RESEARCH ARTICLE

Extractability of water-soluble soil organic matter as monitoredby spectroscopic and chromatographic analyses

Ezzhora Nkhili & Ghislain Guyot & Nathalie Vassal &Claire Richard

Received: 22 November 2011 /Accepted: 6 January 2012 /Published online: 17 January 2012# Springer-Verlag 2012

AbstractPurpose Cold and hot water processes have been intensive-ly used to recover soil organic matter, but the effect ofextraction conditions on the composition of the extractswere not well investigated. Our objective was to optimizethe extraction conditions (time and temperature) to increasethe extracted carbon efficiency while minimizing the possi-ble alteration of water extractable organic matter of soil(WEOM).Method WEOM were extracted at 20°C, 60°C, or 80°C for24 h, 10–60 min, and 20 min, respectively. The differentprocesses were compared in terms of pH of suspensions,yield of organic carbon, spectroscopic properties (ultraviolet–visible absorption and fluorescence), and by chromatographicanalyses.Results For extraction at 60°C, the time 30 min was optimalin terms of yield of organic carbon extracted and concentra-tion of absorbing and fluorescent species. The comparisonof WEOM 20°C, 24 h; 60°C, 30 min; and 80°C, 20 minhighlighted significant differences. The content of total or-ganic carbon, the value of specific ultraviolet absorbance

(SUVA254), the absorbance ratio at 254 and 365 nm (E2/E3),and the humification index varied in the order: WEOM (20°C,24 h)<WEOM (80°C, 20 min)<WEOM (60°C, 30 min). Thethree WEOM contained common fluorophores associatedwith simple aromatic structures and/or fulvic-like and com-mon peaks of distinct polarity as detected by ultra perfor-mance liquid chromatography.Conclusions For the soil chosen, extraction at 60°C for30 min is the best procedure for enrichment in organicchemicals and minimal alteration of the organic matter.

Keywords Soil WEOM . Extraction method .

Characterization spectroscopy . Chromatography analyses

1 Introduction

Soil organic matter (SOM) plays a key role in soil majorbiogeochemical. Conventionally, SOM is a complex mix-ture of aromatic and aliphatic hydrocarbon structures thathave attached amid, carboxyl, hydroxyl, phenoxyl ketone,and various minor functional groups. Therefore, the hetero-geneous molecular aggregate in natural soils increase SOMcomplexity.

SOM in natural and agro-ecosystems has chemical prop-erties that enable it to interact with other organic compoundsand metal ions through π bonding, hydrogen bonding, li-gand exchange reactions, dipole–dipole interactions, andcovalent bond (Northcott and Jones 2000). SOM can thereforebe involved in biotic and abiotic reactions.

The fraction of SOM extractable by water [water extract-able organic matter (WEOM)] represents the most activeand mobile form of SOM. The importance of WEOM prop-erties as indicators of change soil quality has received greatattention (Chantigny 2003; Corvasce et al. 2006; Zsolnay

Responsible editor: Philippe Garrigues

E. Nkhili :G. Guyot :C. RichardLaboratoire de Photochimie Moléculaire et Macromoléculaire(LPMM), Clermont Université, Université Blaise Pascal,BP 10448, 63000 Clermont-Ferrand, France

N. VassalVetAgro’Sup, Campus Agronomique de Clermont,Clermont-Université, UR AFOS,63370 Lempdes, France

C. Richard (*)CNRS, UMR 6505, LPMM,BP 80026, 63171 Aubière, Francee-mail: Claire.RICHARD@univ-bpclermont.fr

Environ Sci Pollut Res (2012) 19:2400–2407DOI 10.1007/s11356-012-0752-0

2003). In particular, several recent studies have suggested avital role for WEOM in the numerous biogeochemical pro-cesses that mediate our environment because of its solubilityin water. It is able to bind and transport anthropogenicmaterials (Kalbitz et al. 2000). In addition, it is the mainenergy source for soil microorganisms, and it influences theavailability of metal ions in soil by forming soluble complexes(Zhang et al. 2006).

WEOM is known to provide an analytical challenge toresearchers seeking molecular-level information. This ismainly due to extreme complexity, low concentration, andhigh polarity. The current interest in positive effect of SOMstimulated the development of simple processes for extrac-tion of WEOM fraction. Indeed, the quality of WEOM notonly depends on starting materials (geography origin, andclimatic and storage conditions) but also on the processesused for their extraction. In particular, this latter must bedone under mild and efficient condition that minimizes thepossible alteration of SOM.

Various extraction protocols have been used to extractorganic matter from soil. In general, alkaline extraction ofhumics substances and their fractionation, in humic acid(HA) and fulvic acid (FA) by acid soluble/insoluble propri-eties (Stevenson 1994), have been studied by specialists insoil geochemistry with reasonable success. SOM is betterextracted by alkaline system than by water at room temper-ature (Ussiri and Johnson 2002). However, the use of NaOHcan be risky and alter the chemical structure of SOM. Thesoxhlet extraction with different solvents (Corradini et al.2006; Marinari et al. 2010) and cold extraction (Kalbitz etal. 2003; Kaiser and Ellerbrock 2005) are widely used in soilchemistry, but they are limited by long extraction period andlow extraction efficiency. Extraction with water at 80°C hasalso been investigated to measure labile C (Ghani et al.2006) and plant available N and S (Curtin et al. 2006).Previous works have shown that amount of organic matterrecovered using water at 80°C exceeds the one obtainedusing cold water by a factor of two (Landgraf et al. 2006;Gregorich et al. 2003). However, the influence of the heat-ing mode on the composition of WEOM may be stronglymodified by an increase of the temperature because WEOMis thermally unstable.

The time and temperature of the extraction procedure canprobably potentially affect the efficiency extraction as wellas the quality of the extracted organic matter. To our knowl-edge, this aspect has been poorly explored. In this work, weused different extraction procedures by varying the watertemperature and the time of extraction. The efficiency andselectivity of these processes were compared in terms oftotal organic carbon (TOC) recovered, spectroscopy proper-ties [ultraviolet–visible (UV–Vis) and fluorescence] of theextracts, and chemical composition [ultra performance liquidchromatography (UPLC) analysis].

2 Experimental

2.1 Soil sampling

Experiments were conducted on the black soil of Limagne(Auvergne, France) which shows similarities with cherno-zem soil, according to French soil taxonomy; these soilscontain a large amount of organic matter and are fertile. Soilwas taken in May 2010 in the agricultural parcel of theLycée Agricole of Marmilhat, Auvergne, France at 45°46′42″N and 3°10′45″E. Soil was collected between 5 and20 cm in four different places (1 kg for each). The soilsample was air-dried at 25°C for 15 days. After this time,the loss of water was of 20%.

2.2 SOM extraction

WEOM was extracted using the following procedure. Sus-pensions of soil were prepared in Milli-Q water in a ratio(soil/water01/10). Each sample was constituted by 12 g ofsoil in 120 ml of water. Extractions were conducted at roomtemperature, 60°C, or 80°C. For cold extraction, the sampleswere mechanically shaken for 24 h. For extractions at 60°Cand 80°C, samples were put in a flask equipped with acooled column and were mechanically shaken. At 60°C,samples were heated for 10, 20, 30, 40, 50, or 60 min. At80°C, they were heated for 20 min. After extraction, thesamples were centrifuged at 5,600 rpm for 60 min. Finally,the supernatant was lyophilized and stored in amber bottles.Each extraction was repeated three-folds. The yields ofextraction were calculated as the ratio of themass of recoveredWEOM to the mass of soil used for the extraction.

2.3 TOC analysis

A Shimadzu 5050 TOC analyzer was used for organiccarbon analysis. The instrument was calibrated with potas-sium hydrogen phthalate standards at concentration thatranged from 0 to 100 mg L−1. The TOC content was calcu-lated as the difference between the total carbon and theinorganic carbon. The measurements were conducted onsolutions containing 100 mg L−1 of lyophilized WEOM,prepared in buffer phosphate 10−3 M.

2.4 Spectroscopy analyses

UV/Vis absorbance and fluorescence spectra were recordedusing diluted samples (TOC≈10 mg L−1) buffered at pH 6.5using phosphate buffers (10−3 M). Spectrophotometric anal-yses were conducted on a Varian Cary 3 UV–Vis spectro-photometer. The kinetic study at 60°C was conducted bymonitoring the absorbance at 254 and 365 nm. Specificultraviolet absorbance at 254 nm (SUVA254 in liter per

Environ Sci Pollut Res (2012) 19:2400–2407 2401

milligram per meter) was determined as the absorbancedivided by TOC concentration (milligram per liter). Theratio E2/E3 (ratio of the absorbances at 254 and 365 nm)was also calculated. It is often used as an indicator forhumification (Matilainen et al. 2011).

Fluorescence emission spectra of WEOM extracts wererecorded in a 1-cm path length quartz cell using a Perkin-Elmer LS-55 Luminescence spectrometer (λex, 254 nm, slit5 nm; λem 300–600 nm, slit 10 nm; and scab speed1,200 nm min−1). To limit second-order Raleigh scattering,a 290-nm cut-off filter was applied for all samples. Humifi-cation index (HIX) was defined as the ratio of the peak areain the upper quarter (Σ435–480 nm) of the usable fluores-cence emission spectra to the one in the lower usable quarter(Σ300–345 nm; Zsolnay et al. 1999); it represents the degreeof condensation of WEOM (Zsolnay 2003).

2.5 UPLC analyses

UPLC analyses were carried out at 25°C on an Alliance(Waters) apparatus equipped with UV–Vis dual absorbancedetector (Waters 2437), two pumps (Waters 2695), and amodel gradient controller. Separation was achieved on a C18

reversed phase column (Phenomenex, Synergi 4 μ Hydro-RP 80 Å, 250×2 mm2 with security cartridge C18 4×2 mm).The binary solvent system used was composed of solvent Awith 100% water Milli-Q and solvent B (MeOH). The bestseparation was obtained with the following gradient: at 0 min,100% A/0% B; at 20 min, 0% A/100% B; and at 50 min, 0%A/100% B. The solvent flow rat was 0.4 mL min−1. Thevolume injection was 10 μl. The spectroscopy detector wasset at 254 nm. All runs were acquired and processed using theEmpower software.

3 Results and discussion

The present study aimed at developing a simple method forthe extraction of soil WEOM. For this purpose, heatingwater extractions at 60°C were optimized in terms of extrac-tions kinetic, of amount of WEOM recovery, and of absorp-tion and fluorescence characteristics. The efficiency of

extraction at 60°C was compared to those of cold extractionat room temperature during 24 h (Kalbitz et al. 2003; Kaiserand Ellerbrock 2005) and of heating extraction at 80°C for20 min (Curtin et al. 2006; Gregorich et al. 2003; Xiaoli etal. 2010).

Some physical and chemical properties of the soil arereported in Table 1. The black soil of Limagne has thefollowing characteristics: (1) variability in the particle sizedistribution with predominance of clay fractions (63.5%);(2) alkaline pH (pH 8.1); (3) a relatively important TOC(34.5 gkg−1), and (4) a high exchange capacity.

3.1 Yield of extraction and TOC content

The yield of extraction was function of extraction time andtemperature. Highest values were obtained at 80°C (0.41±0.08%) and lowest at 20°C (0.23±0.05%; Table 2). At 60°C,the yield increased with time, reaching a maximum value of(0.36±0.11%) at 40 min and then slightly decreased (Fig. 1).For determining the organic compounds content, we mea-sured the TOC. For WEOM extracted at 60°C, TOC in-creased with extraction time, varying from 25.7±2.0 mg L−1

at 10 min to 41.4±2.3 mg L−1 at 40 min and then slightlydecreased (Fig. 1). The TOC values obtained at 80°C and20°C were lower (Table 2). These results are in line with theliterature which shows that the organic matter extraction isbetter in hot than in cold water (Landgraf et al. 2006; Curtinet al. 2011). However, too high temperature may inducesome decomposition and the fact that TOC is smaller at80°C than at 60°C is likely due to the thermal decomposi-tion of organic matter into low molecular weight with CO2

formation.

3.2 pH of extracts

After each extraction of WEOM, the pH of the suspensionwas measured using a glass electrode. In 60°C conditions,Fig. 2 shows that the pH suspension tended to decrease withincreasing heating time (from 8.2 to 7.3). At 80°C, the pH ofthe suspension was significantly more acid (6.3).

To better understand the relationship between pH suspen-sions and organic carbon content, pH was plotted as a

Table 1 Main chemical and physical properties of the Limagne black soil

pH ECa (meq kg−1) TOCb (g kg−1) Total N (g kg−1) C/N atomic ratio Soil texture (%)c

Clay Slit Sand

8.1 443.0 34.5 3.2 12.6 63.5 21.6 14.9

a Exchange capacity expressed in milliequivalent per kilogramb Total organic carbon in the soil (grams per kilogram)c On decarbonated soil

2402 Environ Sci Pollut Res (2012) 19:2400–2407

function of the yield of C (inset of Fig. 2), independently ofthe extraction time and extraction temperature. Except forthe data related to the sample extracted at 80°C, the pH ofsuspension is negatively correlated to the C content(R200.6325). This result is connected to hydrolysis ofmacromolecules into lower molecular weight moleculeswith formation of acidic compounds such as benzoic acid,amino acid, and sugars in the hot water extraction or to thereplacement of cations by protons (Curtin et al. 2011).

3.3 UV–visible absorbance

Absorbances measured at several wavelengths in the UV–Vis range have been proposed to characterize the spectro-photometric profile of soil WEOM. In this study, we col-lected the data of absorbance at 254 and 365 nm, whichserve as a rough indicator of overall WEOM concentration.In particular, the absorbance at 254 nm reflects the contentof aromatic compounds and is related to the degree ofpolycondensation in WEOM samples (Zsolnay 2003). The

365-nm absorbance could principally be related to extendedconjugation in aliphatic or polyaromatic structures as well asto the presence of metallic complexes.

Figure 3 shows evolution of absorbances at 254 and365 nm with heating time. For both curves, a maximumwas observed after 30 min. The decrease between 30 and60 min suggested that WEOM were prone to degradation.The ratio E2/E3 was chosen to monitor the change ofWEOM composition during extraction time. According toPeuravuori et al. (Peuravori and Pihlaja 1997), there is arelationship between the ratio E2/E3 and both the degree ofaromaticity and the molecular size. Higher E2/E3 ratios areusually associated with lower molecular weight and lowerdegree of aromaticity. For WEOM obtained at 60°C, thelowest E2/E3 value is found at 30–40 min (see Fig. 3). Thismeans that the content of aromatic structures substitutedwith, probably, polar groups such as hydroxyl and carbox-ylic groups is maximum at this extraction time. The increaseof the ratio at higher extraction time would confirm thatsome thermal degradation of WEOM takes place, leading tosmaller molecular size molecules. The WEOM extracted at

Table 2 Yield of extraction and physicochemical parameters for different extraction methods

Extraction method Yielda in % TOCb in mg L−1 pH E2/E3c SUVA254

d HIXe

60°C, 30 min 0.34±0.01 40.07±2.04 7.62±0.03 7.23±1.12 7.18±0.36 5.93±0.29

80°C, 20 min 0.41±0.01 36.91±1.16 6.38±0.01 7.84±1.06 6.42±0.19 6.85±0.34

20°C, 24 h 0.23±0.03 31.03±1.96 7.71±0.06 9.50±0.08 6.55±1.03 5.28±0.18

Results are presented as the mean (n03)±SDa The yield of extraction is obtained by dividing the mass of recovered WEOM by the mass of soil used for the extractionb Total organic carbon in WEOM (100 mg L−1 )c Ratio of the absorbances at 254 and 365 nmd Specific ultraviolet absorbance at 254 nm (liter per milligram per meter)e Ratio of the peak area in the upper quarter (Σ435–480 nm) of the usable fluorescence emission spectra to the one in the lower usable quarter(Σ300–345 nm). Excitation at 254 nm

10 20 30 40 50 60

0.15

0.20

0.25

0.30

0.35

0.40

Heating time/min

Yie

ld o

f ext

ract

ion/

%

25

30

35

40

45

TO

C/m

g.L-1

Fig. 1 TOC content of WEOM extracted at 60°C (closed circles) andyield of extraction (open circles) as a function of the heating time. TOCwas measured in 100 mg L−1 solutions

10 20 30 40 50 60

7.4

7.6

7.8

8.0

8.2

o

2 4 6 8 10 12 14 16 186.0

6.5

7.0

7.5

8.0

8.5

y=-0.0468x+8.194R 2=0.633

pH

% of extracted C

pH

Heating time/min

Fig. 2 Variation of pH suspensions as a function of heating time. Inset:variation of pH as function of percentage of C extracted; the label (“o”)corresponds to the sample 80°C, 20 min

Environ Sci Pollut Res (2012) 19:2400–2407 2403

80°C and at 20°C show also a E2/E3 ratio lower than that at60°C (Table 2).

To better understand the temperature effect on composi-tion WEOM, we compared UV–Vis spectra of the WEOMextracted at 60°C, 30 min; 80°C, 20 min; and 20°C, 24 h.Absorption spectra are quite similar (Fig. 4). First, we notethat the spectra associated with WEOM from fresh soil aredifferent from those of other reported humic substances. Thespectra present two main absorption bands: the first onebeing located between 210 and 240 nm and the secondone between 250 and 350 nm. The shoulder at 280 nm,which is generally small in humic substances, is quite im-portant here, and the weak absorbance increase between 400and 240 nm contrasts with the sharp increase below 240 nm.These results suggest that the aromatic structures in thisWEOM of black soil present a high degree of substitutionwith oxygen containing functional groups. The absorptionband below 240 nm is less intense for WEOM extracted at80°C. This means that the compounds in this UV regionwere impacted by temperature extraction. The absorbanceband between 200 and 240 nm is associated with benzenoidband (Xiaosong et al. 2011; Fuentes et al. 2006). This lower

absorbance might be connected to the lower TOC contentand slightly higher E2/E3 value and may be due to thedegradation of carboxylic compounds linked to aromaticrings with formation of CO2. Recently, there has been agreat attention to use hot water (≥80°C) for extracting labilefractions of SOM. Thermal decomposition studies con-firmed that hot water-extractable organic matter is easilydegraded into low molecular weight (Gregorich et al.2003). Concerning the second absorption band of the spec-tra, some differences among samples can be observed too.The absorbance varied in the order: WEOM (20°C, 24 min)<WEOM (80°C, 20 min)<WEOM (60°C, 30 min). It isreported that absorption at wavelengths between 240 and400 nm refer to as electron-transfer (ET) bands and that thepresence of aromatic rings substituted by polar functionalgroups, such as hydroxyl, carbonyl, carboxyl, and estergroup, increase the intensity of this ET band (Xiaosong etal. 2011; Fuentes et al. 2006). The absorbance intensity isthe highest in WEOM (60°C, 30 min). This shows that thissample is enriched in aromatic structures with a high degreesubstitution with oxygen containing functional group com-pared to the other extracted samples.

The SUVA254 is another parameter that has been widelyused as index for the abundance of aromatic carbon inWEOM (Fuentes et al. 2006). Values are higher for WEOM(60°C, 30 min) than for WEOM (20°C, 24 h) and WEOM(80°C, 20 min) (Table 2). This confirms that the WEOMextracted in (60°C, 30 min) presents the highest aromaticitystructure.

3.4 Spectroscopy fluorescence

Fluorophores are also associated to soil WEOM. A numberof studies have shown the suitability of fluorescence spec-troscopy to discriminate between different types of humicsubstances. Emission spectra of WEOM extracted at 60°Cobtained upon excitation at 254 nm are given Fig. 5; theyhave been recorded in solutions showing a TOC value of10 mg L−1. They all show a similar shape which typicallyconsisted in a broad band emission with a maximum cen-tered around 440 nm and shoulders at 350, 400, 500, and550 nm. This means that they contain the same fluoro-phores; only the concentrations seem different. The fluores-cence intensity at 440 nm varied in the order: 10<20<60<40<30 min. We observed an increase between 10 and30 min, while after 30–40 min, the fluorescence intensitydecreased with increasing extraction time. This result isagain consistent with a progressive molecular decomposi-tion of WEOM after 30 min at 60°C. In other words, thetime and temperature extraction can affect the fluorophoreof soil WEOM.

The common peak at λexc/λem0254/440 nm may beassociated to simple aromatic units such as phenolic-like,

1 0 20 3 0 4 0 50 6 00 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

0 .2 5

0 .3 0A

bsor

banc

e

Heating time/min

0

2

4

6

8

1 0

E2/E

3

Fig. 3 Absorbances at 254 (closed circles) and 365 nm (closed tri-angles) and ratio E2/E3 (open circles) of WEOM extracted at 60°C as afunction of the heating time. TOC content010 mg L−1, pH 6.5

2 00 300 40 0 50 0 6000 .0

0 .2

0 .4

0 .6

0 .8

1 .0

1 .2

Abs

orba

nce

Wavelength/mn

Fig. 4 UV–visible spectra of WEOM (60°C, 30 min; dark line),WEOM (80°C, 20 min; blue line), and WEOM (20°C, 24 h; red line).TOC content010 mg L−1 in pH 6.5

2404 Environ Sci Pollut Res (2012) 19:2400–2407

hydroxyl-substituted benzoic, cinnamic, and coumarinderivatives (Traversa et al. 2010).

The HIX, which is derived from emission fluorescencespectra (HIX), is well suited for determining the extent ofhumification of WEOM (Corvasce et al. 2006; Xiaoli et al.2010) and describes not only the content of aromatic-likespecial UV absorption but also the molecular structure ofWEOM (Senesi et al. 1991). During extraction, the HIXincreased progressively with increasing extraction time(Fig. 6). This result can be associated to a better extractionof humified compounds with increasing heating time. Con-trarily to the other parameters which decreased after 30–40 min of extraction, HIX is increasing with time. More-over, HIX is higher for the extract at 80°C than for theextract at 60°C, while the other parameters except yield ofextraction were lower at 80°C than at 60°C. It must bereminded that fluorescence is a very specific property aris-ing from a particular group of molecules and not from allchromophores; consequently, emission variations must notbe related to modifications of the entire pool of molecules.On one hand, comparing emission intensities brings somequantitative data. Emission intensities increase with theabsorbance at 254 nm showing that fluorophores content is

linked to this absorbance, but the fraction of chromophoresyielding fluorescence remain unknown. On the other hand,HIX index informs on qualitative aspects. The proportion oflong wavelength- to shorter wavelength-emitting constitu-ents increases with extraction time. In the case long heatinginduces some thermal degradation, this degradation affectsshorter wavelength-emitting constituents a little more thanlonger ones. Alternatively, it might happen that longwavelength-emitting constituents are produced on heating.

To resume, WEOM (60°C, 30 min) is the more fluorescentsample and should contain more aromatic and/or heterocycliccompounds, and less readily degradable compounds such ascarbohydrates, amino acids, and carboxylic acids (Xiaoli et al.2010). Consequently, the information brought by fluorescenceis in agreement with those given by UV–Vis spectroscopy.

3.5 UPLC analyses

To date, there is only little scientific data concerning themolecular composition of soil WEOM. This is mainly due tolow concentration of water-soluble organic compounds in soiland to the mixture complexity. A deeper investigation byUPLC on the molecular composition of soil WEOM is need-ed. The soil compounds extracted by water at 60°C, 30 min;80°C, 20min; and 20°C, 24 h were subject to chromatographyseparation by RP-UPLC, using a linear gradient of increasingmethanol concentration in water, as described in Section 2.The chromatography column material, Hydro-RP 80 Ǻ, waschosen because it is stable in a 100% aqueous mobile phaseand is therefore suitable to resolve highly polar compounds.Therefore, the hydrophobicity of each separated compoundwas estimated on the basis of the volume percent of methanolin the mobile phase and to the retention time.

The UPLC chromatograms obtained at 254 nm weresimilar for all WEOM whatever the temperature of extrac-tion. For each, we distinguish eight peaks (Fig. 7a). Analysisof UV–Vis spectra of individual peaks gives two types ofabsorption spectra (Fig. 7b). Peaks 1 to 4 were characterized

250 300 350 400 450 500 550 600 6500

50

100

150

200

250

300

350

400

450

500

550

60 min

40 min

30 min

20 min

Flu

ores

cenc

e In

tens

ity/a

.uWavelength /nm

10 min

Fig. 5 Fluorescence spectra ofWEOM extracted at 60°C as afunction of the heating time.Intensities were normalized onthe TOC content

1 0 2 0 3 0 4 0 5 0 6 00

1

2

3

4

5

6

7

Hum

ifica

tion

Inde

x (d

imen

sion

less

)

Heating time/min

Fig. 6 Variation of humification index with extraction time for theWEOM extracted at 60°C

Environ Sci Pollut Res (2012) 19:2400–2407 2405

by an exponential decay type without maximum but extend-ing until 400 nm. Only a shoulder around 280 nm appearedon peaks 3 and 4. This feature may be due to a sum ofelectron-delocalized systems to a system of delocalizedelectrons or by a sum of charge transfer complexes (DelVecchio and Blough 2004). Peaks 6 and 8 showed absorp-tion spectra with well-defined maxima (254 nm for peak 6and 262 nm for peak 8) and not absorption above 300 nm.Most probably, they correspond to simple organic com-pounds with conjugated doubles bonds. Peak 5 presentsboth the exponential decay and a maximum around250 nm. Peak 7 was not taken into account because it didnot show real absorption spectrum.

In order to have a rough evaluation of the relative abun-dance of these peaks, we gathered them in three fractions:fraction 1 gathers peaks 1 to 4; fraction 2, peaks 5 and 6; and

fraction 3, peak 8. We measured the area percent of thedetected peaks and used these values as indicative of therelative abundance of each fraction of WEOM. Table 3shows the relative fractions abundances. In general, differ-ences among samples are not very important. All WEOM

Fig. 7 UPLC chromatogram ofWEOM (60°C, 30 min). aChromatogram with UVdetection set at 254 nm. Untilretention time of 10 min, theeluent was pure water. Then,the methanol proportion wascontinuously increased to100%. b UV absorption spectraof eluted compounds obtainedfrom the diode array detector

Table 3 Percentage area of the different UPLC fractions

Fraction WEOM (20°C,24 h)

WEOM (60°C,30 min)

WEOM (80°C,20 min)

F1 70 65 79

F2 29 33 17

F3 1 2 4

F1 corresponds to peaks 1–4, F2 to peaks 5 and 6, and F3 to peak8 (see Fig. 7)

2406 Environ Sci Pollut Res (2012) 19:2400–2407

have the highest percentage in the fraction, which is themore hydrophilic one, comprised between 65% and 79%.However, WEOM (60°C, 30 min) shows a higher abun-dance of hydrophobic constituents (35%) than WEOM(80°C, 20 min) and WEOM (20°C, 24 h). This means thatextraction at 60°C improved the solubilization of hydrophobicconstituents and minimizing decomposition.

4 Conclusion

This study suggests that fundamental concepts regarding thecontrols of SOM extraction technique need to be revisited.Taking a soil enriched in clay, we show that the efficiency ofthe extraction protocol is potentially affected by the time ofextraction and the water temperature. The combination ofkinetic study, chemical analysis, spectroscopic investiga-tion, and chromatographic separation of recovered WEOMdemonstrated significant qualitative and quantitative differ-ences in the composition and molecular structure of theWEOM samples. We observed that extraction at 60°C for30 min was the best procedure in terms of yield, enrichmentin organic chemicals, and minimal alteration of soil WEOM.

Acknowledgements The authors wish to thank the Region Auvergneand the European Regional development fund for their financial sup-port and Guillaume Voyard, Assistant Engineer CNRS, for excellenttechnical assistance.

References

Chantigny HM (2003) Dissolved and water-extractable organic matterin soil: a review on the influence of land use and managementpractices. Geoderma 113:357–380

Corradini D, Grego S, Liburdi K, Marinari S, Tonci M (2006) A com-bined approach employing soxhlet extraction and linear gradientelution reversed-phase HPLC for fingerprinting of soil organicmatter according to hydrophobicity. Chromatographia 63:S11–S16

Corvasce M, Zolnay A, D’Orazio V, Lopez R, Miano ML (2006)Characterization of water extractable organic matter in a deep soilprofile. Chemosphere 62:1583–1590

Curtin D, Wright CE, Beare MH, McCallum FM (2006) Hot water-extractable nitrogen as an indicator of soil nitrogen availability.Soil Sci Soc Am J 70:1512–1521

Curtin D, Chantigny MH, Grennfield LG (2011) Controls on theextractability of soil organic matter in water over the 20 to 80°Ctemperature range. Soil Biol Biochem 75:1423–1430

Del Vecchio R, Blough NV (2004) Optical properties of humic substan-ces and CDOM: relation to structure. Environ Sci Technol 38:3885–3891

Fuentes M, Gonzalez GG, Garcia-Mina JM (2006) The usefulness ofUV-visible and fluorescence spectroscopies to study the chemicalnature of humic substances from soils and composts. Org Geochem37:1949–1959

Ghani A, Dexter M, Perrott KW (2006) Hot-water extractable carbonin soil: a sensitive measurement for determining impacts of fertil-ization, grazing and cultivation. Soil Biol Biochem 35:1231–1243

Gregorich EG, Beare MH, Stoklas U, St-Georges P (2003) Biodegrad-ability of soluble organic matter in maize-cropped soils. Geoderma113:237–252

Kaiser M, Ellerbrock RH (2005) Functional characterization of soilorganic matter fractions different in solubility originating from along-term field experiment. Geoderma 127:196–206

Kalbitz K, Solinger S, Park J-H, Michalzik B, Matzner E (2000)Controls on the dynamics of dissolved organic matter in soils: areview. Soil Sci 165:277–304

Kalbitz K, Schmerwitz J, Schwesig D, Matzener E (2003) Biodegra-dation of soil derived dissolved organic matter as related to itsproperties. Geoderma 113:273–291

Landgraf D, Leinweber P, Franz M (2006) Cold and hot water extract-able organic matter as indicators of litter decomposition in forestsoils. J Plant Nutr Soil Sci 169:76–82

Marinari S, Liburdi K, Corradini D, Grego S (2010) Reversed-phasehigh performance liquid chromatographic profile of organic frac-tions extracted by solvents with different polarity as a tool toevaluate the hydrophobic character of soil under differentmanagement. Soil Till Res 109:36–40

Matilainen A, Gjessing ET, Lahtinen T, Hed L, Bhatnagar A, SillanpääM (2011) An overview of the methods used in the characterisationof natural organic matter (NOM) in relation to drinking watertreatment. Chemosphere 83:1431–1442

Northcott GL, Jones KC (2000) Experimental approaches and analyt-ical techniques for determining organic compound bound residuesin soil and sediment. Environ Pollut 108:19–43

Peuravori J, Pihlaja K (1997) Molecular size distribution and spectro-scopic properties of aquatic humic substances. Anal Chim Acta337:133–149

Senesi N, Miano TM, Provenzano MR, Brunetti G (1991) Character-ization, differentiation and classification of humic substances byfluorescence spectroscopy. Soil Sci 152:259–271

Stevenson FJ (1994) Humus chemistry, genesis, composition, reac-tions, 2nd edn. Wiley, New York

Traversa A, Loffredo E, Gattullo CE, Senesi N (2010) Water-extractable organic matter of different compost: a comparativestudy of properties and allochemical effects on horticulturalplants. Geoderma 156:287–292

Ussiri DAN, Johnson CE (2002) Sorption of organic carbonfractions by spodosol mineral horizons. Soil Sci Soc Am J61:253–262

Xiaoli B, Linin W, Wenbao M, Xingna Y, William HMD, Honghua R(2010) Spectroscopic characterization of hot-water extractableorganic matter from soils under four different vegetation typesalong an elevation gradient in the Wuyi Mountains. Geoderma159:139–146

Xiaosong H, Beidou X, Zimin W, Xujing G, Mingxiao L, Da A,Hongliang L (2011) Spectroscopic characterization of water ex-tractable organic during compositing of municipal solid waste.Chemosphere 82:541–548

Zhang J, Song C, Yang W (2006) Land use effects on the distributionof labile organic carbon fractions through soil profiles. Soil SciSoc Am J 70:660–667

Zsolnay A (2003) Dissolved organic matter: artefacts, definitions andfunctions. Geoderma 113:187–209

Zsolnay A, Biagar E, Jimenez M, Steinweg F (1999) Differentiatingwith fluorescence spectroscopy the sources of dissolved organicmatter in soil subject to drying. Chemosphere 38:45–50

Environ Sci Pollut Res (2012) 19:2400–2407 2407