Control of volunteer soybean plants in sunflower crop 1

Sunflower is an oil crop with important agronomic characteristics. Its main product is oil with excellent nutritional quality. Furthermore, the grain can be used for human consumption and animal feed, as well as raw material for the production of biodiesel. Also, sunflower is an option for succession or rotation crops in major grain regions of Brazil. However, there have been many questions about the presence of volunteer soybean that emerges in successive crops. This plant comes from seeds that fall on the ground by natural threshing of the pods (Bond & Walker 2009) or due to losses during crop harvest (Toledo et al. 2008). ABSTRACT RESUMO


INTRODUCTION
Sunflower is an oil crop with important agronomic characteristics.Its main product is oil with excellent nutritional quality.Furthermore, the grain can be used for human consumption and animal feed, as well as raw material for the production of biodiesel.Also, sunflower is an option for succession or rotation crops in major grain regions of Brazil.
However, there have been many questions about the presence of volunteer soybean that emerges in successive crops.This plant comes from seeds that fall on the ground by natural threshing of the pods (Bond & Walker 2009) or due to losses during crop harvest (Toledo et al. 2008).

ABSTRACT RESUMO
The control of this plant is required and regulated by law in several Brazilian States, based on the creation of a host-free period (Seixas & Godoy 2007).Except for the season traditionally used for soybean sowing, the action defines the period of the year when the presence of living soybean plants is not allowed on fields.This practice is considered one of the main strategies for controlling Asian soybean rust (Phakopsora pachyrhizi), preventing the survival and spread of the fungus.
Another issue is the emergence of transgenic soybean resistant to glyphosate, demanding considerable changes in the management of volunteer soybean, given that glyphosate is no longer an alternative way of control (Dan et al. 2009).This Sunflower (Helianthus annuus) sown offseason, after soybean crop (Glycine max), is affected by the competition imposed by volunteer plants.Two experiments were carried out to evaluate the control of volunteer soybean plants in sunflower crops.The sulfentrazone herbicide (75 g ha -1 , 100 g ha -1 and 250 g ha -1 ) causes phytotoxicity to sunflower immediately after application, however, plants recover, with no yield losses.These doses do not cause the total death of volunteer soybean plants, but temporarily paralyzes their growth, avoiding the competition with the sunflower crop.The glufosinate ammonium and ametryn herbicides are effective in controlling volunteer soybean plants, however, symptoms of phytotoxicity in the sunflower crop are high, reflecting in losses of dry weight biomass and crop yield.The other treatments do not provide satisfactory control of volunteer soybean plants and even reduce the sunflower dry weight biomass and yield.KEY-WORDS: Helianthus annuus; Glycine max; herbicides.

PALAVRAS-CHAVE:
Volunteer soybean control becomes even more complex due to the lack of effective and selective herbicides to control broadleaf weeds in sunflower (Santos et al. 2012).Although the most effective and widely used method is the chemical control, there are no herbicides registered for controlling volunteer soybean in sunflower crops.Alachlor, trifluralin, s-metolchlor and fluazifop-p-butyl are registered in Brazil to sunflower crops (Brasil 2014), however, none of them is efficient in controlling volunteer soybean plants.
This study aimed to evaluate the control of volunteer soybean plants in succession of soybeansunflower by using contact and systemic herbicides.
Both experiments were laid out in a randomized complete block design, with three and four replications for experiments 1 and 2, respectively.Experiment 1 was conducted in vases, with a single vase per plot, while experiment 2 was carried out in the field, with each plot consisting of five rows of 5 m long, with an area of 6 m 2 (1.5 m x 4.0 m).
Pots with 3 kg capacity were filled with substrate composed of a mixture of soil, manure and sand (1:1:1).Seeds of sunflower (BRS 323) and soybean (Vencedora BRSMG68) were sown in the same pots, maintained on benches exposed to full sun.The pots were watered regularly, keeping the soil moisture close to field capacity.The plants were thinned after emergence, in order to keep four plants of each species per pot.The average values of rainfall and air temperature during the experiment 1 are shown in Figure 1a.
Treatments were applied on July 7th, 2013, with a backpack sprayer pressurized with carbon dioxide and kept at constant pressure (196 kpa), in order to deliver a spray volume of 150 L ha -1 .The sprayer bar (0.5 m length) consisted of two flat-fan nozzles (Magnojet BD 110.02), spaced 0.5 m apart.The vases of each treatment were separated from the rest of the experiment for application of the  respective herbicide.Climatic conditions during the herbicide application were as it follows: temperature of 28 ºC, relative humidity of 67 % and wind speed of 2.5 m s -1 .
The phenological stages at the application time were V 2 (Castiglioni et al. 1997) and V 1 (Fehr et al. 1971), respectively for sunflower and volunteer soybean plants.Phytotoxicity percentages on sunflower plants and the percentage of control on volunteer soybean were evaluated at 10 and 20 days after herbicide applications (DAA).It was used a visual scale, with zero corresponding to no visual injury symptoms on sunflower or no soybean control and 100 % corresponding to plant death of sunflower and soybean (SBCPD 1995).
The dry matter production of sunflower and soybean plants were obtained at 21 DAA.The plants of both species were cut in each pot.Fresh matter of roots and shoots were placed in kraft paper bags and dried in a forced ventilation oven at 65 ºC, for 72 hours.The dry matter was measured with a graduated scale.
Sunflower was sown in an area where soybean had been harvested.Each plot consisted of five rows of 5 m long (10 m 2 ), with a net area of 6 m 2 (1.5 m x 4.0 m).The sunflower hybrid BRS 323 was sown in rows 0.5 m apart, with a plant stand of approximately 55,000 plants ha -1 .The fertilization at sowing time consisted of 400 kg ha -1 of NPK (08-20-18).Nitrogen (50 kg ha -1 ) and boron (1.2 kg ha -1 ) side dressing were performed at 25 days after sowing (DAS).The average values of rainfall and air temperature during the experiment 2 are shown in Figure 1b.
Treatments were applied on March 18th, 2014, using a backpack sprayer pressurized by compressed CO 2 (196 kpa) with a bar of 1.5 m length, equipped with four flat-fan nozzles (Magnojet 110.01 BD), spaced 0.5 m apart, and a spraying volume equivalent to 80 L ha -1 .
Environmental conditions during the pulverization were: wind speed of 3 m s -1 , temperature of 29 ºC and relative humidity of 65 %.The phenological stages of sunflower and soybean plants at the application time were respectively V 4 (Castiglioni et al. 1997) and V 3 (Fehr et al. 1971).
The phytotoxicity percentages on sunflower plants and the percentage of volunteer soybean control were evaluated at 7, 14 and 21 DAA, with zero corresponding to no visual injury symptoms on sunflower or no soybean control and 100 % corresponding to plant death of sunflower and soybean (SBCPD 1995).
Volunteer soybean plants were cut inside a square of 0.5 m x 0.5 m (0.25 m 2 ), at 120 days after sowing, and dried in a forced air ventilation oven at 65 ºC, during 72 hours, to constant mass.The number of achenes per plant, plant height and weight of one thousand achenes were determined at harvest time.Plant stand and sunflower yield were obtained in an area of 6 m 2 , with results turned into plants ha -1 and kg ha -1 , respectively.Data obtained for dry matter of volunteer soybean plants were subjected to transformation (log x), in order to complete the analysis of variance.
Data were submitted to Anova and means compared by the Scott-Knott test (p ≤ 0.05), for both experiments (Ribeiro Júnior 2001).

RESULTS AND DISCUSSION
The percentages of phytotoxicity on sunflower plants and control of volunteer soybean at 10 and 20 DAA, as well as dry weight biomass of sunflower and soybean plants, concerning the treatments for the experiment 1, conducted in Coronel Pacheco, are shown in Tables 1 and 2.
Glufosinate ammonium, ametryn and carfentrazone provided high phytotoxicity values in sunflower plants, while the other treatments caused initial phytotoxicity with subsequent recovery of plants (Table 1).
The best levels for controlling voluntary soybean plants were obtained with the application of glufosinate ammonium, ametryn and oxyfluorfen (Table 1).
Sulfentrazone (250 g ha -1 ) had no effect on sunflower dry weight matter and provided reduction on volunteer soybean biomass ( The selection of herbicides to the field experiment 2 was performed according to the results of the experiment 1.It was observed that contact herbicides, such as sulfentrazone, cause the death of sunflower leaves.However, the apical bud was not injured and there was plant recovery.Thereby, besides selecting the best treatments described in experiment 1, other contact herbicides (MSMA and saflufenacil) were added to the field study (experiment 2).Tembotrione and triclopyr were also applied and, even though they are not contact herbicides, they are effective in the Fabaceae weed control.In addition, tembotrione and saflufenacil are registered in Brazil to control volunteer soybean plants (Brasil 2015).
The phytotoxicity percentages on sunflower plants at 7, 14 and 21 DAA, for the experiment 2, are shown in Table 3.
Treatments with glufosinate ammonium caused the greatest injury to sunflower plants.This herbicide acts by contact, changing the metabolism of ammonia (Rodrigues & Almeida 2005).The means obtained in treatments with two doses of this herbicide reached values ranging 72-77 %, at 7 DAA.A significant increase of symptoms was observed on the second visual evaluation, characterized by tissue necrosis (79-80 %).However, there was a slight recovery at 21 DAA (54-60 %).
Sulfentrazone inhibits the activity of the protoporphyrinogen oxidase (Protox) enzyme, accumulating protoporphyrin IX.This molecule interacts with oxygen and light, forming the reactive oxygen, which is capable of causing peroxidation of lipids in cell membranes.Treatments with sulfentrazone caused tissue necrosis, reaching values of 20 % and 45 %, at the first evaluation.However, the apical meristems of sunflower plants were not injured by the herbicide.There was a recovery of the plants, with emergence of new leaves free of the effects caused by sulfentrazone.
The tembotrione herbicide inhibits carotenoid synthesis.The pigment-inhibiting herbicide acts by blocking the formation of carotenoids, resulting in destruction of the chlorophyll by light energy.The new leaves turn white, since there is no pigment synthesis (Oliveira Júnior 2011).The sunflower plant leaves turned white, evolving to dry and dead tissue.The percentage of phytotoxicity was 34 %, at 7 DAA.However, it was observed a reduction on injury symptoms at 21 DAA (24 %).
The application of carfentrazone provided high values of phytotoxicity at 7 DAA (78 %).This herbicide acts by contact, being also a Protox inhibitor.However, there was a recovery of plants, with emergence of new leaves.The phytotoxicity dropped to 34 %, at the last evaluation.The saflufenacil also inhibits the synthesis of protoporphyrinogen oxidase.The two doses of saflufenacil provided percentages of phytotoxicity ranging from 47 % to 76 %, at 7 DAA.There was a recovery, with emergence of new leaves.The percentages of phytotoxicity decreased to 20-25 %, at 21 DAA.
The triclopyr belongs to the chemical group of growth regulators.This product causes epinasty, inhibition and chlorosis of the growing meristems (Rodrigues & Almeida 2005).The sunflower plants turned twisted and chlorotic, achieving percentage of phytotoxicity of 36 %, at 7 DAA.The values of injury symptoms increased at 14 DAA, reaching 45 %.However, a slight recovery of plant growth was observed at 21 DAA (35 %).
The MSMA is a contact herbicide that belongs to the group of organic arsenicals.However, its mechanism of action is not well defined (Oliveira Júnior 2011).The characteristic symptoms are related to rapid desiccation of leaves and destruction of cell membranes.In the first two evaluations, values ranged from 18 % to 22 % (Table 3).There was a beginning of recovery of plants (19 %), with the emergence of new leaves, at the last evaluation.
The mean values of volunteer soybean control after 7, 14 and 21 DAA and dry weight biomass related to the experiment conducted in Rio Verde are shown in Table 4.
Glufosinate ammonium was the most effective molecule to control volunteer soybean plants.The two doses administered led to control percentages ranging from 84 % to 87 %, at 21 DAA.This herbicide provided excellent levels of control, when applied in the early development of volunteer soybean (Bond & Walker 2009).The glufosinate ammonium, at doses of 300 g ha -1 , 400 g ha -1 and 500 g ha -1 , led to complete death of volunteer soybean plants, when applied on post-emergence of cotton crop (Braz et al. 2013).
Concerning the differential sensitivity to herbicides in soybean and sunflower, soybean is more tolerant than sunflower.Thereby, herbicides that control soybean plants also eliminate completely sunflower plants.
The two doses of sulfentrazone led to soybean control of 16 % and 21 %, in the last evaluation.Those treatments did not cause death of volunteer soybean plants.However, there is a temporary Tembotrione, carfentrazone, saflufenacil, triclopyr and MSMA were not efficient in controlling volunteer soybean plants.Low percentages of control were also observed when tembotrione was applied on a different volunteer soybean cultivar (Valiosa RR ® ) (Dan et al. 2009).
MSMA did not control volunteer soybean at the dose of 925 g ha -1 (York et al. 2005).Even though there was extensive necrosis of the leaves that had received the product, the emergence of new leaves did not present symptoms of injury.
The application of glufosinate ammonium and sulfentrazone doses led to the lowest dry matter values of volunteer soybean plants (Table 4).The herbicides were applied when volunteer soybean plants were at the V 3 growth stage.Probably, the control levels of these treatments could be higher if the applications were performed in more juvenile stages of the soybean cycle.
Treatments with tembotrione, carfentrazone, saflufenacil, triclopyr and MSMA did not provide efficient control, although a reduction of dry biomass of volunteer soybean was achieved.
The mean values for number of achenes per plant, plant height, weight of one thousand achenes, plant stand and sunflower yield for the experiment conducted in Rio Verde (experiment 2) are shown in Table 5.

Table 1 .
Phytotoxicity percentage on sunflower plants and volunteer soybean control at 10 and 20 days after application (DAA), for treatments in the experiment 1(Coronel Pacheco, Minas Gerais State, Brazil, 2013).

Table 3 .
Phytotoxicity percentage on sunflower plants at 7, 14 and 21 days after application (DAA), for treatments in the experiment 2 (Rio Verde, GoiásState, Brazil, 2014).Control of volunteer soybean plants in sunflower crop This fact mitigates the effects of competition between soybean and sunflower.In addition, smaller leaf area provides worse conditions for soybean Asian rust survival (Calaça 2007).

Table 4 .
Percentage of control at 7, 14 and 21 days after application (DAA) and dry weight biomass of volunteer soybean, for the treatments in the experiment 2 (Rio Verde, GoiásState, Brazil, 2014).Means followed by the same letter in each column are not statistically different by the Scott-Knott test (p ≤ 0.05). 1

Table 5 .
Number of achenes per sunflower plant, plant height, weight of one thousand achenes, sunflower stand and yield, for the treatments in the experiment 2(Rio Verde, Goiás State, Brazil, 2014).