WEBVTT 1 00:00:02.690 --> 00:00:24.960 Mark Kushner: Welcome to this week's multi seminar. It's my pleasure to introduce Dr. Elaine Petrol. Our is our seminar speaker. Elaine is an assistant professor and mechanical and aerospace engineering at Cornell University, and director of the Astral Lab, which is focusing on sustainable space propulsion. 2 00:00:24.970 --> 00:00:37.020 Mark Kushner: Elaine received her Vs. Ms nphd. And aerospace engineering at University of Maryland, where her Phd. For thesis focused on water plasmas for propulsion. 3 00:00:37.920 --> 00:00:44.460 Mark Kushner: She was a Postdoc researcher at Mit, studying electro spray thrustered technology 4 00:00:44.690 --> 00:01:00.359 Mark Kushner: and acknowledge meant of her accomplishments. She has received the 2,023. If it was our young investigative program award with name, an Arcs scholar, and with the National Science Foundation and Amelia Earhart fellow. 5 00:01:00.570 --> 00:01:08.589 Mark Kushner: and finally was also recognized as one of aviation week and space technology, twenty-something emerging leaders 6 00:01:09.140 --> 00:01:18.010 Mark Kushner: prior to graduate studies at University of Maryland, elaine worked on the Maven Mars Orbiter, and Jean's Web Telescope and projects. 7 00:01:18.500 --> 00:01:24.799 Mark Kushner: Elaine is a founding member of the National women of Aeronautics, Astronautics organization. 8 00:01:25.080 --> 00:01:26.190 Mark Kushner: Then 9 00:01:27.120 --> 00:01:40.260 Mark Kushner: we very much appreciate you coming and making the the trek, and we present present you with the famous Nipsy nug. Oh, and we need to record set event. 10 00:01:42.110 --> 00:01:52.549 Mark Kushner: Thank you so much for my favorite bug, Chicago and related presentation and field modeling molecular on engagement and 80 surface interactions. 11 00:01:52.840 --> 00:01:54.150 Mark Kushner: Thank you so much. 12 00:01:54.470 --> 00:02:03.159 Mark Kushner: Alright. Thank you all for coming to my talk and for inviting me here. It's really been a pleasure so far to speak with everyone that I've 13 00:02:03.350 --> 00:02:17.980 Mark Kushner: met today, and to see such a vibrant plasma, physics and engineering community at am I at Michigan, and be here to to learn from you all and talk to you about some of the problems that we're trying to solve in our research group. 14 00:02:18.000 --> 00:02:32.869 Mark Kushner: so with that, I will jump into the presentation. I'm going to talk mostly about the modeling work, but if we have time I might talk a little bit about experiments that we have ongoing as well. And just to see 15 00:02:35.060 --> 00:02:36.950 Mark Kushner: that's my sighted answer. 16 00:02:37.420 --> 00:02:45.500 Mark Kushner: That wanna work just to keep track of time we have until 40'clock. Is that right. But I should end a little early for questions. 17 00:02:46.190 --> 00:02:50.030 Mark Kushner: Sorry? Or we're earlier than that. Florida. 18 00:02:50.080 --> 00:03:08.350 Mark Kushner: Okay. 19 00:03:08.350 --> 00:03:20.509 Mark Kushner: talking a lot about the modeling work that we've been doing in my group over the last couple of years. I really love doing experiments and the experimental part of understanding processes and validating models. 20 00:03:20.510 --> 00:03:50.420 Mark Kushner: 2 pictures here of this is of setting up our when our first vacuum chamber in my new lab at Cornell, which was founded in 2020, so you can see the remnants of Covid having to wear a mask when we were first installing that chamber, and then some you know where kind of my work. The lecture spray started in the in the bottom picture at Mit with the chamber that we had fitted with quite a few diagnostics to try to understand the dynamics and processes and 21 00:03:50.420 --> 00:03:52.290 Mark Kushner: electric spray ion blooms. 22 00:03:53.390 --> 00:04:12.419 Mark Kushner: So I thought, I'm gonna talk about the applications of electro sprays and the challenges of electrosprays, but I thought I would start by just introducing what an electro spray ion floom is. So I think electrospray maybe, is not the only molecular ion beam, but all electrosprays are molecular ion beams. 23 00:04:12.420 --> 00:04:25.249 Mark Kushner: With certain shared characteristics. So the first is we start with a liquid from which we form our ion theme. So this liquid, before it's electrically stressed, will form some 24 00:04:25.250 --> 00:04:40.969 Mark Kushner: stationary meniscus geometry. We like to start with this meniscus on some sort of geometry that helps enhance our electric field and reduce the voltage at which we need to form the ion plume. So we typically use some sort of needle or sharp 25 00:04:40.970 --> 00:04:59.809 Mark Kushner: sharply pointed geometry. And so that liquid is going to be on the top of the needle. But it's also going to be connected to a reservoir which will allow for a long term operation. So in this case I have shown a porous needle where the reservoir is actually the material that is connected through the pores of the substrate. 26 00:05:00.340 --> 00:05:11.690 Mark Kushner: indicating to. So these systems in in general require, on the order of a little less than a kilovolt to a few kilovolts to fire, depending on the characteristics of the liquid. 27 00:05:12.920 --> 00:05:14.370 Mark Kushner: And 28 00:05:14.920 --> 00:05:38.760 Mark Kushner: do this, and so once you get to the critical voltage required to to stress the liquid to a an equilibrium point. You've created enough electrical pressure that you can. The ions can overcome the surface tension and start to form a plume that's highly directional and relatively focused. 29 00:05:39.280 --> 00:05:58.070 Mark Kushner: Now, all electrospray plumes tend to be poly dispersed, although the poly dispersy of the plume depends on the conditions and on the propellants. And that's something that we're really interested, especially from our propulsion, perspective and understanding. So that means we have ions of different charge to mass ratios, different mobilities. 30 00:05:58.960 --> 00:06:17.089 Mark Kushner: and something that's true of all electric spray eye on plumes is that we have extremely large density gradients. And you know, I would say, for plasma processes, at least relatively large velocity in gradients. So this definitely isn't a homogeneous or a uniform 31 00:06:17.230 --> 00:06:35.190 Mark Kushner: plasma across our system. So we have density gradients anywhere from quite high for plasma, as I would say, if you're above 10 to the 20 to quite low in the far field regions of the phone, down to 10 to the 13, and then velocity gradients. We have, you know. 32 00:06:35.190 --> 00:06:47.430 Mark Kushner: a slow thermal particles anywhere from like 100, or maybe even directionally, less than hundreds of meters per second to tens of thousands of meters per second for our highest 33 00:06:47.730 --> 00:06:49.890 Mark Kushner: speed particles in the bloom. 34 00:06:50.230 --> 00:07:10.059 Mark Kushner: and then electrospray ion sources can be implemented on a wide range of length scales, so you can go all the way to nanoscale electric sources, where this whole picture is contained in less than a millimeter of volume and every dimension to say greater than 10 cm for other 35 00:07:10.270 --> 00:07:12.299 Mark Kushner: examples. 36 00:07:12.880 --> 00:07:42.120 Mark Kushner: So I wanna talk about some applications of electrospray ion beams. And I thought I would start with what I think I imagined by the numbers has gotta be the most common use of electrospray ion sources. And that's for electro spray ionization, mass spectrometry. So this has been a key technology and biochemical analysis since the 19 eighties. And it's actually so impactful that a Nobel Prize was awarded to John in the 19 nineties for his 37 00:07:42.180 --> 00:08:08.760 Mark Kushner: development of the electrospray ion technique to isolate large biomolecules. So to do mass spectrometry, you have to ionize your molecule. And most ionization processes are destructive to large molecules. So you want to put energy into knocking off an electron, but energy goes simultaneously into breaking the bonds of the molecule. You've no longer preserved it in its full structure. 38 00:08:08.830 --> 00:08:14.389 Mark Kushner: dissolving your biomolecules in in some kinds of fluid that could be electro sprayed 39 00:08:14.670 --> 00:08:39.460 Mark Kushner: and in process in particular ways, they could be entrained in the spray and detected by some sort of detector downstream, coupled with that electro spray ionization source. So this plot that I have shown maybe see my laser hole. I can't point at both sides at the same time. Maybe I have a mouse. So this plot here 40 00:08:39.460 --> 00:08:54.910 Mark Kushner: shows how, Esi, if you couldn't read it, maybe the the text is kind of small kind of took. This is it's basically size of the molecule on the y-axis. So it took the size of molecules that are detectable apply orders of magnitude from other 41 00:08:54.950 --> 00:09:11.190 Mark Kushner: other processes that were used at the time. And even this plot is this particular to one particular, like one specific device made by specific manufacturers. So the electro spray process in general extends beyond what's shown in that plot up to even being able to detect. 42 00:09:11.230 --> 00:09:19.860 Mark Kushner: fully intact biological organisms, such as a virus, which is a million amu. 43 00:09:21.290 --> 00:09:27.160 Mark Kushner: So I thought I would show you know, some more applications of electrospray ion beams on these axes, which 44 00:09:27.160 --> 00:09:51.760 Mark Kushner: kind of maybe determine to some degree the considerations that we would take for modeling the electro spray plume. So on the Y axis I have shown current, and on the X axis X axis I have shown theme energy, so electro spray ion sources are somewhere down relatively low current, single single emitter with just one ion spray. And you don't need. It's super high energy, for 45 00:09:51.760 --> 00:09:55.110 Mark Kushner: you know, super high energy or voltages for these systems. 46 00:09:55.500 --> 00:10:20.469 Mark Kushner: Another application which is the one that brought my, you know, brought electric sprays into my area. A research is satellite propulsion. So this is a picture of a commercial satellite propulsion system that's on the market today. This is relatively new technology. The company is called Axe on and they build electrospray ion sources for the purpose of space propulsion for small satellites. 47 00:10:20.690 --> 00:10:29.140 Mark Kushner: So this system is size that would go on the Cubesat. So we're talking about. Maybe something about a foot 48 00:10:29.310 --> 00:10:54.679 Mark Kushner: actually have a picture of this to pictureize the dimensions in a couple of slides. You may be like a foot high, and you know maybe half foot why. And this system is higher current than the single emitter electrospray ion source, because we operate hundreds or thousands or tens of thousands of electrospray ion sources in parallel to get enough currents, and therefore enough thrust to appreciably move a satellite. 49 00:10:54.900 --> 00:11:23.239 Mark Kushner: There's another example of this more commonly referred to as the the feed approach. So this is just the difference between these 2 systems is the type of fluid that you're generating your ion from. So in this case. You see glowing purple. It's a liquid metal indium specifically, and we have a little bit of a higher current density and higher operating voltage per emitter. But we still need to operate several of those emitters in parallel 50 00:11:24.710 --> 00:11:50.320 Mark Kushner: other applications of electrospray that I've come across in the literature are surface deposition of coatings for different applications. So one example that I borrowed this image from is this is, they were depositing some coding that was actually antimicrobial on the surface. But I've also seen examples of where you might be specifically depositing some sort of biomaterial on the surface as well. 51 00:11:51.290 --> 00:12:11.870 Mark Kushner: Another application of the electric spray. I am, process is focused ion the microscopy. So here we have a relatively low current, maybe just one ion theme, but a very high voltages to get certain interactions with the surface that allows you to look at the 52 00:12:11.870 --> 00:12:34.880 Mark Kushner: electrons generated in the collision with the surface and use that for microscopy, and then on an even higher. The higher energy end of the spectrum is the use of electric sprays for actually milling and eroding a surface so focus eye on them. Milling will operate up to hundreds of kilovolts of energy, at least for impact with your surface. 53 00:12:36.470 --> 00:12:49.820 Mark Kushner: And then these systems can generally be broken into 2 categories based on the liquid that you're spraying from. So on. These, like higher energy side of the spectrum is typically done with liquid metal. 54 00:12:50.290 --> 00:12:55.600 Mark Kushner: Liquor metals in the lower energy part of the spectrum is typically polar solvents. 55 00:12:56.240 --> 00:13:02.870 Mark Kushner: But the physics and the Plume generation evolution process are quite similar between the 2. 56 00:13:03.640 --> 00:13:31.010 Mark Kushner: So again, my introduction into this area of work has been inspired by, you know, questions and interest in this technology from space propulsion. So the models that I'm gonna be talking about today. And we've been developed have been developed with the priorities and research questions of that community in mind. But I'm hoping, you know, as we expand upon this work, that we could find some applicability of our models to some of these other areas of work. 57 00:13:31.940 --> 00:13:47.309 Mark Kushner: And maybe, conversely, you know, models from other areas that can inform maybe best. You know the best methods for us to to capture some of these new processes that are that the electrospray plume introduces. 58 00:13:47.310 --> 00:14:04.560 Mark Kushner: So what I'm gonna talk today covers the you know, anywhere down from single particle impacts with surfaces. I would I would consider that to be very low current to up to the operating currents of save these modern day. 59 00:14:04.730 --> 00:14:19.280 Mark Kushner: propulsion systems that have been built from space and energies on the order of very low energies. Looking at the low, the slowest particles in our plume up to the, you know, firing potential or operating potential of our ion source. 60 00:14:19.450 --> 00:14:33.109 Mark Kushner: So I just want to talk a little bit more about the electrospray thrusters since that's been the foundation of our model. So back to this picture of the electric spray 61 00:14:33.220 --> 00:15:01.850 Mark Kushner: in this diagram, so I have, you know, like this cartoon of one electrospray in the thruster here, which is one example of thrusters that are entering the market today. So all of the I don't know if it's, you know, kind of obvious, just from looking at it. But on the surface this is called like the tile design. So on the surface, there's all of these one square centimeter, you can see they have kind of gold outlines, arrays of emitters. 62 00:15:01.850 --> 00:15:11.330 Mark Kushner: So each one of these gold squares is a, you know, postage, stamp size, or one square centimeter, actually have a ruler here to give a scale, a size 63 00:15:11.490 --> 00:15:28.510 Mark Kushner: scale of the system. And so the state of the art is to pack something like 500 emitters per square centimeter, and then to put many of these emitter chips on a surface, and so the thrust the ion plumes would be coming 64 00:15:28.550 --> 00:15:42.990 Mark Kushner: away from, you know, coming off of this top surface, and you can operate these and either positive or negative polarity, and I'll talk about that a little bit more. So you can. You can emit positive or negative ions from the surface. 65 00:15:43.040 --> 00:16:06.950 Mark Kushner: So like all of the kind of interesting physics. All of the plasma physics happens in this top less than a millimeter of surface of this system. So these systems are designed to be very compact. It's it's nice for space propulsion cause we don't have to store a compressed gas. We don't have any moving parts for fluid delivery, and you know, to kind of, I mean, this is has been revolutionary for our field to take. 66 00:16:06.950 --> 00:16:13.759 Mark Kushner: You know, the whole complexity of a plasma system and kind of get the same. 67 00:16:13.760 --> 00:16:24.180 Mark Kushner: you know, essential final product, which is a plume of highly directional ions, but have that all generated within, say, like a centimeter of volume. 68 00:16:25.360 --> 00:16:35.120 Mark Kushner: when that you know now that produces new challenges for us and modeling these very small-scale processes. 69 00:16:35.210 --> 00:16:44.690 Mark Kushner: I wanted to define a little more of the properties of the propellants that we work with, which ultimately form the ions in our eye on bloom. 70 00:16:44.690 --> 00:17:14.679 Mark Kushner: So we're working with organic polar solvents or molecular salts. Specifically, we're working with the class of molecular salts called ionic liquids. These are you know, I guess, relatively exotic, but also relatively common substances. Now that have been studied for a lot of different applications from electrochemistry to materials processing and happen to be a really nice choice for space propulsion. So. 71 00:17:14.740 --> 00:17:37.170 Mark Kushner: and these salts are composed of mixtures of cations and anions. So your molecules are already in an ionized state, and they also have the property that they're have a very low vapor pressure. So they are. They remain stable under vacuum. So you can expose this liquid to vacuum, and you won't lose any of your propellant. 72 00:17:37.240 --> 00:18:01.830 Mark Kushner: And so most of those different Ionic liquids that have been, you know, studied in our continue to be studied today for use in electrospray propulsion systems. One of the ones that's been very well characterized is Emi B at 4, which is the one I have shown here. So on the left is the Emi plus cat, Ion. You can see the little plus and the molecules that make up that 73 00:18:02.170 --> 00:18:20.239 Mark Kushner: the atoms that make up that molecule are carbon, hydrogen and nitrogen. And then it's anion is BA 4 minus so just one boron and 4 fluorine atoms. These sources typically operate around a few kilovolts. 74 00:18:20.270 --> 00:18:42.340 Mark Kushner: And here's just a a kind of nicer view of some of the molecules that you'll find in the plume. So on the left side of this chart is the cat ions. So you have the monomer cat, ion emi, plus. It's just extracting the cat ion or the monomer anion on the right side via 4 minus 75 00:18:42.340 --> 00:19:10.590 Mark Kushner: unfortunately, because it's usually, I'd say, unfortunately, because it usually represents a performance loss and certainly a source of complexity. We're not able to control the emission process such that we only extract the monomer ions. So instead, you'll get some solvated clusters coming along for the ride. And that can be anywhere. But you know that spans the spectrum, depending on how you've designed your ion source 76 00:19:10.590 --> 00:19:38.750 Mark Kushner: from very low molecular weight, clusters all the way up to large, almost macroscopic droplets, which is typically not what we want for the case of propulsion, because that heavy droplet is gonna have a low velocity and a low mass utilization efficiency. So shown here just what we call the or you know what are called the dimer species. So that's they still have a charge of just plus or minus one. But now there's 77 00:19:39.010 --> 00:19:52.999 Mark Kushner: 3 total molecules. So you have about 3 times the molecular weight in this dimer species, as in your monomer species, and you go up from there trimers, tetramers, Pentamers. 78 00:19:55.420 --> 00:20:13.280 Mark Kushner: So of course, our you know, our development of models is really is in part driven by wanting to better understand these processes and wanting to be able to explain and mitigate failure modes that we observe in testing. So you know, somebody asks 79 00:20:13.280 --> 00:20:26.529 Mark Kushner: today, actually, it was in the video at lunchtime. What are the key challenges for electrospray propulsion devices? And I said, well, the key challenge is really lifetime for all of our electric propulsion devices. We need them to operate for hours on orbit 80 00:20:26.720 --> 00:20:54.440 Mark Kushner: hours. Sorry, not even hours, tens of thousands of hours on orbit, months, years. And so the electrospray Ion sources being less mature than other systems, or have not yet met the lifetime goals that we have for full utility in space. And so this picture on the right, I have shown is the result of a long duration test that was done with an array of electric spray sources, and what you can see is there's been propellant that's accumulated 81 00:20:54.510 --> 00:21:12.039 Mark Kushner: on this extractor electrode, which has a hole for every ion plume to come through. So this is these darker parts that you see on the grid, and eventually enough propellant builds up that you can have arcing, and then electrochemical degradation. 82 00:21:12.040 --> 00:21:26.150 Mark Kushner: electrochemical degradation of the material that's built up, but also structural degradation of the emitter tip that was forming the Ion plume. And eventually these kinds of processes can lead to catastrophic failure of the device. 83 00:21:26.150 --> 00:21:52.220 Mark Kushner: But we had really no no models, and not very good physical intuition to the, you know, the processes that were causing this deposition to happen. So we don't measure very high current interception. So we, you know there the conclusion, there are processes that you know are not currently in in kind of our understanding. Maybe. First order understanding of the dynamics of the beam that could be leading to this deposition. So 84 00:21:52.530 --> 00:22:21.960 Mark Kushner: that was one question that we hope to shed light on in developing models. Other questions are questions that are pertinent to the operation of any plasma system, and certainly a plasma propulsion system. So what kind of secondary species are generated, and what impacts do those secondary species have on the operation of our system? So some examples I have this cartoon here showing a subset of the 85 00:22:22.150 --> 00:22:35.680 Mark Kushner: outcomes that we anticipate would arise from these molecular ions hitting a surface, whether it's a surface of the thruster itself, or whether it's a surface in your experimental test chamber. 86 00:22:36.520 --> 00:23:04.999 Mark Kushner: So we wanna know, you know, like, when these ions hit the chamber at the range of energies. What are the impacts? You know, what are the outcomes of those collisions? And where do those byproducts go? And then we also are curious, you know. Are there collisions happening within our plume that could be creating byproducts there? What are those byproducts? And how does that affect the efficiency. And then, finally, as people have been, you know, doing better 87 00:23:05.000 --> 00:23:34.089 Mark Kushner: experimental characterization of the system they found. This is a study here at Madison, out of Afrl that showed that if you, you know, operate the electrospray thruster, and you measure all of the current that's been emitted, and then you measure the mass change of the device. There's about like 50 of the mass has been lost and is unaccounted for in the current measurements. So 88 00:23:34.430 --> 00:23:43.419 Mark Kushner: there's still kind of an open question of where that mass is going, and we're hoping that you know, by developing these 89 00:23:43.570 --> 00:23:48.370 Mark Kushner: more highly resolved numerical models that we can shed some light into that 90 00:23:48.700 --> 00:24:12.380 Mark Kushner: for for propulsion, missing half of your maxes or half of your masses, would be unacceptable, and would have a huge impact on the design of the system. We like to design our spacecraft so that we know exactly how much mass they're going to use. And that's there's not a lot of margin there for the performance of your spacecraft. 91 00:24:12.900 --> 00:24:15.320 Mark Kushner: And so, 92 00:24:15.490 --> 00:24:28.750 Mark Kushner: you know, question is like, Where is all of this? Missing mass going? And could it be going into processes in the plume that are not detectable by traditional ion current measurement? 93 00:24:29.770 --> 00:24:46.939 Mark Kushner: So now I want to switch to talking about a little bit of the state of the modeling both current and what led us to develop the site. Sorts of models that we've been working on in our lab. So I'll start. So we've talked about how the electric spray process is. 94 00:24:47.560 --> 00:25:16.640 Mark Kushner: multi scale, and I'll start at the smallest. Well, not quite the small scale, I would say the smallest scale is the molecular scale, but the smallest system scale is the scale of the mission. So it's been known for some time, especially in the conditions that we like to operate electrospray sources for propulsion, that the emission region itself is quite small. So this meniscus, you can see the meniscus on an emitter with your eyes, but you can't actually see 95 00:25:16.710 --> 00:25:23.239 Mark Kushner: the exact location where the ion plume is coming from. So you know, the picture 96 00:25:23.370 --> 00:25:49.470 Mark Kushner: of what's happening is when you electrically stress this surface that's allowed to deform, due to the charged particles deform to allow for a balance of stresses along the surface, that that deformation really happens at the very smallest scale, at the tip of the meniscus with many 97 00:25:49.470 --> 00:26:11.849 Mark Kushner: optical methods. And so this depends on the the fluid you're using. But with the fluids that we typically use for electric sources, this is predicted to be as small as 50 nanometers. So that's both a relatively small region for us to resolve in the context of wanting to model an overall system that is, centimeters in scale. 98 00:26:12.000 --> 00:26:17.200 Mark Kushner: and so that presents certain certain challenges. So there, you know. 99 00:26:17.240 --> 00:26:37.279 Mark Kushner: what I'm kind of showing. On the left side of this plot are 2 different electro-hydrodynamic models that were created to model that emission process. So they've been pretty good, actually, in different operating regions at predicting the current versus voltage. So we have pretty good confidence that they're 100 00:26:37.530 --> 00:26:52.740 Mark Kushner: pretty closely approximating the emission conditions, because we predict the currents that we measure at voltages, that sources are designed for with the representative geometries and fluid conditions. 101 00:26:53.690 --> 00:27:02.929 Mark Kushner: and so that sort of like in the last couple years. So these are from like 2,018 to 2,022. These models have been developed. 102 00:27:03.030 --> 00:27:26.100 Mark Kushner: People have also looked at using molecular dynamics to model this emission process. And so I would say, there's pros and cons to both of these approaches. Molecular dynamics is, is a more highly resolved model. But however, the volumes that people have been able to tractably 103 00:27:26.250 --> 00:27:56.140 Mark Kushner: capture with these systems are on the order of you. Can you probably can't read the text, but this one I have shown the leftmost plot is a few 100 angstroms, so that's large enough to just barely resolve the emission. But it really can't capture the broader electric fields of the systems, and it can't at all capture the fluid, the fluid mechanics, or the fluid pathways which are important is, you know, it's really important to understand 104 00:27:56.140 --> 00:28:10.449 Mark Kushner: full fluid profile all the way through the reservoir. So these systems, like, while they can present an interesting picture of the particle morphology. They really can't match the boundary conditions of a realistic system. 105 00:28:10.720 --> 00:28:27.969 Mark Kushner: So that's kind of the picture of the state of the art right now. And when we when, as we've been working through this problem for modeling the emission. then the next stage of modeling is to model the expansion of the plume. And so when I started this 106 00:28:27.970 --> 00:28:55.910 Mark Kushner: project there were kind of 2 approaches that people had been starting to take, so the first was to try to. So in our community, and a lot of plasma physics community. One of the main modeling tools is the particle and cell technique. So some people researchers have been using particle and cell models to monoelectric propulsion systems for a long time started applying these to electrospray systems. 107 00:28:56.210 --> 00:28:58.980 Mark Kushner: I found that there was a key issue with the particle 108 00:28:59.020 --> 00:29:19.019 Mark Kushner: using the particle and cell approach, and that was so particle and cell requires having a spatial grid to solve your both your Laplacian electric field and your electric field. And that's the utility of particle and cells. You could simplify those effects over a spatial grid 109 00:29:19.020 --> 00:29:40.750 Mark Kushner: and people have come up with ways to develop spatial grids for systems with density gradients. But they don't quite work very well when you need to get down to this small of a spatial scale. And with this, these stream of these extreme of density gradients. So the best studies that I had seen in our field. Only got down to about one micron resolution. So that's not 110 00:29:40.750 --> 00:29:51.769 Mark Kushner: highly resolved enough to capture the emission dynamics, and your plume is evolving so quickly over those first couple of microns that you're really smoothing out of a lot of important physics. If you take that approach. 111 00:29:52.260 --> 00:30:17.450 Mark Kushner: Then I was really excited when I saw some people starting to use in body or lagrantian methods. So here, you know, this is very like the most naive way you would go to, you know, naive approach. You would take to modeling a plasma or any system you just, you know, solve the equations of motion for every single particle, and and calculate the effects at the low. You know the the electric 112 00:30:17.450 --> 00:30:42.010 Mark Kushner: forces at the location of the particle due to all other particles and due to your boundary conditions and then propagate it. So people had shown that this was tractable, maybe not at first, for a realistic electrospray system. But it sort of got us thinking about using this approach, but maybe a caveat of where this started is. It was missing the background Laplacian electric field which is 113 00:30:42.140 --> 00:30:48.670 Mark Kushner: just as, or, more important to the evolution of the plume as the interparticle forces. 114 00:30:48.840 --> 00:31:03.230 Mark Kushner: and both of these both of these approaches had the downside, that they had unrealistic injection conditions, so nobody had yet really matched the realistic emission models result highly resolving that geometry 115 00:31:03.450 --> 00:31:08.359 Mark Kushner: of the emission site with the initial conditions to these particle models. 116 00:31:09.490 --> 00:31:21.120 Mark Kushner: so the foundation of our efforts has been trying to bridge those kind of issues and develop a new approach which would 117 00:31:21.690 --> 00:31:24.020 Mark Kushner: which would be able to overcome. 118 00:31:24.620 --> 00:31:25.940 Mark Kushner: You know these 119 00:31:26.990 --> 00:31:50.869 Mark Kushner: these challenges. So we start by using one of these electro hydrodynamic models. So one of my collaborators, Timoth kaludchess, spent his Phd developing these models? Specifically, for they they work very well for organic liquids in pure ion emission, although there are many nice models out there. That could. 120 00:31:51.050 --> 00:32:03.900 Mark Kushner: you know, replace or could be used in tandem with this? So what the elect, the electro hydrodynamic model does, is it? You know it solves the steady state geometry 121 00:32:04.230 --> 00:32:18.540 Mark Kushner: and current emission for electrically stress meniscus. The result is the current as a function of the voltage that you apply to the system, which is really nice, because those are the actual parameters that you would tune 122 00:32:18.670 --> 00:32:46.550 Mark Kushner: in the physical system. This tells us not only the steady state geometry and the steady State electric field, or the Laplacian field in the domain, but it also tells us exactly where the electric field is strong enough to overcome the surface tension, so that we know exactly where on the meniscus the particles are starting from, and that gives us, the, you know, pretty 123 00:32:46.980 --> 00:32:52.100 Mark Kushner: a pretty accurate or highly resolved picture of initial conditions for our bloom model. 124 00:32:53.250 --> 00:33:09.209 Mark Kushner: And then we simply inject these particles into our domain and integrate the forces that are acting on them, which is the electric field. From the background Laplacian 125 00:33:09.680 --> 00:33:17.950 Mark Kushner: solution, which is interpolated at the position of the particle and the field due to interactions between particles in the bloom. 126 00:33:18.220 --> 00:33:26.039 Mark Kushner: and so that general approach is called the embody approach. It's been used in other contexts in plasmas. When I started this 127 00:33:26.040 --> 00:33:50.510 Mark Kushner: work. I found a really nice open source code that I didn't build off of, but just kind of like inspected to see if it was possible that I can't remember who created at the time, but it was created for dusty plasmas, and they were looking at interactions at the scale of maybe a dust particle in a plasma. So I thought, Ok, we can probably do this for electrosprays. And we also got a lot of inspiration from the field of 128 00:33:50.510 --> 00:33:58.510 Mark Kushner: cosmology, where embody massive embody simulations are performed for modeling the evolution of the universe 129 00:33:59.730 --> 00:34:02.410 Mark Kushner: with the same. You know, one over r squared force. 130 00:34:02.690 --> 00:34:29.209 Mark Kushner: So I have here. Just I'm gonna show the results of our simulations, and I've kinda already walked through the the steps for for completely in the simulation. But there's one thing that I haven't yet talked about, which was the first kind of molecular effect that we included, and that's the accounting for the stability of these cluster species. So these you know, when the ion comes out with the 131 00:34:29.210 --> 00:34:34.580 Mark Kushner: a neutral pair attached to it, and when these solvated clusters is only a metastable state. 132 00:34:34.580 --> 00:34:50.820 Mark Kushner: and that time scale of its stability is on the same timescale as it's transit time through the system. So we observe in our experiments, and it has become pretty well characterized what the rate constants for these 133 00:34:50.929 --> 00:34:54.309 Mark Kushner: these species breaking up is, and understanding 134 00:34:54.400 --> 00:35:11.670 Mark Kushner: they're understanding their stability and being able to model their stability while they're being accelerated is really important to understanding both the mass flux and the velocity or energy distribution function of ions in the plume 135 00:35:11.670 --> 00:35:31.619 Mark Kushner: calculation loop at every time. Step to interrogate. You know both the time. So it's injection to check against a a probabilistic model for its stability, but also to check the local electric field which has been shown to have a a really important enhancing effect on the breakup of these 136 00:35:31.620 --> 00:35:34.719 Mark Kushner: species, and then we 137 00:35:35.180 --> 00:35:48.649 Mark Kushner: we propagate each particle forward like. If it contains no charge, as of right. Now, it just continues on a ballistic trajectory, and that the species that's left with the charge continues to be accelerated by the electric fields. 138 00:35:50.080 --> 00:35:58.450 Mark Kushner: So I have the results of one of our simulations shown here. Hopefully, it's rendering, okay, so we 139 00:35:58.500 --> 00:36:20.929 Mark Kushner: you can see, like the different colors, red, green, and blue are the 3 primary ions that are injected in our beam from monomers up to trimers, and then, as those dimer and trimer ion clusters fragment, they form neutrals, and those are all the yellow species that are shown in the bloom. 140 00:36:21.030 --> 00:36:24.530 Mark Kushner: That you know the the time scales of these neutrals 141 00:36:24.550 --> 00:36:36.010 Mark Kushner: is their their velocities and their densities are quite different than the ion and velocity density, so that presents you know, a modeling challenge to capture, each 142 00:36:36.320 --> 00:36:50.240 Mark Kushner: evolving to steady state the ions evolved the steady state in probably about 10 nanoseconds, but we require at least 2 orders of magnitude longer on the order of a microsecond for this neutral 143 00:36:50.370 --> 00:37:11.290 Mark Kushner: plume to evolve, to study state, and so you can see it continuing to fill the domain. And if you I don't know if you can see it on the screen, but you can kind of track some of the keep your eye on some of the really slow moving particles, and and get a sense of what the what's driving the lower timescale. 144 00:37:11.520 --> 00:37:15.699 Mark Kushner: or the longer timescale processes in the system. 145 00:37:19.210 --> 00:37:33.120 Mark Kushner: so here, just as kind of a summary of some of the predictions we're able to make about the Ion plume. I'll also maybe just summarize kind of the computational complexity of the problem. 146 00:37:33.180 --> 00:37:48.849 Mark Kushner: so we have, and we're able to do single particle tracking, because in a single ion plume we have only on the order of 10 to the 4 to 10 to the 5 particles in this key region of interest where things are 147 00:37:48.990 --> 00:38:13.890 Mark Kushner: reaching steady state. And that's attractive will number. These kinds of simulations can be run in less than a day with minimal parallelization. And we're not using extreme computational resources. Parallel supercomputing clusters at this point, which also tells me. We probably 148 00:38:13.890 --> 00:38:21.139 Mark Kushner: could. Use these methods. And I think there are reasons. You know, we might want to use methods like these. 149 00:38:21.140 --> 00:38:29.859 Mark Kushner: To capture the multi-scale nature of processes in these themes that may be at at the array scale. But we're not quite there yet. 150 00:38:29.870 --> 00:38:50.240 Mark Kushner: So so yeah, with this with this model, we're able to predict things like the ion density and the neutral density which had not yet not been predicted. Before these kinds of models were developed in their evolution across the beam. And an interesting observation that we make. 151 00:38:50.240 --> 00:39:08.410 Mark Kushner: you know, with having bringing all these physics together is that we predict there to be this low energy kind of cloud, or, you know, wider angle cone of neutral species in the electrospray ion plume that really hadn't been detected before. Because they're not. They don't show up in our current 152 00:39:08.410 --> 00:39:09.810 Mark Kushner: measurements. 153 00:39:11.470 --> 00:39:13.709 Mark Kushner: And just to kind of 154 00:39:13.790 --> 00:39:36.759 Mark Kushner: this plot here also can be generated or was generated with the this simulation approach we've used. This is a little bit of convoluted way of trying to just show this information, but the top plot shows the number of fragmentation events. Per nanosecond. So this is the fragmentation rate 155 00:39:36.760 --> 00:39:54.560 Mark Kushner: as a function of the local electric field. And you can. You know, basically highlight it. And then the the plot below it shows the evolution of the electric field. With the Y axis of the bottom plot being the axial distance from the 156 00:39:54.560 --> 00:39:56.550 Mark Kushner: onset of the Ion plume. 157 00:39:56.930 --> 00:40:20.649 Mark Kushner: So what? What I'm you know, trying to highlight in this plot is, I've boxed in the first 5 microns of the evolution of the beam. And you can see that there's this, this spike in this really strong enhancement of fragmentation which can't be resolved, and, you know, can't be modeled if you're not resolving at the scale, and also the really extreme 158 00:40:21.200 --> 00:40:30.239 Mark Kushner: electric gradients and electric fields and and strong electric fields that are present in these first 5 microns. 159 00:40:32.450 --> 00:40:51.630 Mark Kushner: Okay, so we're at 3 45. So I probably try to, you know, spend about 10 more minutes going through some of the results we've generated. I probably won't be able to go in through so much detail and everything. But I'd like to save time for QA. So I could talk about the things that are most of interest. And maybe, you know, get your 160 00:40:51.870 --> 00:41:16.640 Mark Kushner: get your feedback. So this chart is just to say that we've been working on experimentally validating this code, and we have pretty good. We're able to pretty well predict the shape of the ion theme. But there's some processes that are leading to divergence of the imbe that's wider than we predict. So I think there's still, we know that there are processes 161 00:41:16.640 --> 00:41:22.169 Mark Kushner: that we're not modeling that we would like to, and we're not sure yet which ones explain the wider divergence. 162 00:41:22.170 --> 00:41:32.480 Mark Kushner: But we have a pretty good match with the ion energy distribution function. That's experimentally observed in these systems. With this approach. 163 00:41:33.190 --> 00:41:54.659 Mark Kushner: so I kinda talked about. Why pick wasn't the right approach for this expanding ion theme. But we found is once, once you're wanting to model systems that are the scale of an actual spacecraft compulsion system. Where you have these hundreds or thousands of individual 164 00:41:54.730 --> 00:42:13.430 Mark Kushner: I am games intersecting. You actually can get a pretty nice picture and model of the ion bloom with the pick approach. So you know, at the scale of just one of those, one by 1 cm arrays that people are developing. We have a 10 to the 5 165 00:42:13.430 --> 00:42:33.030 Mark Kushner: increase in ions. We have a 10 to the 9 increase in Domain volume, and that 10 to the 2 increase in the time that we need to run the simulation steady state. So I'm still kind of I'm still interested to push our our single particle tracking method 166 00:42:33.030 --> 00:42:52.840 Mark Kushner: to something like this domain. But I think a better starting point for us, and getting some insight into the important processes was to use particle and cell. And again, we're able to resolve the link scales and the timescales of importance at the array with particle and cell is what we found. 167 00:42:52.840 --> 00:43:05.489 Mark Kushner: so I probably won't go into a lot of detail here. We didn't build a particle and cell code from scratch for this. Luckily we didn't have to. One thing that we did have to do for the electrospray 168 00:43:05.490 --> 00:43:28.830 Mark Kushner: ion plume, which is different than traditional plasma based propulsion systems is implement a multi-grid poisson solver. So here we're not doing for people who are familiar with pick when we don't have electrons, or at least a substantial electron. We don't have a quasi neutral plume. So we're we're solving a pick. 169 00:43:29.100 --> 00:43:43.729 Mark Kushner: you know, system. We don't have a standard Boltzmann inversion. But we actually have to solve the Poisson equation on the grid. So we implemented a multi-grid poisson fast Poisson solver for the system which we found to work pretty well. 170 00:43:43.870 --> 00:44:11.520 Mark Kushner: maybe one of the key takeaways to discuss from the pick work. Is that a question that we wanted to investigate with pick is, what is the importance of the evolution of the electrospray ion beam? When you have many Ion plumes clustered in an array, and our initial results are showing us that space charge 171 00:44:11.520 --> 00:44:18.760 Mark Kushner: does induce spreading out to 30 degrees, which is on the order of what's observed experimentally. So 172 00:44:18.920 --> 00:44:32.329 Mark Kushner: it's like 3 plots here. The right most one shows what if you just superimpose, are electro spray, ion plumes, what your resulting beam would look like in terms. 173 00:44:36.370 --> 00:44:44.529 Mark Kushner: seems to be an echo. 174 00:44:47.370 --> 00:44:54.589 Mark Kushner: I mean, my speaker must have come on. I was like I was still muted so I couldn't understand why. 175 00:44:55.110 --> 00:44:57.790 Mark Kushner: So we this 176 00:44:57.970 --> 00:45:02.350 Mark Kushner: plot shows, looking at different numbers. 177 00:45:02.810 --> 00:45:22.009 Mark Kushner: simulating different numbers of electrospray emitters clustered together anywhere from rows of 10 by 10 shown in the orange line up to 24 by 24, which gets you up to about 500 emitters in the brown line in the brown curve, and we see 178 00:45:22.530 --> 00:45:44.829 Mark Kushner: an increase in the divergence not just due to the geometry, but specifically due to the space charge or the repulsion of ions of like species, and characterizing the divergence of the beam is really important in the context of space propulsion, because any additional divergence you get is a loss and thrust and efficiency of your system. 179 00:45:46.140 --> 00:46:14.670 Mark Kushner: Alright, we're at about 3 50, so I don't have time, unfortunately, to to cover all of the things I wanted to talk about. But I guess I'll just say, another aspect of this work that we've been investigating is what, looking down at the molecular scale and trying to understand what the chemical stability of these species is. And we predict we're very confident that there are collisions that are changing the overall 180 00:46:14.920 --> 00:46:33.379 Mark Kushner: chemical balance of the system, both due to collisions with walls which these simulations are investigating, using molecular dynamics and understanding what the products are as a function of that collision energy. 181 00:46:33.530 --> 00:46:42.150 Mark Kushner: So here, you know, basically, once you get to even modestly, like not even high energies, but like modest energies 182 00:46:42.150 --> 00:47:06.869 Mark Kushner: on the order of tens of ev collisions. Our beam is that one Kev, we start to break the covalent bonds and introduce new species into the system. And at one key, Kev, we've broken many covalent bonds, and we're left with mostly light fragments, which are, many of them are free radicals, and are probably likely to interact with each other, interact with other materials that they come into contact with. 183 00:47:07.410 --> 00:47:23.270 Mark Kushner: We've also been modeling and identifying the outcomes of collisions that happen with ions and other ions in the plume. So this is a subset of those 184 00:47:23.480 --> 00:47:27.459 Mark Kushner: collisions that you would expect to see. 185 00:47:27.750 --> 00:47:48.100 Mark Kushner: and most plasma processes happen here with electrospray ions, with some with a lot of similarities. So anything from momentum exchange. To charge exchange, which looks a little, you know, means something a little bit different with electrons, but still happens all the way up to just breaking all the covalent bonds. 186 00:47:48.160 --> 00:48:10.259 Mark Kushner: So what we've been doing is taking that, these data sets that are generated at the small scale with molecular dynamic simulations and converting those into cross-sections which can be used in more traditional plasma modeling techniques. We can either implement those in our embody code or in in the pick code. So here we have cross-sections shown by 187 00:48:10.340 --> 00:48:22.129 Mark Kushner: bicolision type on the left, and then specifically cross-sections for the formation of these byproducts on the right, which we call covalent fragmentation. 188 00:48:23.810 --> 00:48:30.869 Mark Kushner: Maybe I'll just kind of stop here and 189 00:48:31.170 --> 00:48:54.010 Mark Kushner: you know, you know, just summarize the multi scale modeling that we've done. And then I have, you know, a few slides on some of the experimental work that we're doing in in case like, I'd rather like leave time for conversation. So you know, in summary, we've created a you know, a new, maybe not a totally new methodology, but developed a new 190 00:48:54.150 --> 00:49:06.390 Mark Kushner: approach for simulating electrospray ion plumes which can resolve all the way down to the small scales of the ion formation from the liquid surface up to. 191 00:49:06.460 --> 00:49:31.179 Mark Kushner: you know, through a single emitter all the way up to a pick simulation of a full array. We've identified a lot of the key processes that we think might be missing in the picture of the electrospray plume. And what we're doing right now is working on taking those small scale processes that are missing and putting them into our plume model. So we can predict where these 192 00:49:31.180 --> 00:49:40.080 Mark Kushner: clouds of neutral species go and what their energy distributions are. So thank you for your attention and look forward to some questions. 193 00:49:45.430 --> 00:49:48.510 Mark Kushner: Thank you very much. Are there any questions? 194 00:49:49.110 --> 00:50:00.930 Mark Kushner: Thanks for this talk, Professor Petro. It's really good. So my first question is, so you shown that sort of broad cloud neutrals. 195 00:50:00.930 --> 00:50:20.199 Mark Kushner: and I wanted to inquire as to like the main physical mechanism that your simulations suggest is responsible? Is it like near field fragmentation before the collimating effect of the field? Or is it more like collisionality? Right now? The only thing we're accounting for is near field fragmentation. So my guess is, once we 196 00:50:20.370 --> 00:50:32.669 Mark Kushner: incorporate collisionality, they'll just be more species and more fragmented species. But you know I don't know. I mean I still don't have an intuition of 197 00:50:33.090 --> 00:50:42.650 Mark Kushner: I haven't done the calculation to ask me, because it's not super straightforward to do what's going to be the driving 198 00:50:42.720 --> 00:51:00.949 Mark Kushner: con contributor, I think. Certainly in ground testing. And the other thing we're not measuring that is, contributing clouds and neutrals is collisions with the wall. But you know, for space applications, you may be more looking at whether it's the fragmentation due to the high electric fields or just the fragmentation due to collisions. 199 00:51:01.180 --> 00:51:17.180 Mark Kushner: Yeah? And then maybe a related follow-up question is, you got a comment at some point about it being a non-neutral beam, and given that, you know, some of these commercial concepts have had trouble on orbit with trying, like just having a non neutral beam and alternating polarity or something, and I may be moving to 200 00:51:17.180 --> 00:51:46.899 Mark Kushner: having a path, though, to neutralize the beam. Do you expect that to significantly change some of your predictions for what the you know. More array scale being dynamics look like? That's a good question. I yeah. Would you have a lower divergence? I imagine you, you likely would, and then we'll just have to introduce some electrons into our pick model, which is something that you know, I think will be really interesting to do in a whole nother kind of can of worms. Cause. Then we'll start to need to answer the question of what the products of electronics. 201 00:51:46.910 --> 00:51:53.230 Mark Kushner: electrospray ions interactions are. And and then also, I mean, I 202 00:51:53.730 --> 00:52:10.039 Mark Kushner: you. Then you have the dynamics of the the electron dynamics. But I think you can probably use all of the same. you know, approaches that we've used for traditional propulsion systems, at least at the array scale. I think, thinking about how to introduce 203 00:52:10.160 --> 00:52:20.470 Mark Kushner: electrons to the end body and the single emitter plume still probably has some open questions that would be interesting to resolve. So yeah, that's the question. 204 00:52:22.740 --> 00:52:32.819 Mark Kushner: This explain the missing mass, or because it's fragmentation, it would still show this current being admitted. 205 00:52:33.120 --> 00:52:40.410 Mark Kushner: I think. I. So I think because it's happening so close. You have this like 206 00:52:40.440 --> 00:52:47.899 Mark Kushner: really strong enhancement of fragmentation right near the emission site, you also there the ions have not been accelerated 207 00:52:48.040 --> 00:53:04.490 Mark Kushner: to nearly their acceleration potential. So I think you could use you could lose a lot of mass in that region that would probably just be in the error bars of your experimental measurements. II mean, I don't know if it explains 50 of the mass, but I certainly think that like. 208 00:53:04.870 --> 00:53:06.079 Mark Kushner: I don't have a 209 00:53:06.170 --> 00:53:29.679 Mark Kushner: quantification for this. Yeah, but I think we're, you know, very close to being able to to do that. You know. Say, you have like this 1,000 ev say you have like one resolution, or of that, then that still means like, if you lose particles within that first 10 ev of acceleration, you're not Gonna see it in your energies. 210 00:53:29.680 --> 00:53:45.880 Mark Kushner: Spectrum. So you're not going to see that that monomer Ion actually started as primary that like now lost neutral that's flying off to the side. So I think we I think we're at the point now that we could make some quantitative estimates of 211 00:53:46.000 --> 00:53:49.429 Mark Kushner: what contribution that might be. But I don't have it today. 212 00:53:51.220 --> 00:53:56.020 Mark Kushner: But hopefully the students are listening and starting to work on it. 213 00:53:59.090 --> 00:54:00.690 Mark Kushner: Any other questions 214 00:54:01.060 --> 00:54:08.400 Mark Kushner: I'll take more. So when you're in body simulations. When you go to inject the particle beam. 215 00:54:08.400 --> 00:54:33.309 Mark Kushner: do you have to bootstrap the poly dispersy from experimental like time of flight, spectra, or rpa spectra? That is what we do right now. So these you know, kind of thinking back to the contrast between the fluid models for emission and the Md. Models, the Md. Model would give you a charge. The charge, you know, prediction of the charge of mass ratio. To be honest, I don't know how accurate those are with what's observed in time. 216 00:54:33.310 --> 00:54:35.359 Flight spectrum. My guess is they're not 217 00:54:36.030 --> 00:55:00.529 Mark Kushner: perfect, because that you're not boundary condition matching anything else in the boundary conditions matter a lot, we know from experiments for the poly dispersed charge to mass ratio of your species in your eye on plume. So I mean right now, that's an open, you know. Maybe an open problem for somebody to solve is that the 218 00:55:00.590 --> 00:55:21.070 Mark Kushner: These fluid models can't predict the morphology of particles coming off the surface and can't predict what I would expect to be some sort of location dependent morphology. But I'm not even sure that the maybe molecular dynamics give you some insight to that. But again, I'm not sure that it gives you a result that would be more accurate in the end 219 00:55:22.770 --> 00:55:23.610 screen. 220 00:55:24.810 --> 00:55:35.169 Mark Kushner: Yes. when you use the experimental time of flight data, for are you just taking the center line of flight? Or are you also counting for any angular 221 00:55:35.180 --> 00:55:37.660 Mark Kushner: distribution at the top. Yeah. 222 00:55:37.800 --> 00:55:39.019 Mark Kushner: we're just 223 00:55:39.270 --> 00:55:45.540 Mark Kushner: they like to have a time of like data at, like the condition. You know. There, there's like. 224 00:55:45.750 --> 00:56:02.240 Mark Kushner: there's both a wealth of data out there. Lots of people have taken time of data. But there's not a lot of time of flight. And Rpa data sets that are taken at the conditions right now that we have a very controlled experiment, and that we can compare with our numerical models. 225 00:56:02.280 --> 00:56:12.380 Mark Kushner: There's only a handful of studies that have looked at time of flight. Like off axis, and I don't think any of those 226 00:56:12.380 --> 00:56:36.929 Mark Kushner: have been perfect for us to match with our set of numerical conditions. But I do think that is also like an open area of work and something that can inform. I mean, you could just use the experimental data to make some spatial emission non-uniformity based on what you're observing. But I think it still needs to be coupled with a model for the evolution of the plume, because 227 00:56:36.930 --> 00:56:51.759 Mark Kushner: you're not taking that time of flight right from the admission site that that like could be because things are being emitted off access non uniform ways, but it's also convolved with how different particles of different charge to mass ratio evolve 228 00:56:51.930 --> 00:56:55.840 Mark Kushner: before they get to the detector. A meter downstream. so 229 00:56:56.470 --> 00:56:59.169 Mark Kushner: say it's an open area to explore 230 00:57:02.060 --> 00:57:03.270 Mark Kushner: anybody else. 231 00:57:04.270 --> 00:57:08.120 Mark Kushner: Maybe I'll reserve the last question. So is there any 232 00:57:10.170 --> 00:57:14.089 Mark Kushner: any pulsing that's done in electroscreing systems? 233 00:57:14.330 --> 00:57:17.720 Mark Kushner: It's a great question. Okay, so 234 00:57:18.350 --> 00:57:38.889 Mark Kushner: and like, in this idea of operating like switching, you don't want to operate in one polarity for too long. And so maybe a decade or so or 2 decades ago it was established that if you alternate the clarity at one Hertz, that's a very slow pulse rate, you could extract positive ions and then negative ions, and then not have any 235 00:57:38.890 --> 00:58:03.779 Mark Kushner: detrimental effects in your propellant. You don't have any enough charge built up in your propellant that you would break it down. So that's mostly operated in a DC. Manner with the caveat, that if you want to extract positive and then negative ions to maintain charge neutrality in the tank. Then you can do that about on the order of one Hertz 236 00:58:03.780 --> 00:58:14.650 Mark Kushner: utility for propulsion. But I do know from other applications of electrospray people have looked at operating electrosprays, and then A/C. 237 00:58:14.650 --> 00:58:42.360 Mark Kushner: I don't know off the top of my head what the frequencies are, but they're much faster than one Hertz. And what they observe is actually the electrospray, the Taylor cone, which in a DC. Solution is the equilibrium geometry that forms is always 49 degrees. That's like the Taylor cone. In reality, we don't always have a Taylor cone at 49.5 degrees, but a lot of situations you do 238 00:58:42.740 --> 00:59:00.940 Mark Kushner: in a DC operation. But if you switch to A/C the angle of the Taylor cone changes, I don't know. You know. I don't know. People have really thoroughly investigated how that affects the properties of your plume. But I thought that was kind of curious. And what you're saying 239 00:59:01.010 --> 00:59:03.329 Mark Kushner: of pulse power. 240 00:59:03.440 --> 00:59:24.350 Mark Kushner: higher voltage for shorter due to cycles on the average, gives you more efficiency than lower voltages. DC. So is there any any nonlinearity that if you could only operate at higher voltage, you'd be higher. You you'd be happier. 241 00:59:24.530 --> 00:59:28.160 Mark Kushner: Have the answer off the top of my, you know, like I think there's 242 00:59:28.200 --> 00:59:42.460 Mark Kushner: you know, you make one thing better, and then something else is worse. So I don't know if I have like a system skill response to that, you know focusing is better at higher voltages. But these electric sprays have kind of like a happy 243 00:59:42.500 --> 00:59:55.210 Mark Kushner: range of operation, and if you push the voltage to high, you might destabilize operation. But again, I think the dynamics change a little bit, and the in that kind of if you're 244 00:59:55.260 --> 01:00:04.789 Mark Kushner: if you're moving to an A/C system. So I think it's definitely an interesting question to explore. Well, thank you and thank you for the 245 01:00:15.470 --> 01:00:23.640 Mark Kushner: remind everybody that 246 01:00:24.430 --> 01:00:24.940 Mark Kushner: that's my.