Hydration Efficiency of Traditional and Alternative Greenhouse Substrate Components

Wettability is a major factor in determining whether a material can be effectively and efficiently used as a component in greenhouse substrates. Poor wettability can lead to poor plant growth and development as well as water use inefficiency. This research was designed to test the wettability and hydration efficiency of both traditional and alternative components of substrates under different initial moisture contents (MCs) and wetting agent levels. Peatmoss, perlite, coconut coir, pine bark, and two differently manufactured pine tree substrate components (pine wood chips and shredded pine wood) were tested at 50% and 25% initial MC (by weight). The objective of this research was to determine the effects of initial MC and wetting agent rates on the wettability and hydration efficiency of these substrate components. Each component received four wetting agent treatments: high (348 mL·m), medium (232 mL·m), low (116 mL·m), and none (0 mL·m). Hydration efficiency was influenced by initial MC, wetting agent rate, and inherent hydrophobic properties of the materials. Wetting agents did increase the hydration efficiencies of the substrate components, although not always enough to overcome all cases of hydrophobicity. Wettability of a material was defined by Letey et al. (1962) as the ability of a liquid to spread over a material’s surface. In substrates, proper wettability ensures a more even distribution of water (and nutrients) throughout the root environment. Appropriate wettability also improves water-holding capacity, which has been shown to increase plant growth (Plaut et al., 1973). Horticultural substrates often have wettability issues resulting from the nature and high volume of organic matter (OM) components in them. These components, primarily composed of sphagnum peatmoss and pine bark, can become hydrophobic, thus reducing wettability (Dekker et al., 2000a; Michel et al., 2001). The molecules of OM contain many organic acid functional groups on their exterior surfaces, like carboxylic acids and phenolic acids, among others. These acidic functional groups tend to repel water from the particle surfaces when in a balanced state with hydrogen cations bound to oxygen anions (Ellerbrock et al., 2005). As substrates dry, hydrophobicity can intensify, complicating the wetting and rewetting process during plant production (Valat et al., 1991). Thus, many organic substrates can develop hydrophobicity issues that hinder water efficiency (Beardsell and Nichols, 1982). There are several factors that can influence a substrate’s wettability, including, but not limited to, MC (de Jonge et al., 1999; Michel et al., 2001), substrate pH (Gautam and Ashwath, 2012), hydrophobicity of the substrate (Fonteno et al., 2013), and preferential flow (Dekker and Ritsema, 1994). Measurement of substrate wettability has been difficult to assess with the most common method in the literature being the measurement of contact angles (Michel, 2009). Another method for measuring substrate wettability described by Letey (1969) and re-evaluated by Dekker and Ritsema (2000b) is known as the water drop penetration time (WDPT) test. To test WDPT, a drop of water is placed on the surface of a substrate and the time it takes for the drop of water to completely penetrate the substrate is measured. This method is less expensive to perform; however, results can vary as a result of the subjective nature of this test. A more recent method described by Fonteno et al. (2013) for determining the wettability of a substrate is known as the hydration efficiency test. In this method, known quantities of water are passed through a substrate and effluents are collected to determine the quantity of water sorbed by the substrate. Wetting agents (WA) are chemicals (dry or liquid form) that increase the wettability of substrates by enabling substrates to be more uniformly wet during/after irrigation events. Wetting agents are used to change the properties of water by allowing the individual water molecules to break some of their hydrogen bonds and spread out more evenly over the surface of a substrate. Wetting agents, like all surfactants, are chemically composed of two parts, a hydrophilic hydrocarbon tail and a hydrophobic lipid head. The hydrophobic end will adhere to the surface of the substrate particle leaving the hydrophilic end exposed. The water molecules will then bind to the hydrophilic end and spread out across the surface of the particle. This reduces surface energy of the solid particle and promotes a more uniform distribution of water over the surface. Wetting agents are commonly used in many substrates to achieve proper hydration with fewer irrigation events after potting. Hydration efficiency was defined in this study as the ability of a material to capture and retain water in the fewest number of hydration events (water applications). The objectives of this study were: 1) to characterize the wettability of traditional substrate components and compare them with two newer pine tree substrate components; and 2) to determine hydration efficiency of these components. Materials and Methods Substrate components. Substrate components tested were coconut coir (Densu Coir, Ontario, Canada), sphagnum peatmoss (Premier Tech, Canada), aged pine bark, perlite, and two types of hammermilled loblolly pine wood (Pinus taeda L.). To create the pine tree substrate components, fresh loblolly pine wood was hammermilled through a 6.35-mm screen after delimbed pine logs were initially processed through either a wood chipper or a wood shredder. Pine trees processed in the two different types of machinery produce completely different pine tree substrate components even when milled through the same hammermill screen size (Jackson and Fonteno, 2013). The pine logs for chipping were harvested on 9 Dec. 2011 and passed through a DR Chipper (Model 356447; 18 HP DR Power Equipment, Vergennes, VT) on 3 Jan. 2012 and hammermilled (Meadows Mills, North Wilkersboro, NC) on 5 Jan. 2012. The pine logs for shredding were harvested on 12 Dec. 2011 and shredded in a Wood Hog shredder (Morbark , Winn, MI) on 9 Jan. 2012. The shredded pine wood (SPW) was then hammermilled as previously described for the pine wood chips (PWC) on 10 Jan. 2012. All processed wood materials were then placed in 55-L poly bags and stored indoors at 22 C until needed for further testing. Care was taken to monitor the bags to prevent extra drying or increased temperatures. Because of the small sample sizes and storage conditions, the materials did not dry out nor display any increased temperatures as are possible with organic materials. All materials were tested over an 8-week period after tree harvesting and processing (hammermilling) during which time wood materials were stored in bulk bags/ totes under shelter. Particle size distribution of 150-g ovendried substrate samples was determined on three replications of each substrate component with 11 sieves (ranging from greater than Received for publication 11 Oct. 2013. Accepted for publication 23 Jan. 2014. This paper is a portion of a thesis submitted by Jeb S. Fields as required to fulfill degree requirements. Graduate Student. Professor. Assistant Professor. To whom reprint requests should be addressed; e-mail [email protected]. 336 HORTSCIENCE VOL. 49(3) MARCH 2014 6.3 mm to less than 0.063 mm) plus a bottom pan. Sieves and pan were shaken for 5 min with a RX-29 Ro-Tap sieve shaker (278 oscillations/min, 150 taps/min; W.S. Tyler, Mentor, OH) and the particle fractions retained on each sieve and the amount that passed through the smallest sieve and retained by the sieve pan were weighed. Moisture content and wetting agent treatments. Each of the six components were hydrated to an initial moisture content (IM) of 50% by weight. Each component was separated into four subsamples of equal volume (4 L). Each subsample was treated with AquaGro -L (Aquatrols, Paulsboro, NJ) WA at 0 (none), 116 (low rate), 232 (medium rate), and 348 (high rate) mL·m, respectively. The amount of WA required to achieve the four testing levels for each sample was premixed with the water required to bring the sample to 50% IM. The amount of water required to bring each substrate up to 50% IM and the amount of WA required for each respective subsample were mixed in a 4-L SureSpray sprayer (Model 20010; Chapin, Batavia, NY). Substrate components were individually spread out at a depth of 1 cm on a metal tray, and the WA/water mixture was evenly applied (sprayed from sprayer). The solution was thoroughly mixed in each component immediately after application by turning and mixing until the entirety of the solution had been applied. Substrate components were also tested at 25% MC. To do this, half of each treatment was spread on a tray and allowed to air-dry until 25% MC was attained. Once attained, samples were sealed in plastic bags to prevent further water loss while also allowing for moisture equilibrium. There were a total of 48 treatments in this study (six components · four WA levels · two initial MCs). Hydration efficiency measurements. This experiment was conducted following the procedures first described by Fonteno et al. (2013) and displayed in Figure 1. The equipment consisted of a transparent cylinder, 5 cm i.d. · 15 cm·h, with a mesh screen (mesh size 18 · 16; New York Wire, York, PA), attached to one end, using rubber pressure plate rings (Soilmoisture Equipment Corp., Santa Barbara, CA); a 250-mL separatory funnel; a 250-mL beaker; and a 10-mL plastic vial (4 cm diameter), referred to in this work as a ‘‘diffuser.’’ The diffuser had five evenly spaced holes in the bottom (2.38 mm diameter), which enabled it to diffuse the force of water as it is released and falls to the surface of the substrate (Fig. 1). As water moves from the funnel into the diffuser, it slowly drips out of the five holes onto the substrate surface with force similar to a drip irrigation system in a greenhouse production setting. A rubber O-ring is placed around the outside of the diffuser, which allowed it to sit at an adjustable height atop the transparent cylinder. The transparent cylinders were packed with each substrate component to achieve a bulk density within 5% of other samples of the same components. To do this, cylinders were filled with substrate and gently packed by holding filled cylinders 10 cm off a flat surface and tapping three times so the height of the sample in the cylinder was 10 cm, which equated to 200 mL of substrate. Four replications were produced for each of the 48 treatments, totaling 192 different samples. Once packed, cylinders were fitted with a diffuser and placed under a separatory funnel. Water was applied in 10 separate hydration events. Each hydration event consisted of 200 mL water being applied to the substratefilled cylinders at a rate of 3 L·h. As a result of hydrophobicity issues of some of the substrate components at 25% MC, water flow rate into the diffuser was slowed to prevent ponding on the substrate surface. The water was passed from the funnel, through the diffuser, onto the substrate, with a 2.5-cm distance between the bottom of the diffuser and the surface of the substrate. As the water percolated through the substrate, it was either sorbed into or onto the substrate or passed through the substrate and was collected into a beaker below. Ponding on the substrate surface was controlled by keeping hydraulic head to a maximum of 2 cm by adjusting the stopcock at the base of the funnel. After the entire 200 mL of water had been applied and passed through the substrate-filled cylinders, equilibration was allowed until dripping ceased ( 3 min). Effluent water in the beakers was recorded and water retained by the substrate was calculated by subtracting leached water (effluent) from total water applied (200 mL). This procedure was repeated for a total of 10 hydrations for each sample. Container capacity measurements. After the tenth hydration event was completed, cylinders were reweighed and any changes in volume resulting from shrinking or swelling were recorded. The cylinders were then placed into a Bucher funnel with holes as described in the North Carolina State University Porometer Manual (Fonteno and Harden, 1995). Samples were then saturated from below in a stepwise fashion at one-third intervals by adding water to the funnels between the cylinder and the funnel wall until water reached the top of the substrate. After saturating for 15 min, the rubber stopper at the base of the funnel was removed and the water was allowed to drain for 30 min. Samples were reweighed and new sample heights were again measured to observe any changes in volume. Samples were then dried at 105 C for 48 h. Once dry, MC and total water retained were determined. Fig. 1. Hydration efficiency apparatus. (Left, top to bottom) Funnel, separatory funnel with stopcock, water flow diffuser, sample cylinder, beaker. (Top right) Close-up of water diffuser with O-ring above the sample cylinder allowing control of hydraulic head. (Bottom right) Water diffuser with O-ring and five holes in the bottom. HORTSCIENCE VOL. 49(3) MARCH 2014 337 Testing on commercial mixes. A final experiment was conducted on readily available commercially produced mixes for comparison with the substrate components tested. Three mixes were used: 1) Mix A was a commonly used commercial grower mix, composed of Canadian sphagnum peatmoss, processed pine bark, perlite, vermiculite, starter nutrients, WA, and dolomitic limestone; 2) Mix B was commercially available retail mix, composed of Canadian sphagnum peatmoss composted softwood bark, perlite, WA, and starter fertilizer; and 3) Mix C was a standard research and grower control mix, consisting of 3:1:1 peat:perlite:vermiculite (v:v:v), containing dolomitic lime, and WA. No additional WA was added to these mixes. The three mixes were moistened following the same procedures as described in Expt. 1, packed in the cylinders, and 10 hydration events were applied as also previously described. Similar to the substrate components in Expt. 1, the three mixes were tested at 50% and 25% MC. Three mixes · two IMs · four replications totaled 24 samples/treatments in this experiment. Hydration efficiency. Data from the 10 hydration events were used to develop wettability curves for each of the samples tested in this experiment. These curves plot the volumetric water content of the sample after each hydration event. The container capacity (CC), which is the highest volumetric MC attained of each sample after saturation and drainage had occurred, was also plotted on the chart to show relationships between CC and sample water-holding after each hydration event. These wettability curves allow easy visualization of substrate hydration. Using the CC as the maximum hydration obtainable for that treatment, it is easy to compare CC with the curve to determine hydration efficiency of each individual sample. Efficiency values. To provide numerical and statistical comparisons, each treatment had its hydration efficiency described with two values: 1) an initial hydration percentage (IHI) rating; and 2) a hydration efficiency value (HE). Initial hydration was the percentage of CC that was attained in a sample after one hydration event. The HE value was the number of hydration events required to bring the sample to CC. For example, if CC was reached at the first hydration event, an IHI of 1.0 and HE of 1 would be achieved. If the sample did not reach CC until the third hydration event, the HE value would be 3 and an IHI less than 1.0. If CC was not attained in the 10 hydration events, that treatment was given an ‘‘x’’ in Table 1 to denote lack of achievement. Statistics were determined on data using SAS Version 9.2 (SAS Institute, Cary, NC). A Tukey’s honestly significant difference (HSD) test with alpha = 0.05 was used to determine differences and similarities between the components at individual MCs and WA levels. A Tukey’s HSD test with alpha = 0.05 was also used to determine the differences in the HE values across all treatments in the experiment. Container capacity and IHI within each component at individual MCs had regression analysis conducted to determine the effect of the rates of WA on each component at each MC. Both linear and quadratic regression was determined and significance was determined using P values with significance ranging from > 0.001 to 0.05. An analysis of variance test was also conducted to test effects of WA and IM on CC and IHI among all components and within individual components.