughlin,2014).Associations with arbuscular mycorrhizal fungi have been suggested as an alternative for facilitating plant Zn uptake from soil(Watts-Williamset al.,2017;Watts-Williams and Cavagnaro,2018;Coccinaet al.,2019),indicating the importance of the residual effect of Zn applied to soil.

Addition of elemental sulphur(S)to ZnO may promote solubilisation of ZnO.As an acid-producing process,elemental S oxidation in soil causes a lower pH around the granule than in the surrounding soil(Mattielloet al.,2017).Furthermore,ZnO and elemental S-based fertilisers have higher nutrient concentrations and lower costs compared to traditional soluble sources of S or Zn.However,elemental S oxidation in granular fertilisers is a slow process,and the effect on ZnO solubilisation may manifest slowly.

Many studies have demonstrated yield improvements with the application of Zn fertilisers to Zn-deficient soils.However,only a small amount of the applied Zn is taken up by crops,resulting in a low apparent recovery of the applied Zn(Zhaoet al.,2011;Luet al.,2012;McBeathet al.,2013).Most of these studies have focused on only a single crop.However,Zn has a high residual availability(Brownet al.,1964;Boawn,1973;Brennan,1990).Therefore,it is important to assess the long-term effects in order to evaluate the fertilisers,particularly slow-release fertilisers,which may show poor performance in the first crop,but improve in consecutive crops compared to a soluble fertiliser.Understanding Zn availability in soil after Zn fertiliser application is essential for developing efficient crop management systems and increasing soil fertility.Short-and long-term plant availability of applied Zn must be understood as part of a balanced approach to Zn fertilisation under different soil conditions,allowing increased food production in a sustainable and responsible manner.

The objective of this study was to evaluate the effectiveness and residual effects of Zn fertilisers in correcting Zn deficiency in a sequential crop(wheat-ryegrass-corn)cultivation.We evaluated the performance of a co-granulated Zn-elemental S fertiliser in providing Zn to plants,compared to traditional Zn sources.Using Zn stable isotope-enriched fertilisers enabled evaluation of the contribution of Zn fertiliser to plant Zn uptake and obtaining of Zn fertiliser recovery rates over time.

MATERIALS AND METHODS

Fertilisers enriched with Zn stable isotopes

We mixed 20 mg(element weight)67ZnO(89.6%67Znenriched metal,Novachem Pty Ltd.,Melbourne,Australia)with 60 mg ZnO of natural abundance.The mixture was well homogenised with a pestle and mortar to obtain a final powder with 25.430%67Zn and 20.798%66Zn(weight/weight),corresponding to a67Zn/66Zn ratio of 1.223,and divided into three parts for production of three67Zn-enriched fertilisers,ZnO,ZnSO4,and ZnO co-granulated with elemental S(ZnOS0).A labelled ZnO suspension(25.6 g Zn L-1)was prepared by adding 700 μL water to the labelled ZnO(22.4 mg).The labelled ZnSO4solution(25.6 g Zn L-1)was produced by the addition of water(28 μL)and concentrated sulphuric acid(13.7 μL)to labelled ZnO(21 mg).Thereafter,distilled water was added to prepare a 660-μL ZnSO4solution.To produce ZnOS0with the desired composition(80%S0,5%ZnO,7.5%sodium(Na)-based bentonite,5%sugar,and 2.5%inulin,weight/weight),we mixed all of the material with 30%(weight/weight)distilled water,pressed and extruded the material using a syringe,and cut the product into 3-mm pellets.After drying,the pellets weighed approximately 17 mg each(13.6 mg S and 0.68 mg Zn).

Crop growth

A sequential crop(wheat-ryegrass-corn)cultivation was performed to evaluate the efficiency and residual effects of Zn sources in a Zn-deficient soil collected near Monarto(South Australia).The soil was classified as a Chromosol,according to the Australian Soil Classification System(Isbell,2002).Soil samples were collected from the top layer(0–20 cm),air dried,and sieved to＜2 mm prior to characterisation and to＜4 mm before use in the growth chamber.The soil had the following physical and chemical properties:clay,8%;silt,7%;sand,85%;pH(H2O),7.8;pH(CaCl2),7.0;Zn(diethylenetriaminepentaacetic acid,DTPA),0.6 mg kg-1;Zn(CaCl2),0.1 mg kg-1;total Zn,29 mg kg-1;S(CaCl2),0.14 mg kg-1;organic carbon,1%.

Black plastic pots(15 cm in diameter,lined with a plastic bag)were filled with the soil(1.65 kg dry weight per pot).Monoammonium phosphate(MAP)fertiliser granules and Zn fertilisers were applied at the same time,with a minimum separation distance of 3 cm to avoid P-Zn(chemical)interactions.The MAP(containing 52%P2O5and 10%N)fertiliser supplied nutrients that were equivalent to 12 mg N kg-1soil and 27 mg P kg-1soil.The amount of applied P was based on previous experiments using the same soil(data not shown).Enriched Zn fertilisers were applied in five equidistant points,at 1.5 cm below the soil surface,with five ZnOS0pellets applied per pot.The ZnO and ZnSO4fertilisers were applied as a suspension and solution,respectively,in the same position as the Zn pellets,using a volume of 25 μL for each spot.A control(CK,no Zn fertiliser)was included.All treatments were replicated five times,resulting in a total of 20 pots.The pots were then wetted to maintain the soil moisture at 80%of field capacity(FC)and left for 1 day to equilibrate.

Subsequently,six pre-germinated seeds of wheat(Triticum durumL.‘Kalka’)were sown in each pot at a depth of 1 cm.After 1 week,each pot was thinned to the most uniform three seedlings.Further liquid-based nutrients were added once per week as 5 mL of solution per pot,resulting in a 5-week total basal nutrient dose per kg of soil of 73 mg nitrogen(N)as urea and ammonium nitrate,0.7 mg boron(B)as boric acid,2.2 mg manganese(Mn)as manganese sulphate,2.7 mg copper(Cu)as copper sulphate,5.3 mg iron(Fe)as iron sulphate,70 mg potassium(K)as potassium chloride or sulphate,and 22.5 mg S from these sulphate salts.The pots were watered with distilled water every day to maintain the soil moisture at 80%of FC.After 6 weeks,at approximately the V-10 growth stage,flag leaves were sampled,and then the plants were harvested by cutting the stems at the soil surface.The flag leaves and shoots were dried at 70°C for 72 h and weighed.

Forty-seven days after the wheat harvest,10 seeds of ryegrass(Lolium rigidum‘Wimmera’)were sown into the undisturbed pot soil at a depth of 1 cm.After 7 d,each pot was thinned to five seedlings.No fertilisation was applied,requiring the ryegrass crop to use the residual nutrients,which is similar to the field conditions.The pots were watered to a designated weight with deionised water to maintain the soil moisture at 80% of FC.At 51 d after seeding,the aboveground biomass was harvested.

One hundred and five days after the ryegrass harvest,five seeds of sweet corn(Zea mays‘Kelvedon Glory’)were sown into the undisturbed pot soil at a depth of 1 cm.After 7 d,each pot was thinned to two seedlings.Further liquid-based nutrients were added once per week as 5 mL of solution per pot,resulting in a 6-week total basal nutrient dose per kg of soil of 10 mg N as urea,0.8 mg B as boric acid,2.6 mg Mn as manganese sulphate,3.2 mg Cu as copper sulphate,6.4 mg Fe as iron sulphate,70 mg K as potassium sulphate,and 27 mg S from these sulphate salts.Additional N was supplied in two applications of 50 mg N kg-1soil,each in the second and third weeks of growth.Zinc was not applied,requiring the corn to use residual Zn.The pots were watered to a designated weight with distilled water to maintain the soil moisture at 80% of FC.After 49 d,approximately at the V-4 growth stage,shoots were harvested by cutting the stems at the soil surface.The harvested material was dried at 70°C for 72 h and weighed.The dry matter of the crops were ground and then digested overnight using hydrogen peroxide(H2O2)and concentrated nitric acid(HNO3)at room temperature,followed by a temperature increase to 125°C for 4 h.

The experiment was conducted in a constant environment chamber under a photosynthetic photon flux density(PPFD)of 210 μmol m-2s-1,a temperature of 23°C during the day(12 h)and 15°C at night,and a relative humidity of 65%.During the intervals between crops,the pots were left in the growth chamber,open to air,and watered with distilled water to stimulate soil microbial growth.This sequential cultivation resulted in 42,140,and 294 d from Zn application to the harvests of wheat,ryegrass,and corn,respectively.

Determination of Zn isotope ratio and fertiliser recovery

Prior to sample analysis,the inductively coupled plasmamass spectrometry(ICP-MS)measurements were carried out following the procedure described by McBeathet al.(2013).Corrections were performed instrumentally and checked by analysing for background levels of60Ni,69Ga,71Ga,and136Ba.The ICP-MS instrument(Agilent 7700 ICPMS,Agilent Technologies,Inc.,USA),which uses He as a collision gas at a flow rate of 2.5 ml min-1,was optimised to minimise double-charged and oxide interference to＜2%(mass-to-charge ratio(m/z)of 70/140)and＜1%(m/z of 156/140),respectively.A certified67Zn isotope standard(Inorganic Ventures,USA),containing 10 mg L-167Zn,was used in the calibration phase and during the analysis.The plant Zn derived from the fertiliser(Zndff,%)and Zn recovery from fertilisers(Znrff,%)were calculated using the following equations(McBeathet al.,2013):

In the first place you must build me a palace to-night, the roof of purest gold, the walls of marble, and the windows of crystal; all round you must lay out a beautiful garden, with fish-ponds and artistic33 waterfalls

where IRplantis the67Zn/66Zn ratio in the plant tissue,SA66and SA67are the fractional abundance(by weight)of66Zn and67Zn in the67Zn fertiliser spike(0.2543 and 0.20798 for67Zn and66Zn,respectively),NA66and NA67are the natural abundance(by weight)of66Zn and67Zn in the control plants,Znplantis the total plant Zn(mg pot-1),and Znaddedis the amount of Zn applied(mg pot-1).

Statistical analysis

We applied one-way analysis of variance(ANOVA)using the general linear model in the Statistical Analysis System(SAS Institute Inc.,2003).The differences between the treatment(Zn fertiliser)means were evaluated using the LSMEANS procedure with the Tukey’s test and least significant difference(LSD)adjustment atP≤0.05.

RESULTS

Crop growth and Zn uptake

There were no significant effects of Zn applications on the shoot dry matter production of wheat,ryegrass,and corn individually or on the total dry matter produced(Table I).

Significant effects of Zn fertiliser treatments on the shoot Zn concentrations and Zn uptake were observed(Table I).Compared with CK,applying Zn as ZnSO4resulted in a higher shoot Zn concentration in all crops except wheat.The tissue Zn concentrations for the ZnO treatment were also higher than those in CK for all three crops,but significantly lower than those of the second crop in the ZnSO4treatment and lower than those of the third crop in both the other Zn fertiliser treatments.Similar to those of the ZnSO4treatment,the tissue Zn concentrations in the ZnOS0treatment showedno significant differences for the first crop,but were significantly higher for the second and third crops,as compared with CK.

TABLE IShoot dry matter production,flag leaf and shoot Zn concentrations,shoot Zn uptake,Zn isotopic ratio(67Zn/66Zn)in shoots,and percentage of Zn fertiliser recovery in shoots of a sequential crop(wheat-ryegrass-corn)cultivation with 67Zn-enriched fertilisers,ZnSO4,ZnO,and ZnO co-granulated with elemental S(ZnOS0)

Shoot Zn uptake was significantly increased by ZnSO4application compared with CK(Table I).There were no significant differences in shoot Zn uptake between the fertilised treatments,except for the ZnO treatment during the corn cultivation,which showed a lower uptake than the ZnSO4treatment.Similarly,the total Zn accumulation by all three crops(wheat-ryegrass-corn)after 294 d was not significantly different between the Zn sources.

Zinc isotopic ratio and fertiliser recovery

Treatment with ZnSO4,followed by ZnO,increased the67Zn/66Zn ratio in the shoots of wheat and ryegrass compared with CK,while there were no significant effects of ZnOS0applications on the isotope ratio in the first two crops(Table I).The ZnOS0treatment increased the67Zn/66Zn ratio only in corn,planted 245 d after application.In the third crop,both treatments with non-water-soluble Zn had higher67Zn/66Zn ratio in the shoots compared to CK,but still lower isotope ratio than the water-soluble Zn(ZnSO4)fertiliser.

The contribution of fertiliser Zn was calculated from the isotopic ratio values.The wheat and ryegrass Zndfffollowed the order:ZnSO4＞ZnO≥ZnOS0(Fig.1).The contribution of fertiliser Zn to Zn uptake increased over time,and this was most pronounced for the ZnOS0treatment,in which the contribution of fertiliser Zn reached 33%in the third crop(294 d after Zn application),compared to only 0.5%in the first crop.In the third crop,ZnOS0had a similar contribution as ZnO,albeitstill lower than ZnSO4.

Fig.1 Percentages of plant Zn derived from the fertiliser(Zndff)in the shoots of a sequential crop(wheat-ryegrass-maize)cultivation for different 67Zn-enriched fertilisers,ZnSO4,ZnO,and ZnO co-granulated with elemental S(ZnOS0).The least significant difference(LSD)is 2.0%,8.1%,and 24%for wheat,ryegrass,and corn,respectively.Values are means(n=5).Bars with different letters within a crop are significantly different,as determined by the LSD test(P≤0.05).

For the ZnSO4treatment,approximately 0.5%of added fertiliser Zn was recovered in each crop,resulting in a cumulative recovery of 1.5%over the three crops(Table I).For the ZnO treatment,the fertiliser recovery was approximately 0.15%for each crop,with an overall recovery of 0.45%.The recovery in the ZnOS0treatment was very low for the first two crops,but much higher for the third crop,resulting in an overall recovery of 0.33%,which was not significantly different from the overall recovery for the ZnO treatment.

DISCUSSION

Zinc is essential in many fertilisation programs,increasing crop yields and food quality.Water-soluble Zn fertilisers,such as ZnSO4,are usually more efficient for crops in the short term,since their fast release quickly increases the Zn concentration in the soil solution.More available Zn in the soil results in a higher plant uptake,increasing the Zn concentration in shoots and,in some cases,grains(Zhanget al.,2013;Zhaoet al.,2014;Liuet al.,2017).

Our study shows that plants fertilised with ZnSO4derived more Zn from the fertiliser than the ZnO-based fertilisers for all three crops.The percentage of Zn derived from the fertiliser increased between the first and second crops.For the ZnO-based fertilisers,this increase is likely due to the slow solubilisation of ZnO.Interestingly,the Zndfffor ZnSO4also increased between the first and second crops.A possible explanation for this is that the spot application of the ZnSO4source induced precipitation(e.g.,in the form of zincite)due to the high concentration at the point of application,hence resulting in a relatively low contribution in the first crop,followed by solubilisation resulting in higher contributions in the subsequent crops.This finding agrees with the results of Ghosh(1990),who found that the relative efficiency of ZnSO4granules was higher in the second crop than in the first crop,in contrast with a ZnSO4solution mixed through soil,for which the relative efficiency decreased over time.Studies focused on Zn fixation in soil using labelled Zn mixed through soil have shown a decrease in lability of added Zn over time,particularly in high-pH soils(Tilleret al.,1972;Tyeet al.,2003;Buekerset al.,2007).In the present study,the availability of the fertiliser Zn increased over time,because i)the added Zn was not mixed throughout the soil and ii)we used ZnO as the Zn source in two of the three fertiliser treatments.

Both ZnO-based fertilisers showed similar contributions in the third crop.However,the contribution of ZnOS0was lower than that of ZnO for the first and second crops.For the third crop(harvested at 245 d after application),a large contribution of the fertiliser Zn was observed in the ZnOS0treatment,with 33% of the Zn being derived from the fertiliser.This increase in the third crop may be related to a lag phase before the onset of elemental S oxidation.The lower recovery of ZnOS0than ZnO for the first two crops may be related to the method of application.While ZnO was pipetted as a suspension in five application points,ZnOS0was added as five pellets.McBeath and McLaughlin(2014)compared different methods of application and found that the effectiveness of ZnO decreased in the order:mixed through soil＞surface application＞banded,likely due to the faster dissolution when the ZnO was more dispersed.In our experiment,even though ZnO was spot applied,the addition as a suspension would have resulted in more dispersion of ZnO in the soil compared to the solid ZnOS0pellets.

Our data showed that after a 294-d sequential crop(three crops)cultivation,only 1.5%,0.45%,and 0.33%of Zn applied as ZnSO4,ZnO,and ZnOS0were recovered by plants,respectively.A long-term field evaluation of Zn fertilisation using the balance method over a period of 23 years showed less than 1.1% recovery of applied Zn by rice(Haqueet al.,2015).Similarly,less than 1% of added fertiliser Zn was recovered by durum wheat using the balance method(Luet al.,2012;McBeathet al.,2013).Other studies have reported Zn recovery efficiencies from 4.8% to 11.5% in spring wheat(Shivayet al.,2008)and,on average,13%in upland rice genotypes(Fageria and Baligar,2005).Thus,it seems that the efficiency of soil Zn application is generally very low,irrespective of the Zn source.Several factors likely contribute to the low recoveries of fertiliser Zn,particularly i)the high rates applied to soil in comparison to crop demand and ii)the dilution of fertiliser Zn in the labile Zn pool in the soil.Added Zn rates are often at least 10 times higher than the Zn taken up by the crop in one season;therefore,even if all Zn is derived from the fertiliser,recoveries in a single crop would still be low.The dilution of fertiliser Zn in the labile soil pool results in even less fertiliser Zn being recovered in the crop(McBeathet al.,2013).

The very low Zn recovery rates by plants,even when ZnSO4is applied,implies that a large amount of Zn from the fertiliser remains in the soil,and therefore a residual effect of Zn application is expected.Provided that Zn is not leached or irreversibly fixed in soil,it is still available for uptake by subsequent crops,and the overall fertiliser recovery is hence dependent upon the timeframe over which efficiency is evaluated.Thus,short-term evaluations give limited information on the overall efficiency of Zn fertilisers.Studies that have assessed the residual effectiveness of Zn fertilisers have shown that it contributes to crop uptake for many years after application.For instance,Brennan(2001)found that 13 years after application,the contribution of fertiliser Zn was still half of that in the first year.In the present study,we showed that the residual effectiveness of Zn fertilisers may be even higher than that in the first crop.Knowledge on the residual effects of fertilisers is needed to develop long-term strategies for crop management and fertilisation.Therefore,we hypothesised that slow-release Zn fertilisers,such as ZnOS0,may have a potential advantage in the long term,carrying both Zn and S in a more balanced ratio for uptake by plants.However,our results show that this was not the case and that a single application of soluble Zn was the most effective up to 294 d after its application.

All Zn sources had a residual effect on successive crops,and ZnSO4still supplied more Zn for corn(the third crop)compared to ZnO and was the most effective source for all three crops,by a large margin.However,the ZnOS0fertiliser showed a sharp increase in effectiveness between the second and third crops,presumably because elemental S oxidation was proceeding at a higher rate after the initial lag phase.Hence,long-term studies(＞1 year)using isotopically labelled fertilisers are necessary to further understand the mechanisms involving different Zn sources,soil reactions,and Zn uptake by plants.

CONCLUSIONS

Zinc recovery by the test crops was extremely low,regardless of the source of applied Zn.After 294 d of sequential crop cultivation,Zn recovered by plants followed the order:ZnSO4＞ZnO～=ZnOS0.A strong increase in the effectiveness over time was observed for ZnOS0,but it was still less effective than ZnSO4.Long-term evaluations are necessary for assessing the effectiveness of slow-release Zn sources.

ACKNOWLEDGEMENTS

This work was supported by funding from the Coordination for Scientific Support for Post-Doctoral Level Training(CAPES-BEX 1562/14-2),Brazil.We thank Colin Rivers,Bogumila Tomczak,Ashleigh Broadbent,and Claire Wright for analyses and technical assistance and the Mosaic Company for infrastructure support.