Plant Pillow Preparation for the Veggie Plant Growth System on the International Space Station

Arcillite substrate (Turface Proleague, Profile Products, LLC) was obtained at two particle sizes – 600 μm to 1 mm and 1–2 mm. The substrate was sieved using standard mesh sieves collecting the fraction between mesh 18 and mesh 30 (600 μm to 1 mm) and the fraction between mesh 10 and mesh 18 (1–2 mm). The arcillite substrate contained significant levels of dust, which had to be removed prior to flight for both crew safety and to maintain functionality of the hardware by reducing the potential for particulate clogging on the reservoir surface and in the fan. To remove dust, two methods were tested for substrate washing. The original method was used in VEG-01, and an upgraded method was added for VEG-03. For the original substrate washing method, the substrate was washed multiple times in a 30 L substrate washer (modified from a design suggested by ORBITEC, see Figure 1) using deionized H₂O to remove adhering dust. The substrate washer consists of a 30 L Nalgene™ Heavy-Duty Cylindrical LLDPE Tank with a spigot (Fisher Scientific, Pittsburgh, PA), three pairs of well washed standard bricks (Lowe's, Mooresville, North Carolina), and two 25.4 cm standard mesh sieves (Fisher Scientific, Pittsburgh, PA). Using this system, substrate was washed in 1000 cm³ batches, with 500 cm³ in each of two sieves in the substrate washer. For the 600 μm to 1 mm substrate, arcillite was placed in a 600 μm (#30) sieve. For the 1–2 mm substrate, substrate was placed in a 1 mm (#18) sieve. The 30 L tank was filled with deionized water (24 L with displacement from bricks, sieves, and substrate) with water flowing over the surface of the substrate to remove dust, and then the water in the tank was drained. This process was cycled an average ten times until drain water exiting the spigot was visibly clear plus one additional rinse. Differences in substrate dustiness led to different numbers of required rinses per batch of substrate. Visible clarity was assessed by sampling outflow water and viewing it with backlighting. Visible clarity was judged as the time when no turbidity or particles were observed.

Figure 1

The second, upgraded washing method consisted of using the same sieves positioned on bricks over a sink drain with a custom washing unit positioned over the sieves (Figure 2). Deionized H₂O was run through the sieves via two shower head attachments (Aqua Dance 8 Rainfall Showerhead, Target, Minneapolis, MN) composed of plastic and stainless steel to eliminate any potential corrosion products leaching in to the substrate. Shower heads were supported with an 80/20 aluminum and PVC frame allowing heads to be raised or lowered. Each shower head is capable of operating independently, or both can be operated simultaneously using PVC shut off valves. Water was run through 250 cm³ of arcillite substrate in each sieve until visibly clear drainage water was observed using the established protocol (see above), with confirmation made from a catch beaker held under the sieve outflow. The substrate was stirred once a minute during a wash.

Figure 2

Following washing with either method, wet and washed substrate was placed in polypropylene sterilizing pans (14.2 L, Thermo Fisher Scientific, Rochester, NY), covered with aluminum foil, and autoclaved for a 60 min gravity cycle with a 30 min dry cycle to eliminate any biological contamination. Pans were removed from the autoclave and oven dried to dryness in a forced air oven at 70°C for at least 72 h without removing covering foil. Substrate was allowed to cool to room temperature and then, within a calibrated sterile laminar flow hood or microbiological safety cabinet, it was transferred to a sterile reclosable bag. Prepared substrate was stored until needed. Just prior to pillow filling operations, substrate was carefully mixed in a sterile reclosable bag in 1000–2000 cm³ batches with added fertilizer. Substrate was mixed by placing measured volumes of substrates and fertilizer in an unused reclosable bag, sealing, and gently agitating to ensure mixing but to minimize any arcillite degradation and dust formation.

Mixing volumes of 1000–2000 cm³ substrate with fertilizer in this method with careful observation of fertilizer distribution provided uniformity of fertilizer in the plant pillows. Plant growth tests prior to flight confirmed that plants in pillows filled with this method had sufficient nutrient levels. For VEG-01 and VEG-03, the fertilizer utilized was Nutricote controlled release fertilizer (18-6-8, type 180, Florikan, Sarasota, FL) at a rate of 7.5 g/1000 cm³ dry substrate. Fertilizer was not sterilized prior to flight as sterilization procedures could damage the polymer coating. To keep the fertilizer source clean and free of contaminants, the bag of fertilizer was opened only in a controlled laboratory environment and maintained in a controlled clean container.

The appropriate numbers of seeds were counted and ~15% additional seeds were added, and placed in a 60 mm Pyrex Petri dish. A 50 mL Pyrex beaker was inverted in a Ball Jar (regular mouth 8 oz. glass Mason jar, Ball and Kerr, Fishers, IN). Thirty mL of household bleach (5.25% sodium hypochlorite) was added and it was sealed with the two-part jar lid. The jar and the dish of seeds were transferred to a fume hood. Initial (bleach alone) fuming was tested by placing a dish of seeds in the jar on the beaker and sealing it for a test duration (Table 1). Later tests used concentrated HCl added to the bleach to fume the seeds with chlorine gas. For chlorine fuming, 0.5 mL (1 mL was also tested) of concentrated HCl was pipetted into the bleach surrounding the inverted beaker. Immediately after HCl was added, the open dish of seeds was transferred into the jar to rest on top of the inverted beaker. The jar was sealed with the ball jar lid for 1 h (other durations were also tested). After 1 h (or other durations) elapsed, the dish was removed from the jar and immediately covered with the Petri dish lid.

Used solution was disposed of by diluting the bleach-HCl mixture into ~2 L of tap water. The diluted solution was disposed down the sink with copious running water. Under a sterile laminar flow the dish with seeds was opened and allowed to off-gas for at least 30 min. After off-gassing, the dish was sealed with Parafilm® until the seeds were used.

The appropriate number of seeds were counted and ~15% additional seeds were added. Seeds were placed in a 1.5 mL Eppendorf tube and 1 mL of 70% ethanol was added to the tube. The tube was sealed and seeds were maintained submerged for 5 min (10 min was also tested). During the submersion, the seeds were vortex blended at least 10 times for 10 s each time. After the test duration, seeds were allowed to settle and excess ethanol was decanted under a sterile laminar flow. Using a sterile spatula, the seeds were scooped out of the Eppendorf tube and spread upon sterile filter paper in a sterile petri dish. With the petri dish open under laminar flow, excess ethanol was allowed to evaporate off for at least 30 min. The dish containing seeds was sealed with Parafilm® until the seeds were used.

Seed surface sterilization methods were confirmed by testing both seed germination and surface sterilization. For seed germination assessments, surface sterilized seeds and unsterilized controls were planted in Petri dishes on germination paper moistened with tap water until glistening and then sealed with two layers of Parafilm®. Sealed Petri dishes were placed in controlled environment chambers with low light (~50–150 μmol m⁻² s⁻¹) set to 20–23°C, 50–75% relative humidity (RH), and ambient CO₂ at 500 ppm. Seeds were tested immediately upon surface sterilization as well as approximately 1 month after sealed storage at room temperature. Surface sterilization was confirmed using two methods. The first involved direct plating of five treated seeds on Trypticase soy agar (TSA) medium with comparison of untreated controls for microbial growth. The second method used triplicate samples of 20 seeds (both treated and untreated) placed in 3 mL of sterile deionized H₂O with glass beads. These were shaken for 2 min and plated onto TSA agar and Inhibitory mold agar (IMA). TSA plates were incubated at 30°C for 48 h, and IMA cultures were maintained at 25°C for up to 7 days. Incubation was followed by enumeration of colony forming units (CFU). Methods for each seed type were only selected if they sterilized the surface of seeds and did not impede germination.

All procedures were conducted under a sterile laminar flow using sterilized tools and plant pillows that had been constructed and maintained in a cleanroom environment. Pillow wicks were cut from polypropylene wipe material (KIMTECH PURE* W4 Dry Wipers, Kimberly-Clark Professional, Roswell, GA) as rectangles with dimensions of 23 mm × 120 mm. The VEG-01 and VEG-03 series flight experiments used standard CTBs designed to hold 18 pillows. Thus for each flight experiment 36 wicks were required, with an equal number needed for the ground control set of plant pillows. Pairs of wicks were placed together with one on top of the other with rougher or fuzzier sides of the wipe material facing inward. Wick pairs were then inserted through the slots of Veggie plant pillow foam gaskets (ORBITEC) with 18 mm of wick material left extending outside of the plant pillow (Figure 3A). Wick tails were adjusted with forceps inside the plant pillow, with the tails protruding through the interior watering ring and overlapping on the lower interior surface of the pillow (Figure 3B). Each internal tail of the pair of wicks was separated when filling the pillow with the arcillite fertilizer blend, to allow substrate to pass between the wicks inside of the pillow. Each pillow was filled with 250 cm³ of substrate by first adding 150 cm³ of substrate to a pillow, next adjusting the tails of the wicks to keep contact with the lower surface of the pillow and allow separation, and then adding the remaining 100 cm³ of substrate.

Figure 3

Following filling, the pillow open end was folded over a single time and temporarily sealed closed with lab tape leaving a tape tab. For the VEG-03 pillows, air flow through the pillow priming irrigation tube was again tested with a sterile 60 mL syringe and quick disconnect (QD) to make sure that the substrate had not clogged the interior drip line sufficient to create back pressure. Filled pillows were sewn shut with Nomex® thread by removing the tape, double folding the folded edge, and sewing a double seam along the center of the fold.

After pillows were sewn they were stored in sterile reclosable bags (labeled with substrate type) until launch preparations commenced (i.e., shortly before turning them over for flight). The final steps in pillow preparation were to plant the surface sterilized seeds, add labels to the pillows, and seal up the planted pillows. For planting, 300 mg of Guar (a glue-like matrix, Sigma-Aldrich Corp., St. Louis, MO), was mixed with 25 mL of deionized H₂O. The mixture was stirred with a sterile spatula, microwaved on high for 10 s, stirred again, microwaved on high for an additional 10 s, and stirred once more. A 3 mL syringe with a 16 gauge needle was filled with the guar gum mixture. The syringe was allowed to come to room temperature and covered with foil to block light until it was used. Under sterile laminar flow, a sewn pillow was removed from its reclosable bag. With forceps, the paired wicks were opened and one seed was positioned. Seed positioning was performed to allow for radicle emergence into the interior of the pillow, so seeds were oriented with the tapered end pointing inward into the pillow interior (for lettuce or zinnia) or the seed scar (black spot) facing inward (for Chinese cabbage) so that the seed rested halfway into the pillow in the mid-plane of the foam gasket (Figure 3C). A drop of guar was used on the end of the forceps to help position the seed. A small drop of guar glue was used to attach the seed to the wick. The process was repeated with a second seed, positioning 8 to 10 mm from the first seed, with both seeds centered in the pair of wicks (Figure 3C). The wicks were pinched closed and held for approximately 20 s. Each planted pillow was allowed to dry under sterile air flow for 20 min while the planting procedure was repeated with the remaining pillows. Forceps were sterilized between seed types. Flight pillow labels were inserted into Aclar® sleeves on the plant pillows after planting. Label material was selected specifically to be nontoxic and not off-gassing. After allowing guar to dry, each pillow was inserted, QD end first, into a gas impermeable Tedlar® bag (6.5 in × 8 in Tedlar® bags, SKC Inc., Eighty Four, PA), while using care and folding flat the wicks to maintain seed position. The Tedlar® bag was heat sealed after most of the air was squeezed from the bag. The heat sealer used was a Polysealer model P-200 (Fuji Impulse, Deerfield, IL) with a setting of 7.5 for the Tedlar® material. After sealing, a uniform seal without wrinkles was confirmed before the packaged pillow was removed from sterile air flow. Each sealed pillow was weighed and photographed. Figure 3D shows a packaged plant pillow prepared for flight.

Pillows were maintained under dry (<50% RH) and dark conditions prior to flight. Handling was minimized to reduce the capacity for substrate grinding and dust formation. Following sealing in Tedlar® bags, pillows were placed into custom-cut flight foam layered in a CTB and either transferred to a launch vehicle for flight or maintained in controlled laboratory conditions for ground controls.