Understanding the Environmental Behavior of Herbicides: A Systematic Review of Practical Insights

Abstract

Herbicides play a crucial role in weed control in various agricultural and non-agricultural settings. However, their behavior in the environment is complex and influenced by multiple factors. Understanding their fate and retention, transport, and transformation is essential for effective herbicide management and minimizing their impact on ecosystems. This chapter begins by emphasizing the importance of studying herbicide behavior in real-world conditions, considering physical, chemical, and biological amendments in soil. It highlights how these amendments can directly affect weed control efficacy when residual herbicides are applied in pre-emergence. Detailed knowledge of herbicide behavior in the environment enables the adjustment of application rates based on soil type and climatic conditions, which is a key aspect of precision agriculture. The study of herbicide interactions in the environment has experienced significant growth across various subfields, particularly in the last three decades. It can be considered a multidisciplinary subject that encompasses areas such as agricultural, environmental, and biological sciences, as well as technology, physics, chemistry, and biomedicine. Overall, there are over 35,000 papers on herbicide behavior in the environment, and the trend indicates that the number of publications will continue to grow in the coming years.

1. Introduction

The behavior of herbicides in the environment, particularly in the soil, depends on the interaction of various processes that determine their ultimate fate [1]. The processes of transport, retention, and transformation of herbicides are often studied separately, but many of them occur simultaneously at different intensities, depending on the characteristics of the application environment and the physicochemical properties of the products. Therefore, understanding soil properties, the involved climatic factors, and the mechanisms of herbicide interaction with the environment are essential for achieving more effective weed control using herbicides in a technically and economically viable manner, and/or for better understanding the ultimate fate (reduced impact) of these products in the environment [1].

Currently, there are a significant number of scientific publications both internationally and in Brazil that addresses the behavior of herbicides in the soil for weed management. However, almost all of these studies focus on a more scientific and theoretical approach, making it difficult for readers such as students, farmers, and professionals in the field to understand and apply the findings in practical situations.

Thus, considering the substantial volume of material available, there is an opportunity to convert this content into a chapter that aims to enhance and contribute to the practical technological knowledge of the reader, while still utilizing a solid theoretical foundation with concepts, examples, and illustrations. Therefore, the purpose of this chapter is to provide a holistic, up-to-date, and practical approach to understanding the behavior of herbicides in the environment, whether in agricultural settings or in interactions with fauna, flora, and humans.

In the following sections, we will discuss the physicochemical properties of herbicides and the processes that govern their behavior and fate in agriculture, including retention, transport, and transformation. In addition to these initial considerations, the interactions between soil-climate factors and herbicides, as well as their effects on the environment, will be discussed in a clear and objective manner.

2. Physical- Chemical properties of herbicides

Herbicides are compounds with diverse chemical structures and physical-chemical properties, and they can be grouped into different chemical families. Herbicides within the same chemical family have close similarities in their chemical structures and, therefore, similar physical-chemical properties. The physical-chemical properties of herbicides are one of the main factors that affect their initial distribution and behavior in the soil. Therefore, herbicides belonging to the same chemical class will exhibit similar behavior in the environment.

The main physical-chemical properties and equilibrium parameters that control the initial tendency of herbicides in the soil are the acid-base dissociation constant (pK_a/pK_b), water solubility (S_w), octanol-water partition coefficient (K_ow), vapor pressure (VP) or Henry’s law constant (K_H), sorption coefficient (K_d and K_oc), degradation half-life time (in the laboratory) and dissipation (in the field) (t_1/2 or DT₅₀), and residual lifetime in plant (RL₅₀). In a hypothetical scenario where environmental factors and climatic conditions are fixed, the distribution of herbicides among compartments (air, water, and sediment/soil) will depend on their physical-chemical properties [2, 3].

2.1 Acid-base dissociation constant (pK_a/pK_b)

Herbicides can be classified as ionizable and non-ionizable (neutral) compounds, having different reactivities and interactions with the soil. Ionization is the process by which an atom or molecule acquires a positive or negative charge, resulting in an atom or molecule with known electrical charge called an ion (anion or cation). Weak acid or base herbicides can be partially (but not completely) ionized in an aqueous solution, depending on the soil/solution pH. Due to partial ionization, there will be variable concentrations of neutral and ionized species in the solution.

The relative strength (degree of ionization) of the weak acid or basic compounds is measured by the ionization equilibrium constant, pK_a or pK_b, which is dependent on the soil solution pH. It calculates the relative amounts of each species present in the soil solution. When the soil solution pH = pK_a/pK_b, the ionizable compound in question (whether acid or base) will be half (50%) protonated (non-ionized or molecular) and half (50%) deprotonated (ionized).

Thus, there will be equal amounts of ionized and non-ionized forms in the solution. In the soil, ionic herbicides behave differently from non-ionic herbicides. Therefore, it is crucial to know which groups of herbicides are classified as ionizable to have an understanding of their potential behavior once applied to the soil [4]. For example, there are non-ionizable (neutral) herbicides such as diuron, carfentrazone, flumioxazin, S-metolachlor, acetochlor, pyroxasulfone, and clomazone; basic herbicides such as atrazine and hexazinone; and acid herbicides such as metribuzin, 2,4-D, alachlor, ametryn, indaziflam, clethodim, chlorimuron-ethyl, glyphosate, dicamba, diclosulam, imazethapyr, nicosulfuron, sulfentrazone, and glufosinate-ammonium. Diquat is a cation (ion +).

In agronomic soils (pH > pK_a), weak acid herbicides are primarily found in an anionic form (negatively charged), while weak base herbicides are mainly found in a neutral form (without charge). Anionic herbicides will be repelled by soil colloids because soils are predominantly negatively charged [4, 5], which increases their leaching potential due to low sorption, consequently reducing weed control efficacy. In contrast, protonated species of basic herbicides will be more sorbed in highly acidic soils (pH < pK_b) and will have lower leaching in the soil profile. Therefore, under field conditions, knowing the herbicide ionization (especially acid herbicides applied directly to the soil in pre-emergence) is crucial for agronomic effectiveness in weed control, as liming practices to raise the pH for agricultural cultivation directly interfere with the herbicide’s bioavailability in the soil solution [3].

2.2 Water solubility (S_w)

S_w is a measure of the concentration of an herbicide that can dissolve in water at a specific temperature [3]. S_w is usually measured in mg/L. According to Ney [6], S_w is considered low when the value is less than 10 mg/L, moderate when the value ranges from 10 to 1000 mg/L, and high when the value is greater than 1000 mg/L. S_w is strongly affected by polarity. Polarity refers to the uneven distribution of charge in a molecule. Generally, the affinity of a chemical compound with water increases with increasing polarity. The more hydrophilic (more polar) groups have a higher affinity with water (higher S_w) [7].

S_w is one of the most important properties that affect the bioavailability for weed control and the environmental fate of herbicides, mainly through leaching transport. The higher the S_w of the herbicide, the higher its bioavailability in the soil solution, meaning lower sorption to soil colloids. On the other hand, under this condition, the herbicide can be more easily leached and contaminate groundwater. Therefore, it is important to mention that leaching is necessary only at the soil surface (0–20 cm), where the herbicide is absorbed by the weed seed bank. Hexazinone and imazethapyr are herbicides applied directly to the soil and have high S_w; however, pendimethalin is highly sorbed to the soil due to its low S_w.

According to Carbonari et al. [8], the S_w of a herbicide in water plays an important role in increasing or reducing its ability to reach the soil when applied on mulch cover. Silva and Monquero [9] also highlighted the importance of S_w and the octanol-water partition coefficient (K_ow) as the most important characteristics to define the behavior of a herbicide applied on mulch cover, where high K_ow values and low S_w values represent less herbicide transport to the soil, affecting the efficacy of herbicides in weed control in sugarcane crops. Knowledge of S_w is fundamental in recommending weed management in sugarcane (which has many registered herbicides), especially during dry and wet periods in ratoon and plant sugarcane.

2.3 Octanol-water partition coefficient (K_ow)

K_ow is a dimensionless value defined as the ratio between the concentration of a herbicide in the saturated n-octanol phase and its concentration in water at equilibrium at a temperature of 25°C [10]. This coefficient is typically expressed in logarithmic form, as log K_ow. This property is applied evaluative to predict the distribution between environmental compartments in equations for estimating bioaccumulation in animals and plants and predicting the toxic effects of a substance [11, 12]. Log K_ow represents a measure of the tendency of a chemical to shift from the aqueous phase to lipids, for example: (1) Positive values of log K_ow indicate the hydrophobic property of compounds; (2) higher values indicate greater hydrophobicity of compounds; and (3) when the log K_ow value is greater than 3, it indicates that the chemical is considered highly hydrophobic [13, 14]. The environmental significance of this parameter is that it can be used to predict the accumulation, mobility, and persistence of herbicides in the soil. Herbicides with high K_ow values prefer organic (nonpolar) environments over aqueous environments (biota, soil, sediments, etc.).

The log K_ow values of pendimethalin and glyphosate are 5.4 and − 3.02, respectively, indicating that pendimethalin may exhibit a strong tendency for sorption in the dead cover and bioaccumulation in the environment, while glyphosate is highly hydrophilic and does not bioaccumulate [3]. Herbicides with log K_ow values below 2.7 (e.g., paraquat, mesotrione, and imazapyr), between 2.7 and 3 (e.g., atrazine), and above 3 (e.g., alachlor) have low, moderate, and high bioaccumulation, respectively.

Regarding the behavior of herbicides in plants, the interaction between pK_a/pK_b and log K_ow governs the distribution of products applied in post-emergence. For example, when a herbicide has intermediate membrane permeability, it will have some mobility in the phloem. Membrane permeability is estimated by log K_ow values. Herbicides with log K_ow values between −1 and 1 are expected to have phloem mobility after foliar applications. However, there are other acids, such as aryloxyphenoxypropionates (FOPs) (inhibitors of acetyl-coenzyme-A carboxylase—ACCase), that have log K_ow values between 3 and 4.5, which are more lipophilic and therefore have limited mobility in the phloem. Generally, compounds with high polarity (log K_ow < 0) and strong ionization (pK_a < 2), such as glyphosate, are mobile through the phloem, although significant amounts move through the xylem. Translocation solely via the xylem occurs for herbicides that inhibit carotenoids and Photosystem II (PSII). When herbicides have high log K_ow values, they are considered contact (non-systemic) herbicides, such as diquat (inhibitor of Photosystem I—PSI) and protoporphyrinogen oxidase (PPO) enzyme inhibitors [15].

2.4 Vapor pressure (VP)

The VP of a herbicide is a measure of its tendency to transform into vapor (gaseous state) and evaporate from its solid or liquid state. It indicates the volatility of the chemicals and their affinity with the air compartment [16]. Typically, the VP of the pure chemical is given in millimeters of mercury (mmHg) at 25°C. To convert to millipascal (mPa), the given value must be divided by 7.52×10⁻⁶. VP is the primary parameter governing the behavior of herbicide vapor and its potential volatility, without considering the influence of environmental conditions. The herbicide active ingredients, formulation type, ambient temperature, and humidity can influence volatility [17].

The VP of different herbicide families, and even within compounds of the same family, can differ by orders of magnitude [18]. Herbicides can be divided into three categories of potential volatility based on VP values (mPa at 25°C): low volatility with VP below 5.0, moderately volatile with VP from 5.0 to 10, and highly volatile with VP above 10 [19].

Auxin herbicides mimic (such as 2,4-D and dicamba) and carotenoid inhibitors (such as clomazone) are commonly volatilized in the field as they have high VP [3]. Techniques such as nanoparticle encapsulation, formulation types, and soil incorporation of herbicides contribute to minimizing losses through volatilization. Therefore, it is important to mention that herbicide volatilization can also be measured by the Henry’s law constant (K_H).

2.5 Sorption coefficient (K_d)

K_d represents the ratio between the concentration of sorbed herbicide (C_s) in the soil and its concentration in the equilibrium solution (C_e) [20]. This concept, according to Weber et al. [21], involves measuring the amount of herbicide sorbed from a specific concentration per unit mass of soil. The higher the K_d of a herbicide, the greater its sorption capacity in the soil. In light of the above, it is important to highlight that K_d values are usually determined at herbicide concentrations that would occur when compounds are applied at recommended field rates followed by sufficient precipitation to bring the soil to field capacity [22]. As a result, herbicide leaching through the soil profile is inversely correlated with K_d [3]. For example, the K_d values of pendimethalin, atrazine, and metribuzin are 228 (non-mobile), 3.2 (moderately mobile), and 0.874 L Kg⁻¹ (mobile) [19]. These K_d values are used in mathematical models (e.g., Linear, Freundlich, and Langmuir) to predict the mobility of herbicides in soils. Typically, sorption coefficients are normalized by the organic carbon content in the soil (K_oc). In practical terms, it is important to understand the percentage of herbicide sorbed, which, by subtracting the initially applied amount, determines the bioavailable quantity of the product in the soil solution, affecting weed control.

2.6 Degradation/dissipation half-life time (DT₅₀)

Overall, DT₅₀ (Dissipation or Degradation Time) or t_1/2 can be understood as the time required for the concentration of the product to decrease to half of its initial value [20, 23], and it can be considered a physicochemical property but is directly influenced by biological factors in the soil. With the same meaning but under controlled laboratory conditions, DT₅₀ = ln(2)/k (the constant rate per day, k) indicates the time required to reduce the concentration by 50% from any concentration point during the incubation period [23]. Consequently, DT₉₀ = ln(10)/k indicates the time required to reduce the initial concentration by 90%, nearly complete degradation of the herbicide. Thus, it is important to express the rate of decline as a first-order degradation (C_t = C₀*e^–kt), where C_t, C₀, and t represent the total concentration, initial concentration, and time, respectively. The degradation process mainly occurs through microorganisms, although chemical degradation and photodegradation also influence the process. Furthermore, the dissipation half-life time (DT₅₀, described in the same way as degradation) can be measured under field conditions, evaluating not only degradation but also off-target transport [3]. The DT₅₀ of sulfentrazone is 541 days, which is considered highly persistent, the DT₅₀ of diclosulam is 49 days, considered moderately persistent, and the DT50 of glyphosate is 15 days (non-persistent) [19].

The values of herbicide DT50 are analyzed using analytical techniques with the use of solvents for extraction, which extract bound residues and the sorbed portion (non-bioavailable in the soil solution for plant absorption). Therefore, DT₅₀ sometimes does not directly reflect the residual effect of herbicides in the soil. For example, diquat has a DT₅₀ of 5500 days in the field, classified as highly persistent [19], yet there are no reports of carryover in crops sown a few days after the application of this herbicide. On the other hand, tembotrione has a DT₅₀ ranging from 4.2 to 87.2 days, classified as non-persistent [19]; however, there are reports of carryover in potato crops planted 5 months after application in Rio Paranaíba, MG, Brazil [24]. The observed symptoms were tuber cracking at harvest time, indicating that the quantity required to cause negative effects on succeeding crops is extremely low.

2.7 Residual lifetime (RL₅₀)

The bioavailability of the herbicide in the soil is determined by the residual lifetime of the herbicide, which is defined as RL₅₀. RL₅₀ is defined as the residue level at which the active substance of the herbicide disappears in the plant [25]. RL₅₀ is estimated through bioassays, where injuries caused by the herbicide are observed in sensitive plants, indicating the herbicide’s activity. Therefore, RL₅₀ of herbicides is different from DT₅₀ (often confused by researchers), and in most cases, RL₅₀ < DT₅₀. RL₅₀ is used to indicate that 50% of the molecule has dissipated while the other half still has an effect on the plant. RL₅₀ depends on environmental factors, cropping system, herbicide properties, and the sensitivity of the species used as a bioindicator plant [25]. For example, RL₅₀ values for oxyfluorfen ranged from 51 to 59 days, and for linuron from 75 to 149 days in three tropical soils, using sorghum and cucumber as bioindicator species, respectively [26].

3. Herbicide behavior in the environment

The use of herbicides is essential for the current agricultural model, and without the use of this practice, ensuring food security in worldwide would likely be unfeasible. However, these products can have various environmental and social impacts, making it crucial that they are adequately used to preserve the natural resources that sustain the production.

Studies on herbicide behavior in the environment provide a better understanding of their dynamics and fate, primarily concerning retention processes (sorption, desorption, and precipitation; formation of bound residue and remobilization), transport processes (plant absorption, translocation, and metabolism; drift, volatilization, runoff, runin, and leaching), and transformation processes (chemical degradation, biodegradation, photodegradation, and mineralization), which typically interact with each other.

The fate and behavior of a herbicide in the environment are mainly related to the physical-chemical properties of the molecule, the matrix, and the environmental conditions. It is also important to note that when a pre-emergence applied herbicide remains active in the soil solution for weed control, it exhibits a positive residual effect. However, when it affects the crop in rotation/succession, it is known as carryover, a negative residual effect, as previously described.

3.1 Retention process

Typically, herbicides used in crops end up in the soil as their fate. Therefore, retention can be understood as a general process of sorption of herbicides in the soil, preventing the molecules from moving both inward and outward from the soil matrix. In practice, retention is one of the key factors that determine the efficacy of herbicides when applied for weed control, as it allows the prediction of the movement and degradation rate of the molecules applied to the soil.

Thus, it can be said that when a molecule is not retained in the soil colloids, it is exposed to other processes. Retention studies may be associated with the search for new methods of herbicide use, aiming for cost-effectiveness and reduced environmental risk. One possibility is the use of differentiated herbicide doses based on soil properties, similar to the usual practice of precision agriculture with fertilizer quantities.

Retention is the phenomenon by which a molecule is captured on the surfaces of mineral and organic colloids in soils. The term “sorption” is widely used because this capture occurs through adsorption, absorption, precipitation, or hydrophobic partitioning, and often the mechanism is not recognized [27]. A portion of the sorbed solute can return to the soil solution, a process known as desorption [28].

The main soil attributes that influence the sorption and desorption processes of herbicides are soil clay and organic matter (OM) content, cation exchange capacity (CEC), and pH. The physicochemical properties of the molecule (molecular structure, size, shape, solubility, speciation—presence of anionic form at normal pH, hydrophobicity, among others), as well as edaphoclimatic and management characteristics of the area, control this process [28].

One way to observe the sorptive behavior of herbicides is through the construction of sorption-desorption isotherms and the study of the kinetics of these processes in the soil. An isotherm is a curve that describes the retention of a solute in a sorbent at increasing concentrations [29]. In other words, it is the relationship between the concentration of the solute retained on C_s and the concentration of the solute in the C_e. To construct an isotherm, the system must be in thermodynamic equilibrium, so the kinetics of the reactions should always be quantified, and all physicochemical parameters (e.g., temperature) must be constant and specified. The estimation of herbicide sorption is represented by the partition or sorption coefficient (K_d), which can be estimated by the following relationship: K_d = C_s/C_e. Therefore, it is important to highlight that the value of K_d is appropriate to describe the sorption of a herbicide when linearity exists between C_s and C_e in an isotherm. Currently, the mobility potential of a herbicide is given by the logarithm of the sorption coefficient normalized to the organic carbon (OC) content of the soil (K_oc = K_d/OC*100). When K_oc is: < 1.0 (highly mobile), between 1 and 2 (mobile); 2 and 3 (moderately mobile); 3 and 4 (slightly mobile); 4 and 5 (poorly mobile); and > 5.0 L Kg⁻¹ (immobile) [30].

Figure 1A illustrates the number of publications on herbicide retention in the soil from 1950 to 2023, with a current total of approximately 9341 published papers in this field. A consistent upward trend is observed from 1950 until the present day. However, up until 1990, the overall number of publications remained relatively low, with only 380 papers published. Subsequently, starting in 1995, there was a substantial increase in the quantity of publications, with rates surpassing 1000 papers every 5 years between 2005 and 2015. Furthermore, from 2016 to 2020, this rate further increased to 2146 publications. In the most recent three-year period (2021 to 2023), a total of 1676 papers were published, indicating an even more pronounced growth trend in the field of herbicide retention.

Figure 1.

Papers on herbicide retention in the soil published each 5 years (gray columns) and total accumulated (black line) between 1950 and 2023 (A) and indexed in different subject areas (B). Source: Data were obtained from CAPES [31].

Figure 1B illustrates the distribution of herbicide retention papers published until 2023 across different subject areas. The top 10 areas with the highest number of publications were pre-filtered and analyzed. It was observed that broader subject areas such as Science & Technology and Life Sciences & Biomedicine had the highest number of indexed studies on herbicide retention, with 7909 and 5838 publications, respectively. Agricultural sciences also played a significant role, with Herbicides accounting for 4636 publications and Agriculture with 1976. The fields of chemistry and physics were represented by Physical Sciences (3487), Adsorption (3382), and Chemistry (3031). Environmental sciences were also prominent, as evidenced by Environmental Sciences (2965) and Environmental Sciences & Ecology (2910). It is important to note that the subject area of Pollution had over 1000 publications, indicating alternative approaches to retention studies, although not depicted in Figure 1B.

In herbicide retention studies, the term “adsorption” was the most commonly mentioned, appearing 6146-fold in abstracts and keywords. Adsorption is a term frequently used to describe the dynamics of both sorption and desorption processes of herbicides, although it is sometimes incorrectly used to solely refer to the sorption process. The terms “sorption” and “desorption” were mentioned 3577- and 2266-fold, respectively. The term “sorption-desorption,” which typically encompasses studies evaluating both behaviors, was less frequently used. The broader term “retention,” which encompasses these studies more comprehensively, appeared less frequently, with 2171 occurrences.

Some practical studies on herbicide retention in the soil are described below. Using information on C_e and bioavailability (%) after desorption, Lima et al. [32] generated dose maps for indaziflam and metribuzin for the entire study area (17.51 ha) focused on precision agriculture. Indaziflam doses ranged from 4.17 to 6.97 g a.i. ha⁻¹ for Amaranthus hybridus, corresponding to a variation of 67.14% between the lowest and highest applied doses. Eleusine indica required indaziflam doses ranging from 4.24 to 7.08 g a.i. ha⁻¹, representing a variation of 66.98% between the lowest and highest applied doses. The authors also found that metribuzin doses ranging from 57.1 to 66.6 g a.i. ha⁻¹ were sufficient for the control of A. hybridus, corresponding to a variation of 16.63% between the lowest and highest applied doses. The species E. indica required metribuzin doses ranging from 94.3 to 110.1 g a.i. ha⁻¹, representing a variation of 16.75% between the lowest and highest applied doses.

The recommendations of different indaziflam doses for control A. hybridus and E. indica resulted in reductions of 17.56 and 15.70%, respectively, in the total amount of herbicide that would be applied compared to the highest recommended [32]. The reduction in metribuzin doses was equivalent to 9.8 and 9.9% for A. hybridus and E. indica, respectively. The greater variation in indaziflam doses compared to metribuzin is due to the higher sorption capacity exhibited by indaziflam compared to metribuzin. Thus, the spatial distribution of the soil’s physical-chemical properties had a greater influence on the bioavailability after desorption of indaziflam and, consequently, on the recommended doses for weed control.

In a study on the spatial distribution of sorption-desorption of hexazinone and tebuthiuron, Mendes et al. [33] developed maps of herbicide bioavailability in the soil solution. However, there are no weed control efficiency tests conducted. Despite this, the authors emphasized that the application recommendation of pre-emergence herbicides, such as tebuthiuron and hexazinone, considering the soil physical-chemical properties, is an alternative to increase weed control efficiency and reduce the risk of environmental contamination.

Similarly, in a study that considered the soil sorption capacity for recommending cyanazine doses for weed control, Mohammadzamani et al. [34] found that the same control efficiency was achieved with a 13% reduction in the total applied dose. However, the authors did not differentiate the weed community by species. Thus, the aspects discussed here regarding the bioavailability after herbicide desorption and weed control efficiency studies provide a broader understanding of the variables involved in accurate dose recommendations in precision weed management [32].

On the other hand, any addition of organic compounds to the soil can directly affect the behavior of residual herbicides and consequently the weed control efficacy [35, 36]. For instance, the application of biochar to the soil is carried out for carbon sequestration, waste management, increased CEC, soil pH elevation, water retention, and nutrient source, which typically benefits agricultural production [37]. However, soil amended with biochar has a high sorption capacity for herbicides, meaning it decreases the bioavailability of the product in the soil solution [38]. The negative effects of biochars on weed control are more noticeable when it comes to pre-emergence herbicides [35].

In a study conducted by Mielke et al. [39], application rates <1% (w/w) of sugarcane straw biochar sorbed ~23% and desorbed ~15% of metribuzin, similar to unamended soil, for all pyrolysis temperatures. However, soil amended with 10% pyrolyzed biochar at temperatures of 350, 550, and 750°C sorbed 63.8, 75.5, and 89.4%, and desorbed 8.3, 5.8, and 3.7% of metribuzin, respectively. Thus, the authors reported that high pyrolysis temperatures and biochar application rates showed the capacity to immobilize metribuzin and improve soil fertility, which can influence weed control efficacy.

In American soils, Mendes et al. [40] found that amendment with biochar from soybean stover, sugarcane bagasse, and wood chips increased alachlor sorption between 4- and 33-fold compared to unamended soil. Biochar from soybean stover, sugarcane bagasse, and woodchips increased indaziflam sorption in the soil by 7, 55, and 69%, respectively [41]. Based on these results, it is expected that biochar additions will affect the bioavailability of herbicides for transport and degradation in the soil.

When evaluating the addition of bone char in a tropical soil, Mendes et al. [42] found that only 1.4 t ha⁻¹ of the bone char decreased the control of eight weed species by 50%. Wood chips biochar applied at 0.5 kg m⁻² decreased weed control by 75 and 60% with atrazine and pendimethalin, respectively [43]. The addition of wheat straw biochar decreased the control of Lolium rigidum with trifluralin and atrazine, requiring an increase in herbicide doses by 3–4-fold [44]. A similar phenomenon was observed for the control of Echinochloa crus-galli with diuron, requiring twice the dose to be applied [45].

The application of diuron in agricultural areas with anthropogenic soils in the Amazon region (Terra Preta de Índio) may result in inefficient weed control, as these soils can decrease the bioavailability of the herbicide in the soil solution due to high levels of OC, which confer high sorption and low desorption of the herbicide, as well as faster degradation compared to sandy soil [46].

3.2 Transport process

The transport of herbicides in the soil has a significant influence on their performance in the field and their potential for contaminating water resources. Therefore, in addition to agronomic efficiency aspects, field studies aiming to understand the movement of herbicides in the soil are essential to predict the contamination potential of these molecules. Upon reaching the soil, the main pathways for herbicide transport are runoff/runin, volatilization, and leaching.

Papers on the transport of herbicides in the environment have seen a significant amount of research, with approximately 10,200 publications identified between 1950 and 2023 (Figure 2A). Prior to 1995, the frequency of publications was relatively low. However, in the 2000, there was a notable increase, surpassing 1000 publications every 5 years. By 2010, this rate had already increased to 1500 publications, and by 2020, it approached 2000 publications. From 2021 to 2023, an additional 1223 publications were recorded, demonstrating the sustained interest of researchers in investigating the transport of herbicides in the environment.

Figure 2.

Papers on the transport of herbicides in environment published each 5 years (gray columns) and total accumulated (black line) between 1950 and 2023 (A) and indexed in different subject areas (B). Source: Data were obtained from CAPES [31].

Studies on the transport of herbicides in the environment encompass a wide range of subject areas, including technology, agricultural, and environmental sciences, as well as environmental, biological, and pollutants sciences (Figure 2B). The diverse approaches taken by researchers contribute to the distribution across these disciplines. For instance, investigations focusing on the movement of herbicides within the 10–20 cm soil layers for weed control purposes would likely fall under the domain of agricultural sciences. Conversely, studies assessing the risk of leaching and groundwater contamination by the same herbicide would be categorized under environmental sciences or pollutants.

Similar to studies on herbicide retention, the subject areas of Science & Technology and Life Sciences & Biomedicine show the highest number of publications, with 8587 and 7586, respectively (Figure 2B). Including these broader subject areas likely serves as a strategy to broaden the readership and impact of the publications. The subject areas of Herbicides, Agriculture, and Pesticides are also prevalent, with 5456, 3094, and 2335 papers, respectively. Furthermore, Environmental Sciences and Environmental Sciences & Ecology are represented with 3384 and 3371 papers, respectively.

When examining the distribution of studies on herbicide transport in the environment, it is evident that the term “environmental transport” is the most frequently encountered in abstracts and keywords, appearing 5708-fold. This is understandable as it is a broad term that enhances the visibility of the paper, although it does not precisely indicate the specific evaluation conducted in the study. Other terms such as “leaching” and “runoff” are also common, occurring 2385- and 1724-fold, respectively, while “drift” and “volatilization” are mentioned 1209- and 553-fold, respectively. This indicates that studies focusing on herbicide interactions with the soil are more prevalent than those examining interactions with the atmosphere. This difference can be attributed to the complexity and multitude of factors involved in quantifying drift and volatilization compared to leaching and runoff assessments.

3.2.1 Runoff/Runin

The movement of herbicides on the soil surface, from treated areas to untreated areas, after heavy rainfall is referred to as runoff or runin when it occurs subsurface (more challenging to measure). The removed herbicide can be present in the soil solution or sorbed to soil particles. Herbicide runoff is governed by a complex interaction between herbicide properties, soil properties, climatic factors, and specific environmental conditions at the site. The main soil and site properties that affect herbicide concentration in water runoff are as follows: slope, soil texture, soil structural stability, soil moisture content, surface vegetation cover, irrigation, precipitation characteristics, topography, hydrology, and geological characteristics. Therefore, soil management practices, as well as the amount of crop residues, also influence water, soil, and herbicide loss.

An important aspect in reducing herbicide losses via runoff is the conservation of soil and water resources, which allows for better soil structure and increased permeability. Among the key management practices is the adoption of no-till farming, maintenance of vegetative cover, irrigation control, and the use of buffer zones. For example, Vaz et al. [47] reported that sugarcane straw on the soil surface reduced water runoff, sediment, and diuron losses but had little effect on hexazinone losses. Crop residues cannot prevent the runoff of highly soluble molecules such as hexazinone. The authors stated that maintaining 7 t ha⁻¹ of straw on the soil surface was sufficient to mitigate water runoff, sediment, and diuron losses.

Regarding herbicide properties, the type of molecule, sorption capacity, persistence, and water solubility are the most important properties that affect surface runoff. Highly soluble herbicides have a tendency to dissolve in water and are more prone to be carried by runoff. Herbicides with high K_d values, such as glyphosate and its metabolite aminomethylphosphonic acid (AMPA), are transported in runoff by sorption to soil particles [48]. Surface runoff of the herbicide will increase when rainfall or irrigation brings water to the surface more rapidly than it can infiltrate into the soil [49]. However, it is important to check weather conditions before applying herbicides to reduce surface water pollution.

3.2.2 Volatilization and drift

Volatilization is the process by which herbicide molecules transition from a liquid state to vapor form, potentially leading to their loss to the atmosphere. In the atmosphere, herbicides can be transported over long distances, reaching undesired locations. This process can decrease the effective time of the herbicide in the applied area. The main factors that affect volatilization are the properties of the herbicide, soil, and environment. In herbicides belonging to the chemical group of thiocarbamates (such as molinate and thiobencarb) and trifluralin, this process can be so intense that slight incorporation after application reduces vapor losses and increases their efficiency.

However, Carbonari et al. [50] found that the combination of dicamba diglycolamine (DGA) salt with potassium salt of glyphosate and a volatility reducer was the mixture with the lowest volatility and the most suitable combination to recommend to farmers. The volatility reducer was effective in reducing dicamba volatilization alone and DGA in combination with all glyphosate salts (potassium, ammonium, and diammonium salt).

Herbicide drift through sprays can be defined as the movement of the molecule from the target area to non-target areas where the application was not intended. This transport occurs through the movement of spray droplets or vapors, which can cause injury or residue levels in neighboring susceptible plants, reducing productivity and affecting the morphology of these crops [51, 52].

Drift of herbicides, especially synthetic auxins, has gained prominence due to the release of genetically modified maize, soybean, and cotton cultivars resistant to 2,4-D and dicamba, resulting in an increased use of these products. Drift of 2,4-D and dicamba can result in quantitative and qualitative losses in the production of sensitive crops planted in areas adjacent to transgenic events that utilize these herbicides. For example, Brochado et al. [53] reported that “ponkan” tangerine seedlings exhibited injury symptoms after exposure to simulated drift (1/16 of the recommended dose) of dicamba and 2,4-D.

There are two types of drift, namely endodrift and exodrift. Endodrift occurs when there are losses of the product within the cultivation area itself, such as herbicide runoff from leaves to the soil, mainly due to the use of excessively large droplets or excessive spray volume. Exodrift is the displacement of the herbicide molecule outside the treated crop area, either in particle or in vapor form. Vapor drift refers to the movement of the herbicide after it has converted into a gaseous form, while particle drift involves the movement of droplets outside the application area [54, 55].

3.2.3 Leaching

Leaching is described as the downward movement of herbicides within the soil matrix or with soil water, and its intensity is determined by their physicochemical properties as well as soil and climate properties.

In practical terms, the movement of herbicides in the soil has a significant influence on their performance in the field. Limited leaching is desirable as it can make the herbicide more efficient by moving it from the soil surface to where weed seeds are concentrated. However, excessive leaching can contribute to the herbicide being carried into the groundwater, leading to undesirable contamination.

According to Oliveira and Regitano [56], the two most important properties related to the leaching process are K_d and DT₅₀. Sorption determines the bioavailability of a herbicide in the soil solution, while the DT₅₀ reflects persistence in the soil, and both regulate the leaching potential. In this regard, several models have been developed to classify the leaching potential of herbicides. Regional-scale leaching indices are useful in decision-making processes or remediation of contaminated environments, as mathematical models based on numerical simulations typically require a wide range of parameters that can be challenging to obtain [57].

The Groundwater Ubiquity Score (GUS), proposed by Gustafson [58], is an empirical index that classifies compounds based on their leaching tendency: GUS = log DT₅₀ (4 – log K_oc), where GUS represents a dimensionless index, DT₅₀ represents the degradation half-life time of herbicide in the soil (days), and K_oc represents the sorption coefficient normalized for organic carbon content (L Kg⁻¹). Herbicides with GUS < 1.8 are considered non-leachable, while values above 2.8 indicate leachability. Those with values between 1.8 and 2.8 are considered intermediate.

The criteria adopted by the California Department of Food and Agriculture (CDFA), proposed by Widerson and Kim [59], establish that herbicides with K_oc values less than 512 L Kg⁻¹ and DT₅₀ greater than 11 days are classified as leachable products. On the other hand, Cohen et al. [60] established that herbicides with K_oc values below 300 L kg⁻¹ and DT₅₀ values above 21 days are considered leachable, whereas those with K_oc values above 500 L Kg⁻¹ and DT₅₀ values below 14 days are classified as non-leachable products.

The criteria of the Environmental Protection Agency (EPA) involve the following values: S_w at 25°C > 30 mg L⁻¹; K_oc < 300–500 L Kg⁻¹; K_H constant <10⁻² Pa m³ mol⁻¹; speciation (the presence of anionic form at normal pH, between 5.0 and 8.0); DT₅₀ in soil >21 days and in water >175 days; field conditions favoring soil percolation, such as annual rainfall index >250 mm; unconfined aquifer and porous soil. Herbicides that meet these characteristics are considered potentially groundwater polluting agents.

In the calculation of three theoretical criteria (GUS, CDFA, and Cohen) based on the physicochemical properties of herbicides, Inoue et al. [61] assessed the potential leaching risk of herbicides in the state of Paraná, Brazil. The authors found that the classification regarding leaching potential demonstrated that acifluorfen-sodium, alachlor, atrazine, chlorimuron-ethyl, fomesafen, hexazinone, imazamox, imazapyr, imazaquin, imazethapyr, metolachlor, metribuzin, metsulfuron-methyl, nicosulfuron, picloram, sulfentrazone, and tebuthiuron are potentially leachable.

Computational simulation models are valuable tools for evaluating the behavior of herbicides. However, it should be considered that these models simplify the product’s behavior in the environment or in specific environmental compartments, and thus, the results should be interpreted considering the simplifications of the model [57].

When selecting a simulation model, several criteria should be considered, according to Cohen et al. [62]. These criteria include validation and calibration with experimental data, suitability of the model for the specific study, availability of the model and user support, availability of input data, and ease of use.

The quantification of leached herbicides in the soil profile is highlighted by analytical techniques such as chromatography and the use of radiolabeled molecules [57]. In terms of qualitative methodologies, there is the bioassay technique, which allows studying the movement of the herbicide through the soil profile using plants sensitive to the tested products [63]. For example, Pereira et al. [64] studied the leaching of clomazone in soil samples collected from different regions of Brazil and used sorghum for herbicide detection. Greater leaching occurred in soils with lower pH and lower OM content. Other studies using clomazone in different soils also reported an increase in herbicide sorption with soil organic matter, reducing leaching [65]. Regarding herbicide mixtures, Refatti et al. [66] determined the leaching potential of herbicides used in the Clearfield® system for irrigated rice and concluded that mixtures of imazethapyr + imazapic, imazapyr + imazapic, and imazethapyr leached into the soil, reaching depths of up to 25 cm in floodplain soil. However, in a study conducted by Mendes et al. [67], the application of mesotrione, alone or in a mixture with S-metolachlor and terbuthylazine, did not influence the leaching of this herbicide.

On the other hand, adjuvants such as mineral oils are widely used in herbicide application to reduce drift and droplet evaporation, as well as to enhance herbicide absorption by the plant. Thus, the addition of mineral oil (1 and 2% v/v) at the time of pre-emergence atrazine application did not interfere with the transport of this herbicide in the arable soil profile via leaching; therefore, the adjuvant may have a positive effect only on the herbicide-plant relationship [68].

The presence of mulch on the soil brings about significant changes in weed occurrence and also affects the behavior of herbicides applied in these production systems [8]. The mulch acts as a barrier to the herbicide’s arrival in the soil, as it is intercepted and can be sorbed or degraded within the mulch itself. Even after the herbicide reaches the soil, the mulch influences its dynamics by altering water dynamics in the soil, OM content, and microbial activity.

In a study conducted by Araldi et al. [69], two different transposition behaviors of herbicides in sugarcane straw were observed. One behavior was related to metribuzin and hexazinone, which quickly passed through the mulch layer (with ~20 mm of rainfall), while another behavior was observed with atrazine, clomazone, and diuron, which required approximately 60 mm of rainfall to be transported from the plant residue, based on the maximum amount of herbicides removed from the straw.

With the addition of vinasse, Takeshita et al. [70] found that 71% of aminocyclopyrachlor reached the leachate (>30 cm depth), while <50% reached the leachate in the other treatments (filter cake and sugarcane straw). The bioavailability of aminocyclopyrachlor was not reduced with the addition of organic material to the soil, which may favor weed control. However, the presence of vinasse increases the risk of herbicide leaching to deeper soil layers than the weed seed bank.

In the presence of sugarcane straw, Silva et al. [71] reported that aminocyclopyrachlor was distributed in the soil profile, mainly from 0 to 25 cm depth, and that 40% more herbicide was retained in the straw with 20 t ha⁻¹. Thus, the authors concluded that the presence of high amounts of straw retains aminocyclopyrachlor but does not prevent it from reaching the soil; the high S_w of the herbicide and intense precipitation (200 mm) over a short period of time (48 h) are factors that contribute to the herbicide passing through this barrier.

The addition of carbonaceous materials to the soil can also affect herbicide leaching. For example, Mendes et al. [72] showed that soils amended with animal bone char, regardless of particle size (0.3–0.6 and 0.15–0.3 mm), increased the sorption of aminocyclopyrachlor and mesotrione, decreased desorption, and consequently decreased the leaching of both herbicides in the soil. For remediation purposes, this is a great strategy to immobilize herbicides in the upper soil layer, but in agronomic terms, it can directly affect the effectiveness of weed seed bank control.

3.3 Transformation process

The transformation of herbicides can begin with their dilution in water in the application tank and refers to changes in the chemical nature of the molecule through physical processes (photodegradation), chemical processes (oxidation, reduction, hydrolysis, formation of water-insoluble salts, and chemical complexes), or biological processes (biological degradation or biodegradation). In general, transformation is considered complete (mineralization) when it results in the formation of CO₂, H₂O, and minerals, and incomplete (metabolism) when it generates metabolites.

The herbicide must be in the soil solution or weakly sorbed to undergo chemical or biological degradation. However, it becomes unavailable for degradation by soil microorganisms or various chemical reactions when strongly sorbed by soil colloids.

The ideal herbicide is one that remains active in the environment for a sufficiently long time to control weeds in the crop where it was used but not so long that it causes injury to susceptible crops in rotation/succession. Therefore, understanding herbicide degradation in the soil is important, especially for herbicides that exhibit greater persistence in the soil and may result in a process called carryover, affecting sensitive crops grown in rotation.

Overall, lower temperatures, lower precipitation, and shorter crop cycles, especially in the case of soybeans with early-maturing cultivars to advance second-crop planting, are factors that increase the possibility of carryover. In the recent study by Inoue et al. [73], the carryover of various herbicides in the soil is addressed, considering various agricultural crops grown in rotation and succession. The main crops sensitive to herbicides include sunflower, soybean, common bean, coffee, sorghum, corn, millet, cotton, canola, among others.

Among the various types of herbicide behavior, transformation has received the most attention in the scientific literature from 1950 to 2023, with over 16,500 publications (Figure 3A). Initially, there were only a few studies until the 1970. However, exponential growth has been observed since the 1990, with over 1500 papers every 5 years between 2000 and 2015. The highest peak was reached in 2016–2020, with 3522 publications, and an additional 2515 publications between 2021 and 2023. If this trend continues, the publication rate could surpass 4000 every 5 years by 2025.

Figure 3.

Papers on the transformation of herbicides in environment published each 5 years (gray bars) and total accumulated (black line) between 1950 and 2023 (A) and indexed in different subject areas (B). Source: Data were obtained from CAPES [31].

Just like studies on retention and transport, the subject areas with the highest number of publications on herbicide transformation in the environment are Science & Technology and Life Science & Biomedicine (Figure 3B). Science & Technology has 13,764 papers, while Life Science & Biomedicine has 11,391 publications. Similar patterns can be observed in other subject areas such as Herbicides (7517) and Agriculture (3469), as well as Environmental Sciences (5284) and Environmental Sciences & Ecology (5248), in addition to Physical Sciences (4163) and Chemistry (3556). This wide range of subject areas highlights the various approaches that can be taken in studying herbicide transformation. For instance, examining the residual effects of a specific herbicide falls under agricultural sciences, while selecting phytoremediation plant species for herbicide-contaminated soils can be categorized under environmental or biological sciences.

Among the main approaches in studies on herbicide transformation in the environment, the term “degradation” is the most commonly mentioned in abstracts and keywords, with 12,685-fold, while “dissipation” is mentioned 2041-fold. Both terms are broad, but “degradation” is more suitable for evaluations conducted under controlled conditions (such as laboratory or greenhouse), while “dissipation” is more appropriate for less controlled conditions (such as field studies). However, it is worth noting that the term “degradation” is used in the majority of studies, including those conducted in field settings. The term “mineralization,” which is more specific in its scope, appears 2395-fold. Regarding specific types of transformations, “chemical transformation” is mentioned 7124-fold, “biodegradation” 4381-fold, and “photodegradation” 1225-fold.

3.3.1 Photodegradation

Photodegradation is the degradation of a molecule by solar radiation, especially ultraviolet (UV) light [74]. This occurs due to the absorption of UV light, which is the most destructive to the herbicide and causes the excitation of its electrons, breaking certain bonds in the molecules. The extent of photodegradation by sunlight is highly dependent on the UV sorption profiles of the herbicide, the environment, and the emission spectrum of sunlight [1], as the energy required to break the chemical bonds in herbicide molecules typically ranges from 250 to 400 nm [74]. Photolysis can be a direct process, in which a substance is transformed through the absorption of light energy, or an indirect process in which other substances absorb energy, transform, and then alter the primary substance [1]. On one hand, photodegradation can be beneficial by reducing excessive persistence of residues in the soil, but it can also be undesirable in terms of potentially reducing weed control efficacy when it occurs shortly after herbicide application. However, fast-acting herbicides that undergo photodegradation in the top 3-mm layer should be incorporated into the soil at the time of application, as photodegradation can diminish weed control effectiveness [52]. For example, trifluralin, atrazine, bentazon, clethodim, and diquat are herbicides that can rapidly undergo photodegradation, while phenylureas may only photodegrade when exposed to long periods of light. Gozzi et al. [75] demonstrated that ethoxysulfuron and chlorimuron-ethyl are also subject to photodegradation. Therefore, surface-applied herbicides are often degraded (lost), particularly if an extended period of drought follows their application.

3.3.2 Chemical degradation

The chemical degradation of herbicides in the soil occurs through chemical reactions and is common for various molecules. Oxidation, reduction, mineralization, hydrolysis, ester formation, photolysis, polymerization, and dehalogenation reactions are common in abiotic degradation. For example, the preferred degradation pathway of 2,4-D occurs through oxidative dealkylation and aromatic ring cleavage [76]. High temperatures and good moisture facilitate these chemical reactions. Additionally, extreme pH values can increase the hydrolysis of certain herbicides [77].

In this regard, the main transformation pathway of herbicides in the application tank is chemical degradation, especially through the process of hydrolysis. This process also initiates a series of degradative activities that occur in the soil and becomes indispensable in transformation processes. Hydrolysis is an important degradation pathway for sulfonylureas and chloroacetamides through acid-catalyzed cleavage and base-catalyzed contraction/rearrangement [78, 79], but not for imidazolinones [80]. A qualitative kinetic assessment was conducted by Lovell et al. [81] to characterize the sorption of isoxaflutole during its rapid hydrolysis into its bioactive metabolite, diketonitrile (DKN). The authors found that the desorption of isoxaflutole coupled with hydrolysis promotes the reactivation of the herbicide’s function after rainfall and contributes to the compound’s efficacy by replenishing the soil solution with its bioactive metabolite. This herbicide exhibits an unusual behavior compared to others, as it is functionally reactivated by rainfall events, providing weed control. Supporting this statement, Oliveira Jr. et al. [82] found that isoxaflutole exhibited high stability in clayey soil even after three simulated rains of 20 mm, spaced 30 days apart and followed by 120 days of dryness after its application.

3.3.3 Biological degradation

Biological degradation occurs through the action of microorganisms and is considered the main mechanism of herbicide disappearance in the soil. Herbicide degradation in the soil can be accelerated using processes that increase microbial activity, such as adding OM and fertilizers, managing soil moisture content, pH, and temperature, deep plowing, and utilizing adapted microorganisms (bioremediation). In this regard, understanding the influence of OM on various processes related to herbicide behavior in the soil allows for the adoption of more appropriate management practices regarding weed control and remediation of contaminated areas [83].

For example, the application of biochar in the soil is strongly correlated with herbicide degradation processes [84]. The authors stated that if the goal is to apply the herbicide in pre-emergence after adding biochar to the soil, caution should be taken as biochar can either decrease or increase the persistence of the chemical, affecting weed control efficacy over time. On the other hand, if the objective is herbicide remediation in contaminated soils, the interference of biochar in the herbicide’s bioavailability in the soil solution can increase the microbial diversity in the soil, which are the agents responsible for herbicide degradation.

When herbicides are bioavailable, they are subject to degradation by microbial or enzymatic action. For degradation in the soil, herbicides involve algae and actinomycetes, but bacteria and fungi are the most important. Biodegradation involves the use of herbicides as a source of N, C, and S. Herbicide degradation by biotic reactions is generally followed by oxidative processes such as beta-oxidation, C cleavage, C hydroxylation, N oxidation, N demethylation, C cleavage, C reduction, N reduction, hydrolysis, and mineralization. Herbicides, in most cases, lose their herbicidal activity after degradation. Only a few metabolites retain herbicidal activity (e.g., isoxaflutole).

Subsequent applications of the same herbicide, coupled with favorable environmental conditions for microbial development, can result in rapid degradation, leading to reduced product efficacy. For example, the dissipation of diuron was favored in orchard areas where this herbicide was applied repeatedly [85]. This fact was confirmed by the herbicide dissipation in areas with a history of application (12 years), and the authors found a DT₅₀ of 37 days. However, when comparing the results with those obtained from an area where the herbicide was applied for the first time, they observed a DT₅₀ of 81 days. The authors attributed this lower DT₅₀ of diuron to soil microorganism adaptation, thus favoring its biodegradation and consequently reducing the residual effect on weed control.

Soil microbiota can utilize herbicides in two ways: as a food (substrate) for their growth and, on the other hand, herbicides can also influence the microorganisms responsible for degradation [86]. Microorganisms exhibit five different strategies for herbicide metabolism, which are catabolism, cometabolism, polymerization or conjugation, accumulation, and side effects of microbial activity, as described below [86].

3.3.3.1 Catabolism

In catabolism, the herbicide serves as a source of energy and nutrients for the growth and development of degrading microorganisms. This process generally leads to complete degradation of the molecule (mineralization). For example, an increase in soil microbial activity with the application of glyphosate has been reported in the literature. It is known that many microorganisms use glyphosate as a source of P in its absence in the environment [87, 88]. However, in soils with oxidic mineralogy, Prata et al. [89] observed increased sorption with decreasing doses of P added to the soil. P competes with glyphosate for sorption sites, and therefore, nutrient deficiency enhances the sorption of the herbicide and consequently increases its degradation.

In a study conducted by Souza et al. [88] on the biodegradation of imazapyr in two soils with different textures and subjected to different herbicide doses, increased microbial respiration was observed in the presence of the herbicide, and they stated that the microbiota was able to utilize imazapyr as a source of C for its growth.

Atrazine can be used as a source of N by soil microbiota, and therefore, its degradation may be favored under nutrient-restricted conditions [90]. The same authors observed a reduction in atrazine degradation when mineral N was added to the soil.

3.3.3.2 Cometabolism

Cometabolism occurs when the herbicide is transformed through metabolic reactions but does not serve as a source of energy for the microorganism. Usually, complete transformation of the molecule does not occur, thus requiring a secondary biodegradable substrate as a source of C and energy [91]. The addition of OM to the soil can demonstrate the cometabolic transformation of herbicides in the soil. The fact that this addition increases microbial activity or biomass suggests that the transformation of many herbicides, such as diuron [92] and ametryn [93], occurred through cometabolic processes.

3.3.3.3 Polymerization or conjugation

Polymerization or conjugation is the process in which the original herbicide molecule or its metabolites combine with natural soil compounds, such as amino acids or carbohydrates, or with another herbicide molecule. The formation of the conjugate typically makes the molecule more polar and thus more hydrolysable. The formation of the metabolite dichloroaniline, characteristic of many urea-derived herbicides such as diuron, isoproturon, monuron, among others, can be considered a result of the polymerization of simpler metabolites; that is, two diuron molecules must be transformed to form a dichloroaniline molecule [94, 95].

3.3.3.4 Accumulation

Accumulation occurs when a herbicide molecule is incorporated into the microorganism without being transformed. This accumulation can occur through active or passive processes and raises significant concerns, as this microbial interference only represents the temporary removal of the molecule [95]. Percich and Lockwood [96] reported the accumulation of atrazine in actinomycetes.

3.3.3.5 Side effects

The side effects of microbial activity occur when the herbicide is transformed as a result of changes in pH, redox potential, among others, promoted by microbial activity. pH elevation, for example, can contribute to the hydroxylation of many molecules, making them more polar and thus susceptible to degradation [92].

In summary, according to Lavorenti et al. [95], biodegradation occurs due to the production of enzymes by herbicide-degrading microorganisms, which, when in contact with these molecules, inside or outside the microorganism cells, participate in a series of reactions such as oxidation, reduction, hydrolysis, dealkylation, decarboxylation, hydroxylation, methylation, dealkoxylation, among others.

Nakagawa and Andréa [97] observed that the mineralization of atrazine (about 15% was mineralized) only occurred under natural conditions (non-sterile soil), indicating that the action of microorganisms was crucial in breaking down the triazine ring of the molecule. In practice, the preferred degradation pathway of the herbicide is important because it indicates how agricultural practices will affect its degradation and persistence in the environment. When degradation occurs through catabolism, the deficiency of the element that the herbicide serves as a source can favor this process.

Lastly, it is important to highlight that bound residues of herbicides in the soil can return to the soil solution (remobilization) and become bioavailable again for plants and be mineralized by microorganisms, which can negatively affect subsequent crops or non-target organisms. Supporting this statement, Viti et al. [98] found that diuron, hexazinone, and metribuzin were remobilized from the soil with the addition of vinasse, filter cake, or sugarcane straw, due to the reactivation of soil microbial activity.

The degradation of herbicides is an important mechanism that tends to control the persistence, activity, and transport of the herbicide in the soil profile [86]. The rate of herbicide degradation in the soil occurs through a combination of processes (photolysis, chemical, and biological) and is dependent on the molecule’s structure, influenced by soil and climate factors, which vary from location to location and from year to year. Therefore, degradation is dependent on various environmental factors that can affect both the population density of microorganisms and their degradation potential.

4. Conclusions

The study of herbicide interactions in the environment has experienced significant growth across various subfields, particularly in the last three decades. It can be considered a multidisciplinary subject that encompasses areas such as agricultural, environmental, and biological sciences, as well as technology, physics, chemistry, and biomedicine. Overall, there are over 35,000 papers on herbicide behavior in the environment, and the trend indicates that the number of publications will continue to grow in the coming years.

Under field conditions, any amendment (physical, chemical, and/or biological) in the soil can directly affect weed control with the use of residual herbicides (applied pre-emergence). Therefore, detailed knowledge regarding the herbicide’s behavior in the environment allows for adjusting the product doses according to soil type and climatic conditions, which is an important aspect of precision agriculture. Additionally, it enables understanding the risks of carryover and drift caused by herbicides in crops, adjusting soil management to mitigate herbicide transport through runoff/runin, implementing remediation techniques for contaminated soils and the use of environmental indicators, employing simulation models to predict herbicide behavior and fate, and conducting ecological and environmental risk analysis for regulatory purposes.

Acknowledgments

The authors thank to National Council for Scientific and Technological Development (CNPq - 404240/2021-6) and Foundation for Research Support of the State of Minas Gerais (FAPEMIG - 2070.01.0004768/2021-2084).