The behavior of phosphine gas during a fumigation is complicated since it’s affected by the multi-variable system of commodity, weather conditions, gas-tightness and insect tolerance. 60 % of the containers used commercially and being fumigated are seriously leaking, leading to fumigation failures. But the new technologies and the new eyes on fumigation are available on the commercial basis and there should be no more excuses for a failures and insect survival.

Dr. Stathis Kaloudis Physicist with PhD on Computational Fluid Dynamics. Centaur Analytics, Inc.

Phosphine (PH3) is a highly efficient fumigant, which has replaced Methyl Bromide that was ozone depleting. As the industry is switching drastically to phosphine, Precision Fumigation methods become important for ensuring its long-term viability against the problems of insect tolerance. A new crop of technologies based on advanced sensors and data analytics, are now ready to face known challenges and ensure high quality, pest-free crops and foods.

PHOSPHINE AND IT’S CHALLENGES Phosphine is the single most relied-upon fumigant to control grain pests, due to its inexpensiveness, ease of application and universal acceptance as a residue-free treatment. Since the use of Methyl Bromide was phased out due to its significant contribution to the destruction of the earth’s stratospheric ozone layer, phosphine is its successful replacement. However, there are several factors that occasionally prevent phosphine fumigations to be successful (e.g. leaky storage structures). Impatience or inefficient quality assurance methods lead to treatment durations which are too short to be effective.

Interestingly, the application of pest control fumigants is mostly done with decades-old monitoring technology and frequently without monitoring at all. Fumigant dosage is seldom correlated to the product being treated, the actual storage micro-climate and the encountered types of warehouse insects.

Improper use leaves the treated commodity susceptible to insects, increasing the possibility of spoilage, but is also known to lead to tolerant strains among key stored product insects throughout the world [Athanassiou et al. 2016; Mau et al., 2012].

Let’s first consider the challenges that fumigators face daily, drawing on scientific research data.

MONITORING Traditional methods of grain monitoring systems employ sensors that are hardwired into the storage structure. Multiplexed signal conditioning is performed outside the structure with the data transmitted to a display and storage device. For fumigation applications, toxic gas sensors that use suction are bulky, hard to use, have low sample frequency and often involve human interaction making them prone to human error. Additionally, since phosphine gas is not distributed uniformly throughout the stored product, accessible sampling areas do not cover all the points of interest. Advances in technology now permit their replacement by wireless sensors, such as the ones developed by Centaur Analytics. These can resolve the shortcomings of human-operated phosphine gas sensors.

SAFETY ISSUES Phosphine is toxic to humans (even at low concentrations such as 1 ppm) and can be absorbed into the body by inhalation. Trained users with protective gear have to enter or approach fumigated areas to monitor phosphine levels, exposing themselves to hazardous environments. Additionally, residual toxic gases should be monitored accurately, to ensure personnel safety while unloading cargo from containers, warehouses and silos.

LEAKAGE Loss of phosphine from an imperfectly-sealed structure can be easily and intuitively visualized (Figure 1), and usually involves diffusion mechanisms driven by the large difference in fumigant concentration between the interstitial and ambient air. Air currents caused by temperature gradients within the grain mass and wind often causes fumigant to rapidly escape from structures [Reed and Pan, 2000]. Leakages are one of the most important factors during a fumigation and a fumigator should be able to identify their existence and quantify them.

DEGASSING RATES OF PHOSPHINE FORMULATIONS It is common to experience incomplete decomposition of phosphine formulations especially in short fumigations of exposure time under dry conditions. Scientific studies show a strong correlation of the degassing rate to air temperature (Tair) and relative humidity (r.h.) and that improper calculation of the degassing rates could lead a fumigation to failure. For example, Xianchang (1994) found that decomposition times of aluminum phosphide tablets ranged from 36 to 204 hours (Figure 2).

SORPTION During fumigation, sorption causes the gas concentration within the enclosure to deplete. As the absorptive capacities of food commodities vary, commodity sorption can be a major factor in determining whether a lethal concentration of fumigant is achieved under sufficiently airtight conditions. For example, paddy rice is 3.5 times more absorptive to phosphine than wheat [Reddy et al., 2007].

SOLUTIONS The behavior of phosphine gas during a fumigation is complicated since it’s affected by the multi-variable system of commodity, weather conditions, gas-tightness (see previous section) and insect tolerance [Athanassiou 2016; Mau 2012]. Even with the assistance of fumigation protocols and guides, a successful fumigation is not a trivial task. Phosphine fumigations pose challenges which can be tackled only with a combination of cutting-edge technology sensors and accompanying software tools.

PRESCRIPTIVE FUMIGATION Fumigators planning a treatment, typically determine the dosage on one of the following options: • Fumigation protocols (e.g. Coresta) that set target concentration for specific exposure time • Guidelines that propose a specific dosage • User defined dosage which is an outcome of personal experience Unfortunately, none of these options incorporate all the complex phenomena described above. To overcome the issue, new algorithms that predict the right amount of phosphine dosage and inform the user for the date of successful completion have been developed involving several parameters like: • storage geometry • stored commodity (witgh sorption automatically calculated) • fumigant physical form (pellets, tablets, etc.) • fumigant chemical composition (Mg3P2, AlP) • storage micro-climate (Tair, r.h.) • expected leakage (none, low, high)

Calculation of phosphine dosage A 20” shipping container (Figure 3) is used for fumigation. Approximately 90% of the container is filled with wheat flour sacks. The container is considered gas-tight, while Tair=20oC and r.h.=70%. The desired concentration is 300 ppm for 6 days of exposure time whereas Mg3P2 plates are available for the fumigation.

The sophisticated algorithm proposes a dosage of 1.8 gr PH3/m3, meaning that 2 Mg3P2 plates (2.0 gr PH3/m3) are sufficient. Additionally, a graph (Figure 4) of phosphine concentration at the far-most location of the container is produced, informing the user that it takes almost a day for the concentration to reach the threshold of 300 ppm. This means that the fumigation will be successful at the end of the 7th day from treatment start.

In this example we used a well-sealed container. In the real world 60% of the containers used commercially and being fumigated are seriously leaking, leading to fumigation failures.

PRECISION FUMIGATION During fumigations, there are some factors that influence the process which are difficult to know beforehand. These parameters include among others the air flow movements inside the storage caused by temperature gradients or the increase of leakage due to strong winds. For that reason, a new computational tool has been developed to perform precision fumigation simulations. The software is based on the Computational Fluid Dynamics approach.

Computational fluid dynamics (CFD) is a branch of fluid mechanics which aims to analyze problems that involve fluid flows. The CFD software can evaluate heat transfer effects originate from the temperature differences between grain temperature and ambient temperature as well as solar radiation. Additionally, it allows for an advanced implementation of a ‘porous medium’ approach for accurately capturing phosphine movement inside the stored product. Furthermore, insect mortality models are integrated, offering a 3-D visualization of the areas with 99.9% insect mortality.

Application example A difficult to predict scenario is that of grain silos, particularly the ones that are subject to weather changes. In our example the steel grain silos shown in Figure 5 are exposed to weather conditions (Figure 6) that are not constant and will influence the distribution of phosphine. In our example wheat is fumigated in these silos using Aluminum Phosphide blankets placed on the surface of the wheat grains.

Simulations (Figure 7) reveal that air moves downwards close to silo walls and upwards in the silo core. The outcome is faster phosphine diffusion on the silo boundaries and an obvious lag in the core region.

In our example the fumigation will turn out to be a failure if recirculation of the fumigant gas is not used (j-system). The main benefit of the CFD approach is its wide applicability on any type of commodity, storage or phosphine formulation.

THE BENEFITS These new predictive capabilities with the simultaneous use of wireless PH3 sensors offer an in-depth knowledge of the phosphine distribution throughout a fumigation treatment bringing unprecedented benefits to its users: • Mitigation and prevention of crop spoilage • cost reduction of the overall pest control application, avoiding excessive chemicals (e.g. overdosing, need to repeat failed fumigations) as well as excess labor • improved quality of finished products (e.g. flour), traceability and defensibility against quality claims from retailers • 100% safe operation and monitoring for the operator

The new technologies and the new eyes on fumigation are available on commercial basis and there should be no more excuses for a failures and insect survival.

Referanslar 1.Athanassiou ve ark., 2016. https://doi.org/10.1016/j.cropro. 2016.08.017 2. Mau ve ark., 2012. https://doi.org/10.1371/journal.pone.0034027 3.Reed ve Pan, 2000 https://doi.org/10.1016/S0022-474X(99)00049-1 4.Xianchang, 1994. http://spiru.cgahr.ksu.edu/proj/iwcspp/pdf2/6/201.pdf 5.Reddy ve ark., 2007. https://dx.doi.org/10.1002/ps.1298