Field-Scale Mobility and Persistence of Commercial and Starch-Encapsulated Atrazine and Alachlor T. G. Gish,* A. Shirmohammadi, and B. J. Wienhold ABSTRACT Recent laboratory studies have shownthat starch-encapsulation (SE) may reduce leachate losses of certain pesticides. This study compares field-scale mobility and per~stenee of SE-atrazine [2-chloro-4-ethylamino- 6-isopropylamino-s-triazine] and alachlor [2-¢hloro-N(2,6-diethyipbenyl)- N-(methoxymethyl)acetamide] to that of a commercialformulation (C10 of atrazine and alaehlor. The research site consisted of four (0.25 ha) fields. Two fields were under no-tillage management (NT) and two were under conventional tmage (CT). One field in each tillage system received SE-formulated atrazine and ulachlor, while the others received CF- atrazine and alachlor. Chemical movement and persistence was deter- mined by analysis of surface samples (--3 cm) taken immediately after application and 1.1-m soil cores collected seven times over 2 yr. No significant difference in herbicide residue levels was observed between NT and CT,but there was a herbicide formulation effect. Soil residue analysis suggests that SE-atrazine was more persistent and less mobile than CF-atrazine. Starch-encapsulated-alachlor was slightly moreper- sistent than CF-alachlor, but no differences in mobility between formu- lations was observed. The differentialfield behavior between SE- herbicides is attributed to thefaster release of alachlor from thestarch granules. Increased atrazine persistence was attributed to thereduction of leachate losses. The reduction in atrazine leaching is likely due to the slow release from the starch granules and subsequent diffusion intothe soil matrix where it is less subject to preferential flow processes. A~ HOUGH ATRAZINE and alachlor are degraded bio- ogically, leaching losses for both can be appreciable (Wehtje et al., 1984; Isensee et al., 1990; Gish et al., 1991a). Additionally, alachlor losses by volatilization can also be significant (Glotfelty et al., 1989). Volatilization losses during application for some formulations of alachlor maybe responsible for low recoveries immediately after application. Helling et al. (1988)foundthat alachlor recov- eries <1 d after application were <57%of that applied when surface broadcast as an emulsifiable concentrate. Starch-encapsulation (SE) is a controlled release tech- nology that may modify the behavior of pesticides in such a wayas to minimizeenvironmental impacts (Gish et al., 1991b; Mills et al., 1991 p. 203-209; Boydston, 1992; Wienhold et al., 1993). Schreiber et al. (1987) proposed that SE herbicides could reduce both volatilization and leaching losses. Gish et al. (1991b)evaluated the mobility of SE-atrazine during frequent high-intensity water inputs and found that after 16 pore volumes, >35% of the ap- plied technical grade atrazine had leached through small soil columns, as compared with <1 to 10% for the three SE formulations studied. In an ecosystem chamber experi- ment, Wienhold et al. (1993) observed that volatilization losses of CF-atrazine were four times greater than SE- atrazine. T.J. Gish and B.J. Weinhold, USDA-ARS Hydrology Lab., Natural Resources Inst., Beltsville, MD 20705; and A. Shirmohammadi, Agricultural Engineer- ing Dep., Univ. Maryland,College Park, MD 20705. Received8 Jan. 1993. *Corresponding author. Published in J. Environ. Qual. 23:355-359 (1994). 355 Rate of herbicide release from the starch granules ap- pears to be a critical factor influencing environmental fate and is governed mainly by a diffusion process (Schreiber and White, 1980). When starch granules are applied to the soil they imbibewater, swell, and the encapsulated com- pound diffuses out of the starch matrix. The degreeof gran- ule swelling and herbicide release is a function of granule characteristics, pesticide chemistry and environmental fac- tors. Environmental factors that strongly influence herbi- cide rate of release include soil water potential, tempera- ture and microbial activity. Undersimilar environmental conditions, alachlor rates of release from starch granules were much faster than those of atrazine (Wienhold and Gish, 1992). Differences in rates of release between atrazine and alachlor were attributed to differences in water solubility of atrazine (32 mg -l) and alachlor ( 240 mg L-l). Starch encapsulation has been suggested as a formula- tion for affecting the environmental fate of agricultural her- bicides. No field study has compared the environmental fate of SE herbicides to that of CF herbicides. The pur- pose of this study was to compare field-scale persistence and mobility of SE-atrazine and alachlor to that of CF- atrazine and alachlor. MATERIALS AND METHODS Field Site and Treatment The experimental site was on a Monmouth sandyloam (clayey, mixed, mesic TypicHapludult)with 2 to 5 % slope, located at the Central Maryland Research and Education Center, near Up- per Marlboro, MD. In 1988, the field site was chisel plowed and divided into four smallfields, each=0.25 ha (Fig. 1). Grass al- leyways and berms were established around each field. Each year, the fields were planted in corn (Zea mays L.) in late May and rye (Secale cereale L.) was planted in the fall as a winter cover crop during 1989-1992. Corn residue was left on all the fields during the winter. Rye was cut in the spring, 1 to 2 wkbefore any tillage practice. Corn and rye residues were soil-incorporated in the conventionally tilled field but remained on the surfaceof the no-tillage treatment.All tillage treatments were initiated in 1989 and consisted of two management practices. Tilled fields were chisel plowed each year in the spring, incorporating any plant residuewhile the onlysoil disturbance on the no-till fields occurred when a no-till corn drill was used for planting in the spring. Pesticide treatments,were surface applied as either (i) CFbroadcast-spray or (ii) granular-broadcastSEformulation Cl’able 1). Atrazine was applied at a rate of 1.7 kgha -t a.i. while alachlor was applied at the rate of 2.8 kg ha -I a.i. Treatments were randomly assigned to each field. Starch-encapsulated her- bicides were prepared usingthe extrusion process(Cart et al., 1991). Starch-encapsulated atrazine granules contained11.1% a.i. and SE-alachlor, 10.1% a.i. Starch granules 0.4 to 1.2 mm in diam. were used. Abbreviations: SE, starch-encapsulation; CF, commercial formulation; NT, non-tillage management; CT,conventional tillage; CV,coefficient of varia- tion; a.i., active ingredient.