Methane Hydrate Crystallization on Sessile Water Droplets

Summary

We describe a method to form gas hydrate on sessile water droplets to study the effects of various inhibitors, promoters, and substrates on the hydrate crystal morphology.

Abstract

This paper describes a method to form methane hydrate shells on water droplets. In addition, it provides blueprints for a pressure cell rated to 10 MPa working pressure, containing a stage for sessile droplets, a sapphire window for visualization, and temperature and pressure transducers. A pressure pump connected to a methane gas cylinder is used to pressurize the cell to 5 MPa. The cooling system is a 10 gallon (37.85 L) tank containing a 50% ethanol solution cooled via ethylene glycol through copper coils. This setup enables the observation of the temperature change associated with hydrate formation and dissociation during cooling and depressurization, respectively, as well as visualization and photography of the morphologic changes of the droplet. With this method, rapid hydrate shell formation was observed at ~-6 °C to -9 °C. During depressurization, a 0.2 °C to 0.5 °C temperature drop was observed at the pressure/temperature (P/T) stability curve due to exothermic hydrate dissociation, confirmed by visual observation of melting at the start of the temperature drop. The "memory effect" was observed after repressurizing to 5 MPa from 2 MPa. This experimental design allows the monitoring of pressure, temperature, and morphology of the droplet over time, making this a suitable method for testing various additives and substrates on hydrate morphology.

Introduction

Gas hydrates are cages of hydrogen-bonded water molecules that trap guest gas molecules via van der Waals interactions. Methane hydrates form under high-pressure and low-temperature conditions, which occur in nature in the subsurface sediment along continental margins, under Arctic permafrost, and on other planetary bodies in the solar system¹. Gas hydrates store several thousand gigatons of carbon, with important implications for climate and energy². Gas hydrates can also be hazardous in the natural gas industry because conditions favorable for hydrates occur in gas pipelines, which can clog the pipes leading to fatal explosions and oil spills³.

Due to the difficulty of studying gas hydrates in situ, laboratory experiments are often employed to characterize hydrate properties and the influence of inhibitors and substrates⁴. These laboratory experiments are performed by growing gas hydrate at elevated pressure in cells of various shapes and sizes. Efforts to prevent gas hydrate formation in gas pipelines have led to the discovery of several chemical and biological gas hydrate inhibitors, including antifreeze proteins (AFPs), surfactants, amino acids, and polyvinylpyrrolidone (PVP)^5,6. To determine the effects of these compounds on gas hydrate properties, these experiments have employed diverse vessel designs, including autoclaves, crystallizers, stirred reactors, and rocking cells, which support volumes from 0.2 to 10⁶ cubic centimeters⁴.

The sessile droplet method used here and in previous studies^{7,8,9,10,11,12} involves forming a gas hydrate film on a sessile droplet of water inside a pressure cell. These vessels are made of stainless steel and sapphire to accommodate pressures up to 10-20 MPa. The cell is connected to a methane gas cylinder. Two of these studies used the droplet method to test AFPs as gas hydrate inhibitors compared to commercial kinetic hydrate inhibitors (KHIs), such as PVP^7,11. Bruusgard et al.⁷ focused on the morphologic influence of inhibitors and found that droplets containing Type I AFPs have a smoother, glassy surface than the dendritic droplet surface without inhibitors at high driving forces.

Udegbunam et al.¹¹ used a method developed to assess KHIs in a previous study¹⁰, which allows for the analysis of morphology/growth mechanisms, the hydrate-liquid-vapor equilibrium temperature/pressure, and kinetics as a function of temperature. Jung et al. studied CH₄-CO₂ replacement by flooding the cell with CO₂ after forming a CH₄ hydrate shell⁸. Chen et al. observed Ostwald ripening as the hydrate shell forms⁹. Espinoza et al. studied CO₂ hydrate shells on various mineral substrates¹². The droplet method is a relatively simple and cheap method to determine the morphologic effect of various compounds and substrates on gas hydrates and requires small amounts of additives due to the small volume. This paper describes a method for forming such hydrate shells on a droplet of water using a stainless-steel cell with a sapphire window for visualization, rated up to 10 MPa working pressure.

Protocol

1. Design, validate, and machine the pressure cell.

1. Design the cell to allow direct visualization of hydrate formation from a water droplet. Ensure that the cell has a main chamber with a see-through sapphire window and four ports for fluid/gas inlet, outlet, light, and wires (Figure 1). Create the final design in engineering design software (Supplemental Figure S1).
2. To check that the pressure cell is safe under working high pressure, conduct a finite element analysis using simulation software.
   1. Input the full-size pressure cell model from the engineering design software into the simulation software.
   2. Assign a Young's modulus of 400 GPa and a Poisson's ratio of 0.29 to the sapphire window.
   3. For all stainless-steel parts, assign stainless steel 316 with a Young's modulus of 190 GPa and Poisson's ratio of 0.27.
   4. In a step-by-step manner, apply 0 to 1, 2, 3, 4 5, 6, 7, 8, 9, and 10 MPa air pressure to the inside of the cell (Supplemental Video S1 and Supplemental Video S2). Treat each loading step as a static problem by ignoring the time-dependent terms in the governing equations and consider only elastic deformation during pressurization.
   5. Use the direct linear equation solver in simulation software to calculate the stress distribution and the deformation of the cell under various pressure conditions (Supplemental Table S1 and Supplemental Table S2).
3. Once the pressure cell design is verified to be safe, have all parts machined based on the engineering design software blueprint.

2. Assemble the pressure cell (Figure 1).

1. Screw the four National Pipe Tapered (NPT) threads into the respective ports on the pressure cell with plumber's tape.
2. Assemble the illumination port using the blueprint design (Supplemental Figure S1, parts C, D, and E) and connect to the top left NPT screw.
3. Connect the pressure transducer to the top port NPT using the branch tee fitting and port connector fitting.
4. Connect the inlet needle valve in the left side NPT screw using a port connector fitting.
5. Install a pressure seal connector into the right-side port of the pressure cell. Insert three K-type thermocouple wires through the pressure seal connector with 3" of slack inside the cell and 3' slack outside the cell.
6. Polish the stage surface with sandpaper (Supplemental Figure S1, Part F).
7. Insert the thermocouples into the respective holes in the stage so that the tips are flush with the top of the stage. Use a small drop of glue in each hole to fix the thermocouples in place and allow them to dry.
8. Fit the acrylic disc on the back wall of the pressure cell to enhance light reflection. Fit the stage in the pressure cell.
9. Install the sapphire window.
   1. Apply vacuum grease to two static sealing O-rings (one 1" and one 1-1/5"). Fit the O-rings into the grooves around the window hole on the pressure cell.
   2. Insert the sapphire window. Cover the sapphire window with a 2-1/4" rubber washer and screw on the stainless-steel washer (Supplemental Figure S1, Part B) using eight M8 stainless steel screws (Figure 2C).

3. Assemble the equipment in a large fume hood (Figure 2).

NOTE: As methane is a flammable gas under pressure, keep all methane-related tubing and vessels away from heat, sparks, open flame, and hot surfaces. Set all equipment up inside a well-ventilated area (e.g., a fume hood). Don safety glasses and lab coat before working with methane gas.

1. Carefully lift the pressure pump into a fume hood large enough for all the equipment to fit (Figure 2A). Place the pump controller on top of the pump base. Connect the pump controller to the pump and plug it into a power strip.
2. Run a high-pressure-rated 1/4" copper pipe from the regulator on the methane gas cylinder to the fume hood next to the inlet of the pressure pump (Figure 2A,B).
3. Place the data logger next to the pressure pump and set the laptop on the data logger (Figure 2A). Plug both into a power strip. Connect the data logger to the laptop via the data logger USB.
4. On the laptop, install the proper software to control the data logger, camera, and pressure transducer on the pressure cell.
5. Set the aquarium beside the data logger and place non-leaching padding in the bottom of the aquarium to limit vibrations to the pressure cell (Figure 2C).
6. Using a new 1/4" copper pipe, coil the copper pipe twice into an oval to fit in the aquarium, leaving room for the pressure cell to sit inside (Figure 2D). Ensure that the coil does not block the sapphire window in the pressure cell. Elevate the pressure cell in the aquarium to view the sapphire window.
7. Place the circulating chiller on the floor near the fume hood (Figure 2A). Fill the chiller with 50/50 v/v ethylene glycol/water.
   NOTE: As ethylene glycol is hazardous, use appropriate safety attire, including gloves, lab coat, and goggles when pouring.
8. Cut two lengths of a 3/8" (inner diameter) plastic tubing to connect the chiller inlet and outlet to the copper pipe ends in the aquarium. Ensure there will be enough slack for the foam pipe insulation to fit before cutting.
9. Slide the plastic tubing through the foam pipe insulation.
10. Connect the insulated plastic tubing from the inlet and outlet on the circulating chiller to the ends of the copper coil inside the aquarium. Secure the seals by wrapping plumber's tape around the metal parts and tightening the connections with worm drive hose clamps. Turn the chiller on and set it to circulate at high speed. Ensure there are no leaks.
11. Apply underwater sealant around the copper coil/plastic tubing connections inside the aquarium. Allow the sealant to cure. Wrap the sealant with duct tape.
12. Install pressure pump tubing (Figure 2E).
    NOTE: Always hand-tighten connections before using tools and never detach the NPT connections with plumber's tape because they will not re-seal well.
    1. Install a 1/8" stainless steel pipe on either side of the pressure pump with the company fittings that came with the pump using plumber's tape (Figure 2F).
    2. With a tube bender, bend the 1/8" pipe forward at a 90° angle, approximately 2" away from the pump, to avoid bending at the connection.
    3. With a tube bender, bend the 1/8" pipe downward at a 90° angle, approximately 2" away from the first bend.
    4. Attach 1/8" to 1/4" adapter fitting to the 1/8" pipe on both sides (Figure 2G).
    5. Attach 1/4" pipe to adapter fitting on both sides.
       NOTE: To affix the valve to the side of the pump, trim the 1/4" tubing so that the attached valve will sit next to the two screw holes.
    6. Install the 1/4" needle valves (Figure 2H). If affixing valves to the pressure pump, machine a steel or plastic plate with two 1/16" holes for screws and one 1/2" hole to secure between needle valve connections. Insert the plate between the valve connections and screw the plate to the side of the pump.
       NOTE: Ensure that arrows on the needle valves point from high pressure (inside the pressure pump) to low pressure (outside the pressure pump).
    7. Connect one end of the 1/4" braided stainless steel flexible pressure-rated hose to the outlet valve on the pressure pump and the other end to the side valve of the pressure cell.
    8. Connect thermocouples from the pressure cell to data logger channels using the data logger multichannel. Connect an additional thermocouple wire to measure the temperature of the tank solution and put the other end in the tank.
    9. Connect the pressure transducer on the pressure cell to the laptop.
    10. Set the pressure cell inside the aquarium, close to the front, for clearer imaging.
13. To insulate the aquarium, wrap the outside of the aquarium with foil-lined fiberglass, with a hole/slit for the camera to view the sapphire window of the pressure cell. Cover the top of the aquarium with insulating material to prevent evaporation during experiments.
    NOTE: Avoid tightly sealing the aquarium top to avoid the buildup of heat from the light source.
14. To prevent the condensation of moist air on the front of the aquarium, run plastic tubing from the closest air valve to the front of the aquarium where the camera will be pointing so that the tubing will not be visible in the photographs.
15. Set the light source unit beside the aquarium and plug it into the power strip.
16. Set the camera in front of the aquarium, with the lens pointing towards the sapphire window. Plug the camera into the laptop and power strip.
17. Elevate all electronics from the hood surface to prevent potential leak damage. Double-check that power is distributed for the power capacity of the outlets.

4. Leak-test the pressure cell with water.

NOTE: To ensure all connections were sealed properly, leak-test the pressure cell with water any time the cell has been reassembled, especially after disconnecting the NPT screws. This is not necessary after removing the sapphire window or top valve. Water is safer under pressure than gas.

1. Open the pressure transducer software on the laptop and start collecting data at a scanning interval of 1 s.
2. Turn on the pressure pump and controller. Press Pump A on the pressure pump controller to monitor the pressure.
3. If there is pressure in the pump, decrease the pressure by pressing Refill on the pressure pump controller while both the pump inlet and outlet valves are still closed.
4. With both pressure cell valves open, open the pump outlet valve slightly by ~1/16" to slowly release the remaining pressure.
5. If connected, disconnect the 1/4" copper pipe from the inlet valve on the pressure pump.
6. Attach 1/4" flexible tubing to the pump inlet valve using a nut and ferrule set. Place the end of the tubing in a gallon of water.
7. Close the pump's outlet valve and open the pump's inlet valve.
8. Press Refill on the pressure pump controller to fill the pump piston with water.
9. Set the pressure cell in a shallow empty container outside of the aquarium.
10. Purge the air out of the pressure cell until water comes out of the top port and fills the pressure cell completely.
    1. Close the pump's inlet valve and open the pump's outlet valve.
    2. Ensure the valves on the pressure cell are still open.
    3. Set the maximum (max) flow to 100 mL/min: on the pressure pump controller, press Limits; press 3 for max flow; press 1 to set max flow; punch in 100; press Enter.
    4. Press D to reach the previous page.
    5. Set the constant flow rate to 100 mL/min: on the pressure pump controller, press Const Flow; press A for flowrate; punch in 100; press Enter. Press Run.
    6. If water does not come out or if the volume in the piston is insufficient, refill the piston again by closing the pump outlet valve, opening the pump inlet valve with tubing in water, and press Refill. Then, purge the air out by closing the pump inlet valve, opening the pump outlet valve, setting the flow rate to 100, and pressing Run.
    7. Once water comes out of the top port of the pressure cell, check for leaks and tighten any leaking connections. Press Stop. Close the pressure cell outlet (top) valve.
11. Pressurize the pressure cell.
    NOTE: Don safety glasses before pressurizing the pressure cell.
    1. Set the max flow limit to 10 mL/min to prevent fast pressurization of the cell: on the pressure pump controller, press Limits; press 3 for max flow; press 1 to set max flow; punch in 10; press Enter.
    2. Pressurize the cell to 100 kPa: on the pressure pump controller, press Const Press; press A; punch in 100; press Enter. Press Run.
    3. Check for leaks. If there is a leak, press Stop on the pump controller, tighten the leaking components, press Run, and repeat until there are no leaks at 100 kPa. Ensure there are no leaks by closing the pump outlet valve and monitoring the pressure cell's pressure in the pressure transducer software.
       ​NOTE: If the pressure decreases consistently and is not normal fluctuation due to room temperature variation, there is a leak.
    4. Increase the pressure in increments of 50 kPa from 100 kPa to 500 kPa, then in increments of 100 kPa from 500 kPa to 1,000 kPa, and finally in increments of ~1,000 kPa from 1,000 kPa to ~10,000 kPa. Do this by changing the Const Press setting as before. Between pressure settings, close the pump outlet valve and monitor the cell's pressure like before to ensure that the pressure is constant. If the pressure drops, carefully tighten the leaking components.
12. Upon reaching 10,000 kPa, close the pump outlet valve and observe how well the pressure cell holds pressure according to the pressure transducer. As a consistent drop in pressure indicates a leak, tighten connections at a lower pressure, ~1,000 kPa.
13. To depressurize, open the pump outlet valve and set the pressure to 100 kPa. Once the pressure plateaus, slightly open the pressure cell outlet valve.
14. To remove water from the pressure pump, close the pump inlet valve, change the max flow and Const Flow settings to 100 mL/min, and press Run until the pump is empty.
15. Disconnect the 1/4" flexible tubing from the pump inlet. Disconnect the braided stainless-steel hosing from the pressure cell. Open both valves and drain the water. Remove the sapphire window to allow the cell to completely dry.

5. Form a methane hydrate shell on the droplet surface.

1. Prepare the equipment.
   1. Connect the methane cylinder regulator to the pump with the 1/4" copper pipe using a new nut and ferrule set. Ensure that the gas cylinder is closed.
   2. Practice droplet insertion technique.
      1. Glue a flexible tip, such as IV tubing, cut at an angle to the end of the cannula to help direct the droplet toward the sapphire window. Attach a 1 mL syringe to the cannula and pull in the desired volume of deionized water (~50-300 µL). Without the needle valve or sapphire window attached, insert the end of the cannula into the top port and practice expelling the droplet onto the center stage. After practicing droplet insertion, remove the droplet and dry the stage.
         NOTE: In this protocol, 250 µL of deionized water was taken into the syringe.
   3. Reattach the sapphire window and washers with M8 screws. Connect the braided stainless-steel hose from the pressure pump to the pressure cell, and double-check that all connections from the gas cylinder to the pressure cell are tight. Open the pressure cell inlet valve (side valve), and set the pressure cell in the aquarium. Insert a fiber optic light source cable into the pressure cell illumination port.
   4. Add 50/50 ethanol/water (v/v) to the aquarium until it is level with the top of the pressure cell, just below the light source connection. Ensure that the hood flow is turned on. When the solution level falls before future trials in the following weeks, add more ethanol. Replace the solution monthly.
   5. Set the chiller to the temperature that will achieve ~0 °C to 3 °C inside the cell (~-4 °C) and start circulating through coils. Turn on the airflow to the front of the aquarium to prevent condensation on the aquarium surface.
   6. Start a temperature log in the data logger software. Set the scanning interval to 30 s. Wait until the temperature inside the pressure cell is stable at 2 °C (~6-24 h).
2. Add a water droplet into the pressure cell using the camera view on the laptop.
   1. Turn on the light source to ~80%. Open the camera software. In live view, focus the camera lens at the cell's inner chamber. Adjust the light source for best imaging.
   2. Start a new temperature log with a 1 s scanning interval.
   3. If attached, detach the outlet needle valve in the top port of the pressure cell. Attach a 1 mL syringe to the cannula and pull in the desired volume of deionized water (~50-300 µL).
      NOTE: In this protocol, 250 µL of deionized water was pulled into the syringe.
   4. Insert the cannula through the top port until the tip is visible in the camera software in live view mode. Expel the fluid droplet from the syringe over the central thermocouple. Reattach the needle valve.
3. Focus the camera on the droplet in the pressure cell. Begin time-lapse imaging every ~60 s.
4. Open the pressure transducer software on the laptop and start collecting data on the chart and the data log at a scanning interval of 1 s (same as the temperature scanning interval). Wait until the droplet temperature is stable between 0-3 °C.
5. Pressurize the pressure cell to the desired pressure.
   NOTE: Don safety glasses before pressurizing the cell.
   1. Turn on the pump and the controller. Close the pressure pump's inlet valve.
   2. Open the pump's outlet valve and the pressure cell's valves.
      NOTE: The pressure cell inlet valve should always be open.
   3. Tare the pump pressure by pressing Zero on the pressure pump controller. Select Pump A on the pressure pump controller to monitor the pressure.
   4. Ensure that the pressure pump is empty if a different fluid other than methane gas was present in the pump. Do this by setting the max flow and Const Flow to 100 mL/min and pressing Run. Leave it running until the pump is empty. Close the pump outlet valve and open the pump inlet valve.
   5. Open the gas cylinder and set the gas cylinder regulator to 1,000 kPa.
   6. Press Refill on the pressure pump controller. When the pump is full and near 1,000 kPa, close the pump inlet valve and the gas cylinder.
   7. Slightly open (~1/16" turn) the pump outlet valve to the cell. Monitor the pressure cell pressure in the pressure transducer software as the pressure may decrease due to the relatively lower temperature in the pressure cell.
   8. Set the max flow to 10 mL/min: on the pressure pump controller, press Limits; press 3 for max flow; press 1 to set max flow; punch in 10; press Enter.
   9. Set the max pressure to 5,000 kPa: on the pressure pump controller, press Limits; press 1; punch in 5000; press Enter.
   10. Set the constant pressure to 1,000 kPa: on the pressure pump controller, press Const Press; press A; punch in 1000; press Enter. Press Run.
   11. When 1,000 kPa is reached, press Stop on the pump controller and close the pump's outlet valve. Monitor the pressure in the pressure cell to ensure there are no leaks. If the pressure drops, use the liquid leak detector to find the leak at the connections and carefully tighten the leaking components.
   12. If the cell is stable, open the pump outlet and set the Const Press to 2,000 kPa. Press Stop and monitor. If stable at 2,000 kPa, set Const Press to 3,000 kPa. Press Stop and monitor. If stable at 3,000 kPa, set Const Press to 4,000 kPa. Press Stop and monitor. If stable at 4,000 kPa, set Const Press to 5,000 kPa. Press Stop and monitor.
   13. If the pressure is stable, close the pump outlet.
       NOTE: If the pump volume runs out, close the pump outlet and slightly open the pump inlet. Slowly open the gas cylinder and set the gas regulator to 1,000 kPa. Press Refill on the pump controller. When the pump is refilled, close the gas cylinder and the pump inlet. Pressurize the pump to match the pressure cell pressure.
   14. Wait for ~12-24 h for the gas to permeate the droplet.
6. Nucleate the hydrate shell using dry ice.
   1. Switch the time-lapse to take images every 2-5 s.
   2. Add dry ice to the top of the cell until the hydrate shell is seen in time-lapse. If the dry ice slides, affix tape around the top of the cell.
7. Observe the progress of the methane hydrate formation through time-lapse photos for ~2-6 h.
8. Depressurize the cell to 2,000 kPa by opening the pump outlet and setting the Const Press to 2,000 kPa. Note when melting occurs.
   ​NOTE: Bubbling in the sessile droplet may occur due to the escape of the dissolved gas.
9. After ~30 min, repressurize the pressure cell to 5,000 kPa to observe the memory effect. Note when a hydrate shell begins to reform. Allow the shell to form for ~30 min to 2 h.
10. Depressurize the cell by opening the pump outlet and setting the Const Press to 0 kPa. If there is residual pressure in the pressure cell, slightly open the pressure cell top valve by ~1/16".
11. Save the pressure and temperature data as .csv files.
12. Remove the droplet by removing the top pressure cell valve as before and extracting the droplet with the syringe/cannula/IV tube. If there is a concern for contamination between trials, remove the sapphire window and sanitize the stage and replace the vacuum grease. Use a suction cup to remove the sapphire window once the pressure cell has warmed to room temperature.

6. Analyze the data.

1. Open the temperature and pressure .csv files.
2. Make a new spreadsheet. Copy the time and pressure columns from the pressure .csv and the time and temperature from the temperature .csv file into the new spreadsheet.
3. Make a scatter plot with time on the x-axis and two y-axes with temperature and pressure (Supplemental Figure S2).
4. Make two more columns for the hydrate stability curve. In the first column, input the temperatures from 273.15 K to ~279.15 K at 0.1 K intervals. In the second column, calculate the pressure by using formula (1) from Sloan & Koh¹³.
   P [kPa] = exp(a+b/T [K]) where a = 38.98 and b = -8533.80 (1)
5. Make a scatter plot of the hydrate stability boundary, with temperature (K) on the x-axis and pressure (kPa) on the y-axis. Add a second series on the scatter plot with experimental temperature and pressure on the x and y axes, respectively (Figure 4).
6. Note on the graphs where a hydrate shell became visible, according to the time-lapse imaging.

7. Maintain the equipment.

1. Top off the tank solution with ethanol before every trial to replace evaporated ethanol. Completely replace the tank solution monthly.
2. Change the o-rings and rubber washer every 2 months of regular use. 
3. Replace port connections if persistent leaking occurs that is not fixed by tightening.

Representative Results

With this method, a gas hydrate shell on a droplet can be monitored visually through a sapphire window of the pressure cell and via temperature and pressure transducers. To nucleate the hydrate shell after pressurizing to 5 MPa, dry ice can be added to the top of the pressure cell to induce a thermal shock to trigger rapid hydrate crystallization. There is a clear morphologic difference upon dry ice-forced hydrate shell formation. The water droplet transitioned from a smooth, reflective surface (Figure 3A) to an opaque hydrate shell with a slightly dendritic surface (Figure 3B). The addition of 100 µg mL^-1 Type I AFP altered the hydrate morphology by inducing ridged edges along the droplet and protrusions from the top of the droplet (Figure 3C,D).

After the hydrate shell developed for ~1 h, the cell was depressurized to 2 MPa (Supplemental Video S3). During depressurization, there was a 0.2 °C to 0.5 °C drop in temperature near the P/T stability curve¹³ (Figure 4) due to exothermic hydrate dissociation. Hydrate dissociation was confirmed by visual melting through time-lapse imaging at the beginning of the decrease in temperature, noted by stars in Figure 4. After complete hydrate dissociation, we repressurized the cell to observe the morphology and melting temperature with the "memory effect"¹⁴, the phenomenon in which hydrate forms faster after hydrate has already formed in the system (Supplemental Video S4). Upon re-pressurization, a hydrate shell reformed within a couple of minutes after reaching 5 MPa, and we observed the same temperature decrease at the stability curve during dissociation.

Negative controls with no droplet and with a droplet that did not form a hydrate shell (Figure 4, Trials 4 and 5) showed no decrease in temperature during depressurization. Upon depressurization below 2 MPa, we observed gas bubbling within the droplet from rapid degassing. Because the apex of each temperature decrease was above the previously established P/T stability curve¹³ (hydrate stability curve #1 in Figure 4), a regression curve was calculated based on the apex P/T of these trials (P [kPa] = EXP(38.98+-8533.8/T [K]), hydrate stability curve #2 in Figure 4).

Figure 1: Pressure cell. The stage on which the droplet sits and the embedded thermocouples are revealed by removing the sapphire window and overlying rubber and steel washers. All parts and connections are labeled. Top left inset: stage shown from above with central and side stage embedded thermocouples. Please click here to view a larger version of this figure.

Figure 2: Methane hydrate experimental setup. (A) The fume hood in which the experimental setup is located. (B) The gas cylinder is connected via a copper coil to the pressure pump. Highlighted from panel (A) are (C) the assembled pressure cell, (D) the 10-gallon (37.85 L) tank without the insulation or solution, (E) the pressure pump, and (F, G, H) zoomed-in images ofpressure pump connections. Please click here to view a larger version of this figure.

Figure 3: Methane hydrate shells. Representative images of the droplet before (A) and after (B) a methane hydrate shell formed on a deionized water droplet and before (C) and after (D) a hydrate shell formed on a droplet containing 100 µg mL^-1 Type I antifreeze protein. Scale bars = 5 mm. Please click here to view a larger version of this figure.

Figure 4: Pressure-temperature stability diagram. Pressure and temperature data during depressurization are shown with P/T stability curves of methane hydrate (#1 from Sloan and Koh 2007¹³ and #2 calculated from taking a regression curve from hydrate melting peaks from this study). Trials with successfully formed hydrate shells on DI water droplets are Trials 1, 2, and 3. Trial 4 was a negative control with no droplet on the stage. The droplet in trial 5 was another negative control in which no hydrate shell was formed. Stars indicate when visual hydrate melting began during depressurization. Trial 1 has a resolution of 30 s (a data point every 30 s); other trials have a resolution of 1 s. Abbreviations: T = trial; M.E. = memory effect; P/T = pressure-temperature; DI = deionized; res = resolution. Please click here to view a larger version of this figure.

Supplemental Figure S1: CAD images for machining the pressure cell. Parts A-F of the pressure cell are labeled with their part letter and dimensions. Abbreviation: CAD = computer-aided design. Please click here to download this File.

Supplemental Figure S2: Pressure and temperature data over time for Trials 2-4. Trials 2 and 3 were regular deionized water droplets that formed hydrate shells. Trial 4 was a negative control in which no droplet was present. The trials are lined up at the first depressurization, which occurs at time zero. A small drop in temperature occurs at the beginning of depressurization due to the gas mixing with the pressure pump. A larger temperature drop occurs due to the hydrate melting after the initial pressure drop, as shown in trials 2 and 3. The temperature fluctuation at the end of trial 4 is due to the opening of the valve leading to complete depressurization, which also occurs at the end of trials 2 and 3. Please click here to download this File.

Supplemental Table S1: Allowable stress (MPa) of the machined pressure cell. Abbreviation: FS = factor of safety. Please click here to download this Table.

Supplemental Table S2: Factor of safety for the machined pressure cell. Abbreviation: FS = factor of safety. Please click here to download this Table.

Supplemental Video S1: Strain. Video of the strain simulation on machined pressure cell. Please click here to download this Video.

Supplemental Video S2: Stress. Video of the stress simulation on machined pressure cell. Please click here to download this Video.

Supplemental Video S3: Trial 3 of hydrate shell dissociation. Time-lapse video of hydrate shell dissociation at 25x speed. Please click here to download this Video.

Supplemental Video S4: Trial 3 of memory effect nucleation. Time-lapse video of hydrate shell formation by memory effect after repressurizing from 2 MPa to 5 MPa at 10x speed. Please click here to download this Video.

Discussion

We have developed a method to form methane hydrate shells on sessile water droplets safely and share this method to machine and assemble a pressure cell rated to 10 MPa working pressure, as well as the pressurizing and cooling systems. The pressure cell is outfitted with a stage for the droplet containing embedded thermocouples, a sapphire window for visualizing the droplet, and a pressure transducer fixed to the top of the cell. The cooling system includes chilled ethylene glycol circulating through copper coils in a tank with 50% ethanol solution, in which the pressure cell is placed. A pressure pump pressurizes the gas from the cylinder to the pressure cell. The hydrate shell forms upon rapid temperature decrease with the addition of dry ice to the top of the pressure cell. We allow the shell to form for 2 h, during which we believe the gas permeates through stochastic cracking of the hydrate shell, and Ostwald ripening over a longer period. Indeed, this device could be used to study these phenomena.

The critical steps for this protocol include: 1) leak-test the pressure cell with water before pressurizing it with gas, 2) practice adding the water droplet onto the stage before inserting the sapphire window, 3) cool the droplet to be stable at ~2 °C before pressurizing, 4) pressurize with a max flow rate of 10 mL min^-1 to 5 MPa in 1 MPa increments, 5) close the outlet valve on the pressure pump (to the cell) to limit gas exchange with the pressure pump, 6) set the temperature, pressure, and time-lapse software to log every 1 s, 1 s, and 5 s (or less), respectively, before adding dry ice, 7) apply dry ice to the top of the cell continuously until a hydrate shell is observed in the time-lapse, 8) allow the hydrate shell to form for at least 1 h, 9) depressurize at the same speed as pressurizing.

During method development, we optimized variables and techniques, including the timing of cooling, pressurizing, depressurizing, droplet size, and the droplet insertion technique. There are a few limitations in using this method. One limitation is the resolution of droplet imaging due to the camera resolution and materials between the camera and droplet (tank, ethanol solution, thick sapphire window). Additionally, while other studies observe the surface droplet on a microscale^7,9,10, this method only allows for macro-scale observations. A microscope lens attachment could be installed if there was interest in micro observations.

Another limitation to this method is not being able to measure the hydrate shell thickness precisely. However, the hydrate thickness can be estimated by subtracting the cross-sectional area before and after hydrate formation and calculating the gas consumption using the change in temperature during depressurization to determine the volume of hydrate formed. Another limitation is that this droplet cannot be viewed in 3D because there is only one side of the pressure cell containing a sapphire window. In contrast, other studies have used cells made entirely of sapphire to observe the droplet from multiple angles⁷. We also did not install a temperature-controlling stage¹⁰ or spectroscopic techniques; however, these could certainly be installed using this setup.

With this method, the morphology, dissociation pressure and temperature, and the change in temperature during hydrate dissociation can be observed with droplets containing additives or alternative stage substrates. This method is relatively cheap, and there are few thorough protocols for forming gas hydrate shells. Because high-pressure systems can be dangerous, we include safety tips for pressurizing and leak testing. Additionally, many setups do not allow the visualization of gas hydrate formation, or do so on a much smaller or much larger scale. Laboratory experiments are a major contributor to the understanding of naturally occurring gas hydrates and natural gas hydrates that can cause lethal gas pipeline explosions. This method can be used to quickly assess the effects of additives on the dissociation temperature and morphology and the ability of additives to eliminate the memory effect. Effective additives could be used as inhibitors in natural gas pipelines or to study the biological activity of deep-sea bacterial proteins^6,15.

Materials

Name	Company	Catalog Number	Comments
CAMERA AND LAPTOP
Camera Body	Nikon	D7200	Name in Protocol: camera
Camera Control Pro 2 Software	Nikon		Name in Protocol: camera software
Laptop	HP Pavilion	hp-pavilion-laptop-14-ce0068st	Needs to be PC with plenty of storage (~ 1 Tb) Name in Protocol: laptop
Macrophotography Lens	Nikon	AF-S MICRO 105mm f/2.8G IF-ED Lens	Name in Protocol: lens
CONSUMABLES
Deionized water			Name in Protocol: DI water
Dry Ice	VWR or grocery store		Buy just before nucleation Name in Protocol: dry ice
Ethanol			Name in Protocol: ethanol
Ethylene Glycol			Name in Protocol: ethylene glycol
COOLING SYSTEM
1/2 in. O.D. x 3/8 in. I.D. x 25 ft. Polyethylene Tubing	Everbilt	Model # 301844	For circulating coolant from chiller to copper coils in aquarium Name in Protocol: 3/8” (inner diameter) plastic tubing
Circulating chiller	Polyscience		Name in Protocol: chiller
Economical Flexible Polyethylene Foam Pipe Insulation	McMaster-Carr	4530K162	3/4" thick wall; 1/2" inner diameter; R Value 3; 6' long Name in Protocol: foam pipe insulation
Plastic tubing			use any tubing that fits the airline connection in the lab and long enough to travel from the airline connection to the front of the aquarium
DATALOGGER
Armature Multiplexer Module for 34970A/ 34972A, 20-Channel	Keysight Technologies	34901A	Name in Protocol: datalogger multichannel
Benchvue or Benchlink software	Benchvue or Benchlink		Name in Protocol: temperature transducer software
Data Acquisition/Switch Unit. GPIB, RS232	Keysight Technologies	34970A	Name in Protocol: datalogger
USB/GPIB interface	Keysight Technologies	82357B	Name in Protocol: datalogger USB
datalogger multichannel
Schott Fostec -Llc 20510 Ace Fiber Optic Light Source	Schott Fostec	A20500	3115PS-12W-B20 115 V ~AC 50/60Hz 5/4.5 W Name in Protocol: light source unit
Schott Fostec light source guide - single bundle	Schott Fostec	A08031.40	Name in Protocol: fiber optic light source cable
METHANE GAS AND REGULATOR
1/4 OD in. x 20 ft. Copper Soft Refrigeration Coil	Everbilt	Model # D 04020PS	For pressurizing ISCO pressure pump. An additional pack is needed for coolant circulation, as listed below. Name in Protocol: high pressure-rated 1/4” copper pipe
Methane cylinder regulator	Airgas	Y11N114G350-AG	Name in Protocol: methane cylinder regulator
Methane gas cylinder	Airgas	ME UHP300	Name in Protocol: methane gas cylinder
PRESSURE PUMP
1/4 in.  flexible tubing, ~ 3 ft.			Connect to pump inlet for leak test Name in Protocol: 1/4"  flexible tubing
260D Syringe Pump W/Controller	Teledyne Instruments Inc.	67-1240-520	Name in Protocol: pressure pump
Controller − Ethernet/USB	Teledyne Instruments Inc.	62-1240-114	Purchase if you would like to install Labview onto computer and control pressure pump remotely. We did not do this.
Smooth-Bore Seamless 316 Stainless Steel Tubing, 1/4" OD, 0.035" Wall Thickness, 1 Foot Long (x5)	McMaster-Carr	89785K824	Name in Protocol: 1/4" pipe
Smooth-Bore Seamless 316 Stainless Steel Tubing, 1/8" OD, 0.02" Wall Thickness, 1 Foot Long (x4)	McMaster-Carr	89785K811	Name in Protocol: 1/8" pipe
Stainless Steel Swagelok Tube Fitting, Reducing Union, 1/4 in. x 1/8 in. Tube OD (x4)	Swagelok	SS-400-6-2	Name in Protocol: 1/8” to 1/4” adapter
PRESSURE CELL
316 Stainless Steel Nut and Ferrule Set (1 Nut/1 Front Ferrule/1 Back Ferrule) for 1/4 in. Tube Fitting (20)	Swagelok	SS-400-NFSET	Used for fitting connections where necessary Name in Protocol: ferrule set
316L Stainless Steel Convoluted (FM) Hose, 1/4 in., 316L Stainless Steel Braid, 1/4 in. Tube Adapters, 60 in. (1.5 m) Length	Swagelok	SS-FM4TA4TA4-60	Connects pressure pump to pressure cell Name in Protocol: 1/4" braided stainless steel flexible pressure-rated hose
ABAQUS	ABAQUS FEA		Name in Protocol: simulation software
Abrasion-Resistant Cushioning Washer for 7/8" Screw Size, 0.875" ID, 2.25" OD, packs of 10 (x1)	McMaster-Carr	90131A107	Name in Protocol: 2.25" rubber washer
Abrasion-Resistant Sealing Washer, Aramid Fabric/Buna-N Rubber, 3/8" Screw Size, 0.625" OD, packs of 10 (x1)	McMaster-Carr	93303A105	Used for illumination port
Acrylic Sheet | White 2447 / WRT31 Extruded Paper-Masked (Translucent 55% (0.118 x 12 x 12)	Interstate Plastics	ACRW7EPSH	Machine a circle of acrylic to fit in the inner chamber of the pressure cell to serve as the background for imaging Name in Protocol: acrylic disc
AutoCAD	AutoCAD		Name in Protocol: engineering design software
Conax fitting	Conax Technologies	311401-011	TG(PTM2/)-24-A6-T, OPTIONAL 1/4" NPT Name in Protocol: pressure seal connector
High Accuracy Oil Filled Pressure Transducers/Transmitters for General industrial applications (x2)	Omega Engineering, Inc.	PX409-3.5KGUSBH	Buy two so there is a backup. Name in Protocol: pressure transducer
HIGH PRESSURE CHAMBER  PARTS	Wither Tool, Die and Manufacturing Company		Machining for pressure cell parts as listed in CAD drawings (Figure S1) Name in Protocol: Part B = stainless steel washer
High-Strength 316 Stainless Steel Socket Head Screw, M5 x 0.80 mm Thread, 14 mm Long (x20)	McMaster-Carr	90037A119	Used for illumination port
High-Strength 316 Stainless Steel Socket Head Screw, M8 x 1.25 mm Thread, 25 mm Long (x20)	McMaster-Carr	90037A133	Name in Protocol: M8 stainless steel screws
Oil-Resistant Hard Buna-N O-Ring, 3/32 Fractional Width, Dash Number 120, packs of 50 (x1)	McMaster-Carr	5308T178	Name in Protocol: 1" o-ring
Oil-Resistant Hard Buna-N O-Ring, 3/32 Fractional Width, Dash Number 128, packs of 50 (x1)	McMaster-Carr	5308T186	Name in Protocol: 1.5" o-ring
Omega Inc. pressure transducer software	Omega Engineering, Inc.		Name in Protocol: pressure transducer software
Polycarbonate Disc	McMaster-Carr	8571K31	Listed in CAD drawings for illumination port, Fig. S1 Part E
Sapphire windows (x3)	Guild Optical Associates, Inc.	Optical Grade Sapphire Window, C-Plane Diameter: 1.811” ±.005” Thickness: .590” ±.005” Surface Quality: 60/40 Edges ground and safety chamfered	Buy three so there are two backups. Name in Protocol: sapphire window
Solid Thermocouple Wire FEP Insulation and Jacket, Type K, 24 Gauge, 50 ft. Length (x1)	McMaster-Carr	3870K32	Name in Protocol: thermocouples
Stainless Steel Integral Bonnet Needle Valve, 0.37 Cv, 1/4 in. Swagelok Tube Fitting, Regulating Stem (x4)	Swagelok	SS-1RS4	Two will be used for the pressure pump as well. Name in Protocol: 1/4" needle valves
Stainless Steel Pipe Fitting, Hex Nipple, 1/4 in. Male NPT (x2)	Swagelok	SS-4-HN	Used for illumination port
Stainless Steel Swagelok Tube Fitting, Female Branch Tee, 1/4 in. Tube OD x 1/4 in. Tube OD x 1/4 in. Female NPT (x2)	Swagelok	SS-400-3-4TTF	Used with pressure transducer Name in Protocol: branch tee fitting
Stainless Steel Swagelok Tube Fitting, Male Connector, 1/4 in. Tube OD x 1/4 in. Male NPT (x4)	Swagelok	SS-400-1-4	Used on top port and side port leading to needle valves Name in Protocol: NPT screws
Stainless Steel Swagelok Tube Fitting, Port Connector, 1/4 in. Tube OD (x8)	Swagelok	SS-401-PC	Use as tube connections between NTP and valve connections Name in Protocol: port connector fitting
TANK
1/4 OD in. x 20 ft. Copper Soft Refrigeration Coil	Everbilt	Model # D 04020PS	For circulating coolant Name in Protocol: 1/4" copper pipe
10 gallon aquarium	Tetra		Name in Protocol: 10 gallon tank
2 oz. Waterweld	J-B Weld	Model # 8277	Name in Protocol: underwater sealant
3 in. x 25 ft. Foil Backed Fiberglass Pipe Wrap Insulation	Frost King	Model # SP42X/16	For wrapping around aquarium Name in Protocol: foil-lined fiberglass
3/8 7/8 in. Stainless Steel Hose Clamp (10 pack)	Everbilt	Model # 670655E	Name in Protocol: worm drive hose clamps
Styrofoam			Name in Protocol: insulating material
TOOLS
1-1/8 in. Ratcheting Tube Cutter	Husky	Model # 86-036-0111
1/2 in. to 1 in. Pipe Cutter	Apollo	Model # 69PTKC001
Adjustable wrench (x2)	Steel Core	Model # 31899	Need two wrenches with jaw at least 1"
Allen wrench set	Home Depot
Duct tape			Name in Protocol: duct tape
Flexible tubing, like an IV line, to fit on the end of grainger probe (canula)			Name in Protocol: IV tube
Grainger 18 gauge probe	Grainger		For inserting droplet Name in Protocol: cannula
High Vacuum Grease	Dow corning		Apply to o-rings before inserting sapphire window Name in Protocol: vacuum grease
Klein Tools Professional 90 Degree 4-in-1 Tube Bender	Klein Tools	Model # 89030	Name in Protocol: tube bender
Snoop liquid leak detector	Swagelok	MS-SNOOP-8OZ	To detect leaks when pressurized when methane Name in Protocol: liquid leak detector
Suction cup	Home Depot		For removing tight fitting sapphire window Name in Protocol: suction cup
Teflon Tape			Name in Protocol: plumber's tape
Temflex 3/4 in. x 60 ft. 1700 Electrical Tape Black	3M	Model # 1700-1PK-BB40	Name in Protocol: electrical tape