Skip to main content
SearchLoginLogin or Signup

Moving Liquids With No Moving Parts

Solid-state microfrabricated devices are in development to replace contemporary heavy mechanical components for use in spacecraft propellant management.

Published onOct 07, 2020
Moving Liquids With No Moving Parts


After several parabolas full of fun flips and weightless pushups, the electrowetting flow regulators were successfully activated!

This is the proof-of-concept operation this project sought to demonstrate. Following this, the devices will be optimized and re-designed to flow additional liquids forward and backwards for use in a spacecraft propulsion system analogue.

Many thanks to the Space Exploration Initiative and Zero-G corp for providing this incredible opportunity!

Ready to Fly!

A wide array of devices have been fabricated, the flight box is done, the camera is ready! After months of minor hiccups with nanofabrication tools, we are only a week away from flying and testing these new electrostatic flow regulators. Lab tests confirmed that the selected coatings and microfluidics features work correctly. This means the only mechanism that can force wetting is the electrowetting effect activated by the power supply. We’ll have to wait and see how things go in free fall!

So Will it Work?

Short answer, yes! First trials with the recently fabricated features showed good fluid transport from droplets placed at the tail end of the feature array.

Open microfluidic channels wetting with water.

This week will be spent preparing all data, designs, and ConOps for the final critical design review next week. This includes preparing footage of the tests, assembling a prototype visual aid, calculating structural loads on the experiment in an emergency scenario, planning each parabola for the flight, and addressing any potential safety concerns. After months of preparation, this experiment is in prime position to provide some very impactful results!

Putting on the Final Touches

Prototype fabrication of the valve test articles in complete. The power supply has arrived along with the digital microscope. The toggle switches for the control array have also been ordered. This lays the groundwork for assembling the flight experiment. Test imaging with the digital microscope proved very successful. The tiny microscope was able to focus on 100 micron details with ease. Meaning the microfluidic devices will be recorded with sufficient resolution to determine flow rate. A flexible mounting arm with 12 in reach radius was also acquired to easily reposition the microscope on different test devices instead of fabricating a separate structure for alignment.

100 micron diameter dot photographed with the digital microscope

With the prototype wafer finished, lab testing can begin next week with the microscope to see what sort of camera and power supply settings are required prior to the flight. Assembly of the fluid tanks has also begun. At least 6 will be assembled, with the possibility of more being based on the lab testing results of flow impedance. This week’s results were a lot to be thankful for. Happy Thanksgiving!

Filming Small for Big Results

Though micro-capillaries and microfluidics have some fascinating physics that allow for liquids to travel through them, filming them is quite difficult. Add in a constantly pitching aircraft with several passengers floating around, and things become complicated. From the outset, this experiment was planned to have minimal reliance on wireless protocols as well as minimal input from the flyer operating the experiment. Luckily, with the advent of universal connectors being adopted in smartphones, a wired digital microscope with 1000x magnification and a working distance that is within the spec of the flight experiment is not hard to find. With the microscope on its way, the next step will be to run through timed scenarios to determine the best way to reliably position it between each parabola. The original plan of a rigid plate that has connecting points above each device to be tested is in the process of being designed for fabrication.

Wafer with microfluidic features ready for dicing

With the power supply ordered and the final etches complete on the test devices, all that needs to be done now is determine the activation voltages for each device type. Unfortunately, the cleanroom fab facilities are closed this week for mechanical maintenance. Being a scheduling pessimist, this sort of delay was accounted for and we are still on track to get some devices tested before the end of the month and film them in a manner similar to flight!

Testing for Flight

With the final etches complete on the devices to be used for flight, it’s time to prep for testing these little microfluidics out in the lab. Shown in the figure below is the final wafer coming out of the etch tool.

Wafer after long-duration etch

The first step to selecting flight devices and a power supply voltage will be to determine the activation voltage for each device geometry. The goal is to find a minimum voltage where each geometry will activate reliably in atmospheric conditions. This will be the voltage set for the flight experiment. Perhaps the most important part of lab testing will be the filming of the devices. While it is obvious when the channels fill with liquid, it is difficult to capture that motion in high detail. Finding the proper lens/camera pairing will be necessary prior to the critical design review and flight. While the digital microscope on hand has excellent video capabilities, it has a very sensitive focus which may cause issues during flight. Ideally, a camera that has onboard memory and is capable of capturing small details either with a macroscopic lens or large optical zoom will be positioned to capture each device’s operation.

The next few weeks will focus on final coatings of hydrophobic layers on the devices, testing, and video trials. Once these are completed, assembly of the flight experiment box can begin.

Operational Focus

The past week has seen a flurry of fab steps to finalize the flight microfluidic devices. The capillaries that permit initial wetting were etched into the wafers along with the features that will carry the fluid to a set destination. In this case, the features are large enough for video recording. The only remaining steps are to etch the back side of the wafer with large border patterns to expose the capillaries etched from the other side and grow an oxide on the entire wafer before dicing. The most recent capillary etch is shown in the figures below.

SF6 plasma etching the micro-capillaries into the silicon wafer

Capillaries etched into silicon with remaining photoresist layer to be stripped after final etch

The big focus of this weeks work was on operating the experiment in-flight. Given the expected rapid wetting of the microfluidics, minimal human interface is ideal to reduce an actuation procedure times. The switch array mentioned previously will make for a simple on=wet off=dry concept test for each device geometry. To best capture these events in the channels, a higher magnification video option will be needed. The lab currently has a long working distance digital microscope which could provide more intimate video of the fluid control capabilities of the devices. The only change that would be needed is a way to quickly reposition the microscope between parabolas to film the device to be tested. The initial concept to achieve this is a 3D printed plate similar to that of a gear shift frame with slots for each respective device placed below. Mounting a clamping structure to the microscope that can quickly be moved to each device without changing the height, will ensure focus is not lost between each run. An initial CAD drawing and 3D test print will ensue shorty to test on the microscope.

Additionally, the final Research Program Package draft is in the works to get approval from Zero-g for the flight. This entails detailed explanation of the experiment components, operation, safety concerns, and proof of structural integrity during maximum loads the plane could see in an emergency scenario. All told, once the devices are diced and tested in the lab, we just need to put all the ingredients together!

Fab Fab and More Fab

Working with the course admins and TA, the construction of the flight structural box is on hold pending the provision of shareable materials. This will reduce cost and enable the possible expansion of video equipment and data collection! A power supply has also been selected and will be ordered as soon as structural component availability is confirmed. In the meantime, fabrication has been ongoing for the new flow controllers.

The first step for photolithography is creating a pattern to expose into the photoresist on the wafers. Shown in the figure below, many small tiles are arranged around the wafer with varying feature sizes imperceptible to the naked eye. These features, once etched, will create a variety of fluid motion behaviors that will give key insights into spaceflight performance.

CAD drawing of photolithography layers

This pattern has already been exposed into the photoresist on the silicon wafers and is ready for etching later this week. Following the etch, the wafers will be cleaned using potent acids and plasmas to remove any organic material and remaining photoresist prior to being coated in an oxide layer to prevent unwanted electrical breakdown. The individual squares will then be diced using a dicing saw to create single flow controllers and valves ready for lab testing and selection for the big flight!

Prepping Flight Components

After a preliminary design review, the focus for this past week shifted to logistics. This includes creating a detailed schedule of work, milestones, and materials needed. Thus far, structural requirements have been figured and a material list for the necessary goods is complete. In addition to the microfluidic devices, propellant tanks must be assembled to feed the valves. The goal for this past week was simply to source and assemble these small tanks for mounting and use on the flight rig. A teflon inner tank hold the propellant while a harder plastic outer shell provides structure and mounting points. A central wick acts as the primary flow path for the propellant to the valves. These assembled are shown in the figure below.

Non-pressurized propellant tank with a single-use electrowetting valve installed atop.

Ordering of parts begins this week. Unfortunately, a slight delay in new valve fab was incurred due to an issue with a tool for depositing oxides on silicon wafers. This has since been fixed and we are back on track for pushing flight component fabrication! Once the new components are ordered, the flight rig can start being assembled.

Design Review Week!

With all the fab work under way, it’s time to take a step back and ensure all the big-picture elements are in place for the actual zero-g flight. This includes fleshing out the details of operation for each parabola during flight, addressing safety concerns, sourcing materials and cataloging them, planning data collection, finalizing component layout, and looking at the criteria for a successful flight.

80/20 aluminum frame rails and plexiglass will be ordered to form the base structure for the experiment. A mid-level baseplate will serve as a mounting point for all hardware to be tested as well as the power supply and data collection devices. Once on hand, these materials will be easy to assemble and strong enough to handle the 9G load requirements set out for the flight.

CAD mockup of experiment layout. Power supply, microfluidic devices, fluid tanks, and a camera are all enclosed in the metal and plexiglass box.

What’s New #3

Designing the initial flight experiment has begun! This means fitting as much good data gathering equipment, microfluidic devices, and necessary structure into the prescribed plane’s footprint. Thus far, a simple 80x20 aluminum box frame will support a baseplate that can be sized to accommodate the needed components. These components will include: microfluidic devices in a hermetically sealed container with accompanying fluid reservoirs, microcontroller, 100 Volt power supply, switch board, and camera. Final layout will be determined by the dimensions of the power supply and camera setup.

Following the final fabrication of the last flow controller design, it was decided that the extremely dense packed layout suitable for a space mission was not necessary to test the underlying conceptual operation. Therefore, a new flow controller layout has been set to a larger 2cmx2cm square chip size. This will allow for thicker features that are more robust during fabrication and flight. Fabrication of these new flight-oriented test articles has begun. To further improve the photolithographic feature resolution and etch-resistance, an oxide mask will be implemented. This is achieved by thermally growing silicon-dioxide on a bare silicon wafer at approximately 1000 degrees celsius. The images below show the darkened wafers with a recent oxide growth completed in the glowing furnace tube. These wafers will be the base for the next fabrication steps.

Oxide growth furnace glowing at 1000 °C

Wafers with dark oxide layer grown on all surfaces

What’s New #2

Final photolithographic and etch steps were carried out. The primary lesson learned had to do with the energy dispersion of the laser in the photoresist. It would seem the features, while captured in photoresist, were not wide enough to withstand the long duration etch to expose the capillaries. The next step will be to re-design the fluid controlling feature geometries to make a more robust test article and ensure a successful fabrication with the available process equipment.

Final photolithography step. Resist is deposited, exposed, and developed before baking to harden.

Final etch step; the exposed pattern in the resist is etched away.

Initial materials were identified for the flight unit as well. These will include a plexiglass enclosure, non-pressurized fluid tank, power supply, micro-controller, and a structural interface for the various microfluidic design iterations to be controlled by the micro-controller. Code for the microcontroller and CAD for the enclosure and experiment structure are now underway. Lastly, with the ability to control liquid in zero-g, there must be some artistic ways to demonstrate these devices. More to come on that. In the meantime, how to best capture the device operation is being sorted out.

What’s New #1

With a strong backing of nanofabrication trials and fluid modeling out of the way, construction of the various flow controller test articles is under way!

Oxide deposition on silicon wafer

All fabrication begins with pure silicon wafers. Thus far, all photolithographic masking procedures were completed to cut the desired fluid-controlling features and geometries into the underlying silicon.

Plasma etching of features into the silicon wafer

Next up, the final photolithographic and etching steps will be completed prior to dicing the wafer into individual flow controllers and valves for testing. Once the best candidates are found, they will be incorporated into a experimental apparatus to be flow on the zero-g flight!


The booming market of smaller, cheaper, and lighter CubeSats has created the need for equally small propulsion systems. Electric propulsion is a very efficient option for most large spacecraft, but small spacecraft typically lack the power needed to operate them. MIT’s Space Propulsion Lab has developed the ion Electrospray Propulsion System to address this operational regime to make CubeSats viable platforms for numerous missions that traditional chemical propulsion cannot enable. To accompany these lightweight and low-power propulsion systems, propellant control must also be low mass and require minimal power. Recent developments in microfabrication techniques have shown that electrostatic fields can effectively initiate the flow of propellant via wicking materials and capillary tubes with no need for heavy mechanical valves and pressure vessels that require more power. These devices are being built

Ion Electrospray Propulsion System

Solid-state electrowetting valve with drop of water before activation

for testing in a zero-gravity environment to conclusively show their operational capabilities for future spacecraft.

No comments here
Why not start the discussion?