**6. Physical/chemical properties in QSAR models: a mechanistic interpretation**

QSAR modelling is a useful technique to predict the activity of chemicals, such as pesticides in a short time and with low cost, establishing a statistical relationship between the activity of chemicals and their structural and physicochemical properties [34–37]. The development of QSAR models with regulatory purposes is based on precaution, while their application facilitates prevention. In this regard, the member countries of the organisation for economic co-operation and development (OECD) have developed a set of five guiding principles, enabling the practical application of QSAR modelling as a reliable tool in the regulatory context, which have been adopted by the European Union and United States [38–40]: (i) a defined endpoint; (ii) an unambiguous algorithm; (iii) a defined domain of applicability; (iv) an appropriate measure of goodnessof-fit, robustness and predictivity and (v) a mechanistic interpretation. The QSAR models are based on the assumption that chemicals are able to reach and interact with the target site by similar mechanism, related to their similar physicochemical properties [40].

*Advanced Sorption Process Applications*

energy requirements [21].

where∼20% of sites accounted for instantaneous sorption on andisol. For the ultisols, most of the sites corresponded to the time-dependent stage of sorption (90%).

The higher value in the overall rate constant, k2, of MSM on andisols with respect to DI in the same soil indicates that this value reflects contributions from the favoured electrostatic interactions considering both a retarded IPD as well as intra-OM diffusion. The way minerals present on VADS are interrelated or chemically spatially distributed, either being freely distributed throughout the soil mass or coating silt and clay grains, is determinant of their chemical role in the whole ion sorption-desorption mechanisms [28]. In this sense, the OM content is the principal component to control the pesticide sorption on andisols, as much by instantaneous equilibrium as by IPD, the presence of kaolinite, halloysite and Al/Fe oxides in Ultisols will be significant in the IPD mechanism. According to this analysis, ultisols present a potential risk of ionisable herbicide transport. The different mineral composition of ultisols impacts on their different physical behaviour, influencing the slowest INIH sorption rate, the sorption mechanism involved and the lowest INIH sorption capacity. All of the above must be taken into account to evaluate the potential leaching of INIH in these kinds of soils.

**5. Sorption of ionisable and non-ionisable herbicides on VADS**

The retention and mobility of herbicides in soil are determined by the strength and extent of sorption reactions [14]. The soil particles (adsorbent) may adsorb herbicides (adsorbate) weakly or strongly depending on the adsorbent-adsorbate interactions. In this sense, herbicides can be adsorbed in soil through different mechanisms such as physical sorption (van der Waals and H bond interactions) and chemical sorption. Physical sorption is fast and usually reversible, due to small

The physical and chemical interactions between herbicides and other organic molecules on the soil particles surface depend on the physical and chemical properties of the soil and herbicide. The nature of surface charge may vary with the chemical properties of VADS. The surface charge amphoteric characteristics will confer to VADS chemical and physical properties absolutely different from those exhibited in soils of constant charge. This surface reactivity of VADS confers to them a particular behaviour in relation to the retention of herbicides, representing an environmental substrate that may become polluted over time due to intensive agronomic uses. There are only a few reports on the behaviour of INIH in VADS despite being important to agricultural systems of many regions (**Figure 1**). A higher sorption capacity of several herbicides has been reported for allophanic soils (**Figure 1**). In this regard, the herbicide sorption on VADS will be affected by soil properties (SOM content, allophane, clay, pH, IS, particle size distribution, moisture content and variable charge) and herbicide chemical properties (molecular structure, molecular size, electrical charge, ionisability, aqueous solubility, hydrophobicity (Kow), volatility, reactivity with soil constituents and longevity in the environment) [29]. Environmental conditions may also affect INIH sorption and mobility on VADS.

**5.1 Sorption of ionisable herbicides and non-ionisable herbicides on VADS**

than 30, regardless of the mineral content [7, 9, 19].

The ion sorption rate in VADS depends on the surface area, CEC, the proportion of Fe/Al oxides or oxyhydroxides present as the surface coating of clay and silt particles [28]. Oxides have also been found to enhance the deprotonation of organic acids and, therefore, increase the activity of the anionic species at a given is expected to be at a maximum if the ratio of the mineral to the OC fractions is more

**114**

### **6.1 QSAR models for sorption of pesticides in soils**

The recent developments in QSAR models for the estimation of soil sorption coefficients of organic pesticides will be reviewed, taking into account the literature related to QSAR and its mechanistic interpretation. The aim of this section is to examine the descriptors used in the prediction of *Koc* values on soil by means of QSAR models, which according to principle (v) should be associated with a 'mechanistic interpretation'. In this regard, two approximations have been used for the mechanistic basis of the models: (i) statistical approach and (ii) mechanistic approach. In the 'mechanistic approach', the descriptors selection can be guided by the modeller's a priori knowledge of the physicochemical properties involved in the mechanism proposed for the studied activity. Thus, presumed mechanistic meaning can be assigned to some molecular descriptor used; then, the modeller selects it personally from a limited pool of potential modelling variables, which are largely employed (for instance: log*Kow*, log*P*, log*D*, pKa, OC, etc.). However, as the sorption mechanisms of pesticides on soils and especially on VADS are quite complex, their understanding is only possible at certain levels of approximation. The a priori selection of one (or more) physicochemical variable(s) for their mechanistic meaning, in relation to one assumed mechanism, is very risky. In this sense, important variables influencing other mechanisms, participating in such a response, could remain ignored.

#### **6.2 Statistical approach: a posteriori mechanistic interpretation**

The European joint research centre (JRC) QSAR models database provides information on the recent developments in QSAR models that can be used for purposes of regulatory assessment of chemicals (e.g. REACH registration). In this database, four detailed QSAR models to predict the sorption partition coefficient are normalised to the OC content of the soil (*Koc*, *Koc* <sup>=</sup> *Kd*/*OC* or *Koc* <sup>=</sup> *Kf*/*OC*). The QSAR Model Reporting Format (QMRF) documentation for all QSAR models is available in the JRC's QMRF Inventory [41–44]. All these models were developed by statistical approach, wherein no mechanistic basis for their descriptors selection was set a priori. The first was published few years ago [44]. In this study, QSAR model was developed using *Koc* values of 142 non-ionic organic pesticides (34 split for the validation set). The *Koc* data for 20 other chemicals were used as external validation set. The QSAR model developed in this study is represented by Eq. (1):

$$\begin{array}{lcl}\text{Log K}\_{\text{ac}} & = & 0.96 \text{ -- } 0.26 \text{ Polarity parameter (AM1)} \text{ (distance} \\ & + & 1.07^{-0.02} \text{ ALFA polarization (DIP)} \text{ (AM1)} \\ & - & 1.99 \text{ } Max \text{ } net \text{ atomic charge (AM1)} \text{ for C atoms} \\ & + & 1.30 \text{ }^{-0.02} \text{ WINSA1 Weight (PNSA 1 \* TMSA)} \text{ (Zefivov)} \\ \end{array}$$

$$\text{IN = 142 (nTr = 108, nPrred = 34); R = 0.75; Q}\_{\text{LMO3096}} = 0.73; \text{s}^2 = 0.445.$$

where '*ALFA polarizability* (*DIP*) (*AM1*)' and '*WNSA1Weighted PNSA*( *PNSA1* <sup>∗</sup> *TMSA* \_\_\_\_\_\_\_\_\_\_\_\_ <sup>1000</sup> ) (*Zefirov*)' are descriptors that quantify the molecular size related to charge distribution. The '*Max net atomic charge* (*AM1*) *for C atoms* and the '*Polarity parameter* (*AM*1)/*distance*' are descriptors related to charges and to charge distribution.

Recently, Mansouri and Williams [41] published a new reliable QSAR model for estimating the *Koc* of heterogeneous organic chemicals. This is freely available as an open-source, command-line application called OPERA (OPEn structure–activity/ property Relationship App) [45]. The model was generated using 729 curated outlierfree experimental *Koc* data, which were divided into training (545 compounds) and

**117**

*Impact of Physical/Chemical Properties of Volcanic Ash-Derived Soils on Mechanisms Involved…*

validation sets (184 compounds). The descriptors' selection consisted of coupling genetic algorithms (GAs) with the weighted *kNN* algorithm, which allows building a model with 12 molecular descriptors, related to *logP* and water solubility. These descriptors were calculated using the free and open-source software PaDEL [46]. The model has potential to correctly predict the Log*Kd* values of organic pesticides since the predictivity-statistics obtained by external validation was quite significant, i.e.

In both QSAR models exhibited above, compounds with large molecular size tend to have higher soil sorption than compounds with small molecular size, due to their lower water solubility. For electronic descriptors related to charges and charge distribution, the presence of active functional group adjacent to carbon allows a high charge on this atom, which together with likewise higher polarity leads to better water solubility. In the first, the minus sign in the QSAR equation (Eq. (1**)**) on these descriptors indicates that the higher the values, the lower the soil sorption. In the last update published by Gramatica et al. in JRC QSAR Model Database, QSAR Eq. (2) [43, 47] and Eq. (3) [42, 47], QSAR models were generated using the *Koc* experimental data of 643 heterogeneous organic compounds obtained from literature. The median of the *Koc* values was used when more than one value was available for a single compound.

*LogKoc* = 0.87 + 0.26 *VP* − 0 − 0.23 *nHBAcc* + 0.08 *nAromBond* − 0.19 *MAXDP* (2)

*LogKoc* = −1.92 + 2.07 *VED*<sup>1</sup> − 0.31 *nHBAcc* − 0.31 *MAXDP* − 0.39 *CIC*<sup>0</sup> (3)

The model described by Eq. (2) and Eq. (3) takes into account two different ways to describe a potential intermolecular adsorbate-adsorbent interaction. First, *nHBAcc* (Eq. (2) and Eq. (3)) is related to electronegative atoms of molecules that form hydrogen bond indicating a potential mechanism. Second, the *MAXDP* descriptor (Eq. (2) and Eq. (3)) related to molecule electrophilicity supposed a possible sorption mechanism of charge transfer between adsorbate and adsorbent. These mechanisms have been suggested for pesticides where amine and/or heterocyclic N atoms (e.g. AT) act as electron donors to acceptor structural groups of humic acid [48]. Moreover, Briceno et al. [14] conclude that the mechanisms involved in the AT soils' sorption are both hydrogen bonding and charge transfer (**Table 1**). The other two descriptors in each QSAR equation, Eq. (2) (*VP* <sup>−</sup> 0 and *nAromBond*) and Eq. (3) (*VED*1 and *CIC*0), are related to molecular size and have positive signs. Models indicate in general that the larger compounds are more sorbed than leached. Consequently, the ability of these models in estimating the *Koc* values of pesticides is restricted to sorption of non-ionised chemicals on permanent charge soils, which were widely represented in the calibration sets as well as in validation sets. Moreover, a mechanistic interpretation of molecular descriptors for non-ionisable pesticides was provided a posteriori, after modelling, by interpretation of the final model in view of an association between the descriptors used and the soil sorption predicted.

**6.3 Mechanistic approach: polyparameter linear free energy relationships**

In the last decade, the concept of polyparameter linear free energy relationships (PP-LFERs) has been widely used for the prediction of sorption coefficients of neutral organic chemical, because of its important mechanistic basis and good

= 0.79; Q2

= 0.79; Q2

LMO30% = 0.79; CCC (concor-

BOOT = 0.79; CCC (concor-

*DOI: http://dx.doi.org/10.5772/intechopen.81155*

where N = 643 (nTr = 93, nPred =550); R<sup>2</sup>

where N = 643 (nTr = 93, nPred =550); R<sup>2</sup>

dance correlation coefficient) = 0.88; RMSE = 0.55.

dance correlation coefficient) = 0.89; RMSE = 0.54

= 0.71 and RMSD = 0.61.

R2

*Impact of Physical/Chemical Properties of Volcanic Ash-Derived Soils on Mechanisms Involved… DOI: http://dx.doi.org/10.5772/intechopen.81155*

validation sets (184 compounds). The descriptors' selection consisted of coupling genetic algorithms (GAs) with the weighted *kNN* algorithm, which allows building a model with 12 molecular descriptors, related to *logP* and water solubility. These descriptors were calculated using the free and open-source software PaDEL [46]. The model has potential to correctly predict the Log*Kd* values of organic pesticides since the predictivity-statistics obtained by external validation was quite significant, i.e. R2 = 0.71 and RMSD = 0.61.

In both QSAR models exhibited above, compounds with large molecular size tend to have higher soil sorption than compounds with small molecular size, due to their lower water solubility. For electronic descriptors related to charges and charge distribution, the presence of active functional group adjacent to carbon allows a high charge on this atom, which together with likewise higher polarity leads to better water solubility. In the first, the minus sign in the QSAR equation (Eq. (1**)**) on these descriptors indicates that the higher the values, the lower the soil sorption. In the last update published by Gramatica et al. in JRC QSAR Model Database, QSAR Eq. (2) [43, 47] and Eq. (3) [42, 47], QSAR models were generated using the *Koc* experimental data of 643 heterogeneous organic compounds obtained from literature. The median of the *Koc* values was used when more than one value was available for a single compound.

*LogKoc* = 0.87 + 0.26 *VP* − 0 − 0.23 *nHBAcc* + 0.08 *nAromBond* − 0.19 *MAXDP* (2)

where N = 643 (nTr = 93, nPred =550); R<sup>2</sup> = 0.79; Q2 LMO30% = 0.79; CCC (concordance correlation coefficient) = 0.89; RMSE = 0.54

$$\text{Log}\,K\_{\text{oc}} = -1.92 + 2.07\,\text{V}\,\text{ED}\_1 - 0.31\,\text{nH}\,\text{B}\,\text{ac} \, - 0.31\,\text{MAXDP} - 0.39\,\text{CIC}\_0 \tag{3}$$

where N = 643 (nTr = 93, nPred =550); R<sup>2</sup> = 0.79; Q2 BOOT = 0.79; CCC (concordance correlation coefficient) = 0.88; RMSE = 0.55.

The model described by Eq. (2) and Eq. (3) takes into account two different ways to describe a potential intermolecular adsorbate-adsorbent interaction. First, *nHBAcc* (Eq. (2) and Eq. (3)) is related to electronegative atoms of molecules that form hydrogen bond indicating a potential mechanism. Second, the *MAXDP* descriptor (Eq. (2) and Eq. (3)) related to molecule electrophilicity supposed a possible sorption mechanism of charge transfer between adsorbate and adsorbent. These mechanisms have been suggested for pesticides where amine and/or heterocyclic N atoms (e.g. AT) act as electron donors to acceptor structural groups of humic acid [48]. Moreover, Briceno et al. [14] conclude that the mechanisms involved in the AT soils' sorption are both hydrogen bonding and charge transfer (**Table 1**). The other two descriptors in each QSAR equation, Eq. (2) (*VP* <sup>−</sup> 0 and *nAromBond*) and Eq. (3) (*VED*1 and *CIC*0), are related to molecular size and have positive signs. Models indicate in general that the larger compounds are more sorbed than leached. Consequently, the ability of these models in estimating the *Koc* values of pesticides is restricted to sorption of non-ionised chemicals on permanent charge soils, which were widely represented in the calibration sets as well as in validation sets. Moreover, a mechanistic interpretation of molecular descriptors for non-ionisable pesticides was provided a posteriori, after modelling, by interpretation of the final model in view of an association between the descriptors used and the soil sorption predicted.

#### **6.3 Mechanistic approach: polyparameter linear free energy relationships**

In the last decade, the concept of polyparameter linear free energy relationships (PP-LFERs) has been widely used for the prediction of sorption coefficients of neutral organic chemical, because of its important mechanistic basis and good

*Advanced Sorption Process Applications*

**6.1 QSAR models for sorption of pesticides in soils**

The recent developments in QSAR models for the estimation of soil sorption coefficients of organic pesticides will be reviewed, taking into account the literature related to QSAR and its mechanistic interpretation. The aim of this section is to examine the descriptors used in the prediction of *Koc* values on soil by means of QSAR models, which according to principle (v) should be associated with a 'mechanistic interpretation'. In this regard, two approximations have been used for the mechanistic basis of the models: (i) statistical approach and (ii) mechanistic approach. In the 'mechanistic approach', the descriptors selection can be guided by the modeller's a priori knowledge of the physicochemical properties involved in the mechanism proposed for the studied activity. Thus, presumed mechanistic meaning can be assigned to some molecular descriptor used; then, the modeller selects it personally from a limited pool of potential modelling variables, which are largely employed (for instance: log*Kow*, log*P*, log*D*, pKa, OC, etc.). However, as the sorption mechanisms of pesticides on soils and especially on VADS are quite complex, their understanding is only possible at certain levels of approximation. The a priori selection of one (or more) physicochemical variable(s) for their mechanistic meaning, in relation to one assumed mechanism, is very risky. In this sense, important variables influencing other mechanisms, participating in such a response, could remain ignored.

**6.2 Statistical approach: a posteriori mechanistic interpretation**

set. The QSAR model developed in this study is represented by Eq. (1):

= 0.75; Q2

(*Zefirov*)' are descriptors that quantify the molecular size related to charge distribution. The '*Max net atomic charge* (*AM1*) *for C atoms* and the '*Polarity parameter* (*AM*1)/*dis-*

Recently, Mansouri and Williams [41] published a new reliable QSAR model for estimating the *Koc* of heterogeneous organic chemicals. This is freely available as an open-source, command-line application called OPERA (OPEn structure–activity/ property Relationship App) [45]. The model was generated using 729 curated outlierfree experimental *Koc* data, which were divided into training (545 compounds) and

where '*ALFA polarizability* (*DIP*) (*AM1*)' and '*WNSA1Weighted PNSA*(

*PNSA*<sup>1</sup> <sup>∗</sup> *TMSA* \_\_\_\_\_\_\_\_\_\_\_\_ <sup>1000</sup> )(*Zefirov*) (1)

= 0.445.

*PNSA1* <sup>∗</sup> *TMSA* \_\_\_\_\_\_\_\_\_\_\_\_ <sup>1000</sup> )

LMO30% = 0.73; s2

+ 1.07−0.02*ALFA polarizability* (*DIP*)(*AM*1) − 1.99 *Max net atomic charge* (*AM*1) *for C atoms*

*LogKoc* = 0.96 − 0.26 *Polarity parameter* (*AM*1)/*distance*

<sup>+</sup> 1.30−0.02 *WNSA*<sup>1</sup> *Weighted PNSA* (

*tance*' are descriptors related to charges and to charge distribution.

N = 142 (nTr = 108, nPred =34); R2

The European joint research centre (JRC) QSAR models database provides information on the recent developments in QSAR models that can be used for purposes of regulatory assessment of chemicals (e.g. REACH registration). In this database, four detailed QSAR models to predict the sorption partition coefficient are normalised to the OC content of the soil (*Koc*, *Koc* <sup>=</sup> *Kd*/*OC* or *Koc* <sup>=</sup> *Kf*/*OC*). The QSAR Model Reporting Format (QMRF) documentation for all QSAR models is available in the JRC's QMRF Inventory [41–44]. All these models were developed by statistical approach, wherein no mechanistic basis for their descriptors selection was set a priori. The first was published few years ago [44]. In this study, QSAR model was developed using *Koc* values of 142 non-ionic organic pesticides (34 split for the validation set). The *Koc* data for 20 other chemicals were used as external validation

**116**

prediction power [49–53]. PP-LFERs are multiple linear regression (MLR) models that employ several solute- or sorbate-specific descriptors as independent variables and their fitting coefficients are denoted as they describe system-specific, soluteindependent properties. In this sense, descriptors and their coefficients quantitatively describe the energetic contribution of different types of sorption coefficients [53, 54]. The major advantages of the PP-LFER approach are its solid mechanistic grounds and the use of uniformly measured calibration data.

In the last decade, different PP-LFER models for organic contaminants sorption on soil estimation have been proposed. Endo et al. [52] proposed two PP-LFER models at environmentally relevant concentrations. However, these models lack reliable PP-LFER descriptors for environmentally relevant chemicals (e.g. pesticides, pharmaceuticals and highly polar compounds, acids, bases, and ionic compounds). This deficiency also has been identified for PP-LFER models developed for high sorbate concentrations previously reported [55]. The PP-LFER models' reviews up to now mainly have been calibrated estimating log*Koc* data of classical pollutants such as PCBs and PAHs and also of organic compounds that have chemical structure comparatively simple than chemicals of current environmental concern. These are often multifunctional or complex organic chemicals like pesticides and pharmaceuticals. The first reliable PP-LFER model for soil-water partitioning was calibrated with data from 79 polar and non-polar compounds that cover a more diverse and wider range of chemical classes than other PP-LFERs published. The model of Bronner and Goss [49] was validated using the experimental data for about 50 pesticides and pharmaceuticals not involved in the calibration set. This has potential to correctly estimate the *Koc* data for multifunctional or complex organic chemicals like pesticides and pharmaceuticals. However, Sabljic and Nakagawa [53] suggest still important drawbacks to the general applicability of the developed model. In view of the scope of this section, we recommend the review made by Sabljic and Nakagawa [53] around this topic.

On the other hand, little attention has been paid to the general applicability of the calibrated PP-LFERs for predicting sorption to soils considering the diversity of soil mineralogy, variable surface charge, OC structures and their interactions [51]. The evaluation of possible applications of PP-LFERs in the study of partitioning of ionic organic chemicals is a subject of ongoing research [56, 57].

#### **6.4 Mechanistic approach: QSAR models for sorption of ionisable pesticides**

In the last decade, different authors developed equations to predict the sorption of ionisable and non-ionisable compounds in soils or sediments [25, 58–61]. Several models have expanded their applicability domain including soil properties and ionisation effects [48, 58, 59]. Franco et al. carried out a surface acidity correction, because the two units proposed by Bintein and Devillers [59] are dependent on soil properties, related to the surface potential of the colloid [25]. These researches suggest a general non-linear equation based on *LogKow* for neutral and ionic species (a fragmentation of *LogD*) and the speciation of monovalent acids, monovalent bases and amphoteric species. Franco et al. aimed to predict pH-dependent *Kd* values of organic acids, considering speciation as a function of soil pH and species-specific partition equilibrium [60]. This modification of their previous models by replacing their constant terms pHopt by a varying pH range allowed that the modified model performs significantly better than the original model for organic acids [25]. The two molecular descriptors, pKa and *logPn*, and the two soil descriptors, OC and pH, used in the model have a major impact on the sorption of ionisable chemicals. Nevertheless, it was not successful to develop the analogous modified model for bases due to the contradictory effect of pH on the total sorption.

**119**

*Impact of Physical/Chemical Properties of Volcanic Ash-Derived Soils on Mechanisms Involved…*

**6.5 Physical/chemical properties of VADS: considerations for a mechanistic** 

In general, VADS are rich in OM, possess high specific surface area, variable surface charge and consequently pH-dependent CEC and AEC (anion-exchange capacity (AEC). In this way, they have significant differences with regard to soils with mineralogy dominated by constant charge minerals [2]. These differences in the physicochemical properties make difficult the *Koc* prediction for organic ionisable compounds using generic QSAR models or also published PP-LFER models. In this regard, the descriptors selection to a mechanistic interpretation of pesticide-VADS system must be related to the type of organic compound (e.g. non-ionisable,

According to physical/chemistry characteristics of VADS, in these soils, the pH is a critical parameter when ionisable pesticide-VADS interaction is considered, because the sign and magnitude of total VADS surface charge depends on pHequilibrium. The total VADS surface charge is defined by the ZPCSoil. The total VADS surface charge, at a given pH, could be negatively charged if pHequilibrium > pHZPC affecting the pesticide-VADS interaction (i.e. andisol and ultisols), which is particularly important for ionisable pesticides. The Fe/Al oxide content is an important variable for ionisable pesticide sorption on positively charged active sites at pHequilibrium < pHZPC and also in the remaining oxide sites that exhibit positive charge even at pH higher than ZPC [4, 17, 48]. Additionally, it is necessary to consider the relationship between these oxides and OM due to blockage of specific sorption sites, and between oxides and pH due to the presence of pH-dependent sorption sites and the speciation of ionisable pesticides [48, 62]. In this regard, Hyun et al. [63, 64] demonstrated that anion exchange in VCS is significant for pentachlorophenol and prosulfuron. Moreover, the extent of anion exchange correlated well to the ratio of pH-dependent AEC to CEC (i.e. AEC/CEC) as well as the ratio of AEC to the total number of soil surface charge (AEC + CEC) (i.e. AEC/(AEC + CEC)). Caceres-Jensen et al. [5] studied the effect of MSM sorption on total VADS surfaces observing a change in total VADS surfaces produced when the highest MSM concentration was sorbed. A displacement of IEPSoil to a higher pH was established for the soils. These results confirmed the contribution of charged surface sites on VADS to the sorption of anionic MSM through electrostatic interactions. Finally, soil composition, mineralogy (e.g. amorphous (hydro)oxides, Fe/Al oxide content), texture (e.g. silt, sand or clay content), surface area of colloids, OM, AEC and its relation with CEC (i.e. AEC/CEC, AEC + CEC, AEC/(AEC + CEC), ZPCSoil and pHequilibrium are potential modelling VADS descriptors; due to these descriptors, the *Koc* magnitude is strongly influenced by physicochemical properties of VADS, sorp-

According to the type of organic compound, special attention has been given to ionisable organic compounds. With changes in the pH, the speciation of soil active sites and of the ionisable pesticide also change, affecting the sorption. For ionisable pesticides, *logD* could be a good descriptor for the variation in hydrophobic interactions. Also, pesticide pKa values are possible descriptors that take into account dissociation in order to describe the interactions for pesticides on VADS considering the soil pH. This molecular property is determinant of hydrophilic interactions for

The surface charge amphoteric characteristics will confer to VADS physical/ chemical properties absolutely different to constant-charge soils, where soil

ionisable, acid, base, etc.) and to physicochemical properties of VADS.

*DOI: http://dx.doi.org/10.5772/intechopen.81155*

tion sites and specific surface area [2, 17, 19, 63].

polar compounds [17, 48].

**7. Conclusions**

**approach**
