Movement of dichlorvos in farm soils: batch and column studies.

11 Soils are the principal environmental fate of pesticides in agricultural areas. Thus, the 12 kinetics, extension, and strength of the adsorption process become critical. Dichlorvos 13 (DDVP) is an organophosphorous pesticide that is used both in agriculture and livestock 14 production. Sorption/desorption assays of DDVP in two agricultural soils (with different 15 textural characteristics) from Pampa Plain (Argentina) were performed in both batch and 16 column systems. From batch studies, kinetics and sorption/desorption equilibrium 17 parameters were estimated. Our results showed that the maxima adsorption is reached 18 after 30 h of time of contact and followed a pseudo-first-order rate. Adsorption/desorption 19 data were well fitted to the Freundlich model obtaining high adsorption constants of 90 20  g (1-1/n) mL (1/n) g -1 and 21  g (1-1/n) mL (1/n) g -1 for the clay loam and sandy loam soil, 21 respectively. The isotherms were non-linear in both cases and the desorption process was 22 unfavourable. Also, positive hysteresis was present for the sandy loam soil. From column 23 studies, breakthrough curves were used to evaluate the mobility of DDVP in the soils at 24 1, 10, and 50 mg L -1 of DDVP. Eluted profiles were asymmetrical as well they presented retardation effects that were in connection with the results in batch conditions. Non- 26 equilibrium sorption was stated for the DDVP movement through columns. Thus, high 27 mobility was observed for DDVP in both soils despite their textural differences.


33
In the last decades, agricultural countries have undergone an important expansion in their 34 activities due to the development of new technologies and an increase in the specialization 35 to attain worldwide food demands. Therefore, there was an increase in the use of 36 phytosanitary products (mainly herbicides, insecticides, and fungicides) to improve crop 37 production yield. However, the excessive application of pesticides has become an 38 important environmental issue due to the high environmental concentrations found and 39 the deleterious effects on non-target individuals [1,2,3,4]. Organophosphate pesticides 40 are used in both agriculture and animal production being dichlorvos (2,2-dichlorovinyl 41 dimethyl phosphate, DDVP) one of the most employed of this family. DDVP is an systems [3,5]. In cultivated regions, soils and streams play an important role in the 48 transport and fate of pesticides used in crops and farms. There, pesticides can undergo 49 several processes such as adsorption, leaching, run-off, volatilization, and degradation 50 3 (biotic or abiotic) affecting the environmental concentrations [6,7,8]. DDVP has a short 51 half-lifetime in freshwater (2.8 days at pH=7 and 20°C) [9] and soils (4 days at pH= 6.2-52 7.4 and 25°C) [10]. However, dissolved organic matter in freshwater could affect the 53 DDVP behaviour reducing both, its photodegradation rate [8] and its biodisponibility 54 [ 11 ]. Moreover, groundwater contamination is not expected, according to its low 55 Gustafson index of 0.69 [12]. According to European regulations, the maximum limit for 56 priority substances in the inland waters is 0.0006 g L -1 according to water policy [13]. 57 However, DDVP was detected ranging from 0.05 to 7.50 gL -1 in freshwater [14,15]. In 58 Argentina, DDVP was the most detected insecticide in fish species collected from farm 59 areas located at the Pampa Plain [16]. 60 Adsorption studies are crucial to predict environmental fate constituting a valuable tool 61 to perform risk assessment evaluations [17]. Sorption studies of pesticides in soils were 62 extensively reported in the literature, however, among the insecticides, DDVP adsorption 63 studies are scarce [18]. Furthermore was quantified by HPLC as was described below (see Section 2.5). From these results,  where V is cumulative outflow volume and V0 is the total water volume into the column.

138
This was obtained gravimetrically at the end of each displacement.  The pseudo-order and the rate adsorption were assessed fitting the experimental data to 151 the kinetics models based on the adsorbent capacity such as the Lagergren equation [21] 152 and the Ho expression [22]. The pseudo-first-order (PFO) model assumes that the rate 153 adsorption is proportional to the driving force representing by: where k1 is the PFO rate constant (h -1 ), qe (g g -1 ) is the adsorbed concentration of DDVP 158 in the soil particles at the equilibrium time and qt (g g -1 ) is the concentration of DDVP 159 adsorbed at soil particles at the evaluated time, t. The integrated expression is: The pseudo-second-order (PSO) model assumes that the adsorption rate is proportional both adsorption and desorption processes as is suggested by the OECD normative [20].

189
For that, the exponential equation was used: The quantity of DDPV desorbed was calculated using the Eq. 9 where is the DDVP 197 adsorbed after a cycle of desorption, is the quantity of DDVP initially adsorbed,  where 1/n was previously estimated from the desorption and adsorption isotherms. An H 208 < 1 value (positive hysteresis) means that DDVP molecules tend to be adsorbed whereas, 209 an H > 1 value (negative hysteresis) suggests that the desorption process is favoured.

210
Additionally, if H = 1 the hysteresis is considered to be absent and the process is 211 considered reversible [23]. Graphics and data analysis were done by using Sigma Plot 11.0 [24]. The physicochemical and textural properties of collected soils are shown in Table 1.  experimental data were adjusted to PFO and PSO models (see Figure 2) and kinetics 249 parameters were obtained (see Table 2).  indicated that the adsorption followed a PSO mechanism whose rate constants ranging 272 from 6.5 to 15.77 g mg -1 h -1 . They attribute these differences to the heterogeneous nature 273 of soil particles and its variable composition [18].  Table 4 summarizes the adsorption parameters estimated in this work.   soils (a high quantity of DDVP is retained after the desorption cycle).

312
Hysteresis calculated from Eq. 10 is associated with the reversibility of the process, 313 comparing indirectly the forces involved in both, the adsorption and the desorption [27].

314
In the present study, similar Freundlich desorption and adsorption coefficients were  (see table 3).

323
Taking into account only equilibrium data could be inferred that MS and 9S had different 324 behaviour in connection with the DDVP movement. In fact, for the maxima concentration 325 evaluated in this work, 500 µg mL -1 , 45% is adsorbed and the rest is lixiviated for MS.

326
Moreover, from the adsorbed quantity, only 2.66% is desorbed after the first desorption 327 cycle. Otherwise, for 9S, at the same concentration, 93.6% was lixiviated and the rest 328 remains adsorbed. From this fraction, 50% is desorbed after the first cycle of desorption.

329
Lixiviation implies a risk of leaching to groundwater sources while, in the opposite, the where ρ is the soil bulk density and θ is the volumetric soil-water content.

358
For non-adsorbed solutes, the adsorption coefficient is equal to zero and R becomes one. In our experiments, this value was around 0.035 -0.03 that is minor to the minimum 374 value to assess equilibrium conditions into the column ( ≥ 5) [30]. In this context, it is 375 20 expected that an eluted profile shows both effects, retardation (due to adsorption 376 processes) and tailing at higher pore volumes (due to non-equilibrium adsorption   Additionally, the pore volume needed to obtain the maximum value of C/C0 was around 392 3.8 and 2.5 -3.0 for Mercedes and 9 de Julio soils, respectively. This result suggests 393 retardation in the DDVP movement [31]. Furthermore, the mobility in 9S soil is around 394 1.25 times higher than in MS that is according to the previous results presented in Table   395 3 were stronger adsorption is estimated for MS.

396
Moreover, from Figure 4A, it is observed a very short shift of BTC at 1 mg L -1 of DDVP.

397
Despite that, for C0 of 5 and 50 g mL -1 , the retardation effect does not seem to be a as a reversible process in MS but non-reversible in 9S. Regarding kinetic assays, it was 418 indicated that both soils followed the kinetics of PSO whose constants were very similar.

419
More important, it was also observed that the equilibrium state was achieved after 30 Availability of data and materials 434 All data generated or analyzed during this study are available upon request.