Development and Demonstration of a New Method of Physical Weed Control

Final Report for LNC10-322

Project Type: Research and Education
Funds awarded in 2010: $174,603.00
Projected End Date: 12/31/2013
Region: North Central
State: South Dakota
Project Coordinator:
Expand All

Project Information


The goal of this project was to design, construct, and test a field-scale implement based upon a newly conceived method of weed control. The new method involved the postemergence application of gritty agricultural materials under high air pressures to selectively control within-row weed seedlings in growing corn, soybean, and other row crops. The gritty agricultural materials can be derived from any of several sources: corn cobs, nut shells, dried distillers grain, N-rich seed meals, ground limestone, poultry wastes, etc. Some of these materials are approved as organic fertilizers, which permits weed control and fertilizer application to occur simultaneously.

A four-row, field-scale, “propelled abrasive grit management” (PAGMan) applicator was designed and constructed by agricultural engineers at South Dakota State University in 2012-13. The applicator subsequently was tested in the field during 2013 and 2014 by SDSU and USDA-ARS plant scientists in Minnesota and South Dakota on both organically certified and conventional fields of corn and soybean. PAGMan design and development was reported formally in a Master of Agricultural Engineering thesis in 2012, and the field-testing will be detailed in a Plant Sciences PhD dissertation in 2015.

Field-testing of PAGMan indicated that for season-long control of within-row weeds, grit must be applied twice. As an example, in corn, the first grit application must occur at the 1-leaf stage of crop growth, and the second application should occur sometime between the 3-leaf and 5-leaf stages of crop growth. Greater than 80% control of within-row weeds results from these tandem applications of grit. (Between-row weeds can be controlled with a conventional interrow cultivator.) Crop yields equivalent to weed-free checks were maintained with two applciations of grit.

Using corn cob grit as an example, rates of grit application were about 500 lbs/acre (500 kg/ha) per pass, or 1000 lbs/acre (1000 kg/ha) when grit was applied twice. With the use of N-rich grit (e.g., seed meals), which contain about 10% N, the equivalent of about 100 lbs/acre (100 kg/ha) of N would be applied to the crop. Corn cob grit can be purchased for about $0.25/lb; thus, season-long control of within-row weeds using this material may cost $250/acre. Many improvements in efficiency are possible in this new system of weed control, which may lower costs. Moreover, recent experiments that succesfully used abrasive grit in vegetable crops indicate that grit costs are a minor expense for overall management of high-value crops.


Context, Background, Rationale, and Need

Context: Only labor availability typically surpasses weed control in its importance to attainment of high yields and positive economic returns for organic growers. Consequently, new forms of weed control are needed to improve organic farming systems. In addition, improved techniques also will ease the problems associated with the transition of conventional farms to organic methods. In other words, as the risk of weed control failure decreases, the fears of some growers to expand organic acreage will be dampened.

This proposal will engineer, field-test, and demonstrate a novel method of postemergence weed control; i.e., compressed air propelling abrasive grit derived from agricultural residues, such as corn cobs and nut shells. Proofs-of-concept by our team have reported weed control in greenhouse and small field plots (Forcella 2009 a, b). Development of demonstration-model prototypes with advice from organic growers, intensive field validation, and on-farm demonstrations now are needed to explore the efficiency and cost-effectiveness of this technique, which is fully expected to meet certified organic standards.

Background: National surveys of organic growers (Walz 2004) and Midwestern experiences (e.g., Posner et al. 2008) repeatedly show that weed management is the primary production-related concern in organic crops. The consistency of this outcome indicates that (1) weeds are a major worry of organic growers, and (2) concern with weed management has not improved over time and continues to be the principal issue related to organic crop production faced by growers.

The book “Non-Chemical Weed Management,” edited by Upadhyaya and Blackshaw (2007) reviews most techniques available to organic growers, such as crop rotation, cover crops, mulches, flaming, steaming, allelopathy, mycoherbicides, classical biological control, etc. Organically-compatible herbicides (Copping and Duke 2007), which are increasingly available, are mostly liquid products in the form of organic acids or oils that are derived from plants or microbes. These chemicals typically are applied in the same manor as conventional herbicides, but they usually are non-selective and expensive (> $200/A). Preemergence products are less common than postemergence chemicals and include corn gluten meal and mustard meal. These seed meals inhibit germination of small-seeded weeds (Webber and Shrefler 2007; Boydston et al. 2008). Prohibitively high costs for corn gluten meal, hundreds of dollars per acre, however, limit its adoption. Most corn gluten meal is sold to home-gardeners.

Regardless of the availability of the aforementioned tactics, weeds continue to be a persistent issue for most organic growers. To date, no single tactic, except tillage, has been adopted extensively in any crop. Organic crop producers typically rely heavily upon soil tillage for weed control (e.g., Cloutier et al. 2007; van der Weide et al. 2008). Soil tillage, however, may not be sustainable in terms of soil health when viewed over the long-term. Consequently, for organically grown crops, a clear impetus exists for weed researchers and others to devise new management tactics that involve neither soil tillage nor synthetic herbicides. Development of new tactics must alleviate concerns with regard to soil degradation and coincide with the regulations of the National Organic Program (

Little doubt exists that new, effective, and reasonably-priced tools for weed control in organic systems are needed and would be adopted quickly. Nørremark et al. (2006), who developed the Pneumat system of using compressed air to kill weeds, also postulated the use of air-propelled abrasive grit to control weeds, but they did not attempt to build and test the proposed system. Grits derived from agricultural residues, such as corn cobs and nut shells, are used in sand blasters, which are powered by air compressors, to strip old paints or oxidized surfaces from walls of buildings, hulls of ships, etc. Far more types of agricultural residues are produced than are processed for grit. Thus, new applications for grits may enhance values of agricultural residues.

Rationale: If agricultural residues or other natural products can be used in sand blasters to shred and kill weed seedlings, this may represent a new option for postemergence weed control in organic agriculture. We have conducted tests in greenhouses and determined that weed seedlings easily can be killed by grit derived from corn cobs and walnut shells under pressures of about 80 psi (Forcella 2009a), and field corn seedlings growing amongst these weeds are largely uninjured (Forcella 2009b). Preliminary experiments in small field plots during 2008 and 2009 using a single-nozzle sandblaster powered by a small air compressor verified the selectivity of the method for field corn and the sensitivity of weed seedlings. Single blasting treatments reduced weeds by about 50%. With two sequential blasting treatments occurring at the 1-leaf and 3-leaf stages of corn, total season-long weed control reached 87%, and this treatment did not injure corn plants nor affect maximum corn yields. Continued weed emergence after the treatments reduced apparent efficacy and suggested that three tandem blasting treatments may be needed. However, timing and frequency of blasting to maximize weed control and crop yield are not yet known.

The simple abrasive grit applicator used to generate the above results was constructed by plant scientists, not engineers, from easily available materials. Although conducted in the field, these experiments were small scale with blasting occurring only with a hand-held single-nozzle applicator with each side of a row treated individually. The nozzle was connected to an air compressor and a grit reservoir mounted on the back of an ATV. The goal of these field experiments was to test the concept rather than to perfect an implement. Whatever the case, this simple system successfully controlled weeds and justified the desire to build field-scale equipment. In other words, we now wish to harness the energy of a tractor, whose size is typical of those on organic farms (ca. 75 to 150 hp), to compress air and eject corn cob grit or other grits through four pairs of nozzles, with each pair directed at either side of a crop row.

The proposed grit applicator is not a panacea for weed management in organic crops. Most notably, it has little effect on large and well-established weeds. Because grit effectively abrades only small weed seedlings (< 4” tall), the need for sequential applications is anticipated due to the often prolonged emergence periods of many weed species. Consequently, timing of grit applications likely will be a critical issue. Unavoidable delays in grit application because of wet weather, labor shortage, etc. will lessen the technique’s efficacy appreciably. Many other possible limitations exist (e.g., energy capacity of tractor, availability of non-GMO grit, etc.), but most of these seemingly have engineering or agronomic solutions. The primary limitations are the timeliness issues mentioned above.

Need. The necessity for development of a grit applicator has been stated above, but for the sake of completeness it will be reiterated briefly here. Organic growers desire new and effective methods for weed control. In response to that need a grit applicator was conceived, and a simple prototype developed and tested successfully in greenhouse and small field-plot settings. A larger field-scale implement is needed to test its practicality and economic feasibility. Several organic growers, many of whom are skilled mechanics, have expressed interest in participating in the development and testing of the field-scale implement.

Project Objectives:



Expected research results: (1) a working field-scale implement; (2) technical knowledge pertaining to physical details, such as nozzle types, angles, distances to row, air pressures, etc.; (3) technical knowledge pertaining to biological details, such as susceptible weeds and stages, tolerant crops and stages, application timing, application frequency, etc.

 Technical papers: (1) Paper in an engineering journal that describes the grit applicator. (2) Paper in an agronomy journal regarding field testing of implement. (3) Paper in weed science journal that examines technical variables, such as nozzle types, etc.

 Educational materials: (1) Extension bulletins from South Dakota State University describing the development and operation of an abrasive grit applicator.

 Field days: (1) Organic Field Day, Yankton, SD, 100 participants; (2) organic Field day, Lamberton, MN, 200 participants; (3) SDSU Field Day, Brookings, SD, 300 participants; (4) Swan Lake Field Day, Morris, MN, 75 participants.

 Involved Organic Farmers: Several organic growers already have provided advice for this project and will continue to do so through the project’s completion. At that point, we expect to have a core group of organic growers who have helped design and field-test the implement. If the implement functions as desired, these growers will be its primary advocates.

 Involved Engineers: Formally trained engineers and engineering-oriented growers are part of this project. Engineering students at SDSU will have completed their Senior Design Projects as integral portions of this venture. An established engineering firm (Crary Co.), will advise students, a small engineering shop (Backman Robotics) will assist with fabrication, and various growers will aid fabrication and testing (e.g., Brakke Farm).


Click linked name(s) to expand/collapse or show everyone's info
  • Dr. Sharon Clay


Materials and methods:

Approach, Activities, Methods, and Inputs.

The project has two phases, an engineering phase and an agronomic phase. The two phases are intimately related, and each has multiple components. Research on some components from each phase can be performed simultaneously. However, some components of the engineering phase must be executed before those of the agronomic phase and, likewise, some aspects of the agronomic phase must precede some from the engineering phase. We ask the indulgence of reviewers while reading the following paragraphs, as descriptions of categories (phase components) and chronologies (work schedules) may be confusing. Additionally, we recognize that the engineering phase of the research is not hypothesis-driven in the classical sense, yet it is an essential precursor to the more hypothesis-driven agronomic research.

Engineering approaches involve (i) harnessing a tractor’s energy to compress air for (ii) propelling crop-derived grit in a directed fashion through nozzles at speeds that abrade small weed seedlings selectively within crop rows, (iii) adapting nozzles for optimum patterns of grit application, and (iv) attaching multiple pairs of nozzles onto a tractor-mounted toolbar for simultaneous multiple-row weed control.

(i) We plan to use a dedicated tractor for the engineering research. The tractor’s minimum size will be 100 hp. The minimum size is required to operate an add-on air compressor, such as the HKL-series of hydraulic-powered compressors from Dynaset Corporation ( or belt-driven VANAIR 130CFM ( The exact compressor to be used will be determined by the engineering students at SDSU in consultation with professional engineers based upon known mechanical specifications and anticipated power needs. (ii) Manifolds will be devised to split the compressed air into eight regulated lines, each tipped by a nozzle. (iii) Nozzles will be designed to provide specific patterns (e.g., flat fan) of grit deposition. Currently available nozzles are characterized by solid cone-type deposition patterns. Nozzle types that work maximally for shredding weeds are not yet known and form part of the agronomic research described below. For experimental purposes, nozzle orifice areas will remain constant, but their shapes will be modified to provide desired delivery patterns of grit. A malleable metal, like copper, can be used to fashion prototype nozzles, but permanent nozzles will be contructed from more resistant materials. (iv) Swivels will be used that allow adjustments regarding the distance of the nozzles from the soil surface, angle of the nozzle relative to the soil surface, and angle of the nozzle relative to the crop row. Optimum distances and angles are not yet known. Components i and ii, above, are interrelated and are expected to represent Senior Design projects for a team of two students. Components iii and iv also are related, but they are simpler technically and represent a combined Senior Design project for a single student. The post doctoral agricultural engineer will be responsible for integrating the components into a functional system, always in consultation with professional engineers.

Agronomic field research entails testing (i) the timing and frequency of application passes, (ii) air pressure or grit air speed requirements, (iii) grit type, size and hardness to affect season-long control of weeds without crop injury in common annual and high-value perennial row crops, (iv) applicator ground speed, and (v) on-farm comparisons and demonstrations.

(i) Factorial experiments with replicated treatments will test for the best times to abrade weed seedlings in terms of crop stages (corn stages V0 through V6) and the frequency of abrasions (1 to 4) needed for season-long weed control. The hypothesis being that multiple abrasion events at critical crop growth stages will control early- as well as late-emerging weed seedlings more effectively than a single event. Although corn will be emphasized in all experiments, supplementary experiments will include soybean and a perennial horticultural crop, possibly brambles.

(ii) Grit air speed probably is the defining variable for injuring small weed seedlings, and this variable can be regulated in part by air pressure and nozzle type and distance to the soil surface. Replicated experiments will test these factors independently in field settings after minimum and maximum boundaries are determined in greenhouse situations.

(iii) Agriculturally-derived grit is available in different forms and mesh sizes. All preliminary experiments employed 20-40 mesh grit, but other sizes may be better suited for weed control. Nozzle orifice diameters determine, in part, appropriately-sized grits. Furthermore, grit is available in differing textures, e.g., hard (walnut) vs. soft (corn cob).

Because almost any material that flows freely can be used as grit, our grower-advisors suggested that pelletized organic manures and other soil amendments be examined for efficacy in weed control with the grit applicator. This intriguing suggestion would affect two agronomic issues simultaneously, namely weeds and soil fertility (soil chemistry). Besides weed control, soil fertility status could be enhanced with materials such as ‘Cluck’ chicken manure and alfalfa meal, and soil pH can be modified with limestone (decreases soil acidity) and cotton seed meal (increases soil acidity). All of these materials are pelletized and amendable to use as grit. We will test and compare grit types, sizes and textures in replicated greenhouse and field experiments. Appropriate experimental designs will be chosen to fit the need of the specific experiment. Weedy and hand-weeded checks will be included where appropriate. Furthermore, in field experiments, we will keep track of grit use rates, tractor fuel consumption and crop yields to calculate costs and cost-effectiveness of using the proposed techniques for weed control.

An especially interesting type of grit may be corn gluten meal (Webber & Shrefler 2007). CGM normally is used as a preemergence herbicide (and N source). Use rates of CGM range from 400 lbs/acre in established turf to 2400 lbs/acre in bare soil (with per acre costs of $100 to $600). Its benefit to weed control may increase if it could be used first to control the initial flush of small weed seedlings via an abrasive grit applicator and, subsequently, to control late-emerging seedlings through its normal preemergence mode of action.

(iv) Tractor/implement speed will affect grit application rate. Simple factorial experiments will examine the interaction between tractor speeds and air pressure for weed control, application rate, and costs. For costs, grit and time will be measured directly, whereas fuel use will be estimated based upon rpm and fuel consumption relationships for specific tractors.

The post doctoral researcher is expected to oversee the agronomic experiments. This individual will be assisted ably by two technical staff members at the USDA-ARS Soils Lab who are permanently assigned to the lab’s weed research program.

(v) Once the prototype field-scale implement is designed and built by our team in 2010-2011, we will test it on-station first. In 2012, at least one organic grower in both SD and MN will be asked to participate in on-farm trials. Several organic growers already have volunteered for this activity. Unfortunately, by 2012 only a single farm-scale implement likely will be available for use. Thus, transportation and timeliness issues are expected if too many on-farm trials are attempted. The implement first will be demonstrated to the growers on their properties, and immediately afterwards the growers will be asked to use the implement on small strip plots within one of their own fields. Their buy-in to the new implement will be critical for generating interest among implement manufacturers as well as other growers. Approximately one to two weeks after on-farm testing, weed control in test strips will be compared to adjacent strips that underwent standard practices as well as small weedy check areas. After assessments are made, the weedy check areas will be hand-weeded.

At this early stage of development of the concept for a new form of weed control, we will not ask growers to dedicate large acreages for testing and, therefore, we have no plans at this time to allocate NC-SARE funds to growers except as land rental fees. Indeed, most organic growers we know are happy to participate without compensation. The reward, in their minds, is knowing that USDA/university-led research is occurring along the lines that they have advocated repeatedly.

Organic growers are intimately involved with the planning and execution of the two annually-held organic field days (Lamberton MN and Yankton SD). Our intention is to garner ideas from growers at such field days in 2010 and 2011 for the types of demonstrations they would like to see in 2012, as well as to solicit additional input regarding implement specifications that would make a prototype implement more useful and acceptable on their farms. Information will also be provided at other field days at other field experiment station locations such as SE Farm, NE Farm, Brookings, and FarmFest in Mitchell, SD and USD-ARS Field Days in Morris.

Research results and discussion:

Construction of the four-row abrasive grit applicator was completed in summer 2012. Construction followed a year of detailed theoretical and practical engineering research, which was the basis of a thesis by Corey Lanoue in the Biosystems and Engineering Department at South Dakota State University. The beta-test implement was examined preliminarily in the field in late-sown corn. As expected, a number of minor defects were observed that summer and corrected during autumn and winter. The final implement was available for testing in early spring 2013. A brief video of the implement, named PAGMan can be viewed at the following web address:

The implment is pulled behind a tractor and is attached via a three-point hitch. The tractor driver regulates air pressure and grit flow through a control panel in the tractor's cab. Pressure is generated and regulated by a large air compressor mounted on the implment, but powered by the tractor's PTO. Air is distributed to four pairs of nozzles, with each pair directed at both sides (3 to 4 inches wide) of a crop row. Grit particles flow to near the tip of each nozzle via suction and are entrained by the pressurized air jet (about 100 psi) and emitted at high velocities toward weed seedlings near the bases of the crop plants.

PAGMan was tested in certified organic corn in 2013 and 2014 near Morris, MN. Field-testing of the implment is the basis of an upcoming PhD thesis by SDSU agronomy student, Mauricio Erazo-Barradas, which is expected to be completed by June 2015. Experimental treatments included grit applications at the following stages of corn development: 1-leaf, 3-leaf, 5-leaf, 1-leaf + 3-leaf, 1-leaf + 5-leaf, and 3-leaf + 5-leaf. Control (no weeding) and hand-weeded plots completed the treatments. All plots were split so that flame-weeding and interrow cultivation also could be examined as supplements to grit application. Four replicaitons were employed for all treatments. Weed control was measured in late August, and crop yields were determined in September.

Season-long weed control of 80% or greater was achieved when grit was applied twice, first at the 1-leaf stage of corn and again at either the 3-leaf or 5-leaf stages. When grit was applied at these times, no losses of crop yield occurred as effects of weed competition. If grit was applied only once, regardless of the stage of crop growth, weed control was less successful and crop yield losses occurred. However, application at the 1-leaf stage was superior to application at other stages. In brief, PAGMan was successfully employed for controlling weeds and maintaining yield in field corn.

However, the research team did observe characteristics of PAGMan that can be improved. These improvements specifically involve sway-stability of the implement. That is, successful and selective in-row weed control demands precise application of grit. Swaying of the implment by even a few inches because of rough terrain or inattentive tractor driving can lead to negligible weed control and destroyed crop plants. Adding two wheels to to the back end of the implement or a single soil-penetrating disk to the middle of the implment could help stabilize the implment appreciably.

Another improvment pertains to row sensors and precision guidance of the tractor that pulls the implement. Modern guidance systems typically involve GPS and real-time kinematics. Such units could be used to assist PAGMan applications. However, older, cheaper, but comparably precise equipment may be sufficient for PAGMan purposes. Such equipment can include tactile sensors, and we currently are exploring this option as an improvment for PAGMan.

Research conclusions:

Our project has struck a chord with other researchers and agricultural and science-based media. Impacts on both of these groups are as follows:

1. Two groups of scientists have been awarded research funds to continue and expand upon this SARE-funded project. The first group is led by Professor Sam Wortman from the University of Illinois. Collaborators include SDSU and USDA personnel. The project is being funded by OREI ($500,000), it started in mid 2014, and its goal is to redesign PAGMan to conform to organic vegetable cropping systems in the Midwest (and elsewhere). The second group is being led by Professor Manuel Ruiz from the University of Seville, with collaboration from staff at the Unicersity of California-Davis and USDA. The project is being funded by the Spanish government ($100,000). The project started in early 2014, and its goal is to provide precision guidance to a PAGMan-type implment for use in organic olive orchards and vineyards in Spain.

2. Numerous media outlets have interviewed project staff and reported information. Perhaps the most interesting and widespread of these was a 7-minute segment of "Living on Earth" with Steve Curwood broadcast around the world Public Radio International in September 2014. Another radio spot was with Minnesota Public Radio in May 2012.

Printed media outlets were numerous and international, and only a few will be highlighted here. These include (a) Popular Science in August 2014, (b) Agricultural Research magazine in July 2014, (c) Organic Broadcaster in November 2014, and (d) the Minneapolis Star Tribune in August 2014.

Farmer Adoption

Many farmers and other weed managers from around the world have expressed interest in this new technique. Vineyard growers in Oregon and State Park personnel in Maryland are constructing their own grit applicators appropriate for their specific needs. The project director (FF) has been invited by a rum (alcohol) producer to the island of Martinique in May 2015 to help with development of an experimental grit applicator for organic sugarcane.

We have not encouraged farmers to invest in grit applicators yet. Although research results are encouraging, we feel more experiments are needed, as are additional modifications to existing implements, before we are comfortble making recomendations that may require expenditure of thousands of dollars by organic growers.

Participation Summary

Educational & Outreach Activities

Participation Summary:

Education/outreach description:


Forcella, F., James, T. and Rahman, A. 2011. Post-emergence weed control through abrasion with an approved organic fertilizer. Renewable Agric. Food Systems 26:31-37.

Forcella, F. 2012. Air-propelled abrasive grit for postemergence weed control in field corn. Weed Technology 26: 161-164.

Forcella, F. 2013. Soybean seedlings tolerate abrasion from air-propelled grit. Weed Technology 27: 631-635.

Forcella, F., Humburg, D., and Clay, S. 2013. PAGMan – Propelled abrasive grit to manage weeds in soybean and corn. WSSA Abstracts 128, Baltimore, MD.

Erazo-Barradas, M., Clay S., and Forcella, F. 2014. Grit application controls weeds in organic crop production. Weed Sci. Soc. Amer. Abstract 268, Vancouver, BC.

Field Days

University of Minnesota, West Central Research & Outreach Centre, July 2012.

USDA-ARS, North Central Soil Conservation Research Laboratory, August 2013.

Outreach presentations

Project participants have engaged in numerous invited and volunteer speaking engagements concerning the research performed on weed control via abrasive grit applications. Increasingly, the information derived from the project is being adopted by other researchers and extension educators in their presentations. The most recent instances of such adoption (that are known to us) include a presentation to a national herbicide resistance summit in Washington, D.C., by Dr Michael Owen (Iowa State University) in September 2014; and a seminar to the annual Kentucky Fruit and vegetable Growers meeting by Dr Shawn Wright (University of Kentucky) in January 2015.

Project Outcomes


Areas needing additional study

Comments in the preceding sections alluded to areas that require additional study. Briefly, these areas are as follows:

1. Redesigned inplements appropriate for vegetable crops, vineyards, and orchards. Such work currently is being pursued at the University of Seville (UdS), University of Illinois, University of California-Davis, South Dakota State University (SDSU), and USDA-ARS.
2. Precision application of grit. This work is being performed at UdS and SDSU.
3. Nozzle design and grit application patterns and efficacies. (SDSU and USDA-ARS)
4. Nozzle positioning and adjustment systems. (SDSU and USDA-ARS)
5. Implement stability (minimum lateral shifting) during application. (SDSU and USDA-ARS)

Currently, funding for these activities are being derived from grants from the Spanish governement (to UdS) and NIFA-OREI (to Univ. of Illinois).

Information Products

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.