3D Printing a Mini Vacuum Pump

Andrew Corselli

Mass spectrometers are extremely precise chemical analyzers that have many applications. But building an inexpensive, portable mass spectrometer that could be deployed in remote locations remains a challenge — mainly due to the difficulty of miniaturizing at a low cost, its necessary vacuum pump.

Now, MIT researchers have utilized additive manufacturing to take a major step toward solving this problem. The team 3D printed a mini version — about the size of a human fist— of a peristaltic pump. The pump can create and maintain a vacuum that has an order of magnitude lower pressure than a dry, rough pump, it doesn't require liquid to create a vacuum, and can operate at atmospheric pressure.

"We are talking about very inexpensive hardware that is also very capable," said senior author Luis Fernando Velásquez-García. "With mass spectrometers, the 500-pound gorilla in the room has always been the issue of pumps. What we have shown here is groundbreaking, but it is only possible because it is 3D printed. If we wanted to do this the standard way, we wouldn't have been anywhere close."

"One of the key advantages of using 3D printing is that it allows us to aggressively prototype," added Velásquez-García. "If you do this work in a clean room, where a lot of these miniaturized pumps are made, it takes a lot of time and a lot of money. If you want to make a change, you have to start the entire process over. In this case, we can print our pump in a matter of hours, and every time it can be a new design."

Here is a Tech Briefs interview with Velásquez-García, edited for length and clarity.

Tech Briefs: What inspired your research?

Velásquez-García: The background that I have is in microelectromechanical systems (MEMS), which is hardware that you make harnessing miniaturization. People are familiar with, let's say the electronics in smartphones and computers, but in addition, there are things like the MEMS accelerometers in phones.

The story, more or less, is that the great majority of this hardware is done in a semiconductor clean room, which is something that is really good for one thing —making semiconductor electronics. So, you are very restricted in the kind of structures and the materials that you can use. The structures look like sandwiches — they’re layered and they’re mostly two-dimensional, and there are just so many materials.

I got interested in using 3D-printed microsystems because 3D printing has shown to be a toolbox that greatly expands the capabilities of what we can make.

That's one part of the story. The second part of the story is the pump. I have been working on mass spectrometry for about 15 years or so, and it has been like the holy grail; a lot of the industries that we have around — like the food industry, pharmaceuticals, oil exploration — rely on mass spectrometry to quantitatively determine the composition of a sample.

Let's say you create a medicine; you need to check that you’re actually creating what you think you’re creating. Or when you’re doing a geological survey, you want to make sure you found the material you thought you found. So, there is a technique you can use called mass spectrometry. The state of the art of these devices is very heavy, very large, very expensive hardware. So, you have to ship the samples to laboratories, and that has a cost.

If you, instead, tried shipping a mass spectrometer, for example, to an oil well; it could shatter. So, what people like me and many others have been trying to do is miniaturizing mass spectrometers. Mass spectrometers require vacuum to operate, so the main challenge is the lack of adequate vacuum sources. Mass spectrometers work by creating a vacuum, but they have to interface with atmospheric pressure. There is not one pump technology that can do the whole range from atmospheric pressure to the high vacuum that you need.

The device we created and that we wrote a paper about, is the first stage. It works with atmospheric pressure, while lowering the torr (pressure) level. And then once you are there, there are other technologies that you can use to further lower the pressure.

Tech Briefs: What were the biggest technical challenges you faced throughout the work?

Velásquez-García: Oh, everything [laughs]. To begin with, for these mass spectrometers, you need clean technology. There are some technologies that use, for example, an oil or some kind of liquid to create the vacuum, and they have problems of contamination. They basically would impregnate everything with that material. So, if you are trying to find a chemical composition of something, you don't want that. The usual solution is something called a diaphragm pump; they look like accordions and have a chamber that expands and contracts. That's how they move pockets of gas from rarified conditions to the atmosphere. That has been extensively studied.

There are a number of particular issues — if you want to get a decent vacuum, you need to stage the pump. Instead of using one, you need to use three, four, five in series, so there's a lot of complexity in that.

So, we adopted a peristaltic pump, which is nature-inspired. (Peristalsis is the motion with which our body processes food.) That type of pump is used in many applications where you want the material to be kept inert, either because you don't want to contaminate it or because it's very reactive. But they have never been useful for making a vacuum. So, the first thing we had to do is to come up with a design that would work.

The reason it hasn't been used before is related to the way they are made — it's just a simple circular tube. But if you engineer that tube you can come up with something that works very well. That was one part, and the second part was how to make it into a working prototype. That required a lot of development of 3D-printing technology.

And finally, even the testing was challenging because this pump is so small and many of the instruments we have available are designed for a significantly larger pump. So, we had to play all kinds of tricks to do it.

Everything was challenging, but it was very rewarding.

Tech Briefs: Can you explain in simple terms how the technology works?

Velásquez-García: These pumps are based on compressing pockets of rarified gas. Imagine that you have a chamber that has rarified gas inside. Somehow you take a volume of that and compress it until it reaches atmospheric pressure and then you release it. And with that you create and maintain a vacuum.

What I described is basically a positive displacement pump, although this pump is a kind of like that, its advantage is that it's not limited by something called the dead volume. In your typical positive displacement pump, there is a minimum volume and a maximum volume of the compression chamber. That ratio pretty much sets how low they can go.

But because of the way this pump operates, it overrides that limit. What we showed with hard models and with experiments, was that you can pretty much reach any vacuum you want, down to probably millitorrs, which is when the assumptions you make about the fluid break down. When you have it under a high pressure like that you assume the gas is a continuum, as in everyday experience. But, when you are in the millitorr range, the gas behaves as individual molecules moving around — what is called molecular flow — it has a granularity. That makes the fluid behave very differently.

So, I think what is really cool about this is that it's a low-cost technology that can drive very low vacuums, beyond what any other positive displacement pump can do. And that it is a dry vacuum, so you can use it for chemical analysis.

Tech Briefs: What is your current work/research and what are your next steps?

Velásquez-García: The part that is in standby is the issue of heating. The heating is important because, at the end of the day, what you want is a pump that can operate for long periods of time. Vacuum systems are on for years and years — in my lab I have pump systems that have been on for 15, 20 years; they’re meant to last. So, the issue of the lifetime is very important. For this particular pump, you need to take care of the heating because that's the mechanism by which the tube degrades.

What you would like is to actuate the pump as fast as possible to reach the lowest vacuum. But if you actuate it fast, it heats up. So, we want to have a solution for being able to actuate fast and then cool down, perhaps by having certain printed parts or having active cooling, like having channels of gas or water. So that is in standby. It's a good idea, but we haven't done it.

The other part that is going very well, and without spilling the beans, I can tell you that there are two other key parts to make a mass spectrometer. We have the pump, which is the hardware that gives the conditions for mass spectrometry to happen. Mass spectrometers cannot operate at atmospheric pressure because when you realize how the particles move in electromagnetic fields unperturbed, if you have pressure, like atmospheric pressure, the particles would collide with the background gas.

The other part, which we already did is an ionizer. It's a device that takes, let's say, a blood sample, and creates ions, which is a version of the material that can be analyzed by the mass spectrometer. So, you need to ionize it — to give it a charge so the particles can be influenced by an electromagnetic field. The ionizer we have is working really, really, well.

The other key part is a mass filter — hardware that sorts out the ions based on electromagnetic fields and based on mass-to-charge ratio —how much mass per unit of charge you have.

That is a physical process, which is why mass spectrometry is so powerful, because it's not based on chemistry, it's based on physics. It's very hard to trick a mass spectrometer. Our version of the mass filter is something called a quadrupole. And, similarly to the pump, the quadrupole is also 3D printed, but 3D printed in ceramic. The ionizer is also 3D printed in some other materials. And they work great.

We’re having some conference papers about that this summer. But I can tell you that these devices work as well as commercial hardware, and they cost a lot less, and they take a lot less time to make. One of our mass filters in quadrupole can be made in about a day and costs on the order of tens of dollars. If you go to buy a commercial mass filter in quadrupole, it probably costs 20 to 30 thousand dollars and takes weeks to machine because it's precision machining. The vision of making a fully 3D printed mass spectrometer is really at hand. My hope is that we can show that within a couple of years — so far so good.

Dry Vacuum Pump

Miniaturized Vacuum Pumps Play Big Roles on Mars and Earth

Tech Briefs: Do you have any advice for engineers aiming to bring their ideas to fruition?

Velásquez-García: I would say that it's a good practice to really question everything. I think a lot of the problems we have in engineering come from grandfathering certain decisions; but things need to be revisited every time there is a new set of rules. For example, every time you have technologies — printing, artificial intelligence, or whatever it is — you should really go down to basics and see how it changes things. Because there are many examples of how we do things in a certain way because they’re limited. For example, most of the manufacturing these days is subtractive, where you remove material from a block to come up with a part.

Another thing is to try to standardize, so for example, you might have a huge inventory of beams of different diameters and plates of different thicknesses. But what if we don't need to do that? What if we don't need to have a stock of a gazillion things and we can make things on the go? That's basically the idea of 3D printing and how powerful it would be for things like in-space manufacturing. It's terribly expensive to send anything to space.

My advice would be don't be shy of challenging the status quo, and also, look for surprising solutions. I think a lot of problems need to be revisited because whatever decision was made back then, might not be true anymore. Engineering should be a constant inquiry for finding better solutions, and we shouldn't leave any stone unturned.

Tech Briefs: Is there anything else you’d like to add?

Velásquez-García: What my group wants to do is to create hardware that has a deep impact and that usually happens by complex functions or multifunctions. There are people that are trying to figure out things like, how can we use 3D printing to replace injection molding. Well, that's a possibility, but what we think will be more impactful is if we say something like: "What would it require us to print, without assembling, in one shot, a smartphone — but not a crappy smartphone, a smartphone that is as good, if not better than, what can be done right now."

We will need to print the batteries, the electrical connections, the transistors, the capacitors, the display, what we call the package — the interface between the device and the exterior. That necessarily happens through the idea of multi-material 3D printing; that has been our focus for a while. Unfortunately, there are just a few technologies that can print multi-materials. In particular, there is one called extrusion, and another called direct-ink-writing, which is basically like using pens to create traces. And then there is something we pioneered, called microplasma sputtering, which has a huge potential because it can create complex hardware in multiple materials with fine resolution.

We can make features that are very small, and that is really the key for that resolution. So, I think there are so many things you can do. For example, I can make a part that maybe is the blade of a turbine. But what if that blade could have electronics there — it could be sensing, it could have actuation — and all that is possible if you involve the right materials.

If you look at a cross section of your hand, for example, you see that there is a stuff that is hard — the bones — and there is stuff that expands and contracts — the muscles. So, the idea that printing with one material doesn't have legs. But the idea of multi-material printing is incredibly challenging because you need to come up with ways to print very dissimilar materials in similar conditions.

So, for example, naively you might say, it's really hard to print something out of metal and something out of polymer because they require such different temperatures for processing. That's true if you use the standard technologies, but if you use something like microplasma sputtering, you can deposit high quality metals at room temperature, and you are able to create them in polymer.

It's going to be very challenging, but I think it's going to be very rewarding because of the kinds of things that we can implement.