矿渣中总汞的测定 Hg(MJ 2014,113,36)

Determination of total mercury in bauxite and bauxite residue by flow injection cold vapour atomic absorption spectrometry
Neetu Bansal a ,⁎,James Vaughan a ,Amiel Boullemant b ,Tony Leong c
a School of Chemical Engineering,The University of Queensland,Australia
b Rio Tinto Alcan Technology QRDC,Australia c
Queensland Alumina Limited,Australia
a b s t r a c t
a r t i c l e i n f o Article history:
Received 9October 2013
Received in revised form 4November 2013Accepted 4November 2013
Available online 13November 2013Keywords:
Mercury determination FI-CV-AAS Bauxite
Bauxite residue
Microwave digestion
A simple method for precise and accurate determination of total mercury in bauxite and bauxite residue was de-veloped by using the flow injection mercury system.Samples of the solid materials were first microwave digested in acidic and oxidising conditions to convert all mercury to an aqueous ionic form.Following filtration and dilution,ionic mercury was reduced to elemental mercury with acidic SnCl 2to produce a cold mercury vapour.The mercury absorbance calibration graph was linear up to 20μg·kg −1with R 2N 0.999.The detection limits were determined to be 23ng·kg −1and 17ng·kg −1for bauxite and bauxite residue respectively.The relative standard deviation for 1μg·L −1mercury standard solution (n =27)was 1.5%.In the absence of a certi fied baux-ite reference material,accuracy of the method was tested with the closest available zinc concentrate reference material.Spiking known amounts of mercury in bauxite and bauxite residue samples was also tested;95–111%recovery was obtained for both samples.The method developed in this paper is recommended for measuring total mercury in bauxite and bauxite residue.
©2013Elsevier B.V.All rights reserved.
1.Introduction
Mercury has long been recognised as a neurotoxic element [1].It is ranked third in the “priority list of hazardous substances ”by the United States Comprehensive Environmental Response,Compensation,and Liability Act [2].The United Nations also finalised the legal document “Minamata Convention of Mercury ”with the aim to reduce mercury emission;140countries are signatories to this convention [3].Recognising the growing international concern over mercury emissions,the International Council on Metals and Mining (ICMM)adopted a Mercury Risk Management Position Statement to position its members in a leadership role on the issue [4].Rio Tinto Alcan and Queensland Alumina Limited began collaboration with the University of Queensland in 2011to develop an improved understanding of mercury in their operations.
Mercury can be found both naturally as well as introduced into the environment by anthropogenic activities [5].Reliable monitoring of mercury release and distribution is challenging due to issues with both sampling and measurement.Many instrumental analytical methods can be employed to determine the trace level of mercury in various samples.The most commonly cited techniques are:atomic
absorption spectrometry [6–11],atomic florescence spectrometry [12–15],inductively coupled plasma mass spectrometry (ICP-MS)[16–19],inductively coupled plasma atomic emission spectrometry (ICP-AES)[20],electroanalysis [21,22],and neutron activation analysis [23,24].
Cold vapour atomic absorption spectrometry (CV-AAS)is a widely used method because of its high sensitivity and selectivity for mercury.This method involves generation of elemental mercury from an acidi-fied solution with a reducing agent (e.g.SnCl 2,NaBH 4)or by other means such as photoreduction,electrochemical (EC)vapour generation,and ultrasound promoted cold vapour generation [25].
The CV-AAS technique requires that the elemental and organic forms of mercury be oxidised into ionic mercury in the feed solution [8].The conversion of complex solids into an aqueous form is accom-plished by heating the samples with concentrated acids.Wet digestion is used most commonly as mercury is lost to the vapour phase during dry ashing.Oxidants are also employed during wet digestion with the most common acids and oxidising agents being HCl,HNO 3,H 2SO 4,HClO 4,KMnO 4,K 2Cr 2O 7and H 2O 2.Elemental mercury,methyl mercury,dimethyl mercury are potential volatile components of the mercury species in a sample.Due to these volatile components,along with the tendency of mercury to adsorb on vessel walls and the requirement for complete digestion of the mercury,care must be taken in the selection of respective digesting agents and overall procedure.The choice of digestion acid depends upon the type of sample and the form of mercury present in the sample.For example,samples
Microchemical Journal 113(2014)36–41
⁎Corresponding author.
E-mail address:neetu.bansal@.au (N.
Bansal).0026-265X/$–see front matter ©2013Elsevier B.V.All rights reserved.
/10.1016/j.microc.2013.11.002
Contents lists available at ScienceDirect
Microchemical Journal
j o u r na l ho m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /m i c r o c
containing organic mercury need strong oxidising agents such as aqua regia and bromine monochloride solution.Samples high in silica may require hydrofluoric acid for complete solid dissolution.Perchloric acid also has strong oxidising power but can be dangerous for the sam-ple high in organic content and it requires heating N200°C to achieve its maximum oxidising power[26].
The sample preparation step in mercury determination is a challeng-ing part of the process,as it consumes time and can result in loss of analyte due to its volatile nature.Pressure microwave digestion is a good alternative to ambient pressure digestion[27].The advantages of pressure digestion are a fast digestion time,greater accuracy,less consumption of reagents,reduced sample size and less contamination [28–31].
Bauxite ore is a sedimentary rock that arises due to weathering of volcanic rocks[32].The minerals are generally well oxidised with alu-minium,iron and silicon being the major components.Metallurgical grade alumina is produced from bauxite ore using the Bayer process which involves digestion of bauxite with hot sodium hydroxide under high pressure at temperatures in the range of130°C to280°C depending on the feed material.Bayer process residue(bauxite residue) is the fraction of the bauxite ore that remains undigested.The mercury content of bauxite ore can vary significantly with values of20–100, 500–700and1200–2000μg·kg−1being reported[33].The mercury varies with the geographical origin of bauxite ore[34]and,within a single deposit,mercury content can also vary significantly.To reliably ascertain the total mercury content bauxite ore and bauxite residue,a robust method was developed using microwave digestion andflow injection mercury system(FIMS)on the resulting solution.The main challenges in the development of the procedure are the presence of trace and variable amounts of mercury and complex matrix of bauxite and bauxite residue.
2.Experimental
2.1.Reagents
Deionised water(greater than1.0MΩcm resistivity)was used to prepare the solutions.All reagents were of analytical grade and checked for trace mercury contamination.A1.2%(w/v)SnCl2reagent was freshly prepared by dissolving12.0g SnCl2.2H2O in30mL HCl and diluted up to1L with deionised water.Mercury stock solution1000mg·L−1in 10%(v/v)HNO3was supplied by Perkin Elmer Australia.Other stock solution was prepared by diluting this mother stock solution with deionised water and1mL BrCl as a preservative for50mL solution. Mercury calibration solutions(0.5to20μg L−1)were prepared from di-lution of the mother stock solution prior to each experiment.Hydrogen peroxide(30%w/w)was supplied by Fluka Analytical,and used as such without further dilution.Bromine monochloride solution was prepared byfirst mixing5.4g of KBr(Sigma Aldrich)in50mL HCl for1h,7.6g KBrO3(Sigma Aldrich)was then added slowly over a period of approx-imately5min to produce the BrCl solution.Hydroxylamine hydrochlo-ride was prepared by dissolving15g NH2OH.HCl in deionised water to a final volume of50mL.Aqua regia(3HCl+1HNO3by volume)was prepared immediately prior to digestion.Argon gas with N99.95%purity (Core gas)was used in all experiments.
The concentrated acids(HCl-37%,H2SO4-95%,HNO3-70(w acid/ w solution)%)were supplied by Sigma Aldrich.They are used with-out dilution unless specified.When diluted,the percentage speci-fied throughout the paper refers to volume percentage:100%∗(V concentrated solution/V diluted solution).
2.1.1.Reference material
Validation of the method described in the present work was performed using zinc concentrate reference material BCR®-109 (Institute for Reference Material and Measurements).2.2.Instrumental
2.2.1.Microwave digester
For the digestion process a speed wave®4(Berghof products)+ instruments GmbH Germany,digester with DAK-100pressure vessels was used.The pressure vessels are made of TFM-PTFE which has a high chemical resistance and hydrophobicity.The vessels are rated up to100bars with a maximum temperature of300°C.To provide a com-fortable safety margin,the maximum temperature applied in this pro-cedure was230°C.The pressure is monitored continuously by a contact free,optical system along with temperature by an external in-frared based estimate.The microwave digestion of bauxite and bauxite residue was carried out in stages,in thefirst stage,the reactor contents were heated to120°C(maximum pressure60bars)for6min with a ramp up time3min followed by heating at200°C for5min with ramp up time of3min.In the third stage,the solutions were heated at230°C for16min with a ramp up time of5min.The total time required for digestion is about45min(including cooling time in the microwave digester).
2.2.2.Flow injection mercury system(FIMS)
A Perkin Elmer FIMS400flow injection mercury system is employed for the determination of mercury content of the diluted digestion liquor. This instrument uses a high performance single beam optical system, solar blind detector and low pressure mercury lamp.Thisflow injection system consists of two peristaltic pumps(P1,P2),a24cm long absorp-tion cell with removable quartz window,electrically heated mantle to maintain the cell temperature at approximately50°C.Theflow injec-tion switching valve hasfive ports with variable length sample loops. The tubing is made of PTFE and the gas liquid separator is a membrane also made of PTFE.FIMS is a combination of theflow injection technique with atomic absorption detection.The hardware was controlled using the Perkin Elmer Win Lab32for AA system software.The instrumental operating conditions are given in Table1.SnCl2/HCl was used as reductant and argon gas was used as a carrier of mercury vapour to the absorption cell.
2.3.Sample collection and nature of samples
Three different bauxite and two different bauxite residue samples were collected from various alumina refineries and mining sites. Samples were collected and placed in plastic containers and transported to the lab.
2.4.Characterization of bauxite residue and bauxite
The main components of bauxite and bauxite residue samples were quantified by XRF(X-rayfluorescence),the bauxite content was 44–53%Al2O3,14–27%Fe2O3,3.8–7%SiO2,2–2.5%TiO2as wt.%and
Table1
Characteristic parameters of FIMS.
FIMS
Wave length253.7nm
Lamp Electrodeless discharge mercury lamp(EDL) Measurement Peak height
FIAS prefill time15s
FIASfill time10s
FIAS inject time15s
Read delay0s
Read time15s
Reductant 1.2%(w/v)SnCl2in3%(v/v)HCl
Carrier3%(v/v)HCl
Carrier gas stream Argon
Carrier gasflow rate50mL·min−1
Cell temperature50°C
Sample loop500μL(can be variable)
37
N.Bansal et al./Microchemical Journal113(2014)36–41
bauxite residue content was15–23%Al2O3,27–30%Fe2O3,10–19%SiO2, 4.7–6.5%TiO2as wt.%.The major mineral phases present in the samples were identified by XRD(X-ray diffraction)using the PDF-22012data-base.The results indicate the major components of bauxite-1,bauxite-2and bauxite-3are gibbsite(Al(OH)3),boehmite(γ-AlOOH),hematite (Fe2O3),and kaolinite(Al2Si2O5(OH)4).XRD patterns for the bauxite samples are similar despite coming from various deposits.The baux-ite residue samples showed the presence of mostly hematite and some boehmite.Certain peaks for the bauxite residue XRD patterns remain unidentified.Of the unidentified peaks,some were matching with sodium aluminium silicate compounds but there was no com-plete match with the compounds in the database.The sodium aluminium silicate is the result of a desilication process which ex-plains the elevated sodium content in the residue[35].Calcium and magnesium contents are also elevated in the residue due to addition of lime during the process and a seawater neutralisation process.The literature shows the presence of aluminium hydroxide as major component and iron oxide,clay,titanium oxide,quartz,water and some other minerals as minor components in bauxite samples[36]. On the other hand the bauxite residue samples contain iron oxide and silica as major component and alumina and some other metal oxides as observed here[37].
2.5.Sample preparation
Sample preparation is a critical step in accurate mercury content determination.As bauxite ore has a non-uniform distribution of mineral phases,the sample preparation step becomes important to obtain a representative sample and measure a sufficient quantity of sample to assess the natural variability.
2.5.1.Mineral processing
20kg bauxite samples were rotary split to about2.5kg fractions and one of the2.5kg fraction was then riffle split to about300g.Before splitting,the coarse fraction of the bauxite was crushed with a mortar and pestle.Finally,bauxite samples were staged pulverised for two cy-cles of5s to minimise mercury loss due to heat produced.The material was then sieved through a300μm mesh screen.The oversized material was re-pulverised for5s until all the material passed through the 300μm mesh.
2.5.2.Digestion of samples
To dissolve the sample matrix,wet digestion was carried out using a combination of acids at high temperature in the microwave digester. The hot,wet digestion can solubilise both inorganic and organic matter. Microwave digestion may enhance contact between acid and particle by fracturing particle that results in providing a new surface for acid attack [38].
Careful selection of acid is required for optimal digestion.A combi-nation of aqua regia and hydrogen peroxide worked well for digesting bauxite and bauxite residue samples leaving behind a silica rich residue. Addition of H2O2in the digestion mixture facilitates the complete oxida-tion of organic matter.For the digestion of bauxite and bauxite residue, closed vessel microwave digestion with varying combinations of acid was systematically evaluated.From an analytical point of view;it is dif-ficult to solubilise bauxite and bauxite residue due to the high silicates/ silica content without using hydrofluoric acid.Also,increasing the amount of acid used can compromise the measurement using FIMS. Some researchers have claimed that digestion with acid can leach all the mercury from the sample into the acid solution and remove silicates byfiltration[39–41].The use of hydrofluoric acid to solubilise silica was avoided,due to its hazardous nature and its adverse effect on the FIMS [42].As mercury has no tendency to form natural silicate material,it was not deemed necessary to dissolve silica in the digestion mixture [43].2.5.3.Bauxite digestion
For the digestion of bauxite,various combinations of acids were test-ed.In each method0.5g of bauxite was digested with single and/or combination of various acids in the microwave digester.
2.5.4.Bauxite residue digestion
Bauxite residue samples were digested as slurry.Before weighing the sample,bauxite residue was stirred for at least1h to make sure it is homogenous.A3.5g of bauxite residue sample was digested with dif-ferent combinations of acids.Moisture content of the bauxite residue was approximately60%that was measured separately after drying a sample in oven at60°C to constant mass.
2.6.Calibration matrix
The mercury standard solution was treated exactly as the sample solutions and underwent every step including microwave digestion with the same reagents.Calibration blanks were also prepared in reagent water containing the exact amount of digestion mixture and preservative with no mercury and treated like a sample.
2.7.Analytical procedure
A0.5g bauxite or3.5g bauxite residue was weighted directly in the TFM-PTFE in-liner vessel.The digestion procedure was carried out in two steps,predigestion and closed vessel microwave digestion.In the pre-digestion step,for bauxite samples,3mL HCl and1mL HNO3 were added followed by drop by drop addition of1mL H2O2over a time period of approximately1min.These samples were kept over-night to ensure the oxidation of all organic matter and total degassing in solution prior to microwave digestion to prevent the build-up of extra pressure due to generation of carbon dioxide gas.For the bauxite residue samples,the same procedure was adopted,except that the amount of acid was3.75mL HCl and1.25mL HNO3.During the pre-digestion step,the PTFE vessels were loosely capped.The following day water was added to the pre-digested sample to a total volume of 15mL.The in-liners were then placed in the pressure vessels and the digestion procedure was carried out.After digestion,the pressure vessels were cooled(to room temperature)for a minimum of2h.
After cooling,750μL of BrCl was added as a mercury preservative and also to assist with the conversion of any remaining organically com-plexed mercury into ionic mercury.After3–4h,the digested samples were transferred to a50mL centrifuge tube along with the washings from the TFM-PTFE in-liner.All the samples were centrifuged for 15min at5000rpm.After centrifuging,the solution was carefully transferred to another centrifuge tube leaving behind a white silica rich residue.The bauxite sample typically had less residue compared to the bauxite residue sample,this was attributed to the relatively high silica content of the residue.The digested decanted sample solution was thenfiltered through0.2μm PTFE syringefilter,and diluted to50mL with deionised water for analysis by FIMS.15min prior to the analysis,hydroxylamine hydrochloride solution was added to neutralise remaining BrCl.
3.Results and discussion
FIMS sensitivity is highly dependent on cold vapour generation which in turn depends on the experimental conditions.To determine the optimum conditions for the given method from the FIMS,physical and chemical variables were studied at10μg·L−1mercury.
3.1.Chemical parameters
3.1.1.Effect of SnCl2
The effect of SnCl2on recovery of10μg·L−1mercury was studied. The concentration of SnCl2was varied from0.01%(w/v)to7%(w/v).
38N.Bansal et al./Microchemical Journal113(2014)36–41
The best recovery(100%)was obtained between1.0and2.0%SnCl2.This is in close agreement with the recommended concentration given by the supplier of FIMS.Reductant concentrations below1.0%and above 2.0%result in decreased recovery of mercury.1.2%SnCl2was chosen as the optimum value.
3.1.2.Effect of HCl concentration in SnCl2
To obtain clear solutions for aqueous SnCl2required dissolution in dilute HCl.Hydrochloric acid concentrations were varied from0.1% (v/v)to7.0%(v/v).Results showed100%recovery when the HCl content of the SnCl2solution was in the range of3.0%to4.0%.3%HCl was selected for experiments.
3.1.3.Effect of HCl as a carrier
The concentration was varied between0.01%(v/v)and10%(v/v). The optimal range was determined to be3–4%HCl for10μg·L−1 mercury and3%HCl was selected for study.
3.2.Physical parameters
3.2.1.Argonflow rate
Argon was used as carrier gas,the effect of gasflow rate was de-termined over the range of30to130mL·min−1(Fig.1).Argon flow rate also plays an important role in the sensitivity of the instru-ment as this carrier gas is responsible for carrying mercury vapour from the gas liquid separator to the absorption cell.Absorbance of the mercury signal was increased with argonflow rate from30mL·min−1 to50mL·min−1and reaches a maximum value at50mL·min−1and again decreases with increasingflow rate.At high argonflow rates the sensitivity decreases considerably due to reduction in residence time and dilution of the mercury vapour by argon gas.50mL·min−1was se-lected as the optimum value.
3.2.2.Reagentflow rate
The maximum absorbance was observed when acid and reducing agentflow rates were10and6mL·min−1respectively.
3.2.3.Effect of sample volume
The effect of the sample volume was studied from20μL to 1000μL for absorbance of10μg·L−1mercury.Absorbance increased with increasing volume size from20–500μL,but from500–1000μL the absorbance plateaued.Results are shown in Fig.1.500μL was se-lected for further parison of oxidising agents
To determine the optimum digestion working conditions for bauxite and bauxite residue samples,different combinations and amounts of digestion mixture were tested for microwave digestion. Total solubilisation of bauxite and bauxite residue was not required as mercury has no tendency to form silicates due to its large ionic radii [44].Initially,different individual acids and combinations of oxidising agent and acid at various amounts were tested.A result showed be-tween4and5mL acid/acid combinations was sufficient and efficient to digest0.5g bauxite in the microwave digester.Next about5mL of acid or combinations of different oxidising agents were tested.This study has been performed on the bauxite-1samples.A single factor Anova test was carried out at95%confidence interval.For one type of study,samples from a specific refinery were selected and taken from one container only.
3.3.1.Aqua regia digestion
In this series of digestions,0.5g of bauxite was mixed with increas-ing amounts of aqua regia.The same procedure was followed as given in Section2.7including addition of BrCl.It is desirable to minimise the aqua regia required for digestion,as large amounts of aqua regia in-creases the detection limit of the assay.The results show(not included) 4mL aqua regia is sufficient to digest0.5g bauxite ore.The optimal value of4mL aqua regia was also verified for a second bauxite sample (results not shown).The lower observed recovery of mercury for higher aqua regia concentration could be related to decreased absorbance in FIMS.
3.3.2.Other digestions
Other methods tested for digestion of bauxite are,4mL aqua regia+1mL H2O2,5mL HNO3,4mL HNO3+1mL H2SO4, 4mL HNO3+1mL HCl, 2.5mL HNO3+2.5mL HCl, 2.5mL HNO3+2.5mL HCl+1mL H2O2.In all these digestion the same pro-cedure given in Section2.7was followed.After testing different diges-tion mixtures,results showed that to digest0.5g bauxite,4mL aqua regia and1mL H2O2gave the best recovery of total mercury.Likewise, for the digestion of bauxite residue samples,different combinations of aqua regia and H2O2were tested.The highest recovery of mercury from bauxite residue samples using3.5mL slurry needed5mL aqua regia and1mL H2O2.
3.4.Effect of sample drying temperature on recovery of mercury in bauxite sample
The effect of drying temperature on bauxite-3was studied for a temperature range of40–240°C.For each temperature,samples
were
0.12
Sample volume (µL)
A
b
s
o
r
b
a
n
c
e
f
o
r
H
g
Argon gas flow rate [ mL. min-1]
sample
volume
Fig.1.Effect of argonflow rate and sample volume on absorbance at10μg·L−1mercury
concentration(error bars
represents the standard deviation of triplicate data point).
[
H
g
]
µ
g
.
k
g
-
1
Temperature in ⁰C
Fig.2.Effect of drying temperature of bauxite-3on mercury recovery(duplicate samples).
39
N.Bansal et al./Microchemical Journal113(2014)36–41
dried for48h in the oven.Fig.2shows that increasing the temperature results in a minor loss in mercury concentration from40to80°C. Further increasing the temperature from80to120°C,there is a sharp decrease in the recovery of mercury that shows a major loss of the volatile mercury that occurs in this region.
3.5.Validation of the method and analytical results of samples
Five point calibration curves with R2N0.999were constructed with the microwave digested mercury standard solution using the same amount of digesting mixture.In the absence of an appropriate bauxite reference material,the method was tested using a zinc ore reference material(Certified reference material BCR109)and by using spike and recovery experiments.The certified value for mercury was 960±120μg·kg−1(mean±standard deviation).Measured value for reference material in this study was973±34(mean±standard deviation)for9replicates.Two statistical tests were conducted to com-pare our results of the certified reference material to the certified value provided by the supplier.A t-test[45]compares the sample mean value
to the standard value.In the t-test,t calc was1.14and t tab was1.83at95% confidence interval for9degrees of freedom.As t tab N t calc,there is not sufficient evidence to reject the null hypothesis,which proves that both of the values are statistically similar.In another test,the confidence interval calculated for the certified value was960±98μg·kg−1at95% confidence interval and the measured value(973μg·kg−1)is within range.
Bauxite and bauxite residue samples analysed by FI-CV-AAS are given in Table2.Fig.3shows the variability of the mercury content. For quality control,a known concentration mercury solution was analysed after every four samples.Accuracy was determined by 80–120%mercury recovery.
To evaluate the variability of mercury concentration within the samples a single factor Anova[45]test was performed on all samples. Results showed in all the samples that there was enough evidence to reject the null hypothesis at≈100%confidence interval as F N F critical, this proves that there is natural variability in the samples.In another attempt an Anova test with Tukey method[46]was performed to compare the mercury contents in all samples.Results showed that the mercury content of bauxite-1and bauxite-2are not statistically differ-ent and the same results were seen in bauxite residue-1with bauxite residue-2.On the other hand the mercury content of bauxite-3is statistically different from bauxite-1and bauxite-2.Mercury present in bauxite is also statistically different and considerably greater than in the bauxite residues.For validation of the method,spike and recovery experiments were done by adding increasing concentrations of mercury to bauxite-1samples and bauxite residue-1samples with digestion of the mixture in the microwave digester.The procedure was repeated for each addition as described in Section2.7.Results are given in Table3.The recoveries for spiked bauxite and bauxite residue samples were between95and111%.3.6.Analyticalfigure of merit
The analyticalfigure of merit of the proposed method was calculated under optimised condition.The calibration curve was linear up to 20μg·L−1with standard solutions of mercury.The precision was calcu-lated as repeatability of the signal.It was determined from the analysis of27replicate sample of microwave digested mercury standard solu-tion containing1μg·L−1mercury with a relative standard deviation of1.5%.The method detection limit(MDL)was calculated by a U.S. EPA method[47].MDL was calculated for the bauxite and bauxite resi-due matrix.Bauxite-1sample was heated to150°C for24h to produce low mercury bauxite.This bauxite sample was used to calculate detection limits as its mercury concentration was reduced between one tofive times the estimated detection limit[47].The detection limit calculated for12samples of the bauxite at95%confidence interval was23ng·kg−1.The detection limit for bauxite residue matrix was determined to be17ng·kg−1at95%confidence interval based on10 samples from the same batch of residue.
4.Conclusion
A simple and accurate method for the determination of total mercu-ry in bauxite and bauxite residue samples by FI-CV-AAS was developed. The method yields lower detection limits than those stated in the liter-ature for the CV-AAS method and is sufficiently low for measurement of mercury in both bauxite and bauxite residue.Satisfactory relative stan-dard deviation values were obtained for standard solution of mercury and the method was also validated using a zinc concentrate mercury reference material.Method performance and validation were checked by quantitative recovery of mercury from spiked bauxite and bauxite residue samples.This method was successfully applied to assess the natural variability of mercury in the samples.Based on the outcomes
Table2
Results of bauxite and bauxite residue by FI-CV-AAS.
Sample Measured value
(μg·kg−1)a No.of samples Tukey method grouping
information at95%
confidence interval
Bauxite-144±1224A Bauxite-246±1216A Bauxite-3135±2118B Bauxite residue-110±326C Bauxite residue-27±18C
a Mean±standard deviation.
b a
u x
i t e
b a
u x
i t e
b a
u x
i t e
b a
u x
i t e
r e
s i d
u e
b a
u x
i t e
r e
s i d
u e
[
H
g
]
(
µ
g
•
k
g
-
1
)
Fig.3.Variability of mercury in bauxite and bauxite residue samples.
Table3
Analytical results of mercury in spiked bauxite and bauxite residue sample.
Samples Hg added
(μg·kg−1)
Hg found
(μg·kg−1)a
Hg recovered
(μg·kg−1)a
Recovery(%)a
Bauxite-1042––
150195152101±3
275339297108±4
650765723111±3 Bauxite residue-109––
556859107±1
12512811995±1
286306297104±1
a n=3,mean±mean deviation.
40N.Bansal et al./Microchemical Journal113(2014)36–41。