Research ArticleFast Determination of the Main Reduced Sulfur Species inAquatic Systems by a Direct and Second-DerivativeSpectrophotometric Method

Sebastien Dupraz 12 Benedicte Menez1 and Franccedilois Guyot3

1Universite de Paris Institut de Physique du Globe de Paris CNRS and Centre de Recherches sur le Stockage Geologique du CO2(IPGPTOTALSCHLUMBERGERADEME) 1 Rue Jussieu 75238 Paris Cedex 05 France2Bureau de Recherches Geologiques et Minieres (BRGM) 3 Avenue Claude-Guillemin 45060 Orleans France3Institut de Mineralogie de Physique des Materiaux et de Cosmochimie Sorbonne-Universite IRDMuseum National drsquoHistoire Naturelle CNRS 4 Place Jussieu 75252 Paris Cedex 05 France

Correspondence should be addressed to Sebastien Dupraz sduprazbrgmfr

Received 13 March 2019 Accepted 10 June 2019 Published 2 July 2019

Academic Editor Patricia E Allegretti

Copyright copy 2019 Sebastien Dupraz et al is is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

e determination of reduced sulfur species in aquatic systems is not an easy and fast task to accomplish regarding the numerouspossible interferences and risks of oxidation that occur with the usual methods of quantification e method presented here is adirect spectrophotometric method that can be used to quantify sulfides sulfites and thiosulfates in a simple and rapid way eprinciple is based on the comparison of second-derivative absorbance spectra of the same sample at different pH (92 47 and 10)and selected absorption wavelengths (250 and 278 nm)is method has been successfully tested and has demonstrated liability to(i) avoid the biases due to absorbance overlaps between the different major chemical species and (ii) keep as a direct method theadvantages over indirect methods on interferences reduction e limits of detections (LOD) reached for total sulfide sulfite andthiosulfate are 137 732 and 192 microM respectively e method displays low accuracy mean and low relative standard deviation(lt4) as well as a good linearity (R2gt0999) Accordingly this method represents a very robust alternative in terms of cost andrapidity for the quantification of reduced sulfur species in different aquatic environments from freshwaters to saline andpolluted systems

1 Introduction

e chemistry of reduced sulfur species plays a predominantrole in numerous applications and fundamental studies re-lated to anoxic environments like for instance undergroundfluids and ecosystems of the subsurface deep water areasswamps sediments decaying organic matter and sewers[1ndash4] It is also the case in applied research dealing withanaerobic bioreactors water purification petrochemicalplants food processing paper mills and tanneries [5ndash8] If weconsider the chemistry of natural systems the major solublespecies for oxygenated waters is sulfates (SO4

2minus) For anoxicmedia the most predominant species are sulfides (HSminus andS2minus) and dissolved hydrogen sulfide (H2S) followed bythiosulfates (HS2O3

minus and S2O32minus) polysulfides (H2Sx HSxminus

and Sx2minuswith x 2 to 9) colloidal elemental sulfur (α-S8) andpolythionates (SxO6

2minus with x 3 to 6) [9ndash11] e otherspecies called intermediate sulfur species generally representa very small fraction (lt001) of the total dissolved sulfurcontent [12] Nevertheless several studies realized in brinesvolcanic lakes and hydrothermal systems have found relativeproportions of thiosulfates and polythionates which couldreach 20 of the total dissolved sulfur content [12 13]Sulfites (HSO3

minus and SO32minus) are less frequently found in

natural environments because of their relatively low stabilityHowever they can be found at high levels in hot springs [14]and are also very commonly used in food and beverage in-dustries as preservatives [15 16]

Due to the variable relative abundance of sulfur speciesin the environment and their implication in numerous

HindawiJournal of ChemistryVolume 2019 Article ID 1039487 12 pageshttpsdoiorg10115520191039487

microbial metabolisms their monitoring is frequently re-quired to assess the nature and level of activities of anaerobicecosystems in fundamental or applied research Checkingthe presence of these species is also needed for qualitativerequirements and security controls like for instance rottenegg flavor and toxicity due to hydrogen sulfide presence[17 18] e main techniques that are generally used forquantification are by order of popularity [19] indirect ul-traviolet (UV)visible spectrophotometric measurements[20 21] potentiometric measurements and separationmethods like iongas chromatography capillary electro-phoresis and flow-based methods that can be coupled tospectrophotometry fluorimetry electrochemical measure-ments absorptionemission atomic spectrometry and flamephotometry for volatile species [7 22ndash25] e majordrawback of all these techniques is their sensitivity to watercomposition that require considerable instrumental pro-cedural and technical needs to be implemented [26] Inparticular improvements in selectivity are still needed toallow measurements of sulfides in saline or highly pollutedwaters [19] Moreover due to the strong reactivity of thereduced sulfur species to oxygen sample handling andpreparation are tedious while it should be as short and directas possible Consequently there is a strong need for a robustmethod that would be simple fast to implement andadaptable to a broad set of waters

Direct spectrophotometric measurements have this po-tential but they are generally not used because of the greatnumber of compounds that share the same absorbance regionsin the UVvisible domain For instance the quantification ofthiosulfates sulfides and sulfites by spectrophotometry ishampered by the fact that these compounds absorb the UV-visible light in the same range of wavelengths Notably sulfidesshow interferences with the sulfite species and consequentlyit is only possible to determine the total sulfide content whenthe amount of sulfites present in the system is known be-forehand Nevertheless improvement in these approaches hasbeen accomplished through specific spectral deconvolutionroutines [27] or by the use of second-derivative absorbancequantification procedures that allow the removal of non-specific linear absorbencies in the spectrum [28]

e method proposed here is also based on directspectrophotometric measurements but in three differentmedia buffered at selected pH Regulation by mean of pH iscommonly used before direct spectrophotometric mea-surements in order to assess the absorption coefficient of achemical species sensitive to pH (ie acid and base con-jugates) With this information it is then possible to de-termine for a given aqueous solution the concentration ofthis species whatever the pH is On the contrary in ourmethod sample measurement is directly achieved in mediawith selected fixed pH and leads to the quantification of allthe acids and bases that derive from a given species Ad-mitting that the acid-base reactions are instantaneous thedirect measurements at adequate pH values can indeed takeinto account all the protonated and unprotonated forms ofthis species In addition a specific absorbance observed inone buffered medium may disappear totally or maintainitself constant in another buffered medium these properties

being explained by the differences between the pKa of thevarious targeted species erefore by subtracting absor-bance spectra of the same sample diluted in media withdifferent pH and by choosing appropriately the character-istic wavelengths to be measured the suppression or iso-lation of a given signal is possible e absorbance overlapscoming from other sulfur species but also from the maininorganic and organic components commonly found innatural environments can similarly be removed Howeverthis procedure prevents the possibility of distinguishing theproportion of different acids bases and ampholytes that arecontrolled by acid-base equilibria Nevertheless and aspreviously stated this relative distribution can be establishedretrospectively by knowing the sample initial pH

Sulfides were selected as the main target of our quan-tification method because as discussed previously theyrepresent the main reduced sulfur species in natural systemsbut they are still difficult to quantifyiosulfates and sulfiteswere also selected because they are among the most solublespecies of the common reduced sulfur compounds and themain intermediate products of sulfide oxidation [2 29] Wepresent here principles and results of the calibration andinterference tests which allowed to assess method liability

2 Materials and Methods

21 Chemicals e method was calibrated with standardsolutions containing various concentrations (up to500mM) of thiosulfate sulfite and sulfide species initiallyunder the following forms sodium thiosulfate (Na2S2O3)from Fisher Chemicals (volumetric solution) sodiumsulfite anhydrate (Na2SO3) from Fisher Chemicals (Lab-oratory Reagent Grade) and sodium sulfide nanohydrate(Na2Smiddot9H2O) from Fluka (32ndash38) All the solutions wereprepared with deaerated ultrapure water (UPW reverseosmosis system PURELABreg Prima from ELGA with aresistivity of 18MΩmiddotcm) and stored under argon atmo-sphere Solutions were also amended with NaCl or NO3

minus inorder to test the impact of salinity and high nitrate levels onthe method liability Yeast extracts (Hy-Yestreg 412 Sigma-Aldrich) and humic acids (purified humic acid technicalgrade solution from Sigma-Aldrich) were also used tomimic naturally occurring organic matter and assess po-tential interferences on the method With the exception ofthese latter organic substances all the other chemicals usedwere of NORMAPURreg quality or certified to meet theAmerican Chemical Society specifications for analysis

22 Spectrophotometric Measurements All the measure-ments were performed in Herasil quartz cuvettes(HELLMAreg N6040 10mm in length) inserted in a UVvisible Shimadzu double-beam spectrophotometer (UV-1650PC deuterium lamp) e UVProbe software (Shi-madzu) was used for data acquisition and processing

23 General Protocol A schematic description of the pro-tocol is given in Figure 1 Samples and aliquots of thestandard solutions were filtered withMillipore filters (02 μm

2 Journal of Chemistry

ISOPOREtrade) and directly added to three dierent pH-buered media of decreasing pH (N AC and H mediarespectively) with (i)N corresponding to NH3 (625mmolL or 4 microLL from a 25ndash30NH3 solution) mixed with NH4Cl(625mmolL) leading to a nal pH of 92 (ii) AC corre-sponding to a mix of Na-acetate trihydrate (625mmolL)and acetic acid (625mmolL) with a nal pH of 47 and (iii)H corresponding to HCl at 01molL with an associated pHof 1 All these media were prepared in UPW and formulatedin order to reach a nal volume of 10mL after sampleaddition Variations in the injected volume of sample in thisformulation allowed dierent dilutions

Spectra were recorded for each medium right aftersample addition using the following parameters spectralacquisition between 220 and 300 nm wavelength scanningspeed (slow 32 nmmin) single acquisition and databunching interval of 01 nm If not specied the blanksassociated with these measurements were made with the

same buered medium (N AC or H) without any additionof the sample Acquisitions were performed at controlledroom temperature (25degC) with thermally equilibratedsolutions

24 UV-Visible Spectrophotometric Characterization of theBuered Media e UV-visible spectral signature of eachmedium (N AC and H) was established by a series of 10measurements For these acquisitions blanks consisted inUPWe pH stability of the media was frequently checkedand care was taken to not let the recipients open to theatmosphere in order to avoid changes in the neutral species(ie NH3 and acetic acid) due to air-solution exchanges

25 Calibration Curves Each of the three species of interest(suldes sultes and thiosulfates) was analyzed three timesin each media (made independently for each measurements)

Sample

Filtration(02 microm)

AC N H

pH = 47 pH = 92 pH = 1

Dilution into thebuffered solutions

UV-visiblespectrophotometry

(220 to 300nm)

Second derivatives NPrime250-ACPrime250ACPrime250

HPrime278

Total sulfideTotal thiosulfateTotal sulfite

Proportional relationships

Figure 1 Schematic representation of the quantication procedure proposed in this study From a ltered sample three direct spec-trophotometric spectra are acquired in three dierent bueredmedia (N pH 92AC pH 47 andH pH 10) in the range 220ndash300 nmSpectra are dierentiated to the second degree (N0 AC0 and H0) and values of second derivative absorbance are extracted at specicwavelengths (N0250 AC0250 at 250 nm andH0278 at 278 nm) ese values can be used to directly quantify total thiosulfate (H2S2O3HS2O3

minusS2O3

2minus) and sulte (SO2HSO3minusSO3

2minus) contents or after subtraction total sulde concentration (H2SHSminusS2minus)

Journal of Chemistry 3

at a concentration of 05mmolL (or 025mmolL for sul-fides in medium N) Each measurement was repeated twotimes (gap of 3min between the onset of each measurement)in the same cuvette in order to detect and quantify anytemporal modification of the signal (eg oxidation by airdioxygen) e individual calibration lines for each specieswere then obtained by measuring 5 different dilutions withfinal concentrations ranging from 500 to 4 micromolL

26 Interferences Tests Tests for specific interferences wereconducted by first mixing the three sulfur species (sulfidessulfites and thiosulfates) at 100 micromolL each with solutionat different salinities (100 146 73 365 and 0 gL of NaCl)in order to estimate the impact of ionic strength increase onthe signal stability then with a synthetic seawater (ie TYAmedium described in [30]) diluted 10 and 40 times in orderto mimic general ionic background of saline and marinewaters and finally with 10mgL of humic acid because itrepresents the main organic constituent of surface aquaticsystems Additionally UV-visible spectra were also acquiredon nitrate- NaClminus and humic acid solutions and on theTYA medium separately mixed with each of the three media(NAC andH) Tests were also performed with yeast extractbecause it provides a large diversity of organic compounds(eg proteins amino acids and carbohydrates) that arefrequently found associated with animal andmicrobial waterpollution All these measurements were at least triplicatedInterference threshold values were defined as the concen-tration that gives a signal variation equal to the corre-sponding smallest limit of detection (see below) In parallela linearity comparison test was also accomplished with thespectrophotometric methylene blue method(SPECTROQUANTreg test kit Ndeg 11477790001 Merck)using natural freshwaters spiked with sulfides Finallymeasurements were also performed on a groundwatersample exhibiting natural occurrences of sulfides sulfitesand thiosulfates and collected in the Dogger formation of theParis basin (Orleans France) In order to validate the presentapproach the same groundwater sample was controlled withstandardized methods Sulfur was measured using themethylene blue coloring method (ISO 10530 1992) Sulfiteswere measured using the iodometric titration method(SANDRE 296) and thiosulfates were quantified with anion-exchange chromatography protocol on ICS-3000 (ISO10304-E)

27 Data Treatment and Statistical Approach All the spectrawere processed using the UVProbe software second-de-rivative function which can be described by equation (1)from [31]

d2A

dλ2

A λprime + Δλ( 1113857minus 2A λprime( 1113857 + A λprime minus Δλ( 1113857

(Δλ)2 (1)

where A is the absorbance at a given wavelength λ λprime is thewavelength at which the second-derivative absorbancevalue is calculated and Δλ is the wavelength interval In thisstudy Δλ was fixed at 8 nm in order to reduce the narrow-band noise effect [28 32] Second-derivative spectra from

media N AC and H were named N0 AC0 and H0respectively

Spectra subtractions were also performed with theUVProbe software Calibration lines were calculated usingthe standard linear least square method from MICRO-SOFTreg EXCEL

In this study the limit of detection (LOD) and the limitof quantification (LOQ) correspond to 3σ and 10σ re-spectively with σ representing the standard deviation of theblanks

σ

1N

times 1113944N

i1xi minus x( 1113857

2

11139741113972

(2)

where N is the total number of measurements i and x themean of the measurement values x

Accuracy was determined by

Accuracy 1

Nprimetimes 1113944

Nprime

i1

mi minus ri

ri

1113888 1113889 times 100 (3)

where Nprime is the number of measurements i to which cor-responds a real (ri) and a measured (mi) absolute valuesOnly values from the calibration test higher than the LOQwere considered

Precision was assessed using the relative standard de-viation (RSD) defined in

RSD 1

NPrimetimes 1113944

NPrime

i1

σi

xi

1113888 1113889 times 100 (4)

where NPrime is the number of measurements sets i for eachdilution during the calibration test Note that only dilutionsthat led to measured values higher than the LOQ wereconsidered

3 Results and Discussion

31 Reduced S Species Distribution in pH-Buffered MediaBy mixing a sample containing sulfides with the three pH-buffered media N AC and H at pH of 92 47 and 1 re-spectively we will form more than 99 of HSminus in N andmore than 99 of H2S in AC andH according to the pKa ofthis species (Table 1) Second-derivative absorbance spectraacquired for the three different media (ie N0 AC0 andH0)show a maximum at a wavelength of 250 nm only for HSminus(ie N0) whereas H2S displays no specific signal in AC0 andH0 (Figure 2(a)) Similarly for sulfites media N and AC willcontain more than 99 of SO3

2minus and HSO3minus respectively

while mediumHwill consist of about 86 of SO2 and 14 ofHSO3minus Second-derivative spectrum H0 acquired in this

last medium presents a minimum for SO2 at 278 nm(Figure 2(b)) e signal for the SO3

2minus and HSO3minusmixture is

not discriminant At last for pH higher than 17 thiosulfateswill be present as S2O3

2minus Second-derivative spectra N0 andAC0 display in accordance the same signal intensity forS2O3

2minus with a maximum in absorbance at 244 nm(Figure 2(c)) while H0 corresponds to a combination

4 Journal of Chemistry

between the HS2O3minus and S2O3

2minus absorbance (estimatedproportion is 6040)

32 Method of Quantification When sulfites sulfides andthiosulfates are mixed together if one wants to isolate thesulfide characteristic signal at 250 nm (N Prime250) from those ofthiosulfates and sulfites the best option is to subtract thesecond-derivative spectrum AC0 from the second-derivativespectrum N0 Proceeding this way the contribution ofthiosulfates which is strictly the same in these two second-derivative spectra is thus cancelled e sulfides are notexpressed in ACPrime and the signal for the sulfite species re-mains low at the wavelength of 250 nm for both N0 and AC0

pH even if the difference between the two second-derivativespectra is non-negligible (Figures 2(a) and 2(b)) For sulfitequantification the second-derivative absorbance value at278 nm and pH 1 (H Prime278) can be directly read on spectrasince sulfides and thiosulfates do not interfere at thesewavelength and pH (Figure 2) In the same way thiosulfatescan also be directly measured at 243 nm at pH of 47 sincethere is no significant contribution of sulfides and sulfites inAC buffered medium (Figures 2(a) and 2(b)) Despite thefact that the maximum of the second-derivative absorbanceappears at 244 nm for thiosulfates (Figure 2(c)) the value at250 nm was preferred (AC Prime250) because the sulfite signalstarts to be significant below 250 nm at this pH(Figure 2(b))

One of the major hindrances for ionic quantificationsby spectrophotometry is sample salinity In brines anal-ysis this implies dilution steps or strong matrix effectsduring measurements [43ndash45] e present method wastested for salinities ranging from 0 to 100 gL Resultsdisplay linear trends for the normalized second-derivativeabsorbance of the three sulfur species (Figure 3) Sur-prisingly the sulfide signal is not significantly affectedand their quantification remains unchanged when salinityincreases For sulfites and thiosulfates the normalizedsecond-derivative absorbance is respectively increasedup to 80 or understated by 30 at the highest testedsalinity (ie 100 gL) and linear correction is thusmandatory Knowing that the solubility limit of NaCl inwater at 25degC is close to 317 gL [46] a 4 times dilution would

reduce all samples salinities below 100 gL and thus allowsperforming a linear correction for sulfites and thiosulfatesin every case

Calibration curves for each sulfur species were ob-tained for various dilutions leading to final concentrationsranging from 0 to 250mM for total sulfide and 0 to500mM for total sulfite and thiosulfate (Figure 4) ey allshow a good linearity according to the dose (with R2 valuesranging from 09992 to 09998) Coefficients of second-derivative absorbances α β and γ (for sulfides sulfites andthiosulfates respectively) were deduced from these cali-bration curves by a least squares regression A fourthcoefficient (δ) was also evaluated in order to provide acorrection for sulfide measurements regarding sulfiteinterferences

α sulfidesN Prime250 minusAC Prime250( 1113857 2389 times 10minus3 Lmmolcm

β sulfitesH Prime250( 1113857 minus 1480 times 10minus3 Lmmolcm

γ thiosulfatesAC Prime250( 1113857 4635 times 10minus3 Lmmolcm

δ sulfitesN Prime250 minusAC Prime250( 1113857 08627 times 10minus3 Lmmolcm

(5)

Once the different coefficients of second-derivativeabsorbance are defined with the help of the equations ofthe corresponding linear regressions it is possible toestablish formulas taking into account the different cor-rections that are required ese corrections correspondto the subtraction of the sulfite contribution to the sulfidequantification and the linear correction for salinity whichis mandatory for sulfite and thiosulfate quantification(with linear regression shown in Figure 3) ose quan-tification equations are described hereafter (Equations (6)to (8)) where concentrations are in mmolL and salinity(S) is in gL

[sulfides] N Prime250minus AC Prime250( 1113857 + 09389 times 10minus5 minus[sulfites] times δ

α

(6)

[sulfites] H Prime278 minus 02748 times 10minus5

(1 + S times 00074) times β (7)

[thiosulfates] AC Prime250 minus 2098 times 10minus5

(1minus S times 00026) times γ (8)

33 Potential Interferences Other reduced sulfur speciesmay also interfere in such kind of approach Above pH 7colloidal sulfur rings which represent the dominant form ofelemental sulfur in polar solvents and are the precursor ofcommon orthorhombic sulfur crystals [46] tend to formyellow-orange solutions of polysulfides by association withHSminus ose polymers have an average size of about 5 sulfuratoms ey absorb in the UV-bc range and could bequantified as Sdeg in Sx2minus by direct measurement at 285 nm forpHs between 8 and 10 with an average molar extinctioncoefficient of 137010minus3 Lmmolcm [47 48] Neverthelessthe kinetics of formation and decomposition at room

Table 1 pKa of frequently encountered sulfur species (in water at25degC)

AcidsBases pKa (s) ReferencesH2SHSminusS2minus 70129 [33]H2S2HS2minusS22minus 5097 [34]H2S3HS3minusS32minus 4275 [34]H2S4HS4minusS42minus 3863 [34]H2S5HS5minusS52minus 3557 [34]H2S2O3HS2O3

minusS2O32minus 0617 [35]

H2SO5HSO5minusSO5

2minus lt093 [36]H2SxO6HSxO6

minusSxO62minus (x 3ndash5) minus23 [37]

H2SO3HSO3minusSO3

2minus 1872 [38 39]H2SO4HSO4

minusSO42minus minus3020 [40 41]

HSCNSCNminus 13 [42]S2O4H2S2O4HminusS2O4

2minus 0324 [33]

Journal of Chemistry 5

240 260 280Wavelength (nm)

ndash1000

ndash500

0

500

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(a)

240 260 280Wavelength (nm)

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

ndash100

0

100

200

300

NPrime

HPrime

ACPrime

(b)

240 260 280Wavelength (nm)

ndash100

0

100

200

300

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(c)

Figure 2 Second-derivative absorbance spectra (N0 AC0 and H0) for (a) suldes (Na2S solution at 235 micromolL) (b) sultes (Na2SO3solution at 490 micromolL) and (c) thiosulfates (Na2S2O3 solution at 500 micromolL) between 220 and 300 nm at 25degC in three dierent bueredmedia (H pH 10 AC pH 47 and N pH 92) Note that H0 and AC0 spectra merge in (a) while N0 and AC0 are superposed in (c)Suldes display a maximum at 250 nm for a pH of 92 (N0) while a minimum is reached for sultes at 278 nm for a pH of 1 (H0) Forthiosulfates maxima are found at 244 nm for the three media (N0 H0 and AC0) Optical path length and Δλ were of 10mm and 8 nmrespectively

6 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

ISOPOREtrade) and directly added to three dierent pH-buered media of decreasing pH (N AC and H mediarespectively) with (i)N corresponding to NH3 (625mmolL or 4 microLL from a 25ndash30NH3 solution) mixed with NH4Cl(625mmolL) leading to a nal pH of 92 (ii) AC corre-sponding to a mix of Na-acetate trihydrate (625mmolL)and acetic acid (625mmolL) with a nal pH of 47 and (iii)H corresponding to HCl at 01molL with an associated pHof 1 All these media were prepared in UPW and formulatedin order to reach a nal volume of 10mL after sampleaddition Variations in the injected volume of sample in thisformulation allowed dierent dilutions

Spectra were recorded for each medium right aftersample addition using the following parameters spectralacquisition between 220 and 300 nm wavelength scanningspeed (slow 32 nmmin) single acquisition and databunching interval of 01 nm If not specied the blanksassociated with these measurements were made with the

same buered medium (N AC or H) without any additionof the sample Acquisitions were performed at controlledroom temperature (25degC) with thermally equilibratedsolutions

24 UV-Visible Spectrophotometric Characterization of theBuered Media e UV-visible spectral signature of eachmedium (N AC and H) was established by a series of 10measurements For these acquisitions blanks consisted inUPWe pH stability of the media was frequently checkedand care was taken to not let the recipients open to theatmosphere in order to avoid changes in the neutral species(ie NH3 and acetic acid) due to air-solution exchanges

25 Calibration Curves Each of the three species of interest(suldes sultes and thiosulfates) was analyzed three timesin each media (made independently for each measurements)

Sample

Filtration(02 microm)

AC N H

pH = 47 pH = 92 pH = 1

Dilution into thebuffered solutions

UV-visiblespectrophotometry

(220 to 300nm)

Second derivatives NPrime250-ACPrime250ACPrime250

HPrime278

Total sulfideTotal thiosulfateTotal sulfite

Proportional relationships

Figure 1 Schematic representation of the quantication procedure proposed in this study From a ltered sample three direct spec-trophotometric spectra are acquired in three dierent bueredmedia (N pH 92AC pH 47 andH pH 10) in the range 220ndash300 nmSpectra are dierentiated to the second degree (N0 AC0 and H0) and values of second derivative absorbance are extracted at specicwavelengths (N0250 AC0250 at 250 nm andH0278 at 278 nm) ese values can be used to directly quantify total thiosulfate (H2S2O3HS2O3

minusS2O3

2minus) and sulte (SO2HSO3minusSO3

2minus) contents or after subtraction total sulde concentration (H2SHSminusS2minus)

Journal of Chemistry 3

at a concentration of 05mmolL (or 025mmolL for sul-fides in medium N) Each measurement was repeated twotimes (gap of 3min between the onset of each measurement)in the same cuvette in order to detect and quantify anytemporal modification of the signal (eg oxidation by airdioxygen) e individual calibration lines for each specieswere then obtained by measuring 5 different dilutions withfinal concentrations ranging from 500 to 4 micromolL

26 Interferences Tests Tests for specific interferences wereconducted by first mixing the three sulfur species (sulfidessulfites and thiosulfates) at 100 micromolL each with solutionat different salinities (100 146 73 365 and 0 gL of NaCl)in order to estimate the impact of ionic strength increase onthe signal stability then with a synthetic seawater (ie TYAmedium described in [30]) diluted 10 and 40 times in orderto mimic general ionic background of saline and marinewaters and finally with 10mgL of humic acid because itrepresents the main organic constituent of surface aquaticsystems Additionally UV-visible spectra were also acquiredon nitrate- NaClminus and humic acid solutions and on theTYA medium separately mixed with each of the three media(NAC andH) Tests were also performed with yeast extractbecause it provides a large diversity of organic compounds(eg proteins amino acids and carbohydrates) that arefrequently found associated with animal andmicrobial waterpollution All these measurements were at least triplicatedInterference threshold values were defined as the concen-tration that gives a signal variation equal to the corre-sponding smallest limit of detection (see below) In parallela linearity comparison test was also accomplished with thespectrophotometric methylene blue method(SPECTROQUANTreg test kit Ndeg 11477790001 Merck)using natural freshwaters spiked with sulfides Finallymeasurements were also performed on a groundwatersample exhibiting natural occurrences of sulfides sulfitesand thiosulfates and collected in the Dogger formation of theParis basin (Orleans France) In order to validate the presentapproach the same groundwater sample was controlled withstandardized methods Sulfur was measured using themethylene blue coloring method (ISO 10530 1992) Sulfiteswere measured using the iodometric titration method(SANDRE 296) and thiosulfates were quantified with anion-exchange chromatography protocol on ICS-3000 (ISO10304-E)

27 Data Treatment and Statistical Approach All the spectrawere processed using the UVProbe software second-de-rivative function which can be described by equation (1)from [31]

d2A

dλ2

A λprime + Δλ( 1113857minus 2A λprime( 1113857 + A λprime minus Δλ( 1113857

(Δλ)2 (1)

where A is the absorbance at a given wavelength λ λprime is thewavelength at which the second-derivative absorbancevalue is calculated and Δλ is the wavelength interval In thisstudy Δλ was fixed at 8 nm in order to reduce the narrow-band noise effect [28 32] Second-derivative spectra from

media N AC and H were named N0 AC0 and H0respectively

Spectra subtractions were also performed with theUVProbe software Calibration lines were calculated usingthe standard linear least square method from MICRO-SOFTreg EXCEL

In this study the limit of detection (LOD) and the limitof quantification (LOQ) correspond to 3σ and 10σ re-spectively with σ representing the standard deviation of theblanks

σ

1N

times 1113944N

i1xi minus x( 1113857

2

11139741113972

(2)

where N is the total number of measurements i and x themean of the measurement values x

Accuracy was determined by

Accuracy 1

Nprimetimes 1113944

Nprime

i1

mi minus ri

ri

1113888 1113889 times 100 (3)

where Nprime is the number of measurements i to which cor-responds a real (ri) and a measured (mi) absolute valuesOnly values from the calibration test higher than the LOQwere considered

Precision was assessed using the relative standard de-viation (RSD) defined in

RSD 1

NPrimetimes 1113944

NPrime

i1

σi

xi

1113888 1113889 times 100 (4)

where NPrime is the number of measurements sets i for eachdilution during the calibration test Note that only dilutionsthat led to measured values higher than the LOQ wereconsidered

3 Results and Discussion

31 Reduced S Species Distribution in pH-Buffered MediaBy mixing a sample containing sulfides with the three pH-buffered media N AC and H at pH of 92 47 and 1 re-spectively we will form more than 99 of HSminus in N andmore than 99 of H2S in AC andH according to the pKa ofthis species (Table 1) Second-derivative absorbance spectraacquired for the three different media (ie N0 AC0 andH0)show a maximum at a wavelength of 250 nm only for HSminus(ie N0) whereas H2S displays no specific signal in AC0 andH0 (Figure 2(a)) Similarly for sulfites media N and AC willcontain more than 99 of SO3

2minus and HSO3minus respectively

while mediumHwill consist of about 86 of SO2 and 14 ofHSO3minus Second-derivative spectrum H0 acquired in this

last medium presents a minimum for SO2 at 278 nm(Figure 2(b)) e signal for the SO3

2minus and HSO3minusmixture is

not discriminant At last for pH higher than 17 thiosulfateswill be present as S2O3

2minus Second-derivative spectra N0 andAC0 display in accordance the same signal intensity forS2O3

2minus with a maximum in absorbance at 244 nm(Figure 2(c)) while H0 corresponds to a combination

4 Journal of Chemistry

between the HS2O3minus and S2O3

2minus absorbance (estimatedproportion is 6040)

32 Method of Quantification When sulfites sulfides andthiosulfates are mixed together if one wants to isolate thesulfide characteristic signal at 250 nm (N Prime250) from those ofthiosulfates and sulfites the best option is to subtract thesecond-derivative spectrum AC0 from the second-derivativespectrum N0 Proceeding this way the contribution ofthiosulfates which is strictly the same in these two second-derivative spectra is thus cancelled e sulfides are notexpressed in ACPrime and the signal for the sulfite species re-mains low at the wavelength of 250 nm for both N0 and AC0

pH even if the difference between the two second-derivativespectra is non-negligible (Figures 2(a) and 2(b)) For sulfitequantification the second-derivative absorbance value at278 nm and pH 1 (H Prime278) can be directly read on spectrasince sulfides and thiosulfates do not interfere at thesewavelength and pH (Figure 2) In the same way thiosulfatescan also be directly measured at 243 nm at pH of 47 sincethere is no significant contribution of sulfides and sulfites inAC buffered medium (Figures 2(a) and 2(b)) Despite thefact that the maximum of the second-derivative absorbanceappears at 244 nm for thiosulfates (Figure 2(c)) the value at250 nm was preferred (AC Prime250) because the sulfite signalstarts to be significant below 250 nm at this pH(Figure 2(b))

One of the major hindrances for ionic quantificationsby spectrophotometry is sample salinity In brines anal-ysis this implies dilution steps or strong matrix effectsduring measurements [43ndash45] e present method wastested for salinities ranging from 0 to 100 gL Resultsdisplay linear trends for the normalized second-derivativeabsorbance of the three sulfur species (Figure 3) Sur-prisingly the sulfide signal is not significantly affectedand their quantification remains unchanged when salinityincreases For sulfites and thiosulfates the normalizedsecond-derivative absorbance is respectively increasedup to 80 or understated by 30 at the highest testedsalinity (ie 100 gL) and linear correction is thusmandatory Knowing that the solubility limit of NaCl inwater at 25degC is close to 317 gL [46] a 4 times dilution would

reduce all samples salinities below 100 gL and thus allowsperforming a linear correction for sulfites and thiosulfatesin every case

Calibration curves for each sulfur species were ob-tained for various dilutions leading to final concentrationsranging from 0 to 250mM for total sulfide and 0 to500mM for total sulfite and thiosulfate (Figure 4) ey allshow a good linearity according to the dose (with R2 valuesranging from 09992 to 09998) Coefficients of second-derivative absorbances α β and γ (for sulfides sulfites andthiosulfates respectively) were deduced from these cali-bration curves by a least squares regression A fourthcoefficient (δ) was also evaluated in order to provide acorrection for sulfide measurements regarding sulfiteinterferences

α sulfidesN Prime250 minusAC Prime250( 1113857 2389 times 10minus3 Lmmolcm

β sulfitesH Prime250( 1113857 minus 1480 times 10minus3 Lmmolcm

γ thiosulfatesAC Prime250( 1113857 4635 times 10minus3 Lmmolcm

δ sulfitesN Prime250 minusAC Prime250( 1113857 08627 times 10minus3 Lmmolcm

(5)

Once the different coefficients of second-derivativeabsorbance are defined with the help of the equations ofthe corresponding linear regressions it is possible toestablish formulas taking into account the different cor-rections that are required ese corrections correspondto the subtraction of the sulfite contribution to the sulfidequantification and the linear correction for salinity whichis mandatory for sulfite and thiosulfate quantification(with linear regression shown in Figure 3) ose quan-tification equations are described hereafter (Equations (6)to (8)) where concentrations are in mmolL and salinity(S) is in gL

[sulfides] N Prime250minus AC Prime250( 1113857 + 09389 times 10minus5 minus[sulfites] times δ

α

(6)

[sulfites] H Prime278 minus 02748 times 10minus5

(1 + S times 00074) times β (7)

[thiosulfates] AC Prime250 minus 2098 times 10minus5

(1minus S times 00026) times γ (8)

33 Potential Interferences Other reduced sulfur speciesmay also interfere in such kind of approach Above pH 7colloidal sulfur rings which represent the dominant form ofelemental sulfur in polar solvents and are the precursor ofcommon orthorhombic sulfur crystals [46] tend to formyellow-orange solutions of polysulfides by association withHSminus ose polymers have an average size of about 5 sulfuratoms ey absorb in the UV-bc range and could bequantified as Sdeg in Sx2minus by direct measurement at 285 nm forpHs between 8 and 10 with an average molar extinctioncoefficient of 137010minus3 Lmmolcm [47 48] Neverthelessthe kinetics of formation and decomposition at room

Table 1 pKa of frequently encountered sulfur species (in water at25degC)

AcidsBases pKa (s) ReferencesH2SHSminusS2minus 70129 [33]H2S2HS2minusS22minus 5097 [34]H2S3HS3minusS32minus 4275 [34]H2S4HS4minusS42minus 3863 [34]H2S5HS5minusS52minus 3557 [34]H2S2O3HS2O3

minusS2O32minus 0617 [35]

H2SO5HSO5minusSO5

2minus lt093 [36]H2SxO6HSxO6

minusSxO62minus (x 3ndash5) minus23 [37]

H2SO3HSO3minusSO3

2minus 1872 [38 39]H2SO4HSO4

minusSO42minus minus3020 [40 41]

HSCNSCNminus 13 [42]S2O4H2S2O4HminusS2O4

2minus 0324 [33]

Journal of Chemistry 5

240 260 280Wavelength (nm)

ndash1000

ndash500

0

500

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(a)

240 260 280Wavelength (nm)

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

ndash100

0

100

200

300

NPrime

HPrime

ACPrime

(b)

240 260 280Wavelength (nm)

ndash100

0

100

200

300

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(c)

Figure 2 Second-derivative absorbance spectra (N0 AC0 and H0) for (a) suldes (Na2S solution at 235 micromolL) (b) sultes (Na2SO3solution at 490 micromolL) and (c) thiosulfates (Na2S2O3 solution at 500 micromolL) between 220 and 300 nm at 25degC in three dierent bueredmedia (H pH 10 AC pH 47 and N pH 92) Note that H0 and AC0 spectra merge in (a) while N0 and AC0 are superposed in (c)Suldes display a maximum at 250 nm for a pH of 92 (N0) while a minimum is reached for sultes at 278 nm for a pH of 1 (H0) Forthiosulfates maxima are found at 244 nm for the three media (N0 H0 and AC0) Optical path length and Δλ were of 10mm and 8 nmrespectively

6 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

at a concentration of 05mmolL (or 025mmolL for sul-fides in medium N) Each measurement was repeated twotimes (gap of 3min between the onset of each measurement)in the same cuvette in order to detect and quantify anytemporal modification of the signal (eg oxidation by airdioxygen) e individual calibration lines for each specieswere then obtained by measuring 5 different dilutions withfinal concentrations ranging from 500 to 4 micromolL

26 Interferences Tests Tests for specific interferences wereconducted by first mixing the three sulfur species (sulfidessulfites and thiosulfates) at 100 micromolL each with solutionat different salinities (100 146 73 365 and 0 gL of NaCl)in order to estimate the impact of ionic strength increase onthe signal stability then with a synthetic seawater (ie TYAmedium described in [30]) diluted 10 and 40 times in orderto mimic general ionic background of saline and marinewaters and finally with 10mgL of humic acid because itrepresents the main organic constituent of surface aquaticsystems Additionally UV-visible spectra were also acquiredon nitrate- NaClminus and humic acid solutions and on theTYA medium separately mixed with each of the three media(NAC andH) Tests were also performed with yeast extractbecause it provides a large diversity of organic compounds(eg proteins amino acids and carbohydrates) that arefrequently found associated with animal andmicrobial waterpollution All these measurements were at least triplicatedInterference threshold values were defined as the concen-tration that gives a signal variation equal to the corre-sponding smallest limit of detection (see below) In parallela linearity comparison test was also accomplished with thespectrophotometric methylene blue method(SPECTROQUANTreg test kit Ndeg 11477790001 Merck)using natural freshwaters spiked with sulfides Finallymeasurements were also performed on a groundwatersample exhibiting natural occurrences of sulfides sulfitesand thiosulfates and collected in the Dogger formation of theParis basin (Orleans France) In order to validate the presentapproach the same groundwater sample was controlled withstandardized methods Sulfur was measured using themethylene blue coloring method (ISO 10530 1992) Sulfiteswere measured using the iodometric titration method(SANDRE 296) and thiosulfates were quantified with anion-exchange chromatography protocol on ICS-3000 (ISO10304-E)

27 Data Treatment and Statistical Approach All the spectrawere processed using the UVProbe software second-de-rivative function which can be described by equation (1)from [31]

d2A

dλ2

A λprime + Δλ( 1113857minus 2A λprime( 1113857 + A λprime minus Δλ( 1113857

(Δλ)2 (1)

where A is the absorbance at a given wavelength λ λprime is thewavelength at which the second-derivative absorbancevalue is calculated and Δλ is the wavelength interval In thisstudy Δλ was fixed at 8 nm in order to reduce the narrow-band noise effect [28 32] Second-derivative spectra from

media N AC and H were named N0 AC0 and H0respectively

Spectra subtractions were also performed with theUVProbe software Calibration lines were calculated usingthe standard linear least square method from MICRO-SOFTreg EXCEL

In this study the limit of detection (LOD) and the limitof quantification (LOQ) correspond to 3σ and 10σ re-spectively with σ representing the standard deviation of theblanks

σ

1N

times 1113944N

i1xi minus x( 1113857

2

11139741113972

(2)

where N is the total number of measurements i and x themean of the measurement values x

Accuracy was determined by

Accuracy 1

Nprimetimes 1113944

Nprime

i1

mi minus ri

ri

1113888 1113889 times 100 (3)

where Nprime is the number of measurements i to which cor-responds a real (ri) and a measured (mi) absolute valuesOnly values from the calibration test higher than the LOQwere considered

Precision was assessed using the relative standard de-viation (RSD) defined in

RSD 1

NPrimetimes 1113944

NPrime

i1

σi

xi

1113888 1113889 times 100 (4)

where NPrime is the number of measurements sets i for eachdilution during the calibration test Note that only dilutionsthat led to measured values higher than the LOQ wereconsidered

3 Results and Discussion

31 Reduced S Species Distribution in pH-Buffered MediaBy mixing a sample containing sulfides with the three pH-buffered media N AC and H at pH of 92 47 and 1 re-spectively we will form more than 99 of HSminus in N andmore than 99 of H2S in AC andH according to the pKa ofthis species (Table 1) Second-derivative absorbance spectraacquired for the three different media (ie N0 AC0 andH0)show a maximum at a wavelength of 250 nm only for HSminus(ie N0) whereas H2S displays no specific signal in AC0 andH0 (Figure 2(a)) Similarly for sulfites media N and AC willcontain more than 99 of SO3

2minus and HSO3minus respectively

while mediumHwill consist of about 86 of SO2 and 14 ofHSO3minus Second-derivative spectrum H0 acquired in this

last medium presents a minimum for SO2 at 278 nm(Figure 2(b)) e signal for the SO3

2minus and HSO3minusmixture is

not discriminant At last for pH higher than 17 thiosulfateswill be present as S2O3

2minus Second-derivative spectra N0 andAC0 display in accordance the same signal intensity forS2O3

2minus with a maximum in absorbance at 244 nm(Figure 2(c)) while H0 corresponds to a combination

4 Journal of Chemistry

between the HS2O3minus and S2O3

2minus absorbance (estimatedproportion is 6040)

32 Method of Quantification When sulfites sulfides andthiosulfates are mixed together if one wants to isolate thesulfide characteristic signal at 250 nm (N Prime250) from those ofthiosulfates and sulfites the best option is to subtract thesecond-derivative spectrum AC0 from the second-derivativespectrum N0 Proceeding this way the contribution ofthiosulfates which is strictly the same in these two second-derivative spectra is thus cancelled e sulfides are notexpressed in ACPrime and the signal for the sulfite species re-mains low at the wavelength of 250 nm for both N0 and AC0

pH even if the difference between the two second-derivativespectra is non-negligible (Figures 2(a) and 2(b)) For sulfitequantification the second-derivative absorbance value at278 nm and pH 1 (H Prime278) can be directly read on spectrasince sulfides and thiosulfates do not interfere at thesewavelength and pH (Figure 2) In the same way thiosulfatescan also be directly measured at 243 nm at pH of 47 sincethere is no significant contribution of sulfides and sulfites inAC buffered medium (Figures 2(a) and 2(b)) Despite thefact that the maximum of the second-derivative absorbanceappears at 244 nm for thiosulfates (Figure 2(c)) the value at250 nm was preferred (AC Prime250) because the sulfite signalstarts to be significant below 250 nm at this pH(Figure 2(b))

One of the major hindrances for ionic quantificationsby spectrophotometry is sample salinity In brines anal-ysis this implies dilution steps or strong matrix effectsduring measurements [43ndash45] e present method wastested for salinities ranging from 0 to 100 gL Resultsdisplay linear trends for the normalized second-derivativeabsorbance of the three sulfur species (Figure 3) Sur-prisingly the sulfide signal is not significantly affectedand their quantification remains unchanged when salinityincreases For sulfites and thiosulfates the normalizedsecond-derivative absorbance is respectively increasedup to 80 or understated by 30 at the highest testedsalinity (ie 100 gL) and linear correction is thusmandatory Knowing that the solubility limit of NaCl inwater at 25degC is close to 317 gL [46] a 4 times dilution would

reduce all samples salinities below 100 gL and thus allowsperforming a linear correction for sulfites and thiosulfatesin every case

Calibration curves for each sulfur species were ob-tained for various dilutions leading to final concentrationsranging from 0 to 250mM for total sulfide and 0 to500mM for total sulfite and thiosulfate (Figure 4) ey allshow a good linearity according to the dose (with R2 valuesranging from 09992 to 09998) Coefficients of second-derivative absorbances α β and γ (for sulfides sulfites andthiosulfates respectively) were deduced from these cali-bration curves by a least squares regression A fourthcoefficient (δ) was also evaluated in order to provide acorrection for sulfide measurements regarding sulfiteinterferences

α sulfidesN Prime250 minusAC Prime250( 1113857 2389 times 10minus3 Lmmolcm

β sulfitesH Prime250( 1113857 minus 1480 times 10minus3 Lmmolcm

γ thiosulfatesAC Prime250( 1113857 4635 times 10minus3 Lmmolcm

δ sulfitesN Prime250 minusAC Prime250( 1113857 08627 times 10minus3 Lmmolcm

(5)

Once the different coefficients of second-derivativeabsorbance are defined with the help of the equations ofthe corresponding linear regressions it is possible toestablish formulas taking into account the different cor-rections that are required ese corrections correspondto the subtraction of the sulfite contribution to the sulfidequantification and the linear correction for salinity whichis mandatory for sulfite and thiosulfate quantification(with linear regression shown in Figure 3) ose quan-tification equations are described hereafter (Equations (6)to (8)) where concentrations are in mmolL and salinity(S) is in gL

[sulfides] N Prime250minus AC Prime250( 1113857 + 09389 times 10minus5 minus[sulfites] times δ

α

(6)

[sulfites] H Prime278 minus 02748 times 10minus5

(1 + S times 00074) times β (7)

[thiosulfates] AC Prime250 minus 2098 times 10minus5

(1minus S times 00026) times γ (8)

33 Potential Interferences Other reduced sulfur speciesmay also interfere in such kind of approach Above pH 7colloidal sulfur rings which represent the dominant form ofelemental sulfur in polar solvents and are the precursor ofcommon orthorhombic sulfur crystals [46] tend to formyellow-orange solutions of polysulfides by association withHSminus ose polymers have an average size of about 5 sulfuratoms ey absorb in the UV-bc range and could bequantified as Sdeg in Sx2minus by direct measurement at 285 nm forpHs between 8 and 10 with an average molar extinctioncoefficient of 137010minus3 Lmmolcm [47 48] Neverthelessthe kinetics of formation and decomposition at room

Table 1 pKa of frequently encountered sulfur species (in water at25degC)

AcidsBases pKa (s) ReferencesH2SHSminusS2minus 70129 [33]H2S2HS2minusS22minus 5097 [34]H2S3HS3minusS32minus 4275 [34]H2S4HS4minusS42minus 3863 [34]H2S5HS5minusS52minus 3557 [34]H2S2O3HS2O3

minusS2O32minus 0617 [35]

H2SO5HSO5minusSO5

2minus lt093 [36]H2SxO6HSxO6

minusSxO62minus (x 3ndash5) minus23 [37]

H2SO3HSO3minusSO3

2minus 1872 [38 39]H2SO4HSO4

minusSO42minus minus3020 [40 41]

HSCNSCNminus 13 [42]S2O4H2S2O4HminusS2O4

2minus 0324 [33]

Journal of Chemistry 5

240 260 280Wavelength (nm)

ndash1000

ndash500

0

500

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(a)

240 260 280Wavelength (nm)

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

ndash100

0

100

200

300

NPrime

HPrime

ACPrime

(b)

240 260 280Wavelength (nm)

ndash100

0

100

200

300

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(c)

Figure 2 Second-derivative absorbance spectra (N0 AC0 and H0) for (a) suldes (Na2S solution at 235 micromolL) (b) sultes (Na2SO3solution at 490 micromolL) and (c) thiosulfates (Na2S2O3 solution at 500 micromolL) between 220 and 300 nm at 25degC in three dierent bueredmedia (H pH 10 AC pH 47 and N pH 92) Note that H0 and AC0 spectra merge in (a) while N0 and AC0 are superposed in (c)Suldes display a maximum at 250 nm for a pH of 92 (N0) while a minimum is reached for sultes at 278 nm for a pH of 1 (H0) Forthiosulfates maxima are found at 244 nm for the three media (N0 H0 and AC0) Optical path length and Δλ were of 10mm and 8 nmrespectively

6 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

between the HS2O3minus and S2O3

2minus absorbance (estimatedproportion is 6040)

32 Method of Quantification When sulfites sulfides andthiosulfates are mixed together if one wants to isolate thesulfide characteristic signal at 250 nm (N Prime250) from those ofthiosulfates and sulfites the best option is to subtract thesecond-derivative spectrum AC0 from the second-derivativespectrum N0 Proceeding this way the contribution ofthiosulfates which is strictly the same in these two second-derivative spectra is thus cancelled e sulfides are notexpressed in ACPrime and the signal for the sulfite species re-mains low at the wavelength of 250 nm for both N0 and AC0

pH even if the difference between the two second-derivativespectra is non-negligible (Figures 2(a) and 2(b)) For sulfitequantification the second-derivative absorbance value at278 nm and pH 1 (H Prime278) can be directly read on spectrasince sulfides and thiosulfates do not interfere at thesewavelength and pH (Figure 2) In the same way thiosulfatescan also be directly measured at 243 nm at pH of 47 sincethere is no significant contribution of sulfides and sulfites inAC buffered medium (Figures 2(a) and 2(b)) Despite thefact that the maximum of the second-derivative absorbanceappears at 244 nm for thiosulfates (Figure 2(c)) the value at250 nm was preferred (AC Prime250) because the sulfite signalstarts to be significant below 250 nm at this pH(Figure 2(b))

One of the major hindrances for ionic quantificationsby spectrophotometry is sample salinity In brines anal-ysis this implies dilution steps or strong matrix effectsduring measurements [43ndash45] e present method wastested for salinities ranging from 0 to 100 gL Resultsdisplay linear trends for the normalized second-derivativeabsorbance of the three sulfur species (Figure 3) Sur-prisingly the sulfide signal is not significantly affectedand their quantification remains unchanged when salinityincreases For sulfites and thiosulfates the normalizedsecond-derivative absorbance is respectively increasedup to 80 or understated by 30 at the highest testedsalinity (ie 100 gL) and linear correction is thusmandatory Knowing that the solubility limit of NaCl inwater at 25degC is close to 317 gL [46] a 4 times dilution would

reduce all samples salinities below 100 gL and thus allowsperforming a linear correction for sulfites and thiosulfatesin every case

Calibration curves for each sulfur species were ob-tained for various dilutions leading to final concentrationsranging from 0 to 250mM for total sulfide and 0 to500mM for total sulfite and thiosulfate (Figure 4) ey allshow a good linearity according to the dose (with R2 valuesranging from 09992 to 09998) Coefficients of second-derivative absorbances α β and γ (for sulfides sulfites andthiosulfates respectively) were deduced from these cali-bration curves by a least squares regression A fourthcoefficient (δ) was also evaluated in order to provide acorrection for sulfide measurements regarding sulfiteinterferences

α sulfidesN Prime250 minusAC Prime250( 1113857 2389 times 10minus3 Lmmolcm

β sulfitesH Prime250( 1113857 minus 1480 times 10minus3 Lmmolcm

γ thiosulfatesAC Prime250( 1113857 4635 times 10minus3 Lmmolcm

δ sulfitesN Prime250 minusAC Prime250( 1113857 08627 times 10minus3 Lmmolcm

(5)

Once the different coefficients of second-derivativeabsorbance are defined with the help of the equations ofthe corresponding linear regressions it is possible toestablish formulas taking into account the different cor-rections that are required ese corrections correspondto the subtraction of the sulfite contribution to the sulfidequantification and the linear correction for salinity whichis mandatory for sulfite and thiosulfate quantification(with linear regression shown in Figure 3) ose quan-tification equations are described hereafter (Equations (6)to (8)) where concentrations are in mmolL and salinity(S) is in gL

[sulfides] N Prime250minus AC Prime250( 1113857 + 09389 times 10minus5 minus[sulfites] times δ

α

(6)

[sulfites] H Prime278 minus 02748 times 10minus5

(1 + S times 00074) times β (7)

[thiosulfates] AC Prime250 minus 2098 times 10minus5

(1minus S times 00026) times γ (8)

33 Potential Interferences Other reduced sulfur speciesmay also interfere in such kind of approach Above pH 7colloidal sulfur rings which represent the dominant form ofelemental sulfur in polar solvents and are the precursor ofcommon orthorhombic sulfur crystals [46] tend to formyellow-orange solutions of polysulfides by association withHSminus ose polymers have an average size of about 5 sulfuratoms ey absorb in the UV-bc range and could bequantified as Sdeg in Sx2minus by direct measurement at 285 nm forpHs between 8 and 10 with an average molar extinctioncoefficient of 137010minus3 Lmmolcm [47 48] Neverthelessthe kinetics of formation and decomposition at room

Table 1 pKa of frequently encountered sulfur species (in water at25degC)

AcidsBases pKa (s) ReferencesH2SHSminusS2minus 70129 [33]H2S2HS2minusS22minus 5097 [34]H2S3HS3minusS32minus 4275 [34]H2S4HS4minusS42minus 3863 [34]H2S5HS5minusS52minus 3557 [34]H2S2O3HS2O3

minusS2O32minus 0617 [35]

H2SO5HSO5minusSO5

2minus lt093 [36]H2SxO6HSxO6

minusSxO62minus (x 3ndash5) minus23 [37]

H2SO3HSO3minusSO3

2minus 1872 [38 39]H2SO4HSO4

minusSO42minus minus3020 [40 41]

HSCNSCNminus 13 [42]S2O4H2S2O4HminusS2O4

2minus 0324 [33]

Journal of Chemistry 5

240 260 280Wavelength (nm)

ndash1000

ndash500

0

500

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(a)

240 260 280Wavelength (nm)

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

ndash100

0

100

200

300

NPrime

HPrime

ACPrime

(b)

240 260 280Wavelength (nm)

ndash100

0

100

200

300

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(c)

Figure 2 Second-derivative absorbance spectra (N0 AC0 and H0) for (a) suldes (Na2S solution at 235 micromolL) (b) sultes (Na2SO3solution at 490 micromolL) and (c) thiosulfates (Na2S2O3 solution at 500 micromolL) between 220 and 300 nm at 25degC in three dierent bueredmedia (H pH 10 AC pH 47 and N pH 92) Note that H0 and AC0 spectra merge in (a) while N0 and AC0 are superposed in (c)Suldes display a maximum at 250 nm for a pH of 92 (N0) while a minimum is reached for sultes at 278 nm for a pH of 1 (H0) Forthiosulfates maxima are found at 244 nm for the three media (N0 H0 and AC0) Optical path length and Δλ were of 10mm and 8 nmrespectively

6 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

240 260 280Wavelength (nm)

ndash1000

ndash500

0

500

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(a)

240 260 280Wavelength (nm)

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

ndash100

0

100

200

300

NPrime

HPrime

ACPrime

(b)

240 260 280Wavelength (nm)

ndash100

0

100

200

300

Seco

nd-d

eriv

ativ

e abs

orba

nce times

105

NPrime

HPrime

ACPrime

(c)

Figure 2 Second-derivative absorbance spectra (N0 AC0 and H0) for (a) suldes (Na2S solution at 235 micromolL) (b) sultes (Na2SO3solution at 490 micromolL) and (c) thiosulfates (Na2S2O3 solution at 500 micromolL) between 220 and 300 nm at 25degC in three dierent bueredmedia (H pH 10 AC pH 47 and N pH 92) Note that H0 and AC0 spectra merge in (a) while N0 and AC0 are superposed in (c)Suldes display a maximum at 250 nm for a pH of 92 (N0) while a minimum is reached for sultes at 278 nm for a pH of 1 (H0) Forthiosulfates maxima are found at 244 nm for the three media (N0 H0 and AC0) Optical path length and Δλ were of 10mm and 8 nmrespectively

6 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

temperature are not fast enough to signicantly interfere intime with a spectrophotometric measurement [49] Similarstatement can be made for polysulfanes (H2Sx) which areadditionally quite insoluble [50] Consequently no newlyformed polysuldes suldes or Sdeg should occur during themeasurement preparation Additionally the contribution ofthe original polysuldes to sulde quantication is cancelledanyway by the second-derivative spectra subtraction asso-ciated with our method To the contrary quantications ofthiosulfates and sultes might be hampered by the presenceof original polysuldes

Polythionates (ie polythiosulfates) display also a strongabsorbance in the UV-c range (with a maximum around215 nm) and have a spectral pattern similar to thiosulfateswith the exception of trithionate that does not absorb lightbetween 220 and 300 nm [23] ese species can be found inacidic hot springs leach solutions and volcano lakes wherethe dominant species are the tetrathionate followed by thepenta- and trithionate [12] Even if these species may slightlycontribute to thiosulfate estimates their concentrations areusually very low compared to thiosulfate content and thusthis should not induce signicant modication of the

y = 2389x ndash 09388

R2 = 09997

0

150

300

450

600

0 005 010 015 020 025 030Total sulfide (mM)

NPrime

250-AC

Prime25

0 times 1

05

(a)

y = ndash1480x + 02748 R2 = 09998

ndash80

ndash60

ndash40

ndash20

00 01 02 03 04 05 06

HPrime

278 times

105

Total sulfite (mM)

(b)

y = 4634x + 2097R2 = 09992

0

50

100

150

200

250

0 01 02 03 04 05 06

ACPrime

250 times

105

Total thiosulfate (mM)

(c)

Figure 4 Calibration curves of the second-derivative absorbance parameters for total sulde (a) total sulte (b) and total thiosulfate (c) inwater at 25degC

y = 00074x + 1016R2 = 09989

y = ndash00026x + 09674R2 = 09664

y = ndash00008x + 10227R2 = 08364

05

1

0 20 40 60 80 100N

orm

aliz

ed se

cond

-der

ivat

ive

abso

rban

ceSalinity (gL)

15

Figure 3 Normalized second-derivative absorbance as a function of salinity values forN Prime250-AC Prime250 (second-derivative values at 250 nm formedia (N and AC respectively)) are represented by diamonds and correspond to total sulde (H2SHSminusS2minus) Values H Prime278 (second-derivative values at 278 nm for mediaH and representing total sulte SO2HSO3

minusSO32minus) andAC Prime250 (second-derivative values at 250 nm for

media AC and representing total thiosulfate H2S2O3HS2O3minusS2O3

2minus) are represented by squares and triangles respectively

Journal of Chemistry 7

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

thiosulfate signal in most cases In addition to those speciesdithionite (S4O6

2minus) thiocyanate (SCNminus) and thiols (iemercaptans) are frequently referred as being the majorsources of selectivity limitation for the quantification ofsulfides [19] Nevertheless mercaptans are generallydeprotonated for pH values higher than 10 In addition theirprotonated sulfhydryl function does not display any sig-nificant absorbance [51] Concerning thiocyanate anddithionite they are both deprotonated for pH equal to orhigher than 47 (ie media N and AC in which sulfides areestimated) Moreover thiocyanate does not show significantabsorbance above 240 nm [52 53] Consequently onlysulfite and thiosulfate measurements are possibly affected bydithionite It should also be added that the occurrence ofdithionite is not natural and that this molecule is unstable inaqueous media

If we exclude the previously mentioned reduced sulfurspecies the major soluble ions which can be found innatural waters (H+ OHminus Clminus Na+ Ca2+ Mg2+ K+ SO4

2minusHCO3

minus CO32minus BO3

3minus NH4+ Sr2+ and Fminus) are optically

inactive between 220 and 300 nm [54ndash57] Iminus and NO2minus are

absorbents but are generally found at very low concentra-tions Only Brminus NO3

minus and Fe2+ may cause interferenceswith sulfites and thiosulfates However as HBr and HNO3are strong acids and thus always deprotonated in aquaticsystems their contributions are cancelled by the second-derivative spectra subtraction inherent to the sulfidequantification step Fe2+ is present in numerous reduced andanoxic waters and it absorbs in the UV range [58] Nev-ertheless as it quickly reacts with sulfides to form ironsulfides its presence as dissolved cation is incompatible withthe presence of sulfides [59] Common dissolved inorganicgases (CO2 N2 CH4 NH3 H2 and noble gas) do not absorbsignificantly in the 200ndash330 nm range except for O2 that canbe detected below 250 nm at high levels [56] Its occurrenceis however not compatible with the presence of reducedsulfur species Similarly carbohydrates and polysaccharidesorthosilicic acid (H4SiO4) aliphatic esters ethers alcoholsand small common organic acids (eg acetic acid and lacticacid) and their corresponding bases do not absorb in thatdomain either [51 60 61] On the opposite humic and fulvicacids proteins and amino acids show generally a strongabsorbance in this area Free acids and hydrolysable com-bined amino acids do represent the major part of dissolvednitrogen-bearing organic matter in natural waters [62] emain optical contribution of these chemical species in theUV range arises from their content in aromatic amino acidsnamely the amount of tryptophan and tyrosine [63]Nevertheless the second derivatization of their absorbancespectra cancels the signal at 250 nm [64] Finally the twomost common natural chromophores (ie carotenoid andchlorophyll) have no specific absorbance below their Soretband which is generally found between 300 and 450 nm[65] To sum up major interferences to sulfide quantificationby this approach may solely arise from humic and fulvicacids

Table 2 shows as a function of the sample dilution LODobtained for total sulfide sulfite and thiosulfate togetherwith the effects of some of the most common interfering

species found in natural waters (including humic acids andorganic matter mimicked with yeast extracts as well as ni-trates and NaCl) It demonstrates that once corrected sa-linity is not interfering anymore for dilutions higher than 10folds Quantifications for weak dilutions (10 fold) also tol-erate relatively high amounts of organic matter (75mgL)and nitrate (200 microM) knowing that surface waters generallydo not exceed few tens of ppm of dissolved organic matterand less than 50 ppm (08mmolL) of nitrates [66] e bestsensitivity of this method is thus reached with weak dilutionNevertheless interferences may be stronger at those weakdilutions and care should be taken to avoid precipitates thatmay form in the pH-buffered measurement medium if thesample is still too concentrated (for instance carbonateprecipitates in the alkaline N medium) Samples that arethemselves strongly buffered or display extreme pHs (iegt10 or lt4) should also be diluted in order to avoid sig-nificant pH shifts in the measurement media Higher levelsof dilution are very interesting for samples for which highquantities of total dissolved sulfur species are expected (iegt1mmolL) In that case interferences are strongly reducedand measurements will have a better accuracy even forhighly polluted systems Finally test of interferences withsulfides sulfites and thiosulfates dissolved in syntheticseawater (TYA medium) or in natural subsurface water didnot show any significant variations compared to calibrationtests performed in UPW (Supplementary Material (availablehere)) is proves the total compatibility of our methodwith marine and terrestrial aqueous systems and moregenerally with a broad range of additional ions

34 Method Liability Comparison of the calculated LODand LOQ couples shows that the method is more sensitivefor sulfides compared to sulfites and thiosulfates (Tables 2and 3) is sensitivity is better than those encountered forion selective electrodes but still low in comparison withother techniques for instance the chromatographic andpotentiometric ones that usually reach sensitivities higherthan 1 micromolL [67] Linearity accuracy and precision of themethod respectively are represented by R2 accuracy meanand RSD display acceptable values (Table 3) and thusdemonstrate the validity of the present quantification pro-cedure Measurements that have been carried out during thecalibration tests with the aim to evaluate the temporalstability of the signal over 10min show no significant var-iations of the recorded spectra and extracted values (data notshown) Contrary to what could have been anticipatedespecially for sulfides which are very reactive to oxygenthere was no evidence of oxidation during the measurementperiod whatever the dilution was

Tables 2 and 3 show that the method is less sensitive withregard to sulfites Nevertheless Figure 3 also demonstratesthat the signal for sulfites is increasing for high salinities Asthe method is compatible with high salinities it is thereforepossible to improve its sensitivity for sulfites by increasingthe ionic strength of medium H In parallel a simplificationof the protocol can be advantageously executed if we con-sider samples with low sulfite amounts (lt1mmolL) andsalinities lower or equal to seawater which may be the

8 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

majority of surface waters Equations (6) and (8) would thusbe simplified by considering S and [sulfites] as null

Several advantages arise from this method compared toother techniques For instance Lawrence et al [19] observedthat traditional indirect spectrophotometric methods weresubjected to incertitude due to oxidative processes and thatmore globally the main hindrance of these approachestoward sulfide quantification was the lack of selectivityespecially with components such as thiocyanate dithionitethiols sulfites and thiosulfates Even if indirect methodsmay be sensitive and reach sub-micromolar detection levels[68 69] they strongly depend on the chemical background(namely salinity and reactive species) of the samples and aresubjected to strong interferences that require at least re-peated and adapted calibration procedures Comparativemeasurement between our method and conventionalmethods has been performed on one natural sample fromthe Dogger Aquifer (Supplementary Data Section 5) eresults showed comparable but not similar results (a dif-ference of 20 and 42 for sulfide and sulfite contents re-spectively) and the conventional method failed to assess thethiosulfate content However such results are difficult tointerpret not knowing if the observed differences are due tovariations in the sample preservation during the proceduresor due to chemical background effects

Comparison tests between the present method and thespectrophotometric methylene blue method have also beenperformed on natural fresh waters spiked with sulfides(Supplementary Figure S1) e results showed a bettercorrelation coefficient for the direct method (R2 0998)than for the indirect method (R2 0993) More recentlyToda et al [70] have developed an innovative method basedon H2S vapor generation after acidification coupled tomicrochannel collection that was tested and adapted toonsite measurements is technique can be considered as adirect method and appears to be very efficient(LOD 04 nM LOQ 10 nM) Nevertheless such methodworks only on trace-level sulfides (10 to 100 nM) and issubjected to strong interferences with Cu(II) Mn(II) andheavy metals For the method presented here we firstdemonstrate that direct measurements were suitable

regarding the risk of oxidation during handling is iscertainly due to the fact that there is neither need of reactiontime nor numerous handling steps Secondly we show thatthiocyanate and thiols could not theoretically induce in-terferences and that sulfites and thiosulfates were selectivelyand independently measured along with sulfides Finallyperforming a direct measurement provides in addition theopportunity to quantify polysulfides in medium N [47] iswould provide a very fast and detailed estimate of the samplereduction state regarding sulfur species

4 Conclusions

e goal of this study was to calibrate and provide a methodthat allows easy to handle fast and reliable quantification ofthe main reduced sulfur species namely sulfides sulfitesand thiosulfates Measurement of total sulfide (H2S andHSminus)for a large set of water qualities especially in pH-bufferedmedia that contain organic matter or have high salinitieswas accomplished successfully Subtraction of second-de-rivative spectra acquired at different pHs has demonstratedits ability to overcome classical interferences induced by thepresence of other reduced sulfur species together with or-ganic and inorganic compounds commonly found in surfaceand subsurface waters Together with the direct quantifi-cation of sulfites and thiosulfates this method gives a globalview of the sulfur speciation in reduced environmentsSensitivity is moderate but remains acceptable for mostquantitative approaches e development of such amethodology could lead to automatized systems and im-provements in sensitivity could be achieved by the use ofother buffered media Moreover it should be possible tosuccessfully develop similar approaches (ie subtraction ofsecond-derivative spectra obtained at selective pH) for thequantification of other pH-sensitive chemical species

Data Availability

e main data used to support the findings of this study areall included within the article and its Supplementary Ma-terial e spectrophotometric raw data are available fromthe corresponding author upon request

Table 3 Main qualitative parameters of the method obtained for the different chemical species limits of detection and quantification (LODand LOQ) linearity (R2) accuracy mean and relative standard deviation (RSD)

LOD (microM) LOQ (microM) R2 Accuracy mean () RSD ()Total sulfide 137 366 09997 023 345Total sulfite 732 2875 09998 minus115 068Total thiosulfate 192 416 09992 minus371 061

Table 2 Limit of detection (LOD) obtained for total sulfide sulfite and thiosulfate as a function of sample dilution salinity andconcentrations of humic acids yeast extract and nitrate for which first interferences with sulfur compounds occur

DilutionLOD (microM) Interferences

Total sulfide Total sulfite Total thiosulfate Humic acidsrange (mgL)

Yeast extractrange (mgL)

NaCl range (gL)withoutsalinity correction NO3 range (microM)

X 10 14 73 19 75 30 gt32012 200X 20 27 146 38 150 60 gt32024 400X 40 55 293 77 300 120 gt32048 800

Journal of Chemistry 9

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

Conflicts of Interest

e authors declare no conflicts of interest regarding thecontent and implications of this manuscript

Acknowledgments

is work was supported by the Centre de Recherches sur leStockage Geologique du CO2 (IPGPTOTALSCHLUM-BERGERADEME) and beneficiated from funds from theFrench Agence Nationale de la Recherche (project CO2FIX)is study contributes to the IdEx Universite de Paris ANR-18-IDEX-0001

Supplementary Materials

Supplementary Material includes in particular the generalprocedure and recommendation to implement the presentmethod as well as comparisons with standardized methodsand commercial kits (Supplementary Materials)

References

[1] D G Maynard Sulfur in the Environment Marcel DekkerNew York NY USA 1998

[2] D E Canfield E Kristensen and B amdrup AquaticGeomicrobiology Elsevier San Diego CA USA 1st edition2005

[3] G Leng Q HuW-F He et al ldquoA simple field method for thedetermination of sulfite in natural waters based on automateddispersive liquid-liquid microextraction coupled with ultra-violet-visible spectrophotometryrdquo Journal of ChromatographyA vol 1584 pp 72ndash79 2019

[4] K Wasmund M Muszligmann and A Loy ldquoe life sulfuricmicrobial ecology of sulfur cycling in marine sedimentsrdquoEnvironmental microbiology reports vol 9 no 4 pp 323ndash3442017

[5] R A Young and M Akhtar Environmentally FriendlyTechnologies for the Pulp and Paper Industry John Wiley ampSons New York NY USA 1997

[6] P Lens L Hulshoff-Pol P Wilderer and T Asano WaterRecycling and Resource Recovery in Industry Analysis Tech-nologies and Implementation IWA Publishing London UK2002

[7] A D Covington Tanning Chemistry Ie Science of LeatherRSC publishing Cambridge UK 2011

[8] P L Almeida T H Figueiredo do Bonfim F A S CunhaK M G Lima J S Aquino and L F Almeida ldquoA rapidsensitive and green analytical method for the determination ofsulfite in vinegars using pararosaniline reaction with imagedetectionrdquo Analytical Methods vol 10 no 4 pp 448ndash4582018

[9] P Lens and L Hulshoff-Pol Environmental Technologies toTreat Sulphur Pollution Principles and Engineering IWAPublishing London UK 2000

[10] J P Amend K J Edwards and T W Lyons Sulfur Bio-geochemistry Past and Present Geological Society of AmericaSpecial Paper Vol 379 Geological Society of AmericaBoulder CO USA 2004

[11] Y Wang F Zan G Guo T Hao J Wang and G Chen ldquoAnovel approach to quantifying elemental sulfur (S0) in en-vironmental samplesrdquo Water Science and Technologyvol 2017 no 2 pp 467ndash472 2018

[12] G K Druschel M A Schoonen D K Nordstrom J W BallY Xu and C A Cohn ldquoSulfur geochemistry of hydrothermalwaters in Yellowstone National Park Wyoming USA III Ananion-exchange resin technique for sampling and preserva-tion of sulfoxyanions in natural watersrdquo GeochemicalTransactions vol 4 no 1 pp 12ndash19 2003

[13] R Gurkan N Altunay and N Gurkan ldquoExtraction pre-concentration and spectrophotometric determination of tracelevels of thiosulfate in environmental watersrdquo Journal of theIranian Chemical Society vol 14 no 5 pp 1033ndash1049 2017

[14] R Hell C Dahl D B Knaff and T Leustek Sulfur Meta-bolism in Phototrophic Organisms Springer DordrechtNetherlands 2008

[15] A Isaac J Davis C Livingstone A J Wain andR G Compton ldquoElectroanalytical methods for the de-termination of sulfite in food and beveragesrdquo TrAC Trends inAnalytical Chemistry vol 25 no 6 pp 589ndash598 2006

[16] R C Wiley ldquoPreservation methods for minimally processedrefrigerated fruits and vegetablesrdquo in Minimally ProcessedRefrigerated Fruits and Vegetables Springer Boston MAUSA 2017

[17] Z Shareefdeen and A Singh Biotechnology for Odor and AirPollution Control Springer Heidelberg Germany 2005

[18] P C Ng T B Hendry-Hofer A E Witeof et al ldquoEfficacy oforal administration of sodium thiosulfate and glycine in alarge swine model of oral cyanide toxicityrdquo Annals ofEmergency Medicine 2019 In press

[19] N Lawrence J Davis and R G Compton ldquoAnalyticalstrategies for the detection of sulfide a reviewrdquo Talantavol 52 no 5 pp 771ndash784 2000

[20] Y Miura M Fujisaki and P R Haddad ldquoSpectrophotometricdetermination of sulfide in the presence of sulfite and thio-sulfate via the precipitation of bismuth(III) sulfiderdquo Ana-lytical Sciences vol 20 no 2 pp 363ndash367 2004

[21] A Perry and D Miles ldquoAn off-the-shelf sensor for colouri-metric detection of sulfiderdquo Tetrahedron Letters vol 57no 51 pp 5788ndash5793 2016

[22] S Motellier K Gurdale and H Pitsch ldquoSulfur speciation bycapillary electrophoresis with indirect spectrophotometricdetection in search of a suitable carrier electrolyte to maxi-mize sensitivityrdquo Journal of Chromatography A vol 770no 1-2 pp 311ndash319 1997

[23] Y Miura and A Kawaoi ldquoDetermination of thiosulfatethiocyanate and polythionates in a mixture by ion-pairchromatography with ultraviolet absorbance detectionrdquoJournal of Chromatography A vol 884 no 1-2 pp 81ndash872000

[24] S-I Ohira and K Toda ldquoIon chromatographic measurementof sulfide methanethiolate sulfite and sulfate in aqueous andair samplesrdquo Journal of Chromatography A vol 1121 no 2pp 280ndash284 2006

[25] L Ferrer M Miro J M Estela and V Cerda ldquoAnalyticalmethodologies for reliable sulfide determinations in aqueousmatrices exploiting flow-based approachesrdquo TrAC Trends inAnalytical Chemistry vol 26 no 5 pp 413ndash422 2007

[26] T R Crompton Determination of Metals in Natural andTreated Water Taylor amp Francis London UK 2001

[27] E A Guenther K S Johnson and K H Coale ldquoDirect ul-traviolet spectrophotometric determination of total sulfideand iodide in natural watersrdquo Analytical Chemistry vol 73no 14 pp 3481ndash3487 2001

[28] R E Eckford and P M Fedorak ldquoSecond derivative UVabsorbance analysis to monitor nitrate-reduction by bacteria

10 Journal of Chemistry

in most probable number determinationsrdquo Journal of Mi-crobiological Methods vol 50 no 2 pp 141ndash153 2002

[29] T Heitmann and C Blodau ldquoOxidation and incorporation ofhydrogen sulfide by dissolved organic matterrdquo ChemicalGeology vol 235 no 1-2 pp 12ndash20 2006

[30] A Postec F Lesongeur P Pignet B Ollivier J Querellouand A Godfroy ldquoContinuous enrichment cultures insightsinto prokaryotic diversity and metabolic interactions in deep-sea vent chimneysrdquo Extremophiles vol 11 no 6 pp 747ndash7572007

[31] J E Cahill ldquoDerivative spectroscopy understanding its ap-plicationrdquo American Laboratory vol 11 pp 79ndash85 1979

[32] A M C Davies ldquoBack to basics spectral pre-treaments-derivativesrdquo European Journal of Mass Spectrometry vol 19pp 32-33 2007

[33] S Motellier and M Descostes ldquoSulfur speciation and tetra-thionate sulfitolysis monitoring by capillary electrophoresisrdquoJournal of Chromatography A vol 907 no 1-2 pp 329ndash3352001

[34] R Steudel ldquoMechanism for the formation of elemental sulfurfrom aqueous sulfide in chemical and microbiological de-sulfurization processesrdquo Industrial amp Engineering ChemistryResearch vol 35 no 4 pp 1417ndash1423 1996

[35] E Wiberg N Wiberg and A E Holleman Holleman-Wibergrsquos Inorganic Chemistry Academic Press San DiegoCA USA 2001

[36] V K Zinchuk and E N Stadnichuk ldquoChemiluminescencedetermination of monopersulfuric acidrdquo Journal of AnalyticalChemistry vol 56 no 10 pp 918-919 2001

[37] G K Druschel R J Hamers and J F Banfield ldquoKinetics andmechanism of polythionate oxidation to sulfate at low pH byO2 and Fe3+rdquo Geochimica et Cosmochimica Acta vol 67no 23 pp 4457ndash4469 2003

[38] Y Miura T Maruyama and T Koh ldquoIon chromatographicdetermination of L-ascorbic acid sulfite sulfide and thio-sulfate using a cation-exchanger of low crosslinkingrdquo Ana-lytical Sciences vol 11 no 4 pp 617ndash621 1995

[39] K Ranguelova and R P Mason ldquoNew insights into the de-tection of sulfur trioxide anion radical by spin trappingradical trapping versus nucleophilic additionrdquo Free RadicalBiology and Medicine vol 47 no 2 pp 128ndash134 2009

[40] I M Kolthoff and P J Elving Treatise on AnalyticalChemistry Interscience Encyclopedia New York NY USA1959

[41] K Gupta D R Roy V Subramanian and P K ChattarajldquoAre strong Broslashnsted acids necessarily strong Lewis acidsrdquoJournal of Molecular Structure THEOCHEM vol 812no 1ndash3 pp 13ndash24 2007

[42] Y Chiang and A J Kresge ldquoDetermination of the acidityconstant of isothiocyanic acid in aqueous solutionrdquo CanadianJournal of Chemistry vol 78 no 12 pp 1627-1628 2000

[43] M Kuhn C Niewohner M Isenbeck-Schroter andH D Schulz ldquoDetermination of major and minor constit-uents in anoxic thermal brines of deep sandstone aquifers inNorthern Germanyrdquo Water Research vol 32 no 2pp 265ndash274 1998

[44] T T Nham and A G Wiseman ldquoA new torch for analysis ofsamples having exceptionally high total dissolved solids byaxially-viewed inductively coupled plasma optical emissionspectrometryrdquo Journal of Analytical Atomic Spectrometryvol 18 no 7 pp 790ndash794 2003

[45] I Ali and H Y Aboul-Enein Instrumental Methods in MetalIon Speciation CRCTaylor amp Francis Boca Raton FL USA2006

[46] Y T Puyate and C J Lawrence ldquoSteady state solutions forchloride distribution due to wick action in concreterdquoChemical Engineering Science vol 55 no 16 pp 3329ndash33342000

[47] W E Kleinjan A de Keizer and A J H Janssen ldquoEqui-librium of the reaction between dissolved sodium sulfide andbiologically produced sulfurrdquo Colloids and Surfaces B Bio-interfaces vol 43 no 3-4 pp 228ndash237 2005

[48] D Zheng D Qu X-Q Yang X Yu H-S Lee and D QuldquoQuantitative and qualitative determination of polysulfidespecies in the electrolyte of a lithium-sulfur battery usingHPLC ESIMS with one-step derivatizationrdquo Advanced En-ergy Materials vol 5 no 16 article 1401888 2015

[49] W E Kleinjan A de Keizer and A J H Janssen ldquoKinetics ofthe reaction between dissolved sodium sulfide and biologicallyproduced sulfurrdquo Industrial amp Engineering Chemistry Re-search vol 44 no 2 pp 309ndash317 2005

[50] J M Shively Inclusions in Prokaryotes Springer HeidelbergGermany 2006

[51] O omas and C Burgess UV-Visible Spectrophotometry ofWater and Wastewate Elsevier Amsterdam Netherlands2nd edition 2017

[52] T R Crompton Determination of Anions a Guide for theAnalytical Chemist Springer Berlin Germany 1996

[53] R M Onorato D E Otten and R J Saykally ldquoAdsorption ofthiocyanate ions to the dodecanolwater interface charac-terized by UV second harmonic generationrdquo Proceedings ofthe National Academy of Sciences vol 106 no 36pp 15176ndash15180 2009

[54] T F Rozan and G W Luther ldquoAn anion chromatographyultraviolet detection method to determine nitrite nitrate andsulfide concentrations in saline (pore) watersrdquo MarineChemistry vol 77 no 1 pp 1ndash6 2002

[55] Malcolm Pirnie Inc Carollo Engineers PC e CadmusGroup Inc K G Linden and P Malley Jr Ultraviolet Dis-infection Guidance Manual Diane Publishing Co Darby PAUSA 2003

[56] B Wozniak and J Dera Light Absorption in Sea WaterSpringer New York NY USA 2007

[57] O omas and C Burgess UV-Visible Spectrophotometry ofWater and Wastewater Elsevier Amsterdam Netherlands2017

[58] S Rajendran B V Apparao and N Palaniswamy ldquoHEDP-Zn2+ a potential inhibitor system for mild steel in lowchloride mediardquo Anti-Corrosion Methods and Materialsvol 47 no 2 pp 83ndash87 2000

[59] A A Abramov and V M Avdokhin Oxidation of SulfideMinerals in Benefication Processes Overseas Publishers As-sociation Amsterdam Netherlands 1997

[60] J C Dias E Suzuki C L de Albuquerque A L FerreiraA R M S Brito and L T Kubota ldquoDetermination of short-chain fatty acids in dietary fiber extracts using ion-exclusionchromatography with suppressed conductivity detectionrdquoJournal of Pharmaceutical and Biomedical Analysis vol 49no 4 pp 1128ndash1132 2009

[61] S R Boethling and D Mackay Handbook of Property Esti-mation Methods for Chemicals Environmental and HealthSciences CRC PressTaylor amp Francis Boca Raton FL USA2000

[62] A Gianguzza E Pelizzetti and S Sammartano Chemistry ofMarineWater and Sediments Springer Heidelberg Germany2002

Journal of Chemistry 11

[63] M G Gore Spectrophotometry amp Spectrofluorimetry Prac-tical Approach Series Oxford University Press Oxford UK2000

[64] H Mach C R Middaugh and N Denslow ldquoDetermining theidentity and purity of recombinant proteins by UV absorptionspectroscopyrdquo in Current Protocols in Protein Science JohnWiley amp Sons London UK 2001

[65] W J Hurst Methods of Analysis for Functional Foods andNutraceuticals Functional Foods ampNutraceuticals Series CRCPressTaylor amp Francis Boca Raton FL USA 2nd edition2008

[66] M van der Perk ldquoSoil and water contamination from mo-lecular to catchment scalerdquo in Balkema Proceedings andMonographs in Engineering Water and Earth Sciences Tayloramp Francis Leiden Netherlands 2006

[67] T R Crompton Determination of Anions in Natural andTreated Waters Spon Press London UK 2002

[68] S E Mylon and G Benoit ldquoSubnanomolar detection of acid-labile sulfides by the classical methylene blue method coupledto HPLCrdquo Environmental Science amp Technology vol 35no 22 pp 4544ndash4548 2001

[69] T V Titova N S Borisova and N F Zakharchuk ldquoDe-termination of sub-micromolar amounts of sulfide by stan-dard free anodic stripping voltammetry and anodic strippingvoltammetric titrationrdquo Analytica Chimica Acta vol 653no 2 pp 154ndash160 2009

[70] K Toda H Kuwahara and S-I Ohira ldquoOn-site measurementof trace-level sulfide in natural waters by vapor generation andmicrochannel collectionrdquo Environmental Science amp Tech-nology vol 45 no 13 pp 5622ndash5628 2011

12 Journal of Chemistry

TribologyAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

International Journal ofInternational Journal ofPhotoenergy

Hindawiwwwhindawicom Volume 2018

Journal of

Chemistry

Hindawiwwwhindawicom Volume 2018

Advances inPhysical Chemistry

Hindawiwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2018

Bioinorganic Chemistry and ApplicationsHindawiwwwhindawicom Volume 2018

SpectroscopyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Medicinal ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

NanotechnologyHindawiwwwhindawicom Volume 2018

Journal of

Applied ChemistryJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Biochemistry Research International

Hindawiwwwhindawicom Volume 2018

Enzyme Research

Hindawiwwwhindawicom Volume 2018

Journal of

SpectroscopyAnalytical ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

MaterialsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

BioMed Research International Electrochemistry

International Journal of

Hindawiwwwhindawicom Volume 2018

Na

nom

ate

ria

ls

Hindawiwwwhindawicom Volume 2018

Journal ofNanomaterials

Submit your manuscripts atwwwhindawicom

[63] M G Gore Spectrophotometry amp Spectrofluorimetry Prac-tical Approach Series Oxford University Press Oxford UK2000

[64] H Mach C R Middaugh and N Denslow ldquoDetermining theidentity and purity of recombinant proteins by UV absorptionspectroscopyrdquo in Current Protocols in Protein Science JohnWiley amp Sons London UK 2001

[65] W J Hurst Methods of Analysis for Functional Foods andNutraceuticals Functional Foods ampNutraceuticals Series CRCPressTaylor amp Francis Boca Raton FL USA 2nd edition2008

[66] M van der Perk ldquoSoil and water contamination from mo-lecular to catchment scalerdquo in Balkema Proceedings andMonographs in Engineering Water and Earth Sciences Tayloramp Francis Leiden Netherlands 2006

[67] T R Crompton Determination of Anions in Natural andTreated Waters Spon Press London UK 2002

[68] S E Mylon and G Benoit ldquoSubnanomolar detection of acid-labile sulfides by the classical methylene blue method coupledto HPLCrdquo Environmental Science amp Technology vol 35no 22 pp 4544ndash4548 2001

[69] T V Titova N S Borisova and N F Zakharchuk ldquoDe-termination of sub-micromolar amounts of sulfide by stan-dard free anodic stripping voltammetry and anodic strippingvoltammetric titrationrdquo Analytica Chimica Acta vol 653no 2 pp 154ndash160 2009

[70] K Toda H Kuwahara and S-I Ohira ldquoOn-site measurementof trace-level sulfide in natural waters by vapor generation andmicrochannel collectionrdquo Environmental Science amp Tech-nology vol 45 no 13 pp 5622ndash5628 2011

12 Journal of Chemistry

TribologyAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

International Journal ofInternational Journal ofPhotoenergy

Hindawiwwwhindawicom Volume 2018

Journal of

Chemistry

Hindawiwwwhindawicom Volume 2018

Advances inPhysical Chemistry

Hindawiwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2018

Bioinorganic Chemistry and ApplicationsHindawiwwwhindawicom Volume 2018

SpectroscopyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Medicinal ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

NanotechnologyHindawiwwwhindawicom Volume 2018

Journal of

Applied ChemistryJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Biochemistry Research International

Hindawiwwwhindawicom Volume 2018

Enzyme Research

Hindawiwwwhindawicom Volume 2018

Journal of

SpectroscopyAnalytical ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

MaterialsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

BioMed Research International Electrochemistry

International Journal of

Hindawiwwwhindawicom Volume 2018

Na

nom

ate

ria

ls

Hindawiwwwhindawicom Volume 2018

Journal ofNanomaterials

Submit your manuscripts atwwwhindawicom

TribologyAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

International Journal ofInternational Journal ofPhotoenergy

Hindawiwwwhindawicom Volume 2018

Journal of

Chemistry

Hindawiwwwhindawicom Volume 2018

Advances inPhysical Chemistry

Hindawiwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2018

Bioinorganic Chemistry and ApplicationsHindawiwwwhindawicom Volume 2018

SpectroscopyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Medicinal ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

NanotechnologyHindawiwwwhindawicom Volume 2018

Journal of

Applied ChemistryJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Biochemistry Research International

Hindawiwwwhindawicom Volume 2018

Enzyme Research

Hindawiwwwhindawicom Volume 2018

Journal of

SpectroscopyAnalytical ChemistryInternational Journal of

Hindawiwwwhindawicom Volume 2018

MaterialsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

BioMed Research International Electrochemistry

International Journal of

Hindawiwwwhindawicom Volume 2018

Na

nom

ate

ria

ls

Hindawiwwwhindawicom Volume 2018

Journal ofNanomaterials

Submit your manuscripts atwwwhindawicom