Below is the text version for the "On the Pathway to Lower-Cost Compressed Hydrogen Storage Tanks Webinar" video, recorded December 17, 2019.
Eric Parker, Fuel Cell Technologies Office:
Hello, everyone and welcome to the U.S. Department of Energy's Fuel Cell Technologies Office Webinar Series. Today we've got a packed presentation from a variety of different presenters on the Pathway to Lower-Cost Compressed Hydrogen Storage Tanks. My name's Eric Parker. I provide program support within the Fuel Cell Technologies Office and I'm the organizer for today's meeting.
And we'll begin in just a second but first I have a couple of quick housekeeping items to tell you about. This call is being recorded and may be posted on DOE's website and used internally. If you do not wish to be part of this recording, please disconnect now. And if you speak during the video connection you're presumed to consent to the recording; however all attendees are on mute so please if you have questions submit them via the Q&A box you should see in WebEx in the bottom-right of your interface.
And then with that we will have a quick Q&A at the end of the presentation. I know we have a lot of ground to cover today but we'll do our best to get to all the questions and so without any further ado, I'd like to introduce today's DOE webinar host, Ned Stetson, who is joining us at DOE headquarters. Hi, Ned.
Ned Stetson, Fuel Cell Technologies Office:
Hello, Eric. Thank you. Hello, everyone. Welcome to another one of the DOE Hydrogen Fuel Cell Technologies Office Webinars. I am Ned Stetson, Program Manager for the hydrogen technologies Program within the office.
One of the key activity areas under the Hydrogen Technologies Program is Hydrogen storage. Onboard vehicles today hydrogen is stored as a high-pressure gas in pressure vessels that are overwrapped by carbon fiber composites. The pressure vessel is very expensive, due in large part due to the cost of carbon fiber composites and actually it doesn't matter if we are discussing hydrogen stored at either 350 or 700 bar or even natural gas at 250 bar, the cost of the carbon fiber is the largest single contributor to the system cost. Thus to get the cost of these pressure vessels down, carbon fiber is an essential element to try to reduce the cost of. Carbon fiber is formed by taking a polymeric precursor fiber and running it through a series of processing steps that result in the graphitization or carbonization of the fiber, leaving essentially only the carbon atoms.
The cost of the carbon fiber is roughly evenly divided between the cost of the precursor fiber and the cost of the conversion processing. For the high-strength carbon fiber used in pressure vessels, the conventional precursor fiber is typically predominately composed of polyacrylonitrile, or PAN, with only a few percent of comonomer or additives. The expense of the PAN fibers are due to the cost of the material, the fact that they are typically produced by a more expensive and capital-intensive solution spinning process, and then the conversion usually results in a relatively low mass yield of 50 percent, meaning it takes roughly 2 kg of PAN fiber to produce 1 kg carbon fiber. In order to try to reduce of the costs of the hydrogen and natural gas storage tanks, a couple of years ago we initiated several efforts on developing lower-cost precursor fibers for high-strength carbon fiber. Today we will hear from three of these projects; two focus on the cost issues relating to the use of PAN precursor fibers and one looking at a lower-cost alternate to PAN fiber.
But before we get to the precursor efforts, we'll hear about the current projections of the baseline costs for several of our hydrogen storage systems for onboard light-duty vehicles. Since 2013, we have periodically reevaluated and updated projected baseline costs based on the status of the current best state-of-the-art. We just posted a most recent update on our website and Dr. Cassidy Houchins of Strategic Analysis will give us a quick update on the current analysis. Following Cassidy, we'll hear from Dr. Matt Weisenberger from the University of Kentucky, Dr. Mike Chung from Penn State University, and Dr. Sheng Dai from the Oak Ridge National Laboratory on the precursor projects and the results they've achieved to date. So if you have any questions, as Eric mentioned, please submit them via the chat function and we'll take as many as we have time for after the last speaker. So with that, I thank all of you for listening and, Cassidy, please go ahead with the update on the cost status.
Cassidy Houchins, Strategic Analysis, Inc.:
All right, thank you, Ned. So as Ned mentioned I'll present an overview of the updates to the 700-bar light-duty compressed-gas storage system and I'll try to get through this fast to motivate the discussion to come on the precursors. So next slide, please?
So I'll hit a couple of high-level storage-system cost results that were discussed in the 2019 program record that Ned had mentioned. I'll talk a little bit about the carbon fiber cost assumptions used in the analysis and point out a couple of specific trends in the precursor. And then finally, I'll present some forward-looking cost sensitivity analyses that suggest a pathway towards achieving DOE cost targets of $8.00 per kWh. Next slide?
So the 2019 program record reflects incremental changes to the storage system. We looked at several design changes, but ultimately the cost boiled down—or the current status and the cost boils down to a couple of key changes as mapped out here in this waterfall chart. The first one is that we adjusted the reporting dollar year from 2007 dollars to 2016 dollars to align the costs with other efforts within EERE. And then we looked at a number of modest changes to the system that reflect the—either current designs or improvements in designs that are reported in the literature. And I'll draw your attention to the ones related to the carbon fiber—so I guess the third green block over, showing a $0.51 cost reduction, reflects the change in the winding pattern that was used as reported by Argonne National Lab and this is what we call a hoop-intensive winding pattern that resulted in approximately five percent reduction in the carbon fiber mass. And then finally there are two blocks next to the final status block each representing $0.35 per kWh reduction that reflect updates to the carbon fiber price.
And these capture both current market prices for Toray T700S as well as projected process improvements reported in the literature, and we'll talk a little about that in subsequent slides. And then finally on this slide, despite the fact that we updated the basis on which we're reporting the system costs from 2007 to 2016 dollars, we continue to see a decrease in the total system cost as shown in the figure on the right. So the top line is the cost reported in 2016 dollars and then the blue line shows the costs reported in 2007 and there's an adjustment based on inflation between those two curves. Next slide?
So when you break down the total system cost, there's no surprise, the majority of 700 bar compressed Type 4 compressed gas storage is in the carbon fiber and the carbon fiber composite. So just drawing your attention to the costs reported or projected out to 100,000 storage systems per year, the carbon fiber is a little bit over 50 percent of the total system cost. And also note that we've made an aggressive assumption about the carbon fiber costs as we'll show in the next slide. If you account also for the resin, the fiber winding, you get nearly 60 percent of the system cost is just in the composite and the manufacturing associated with the composite. Next slide, please?
So we looked at carbon fiber, the cost to manufacture carbon fiber, building up capital costs and operating costs for a 1,500 ton/year processing plant and we come up $25.00/kg of carbon fiber, and this compares with current market price of Toray T700, $26.00/kg carbon fiber. I won't coin that we are doing a bottom-up cost analysis of the T700 fiber. We don't have enough information to say that, but at least the cost projection is roughly similar to the Toray T700 current market price. We've pulled in a couple of process improvements that are reported in the literature for scaling up from 1,500 tons/year to 25,000 tons/year and you see a roughly $3.00/kg carbon fiber cost reduction by moving to a lower-cost precursor—this is reported out of Oak Ridge by David Warren's effort—and then finally there's an oxidation plant scale-up process improvement that was reported in this Kline industry report.
When you take all of those process improvements into our bottom-up cost analysis you see this $19.00/kg of carbon fiber, and then finally we scale—because all of our costs are based on T700, we assume that on a percentage basis those same process improvements would reduce T700 price by 19/25 of the total cost of the T700 carbon fiber. Next slide, please?
And finally we looked sensitivity to a number of —we looked at a pathway for reducing 700-bar light duty vehicle storage system. The first one we folded in was this carbon fiber price target so this, in a funding announcement the DOE announced, that DOE had out in 2017, they had a target of $12.60/kg, so we used that carbon fiber price as a basis in our model, and you see this $3.00 per kWh cost reduction. And we looked at a number of other sort of modest reductions, reducing the safety factor, improving the coefficients and variations of this, variation in the fiber strength and in the manufacturing, and then when you take all of those reductions and you end up around $10.00 per kWh. And in order to meet the $8.00 per kWh it requires an additional and potentially more aggressive either change in the targeted carbon fiber price or then we need to start looking at balance of plant components and reducing the costs there.
So we assume that you could make some sort of a simplification of the balance of plant holding in the regulator and the valve costs and if you do that then you end up with this $1.34 reduction and then you get the $9 per kWh in our projections to high volume. As I mentioned this is our efforts to plot out a forward-looking view of how you might achieve the DOE cost targets. Of course that remains to be seen and require funding and demonstration in the R&D community but this is one possible path that we could see for achieving this cost of it. I believe that's my final slide; there might be a summary in here. Next slide?
Yeah so just quickly to recap, we reported an update to the 700 bar Type 4 light-duty vehicle storage system cost. We made some incremental changes. We looked at the carbon fiber assumptions and the current carbon fiber inputs to our model reflect some aggressive assumptions about scale-up for carbon fiber manufacture and finally to achieve the DOE cost targets, at least according to the pathway that we've plotted out, would require significant decreases in carbon fiber costs and also likely improvements in the balance of plant. And that's it. Thank you.
Eric Parker:
Go ahead, Matt. Thanks, Cassidy.
Matt Weisenberger, University of Kentucky:
Okay, thank you, Cassidy. This is Matt Weisenberger. I'm at the University of Kentucky Center for Applied Energy Research. Can you hear me, Eric?
Eric Parker:
Yes, go ahead.
Matt Weisenberger:
Great. I'm happy to report today on our progress on the development of small diameter hollow PAN precursors for carbon fibers. As Cassidy mentioned there's an aggressive cost-reduction goal here at $12.60 per kWh. Next slide, please?
Okay so in general the way that PAN-based carbon fiber is produced is beginning with the polyacrylonitrile starting material, it's then solution spun into a precursor fiber, which is then thermally converted into the carbon fiber, which is then formed into preforms, perhaps woven or wound into composites. So next?
Well, we're gonna be focusing on the precursor part of this. This is an example of the precursor that we've spun in this project from a low-cost PAN polymer. We have already made the solid versions of the fiber and demonstrated them to perform at the T700 level. Next slide?
You can see that our facility at UK, at the CAER is our fiber-development facility. We're doing air gap solution spinning. If you Google "CAER fiber" you can see a short video of our facilities on YouTube. Next.
The process begins by pressurizing the filter, the spinning solution or the spinning dope into a coagulation process where the fibers then are formed through an air gap and they're phase inverted in a coagulated bath. They're drawn out of the bath at a constant rate through washing baths, then through drawing baths, then a sizing is applied to the white fiber or the precursor fiber, which is dried and then wound up. Next slide.
This slide was to have a video, which I'm not sure if it's going to work, but it demonstrates the air gap spinning, so yeah, it's working. You can see then in the video that the filament's forming in the air gap, which is approximately 5 mm, and then that fiber passes down through the line. At the end it's being taken up at a traverse wound. Our process to make the hollow fiber is utilizing the fundamentals of this existing process, which is the way PAN-based precursors are currently produced. Next slide.
So as Cassidy mentioned the carbon fiber cost is huge for the composite overwrapped pressure vessel and all the composites. The largest costs in the production of the carbon fiber are indeed the precursor but also the oxidation process, which is a bit of a rate limiter in the production of carbon fiber, so it's a slow process. Next slide, please.
So as I mentioned we're developing ways to spin and process a hollow precursor from this low-cost PAN polymer. The second reason we're interested in this is of course we'll use less precursor and we hypothesize we can get higher specific properties or higher modules and strength-per-unit density, which is a driver for the use of carbon fiber, and, moreover, the hollow precursor will allow us to oxidize the fiber much faster. Next slide, please.
There is a skin-core structure in carbon fiber and so looking at this, this is a TEM micrograph of T700 fiber wherein we're looking down the cross section of the fiber. You can see the differences in contrast between the skin and the core in this TEM. The next slide please? Yeah, moreover, we know that from a tensile-strength point of view, which is a big driver for the pressure-vessel application that the strength is limited by the flaws on the carbon fiber and particularly with the flaws on the surface of the fiber, more so than the internal flaws, and, go ahead, next, please?
Moreover, we know now that the skin of the carbon fiber has a higher modulus than the core of the carbon fiber and a hypothesis is that if we can simply remove the core that we can conserve the modulus and can conserve the strength and, in some sense, it's akin to having two springs in parallel, one a very stiff spring and one a very soft spring, and lifting a heavy object. The spring that has the higher ring content, a higher modulus, takes a disproportionate amount of load. Next, please?
So, again, the hypothesis is that we can produce a hollow fiber retaining or exceeding the specific strength of T700. Our approach is to utilize a segmented-arc slip based spinneret, so it's a shaped spinneret, in the upper-right, traditionally used in the melt-spinning area. In this way we have no sacrificial polymer or no bore fluid used to produce the hollow filament and that it's drop-in scalable for doing multifilament tows. Of course to produce carbon fiber we need thousands of filaments. So what we've learned is we have to have very precise control in this air-gap solution spinning to enable the hollow filament to form. It forms by the healing of the two arc slips as shown in the lower left from the melt-spinning. On the right images, those are images from our line of doing solution spinning. Next slide, please?
Our initial results were challenging of course. There wasn't a tendency to make a round or a homogenous fiber at all and with time and trials we've been able to identify the balance needed between the air gap, the spin draw, the polymer dope composition and temperature, and the coagulation bath composition and temperature, to control towards the production of a multifilament, continuous tow-spun, hollow PAN precursor fiber, which I think is shown on the next one.
These are hollow PAN precursor fibers spun in the multifilament tow on the yellow cardboard core. You can see that. And we've tested the modulus and strength of these precursor fibers as a function of the fiber outer diameter, and we note that there's a large increase in strength as we reduce dimensions, and this is to do with the relative orientation of the polymer as it gets smaller and dimensionally more oriented. We're currently at these fibers 56 μm on the outer diameter and 32 μm inner diameter with a density of 0.75 g/cc. Our final target is to get to an OD of 14 μm and an ID of 9.3 μm. Next, slide?
This also indicates our current status, multifilament spun hollow TechPAN precursor where the outer diameter is 46.3 μm and an ID of 20.8 μm. This is just a visual on where we are. And moving on to our next slide, we have begun our initial or preliminary thermal conversion. In the upper-left we have an EDS image from SEM where we can see oxygen in blue indicative of oxidation of the fiber occurring from both the interior and exterior simultaneously in the fiber. This is also shown on the lower left of the polarized optical microscopy and this is what we're really keenly interested to see play out in a path towards faster oxidation of the fiber. If we fully carbonized the fiber, we have observed that we can indeed maintain the hollow shape and the circularity of the fiber. Next slide?
To recap we did make solid carbon fibers from this non-branded polymer and we were able to show that it was capable of producing T700 properties and we're currently developing stable, multifilament continuous tow, hollow fiber precursor spinning. We have multifilament tow where the individual outer diameter of the fiber is 46.3 μm, the ID is 20.8 μm, leaving a wall thickness of approximately 12.7 μm, and our targets are to take the OD down to 14 μm, with the final wall thickness of 2.35 μm. Our trials towards fast oxidation are just now beginning as we realized and produced now the multifilament hollow tow. And in terms of costs, our utilization of a non-exclusive, non-branded, low-cost polymer, our TechPAN polymer, that has a 13.8 percent cost reduction relative to the Warren report. And by going to a hollow structure we believe we can reduce the cost another 15 percent and if we can oxidize those fibers, as predicted, by a diffusional analysis, an oxygen-diffusion-limited process, that we can oxidize it up to 35 times faster, bringing the cost all the way down to $12.21/kg. That's our goal and I think I only have one more slide?
But in sum that's where we are with the project and we're excited to be where we are in testing these hollow fibers. Thank you.
Eric Parker:
Thank you, Matt, thank you very much. Okay, Mike?
Mike Chung, Penn State University:
Okay can you hear me? Okay this is Mike Chung from Penn State. Our research focuses on polyolefin precursor to make a carbon fiber and we also have a supporting team from Oak Ridge National Lab and certainly support by DOE EERE. Next slide? Next?
Okay, you know, these—carbon fiber produced for 60 years and you, 90 percent of carbon fiber is produced by polyacrylonitrile but there are some low tensile strength carbon fiber produced by pitch. And I'd just like to show you on the left-hand side, polyacrylonitrile structure. The chemical structure is a polymer, hydrocarbon polymer, with nitrile group along the polymer chain, and pitch is a polyaromatic small molecule mixture. Okay, next one?
You know I think that I don't have to repeat this. I'd just like to say that there are three steps, the spinning, wet spinning to make a polyacrylonitrile precursor fiber. Then they have to be stabilized in air before carbonization so there are much more steps to produce carbon fiber, which is high cost. Next one?
You know I show the chemistry a little bit, the stabilization. In fact, oxygen is required, like a previous presenter, the 200°–300°C. This takes a little bit longer time because of the chemical reaction. Nitrile group have to be polymerized and oxygen help remove the hydrogen to become conjugated molecule structure and this is like a data type of shift but the real structure is in the bottom one, which is much more complicated because the oxidation involved, the different functionality along the polymer chain, and these have to be removed during the carbonization or the heat [inaudible], at the end, it's only carbon, the graphene structure, 2-dimension structure. So you can realize the chemistry is a long way from 1-dimension to 2-dimension and you have to go through many, many different chemical reactions so that yield is only 50 percent of it all. The next one?
The one important thing is that during the carbonization, because the orientation of the crystal, the polymer, the chain, everything is very important, so the morphology, the high-tensile-strength carbon fiber, the morphology is very specific. The crystal size, which is dense, about 8 nm and there are several graphenes sticking there together, so these have to orient along the fiber direction and there are is a crystalline amorphous ratio and you have very specific, and during the stretching it's very important to remove the structure defect because there are many volatile compounds coming out, so the voids have to be removed and reduce the fiber diameter as well. So the morphology is very specific. Next one?
So with this information we already know, so we'd like to design a new precursor, and our precursor is—you know something, we studied several precursors, and the one I'd like to show you is based on polyethylene. It's a very simple polymer, it's a hydrocarbon. There is no [inaudible] and if we connect the pitch molecule, like I told you, pitch is also a precursor to make carbon, so if we connect this with polyethylene, we believe it's going to become a new precursor. And the advantage is polyethylene can be easily melt-spinning as well as pitch so melt-spinning becomes possible and because it's pure hydrocarbon and based on aromatic, so maybe one step doesn't require oxygen, it's possible to produce carbon fiber. And the key is the pitch molecule helps to serve as an active site, which between this stabilization temperature, I think we believe is 300°-400°C, before the polymer can degrade, we have to stabilize the structure, which involves the cross-linking conjugation and so on and so forth.
In this case we don't need the external reagent, don't need the oxygen, and stabilization doesn't produce any byproducts and if we don't introduce the [inaudible] we believe the carbon yield can be much higher and we believe it can be 80 percent or higher. Next one?
So I'd just like to show you some of the results, which have been observed and you can see the structure is polyethylene. It's CH2-CH2 along the polymer chain but in fact we did differently. We said we'd be able to connect the pitch molecule onto the polymer chain, 10 carbon to polymer chain, and if this pitch molecule is a mesophase, in fact during the carbonization one step under nitrogen can reach 87 percent yield. So we remove the hydrogen and it's quite interesting we observed these results. But you know in this polymer because so far we still find the polymer is very, very high viscosity because the mesophase, the liquid crystalline phase, is much difficult to do the fiber spinning because the polymer with mesophase, and so we're still working on melt-spinning. But the next approach … next one … next?
Yeah, we find now if we don't use the mesophase pitch, we use the isotropic, so whichever these below molecules, these low softening temperatures and, in addition, in this case, we can add the free pitch mixed with the polymer because free pitch, by itself, is a precursor to carbon, at the same time, can serve as a plasticizer and mix quite well. So in this case although the yield is a little bit lower can reach about 70 percent or higher or at least 65 percent and based on TGA results the carbon yield is still much higher than polyacrylonitrile but this mixture can be melt processing. The next one?
I'd like to show you the fiber. Okay the first one I show you is the pure polyethylene with pitch in the mesophase, like I told you it's difficult for melt processing because the viscosity is too high at this point, but we can do the electrospinning. Electrospinning is very convenient to spin into the fiber form and you can see the fiber's surface is very uniform. The size is about 10 μm, you know? It's completely processable and so then we can convert to carbon fiber. The next one?
I'll show you another example. This is PE-pitch with isotropic pitch and in fact the polymer is melt-processable so we can use very simple equipment like the one in our laboratory, and this one can spin the fiber and that will dry inside, you know? There are some examples that we spin several different ratios between the free pitch and the polymer but the fiber diameter is much bigger and so there is no problem to do the melt-spinning. The next one …
So we sent some sample to Oak Ridge National Lab, they certainly have much more professional and better equipment so they can spin into the precursor fiber, and this precursor fiber is about 30 μm in diameter and can continue to spin and to take up the fiber, okay? So this can be fibers melt-spinning, okay? The next one …
I'd like to show, you know, we say this fiber we can produce, so we did the carbonization of this precursor fiber to carbon fiber, but at this point we still don't have enough samples so we just put it in the oven to do the carbonization without stretching. Like I told you the stretching is very important in making carbon fiber but the—we like to see the result, how the chemical evolution happened in the sample. And you can see the different sizes, the solid state carbon NMR to show you how the—mostly we focus on the polyethylene part, the polymer chain—how the polymer chain changes and you can see around about 30 ppm there is a very sharp peak. That sharp peak is due to the polyethylene polymer chain and the below peak, which are around 125 ppm, is carbon, is aromatic carbon, and you can see this aliphatic polyethylene peak hit about 400°C, which reduced the intensity, but at 420°C this peak completely disappeared. In fact this peak, because pitch and the polyethylene become conjugated, becomes conjugated carbons so everything becomes the peak at 125 ppm, and you can see the two fibers on both sides, and so that's typical of the carbon, conjugated carbon structure.
So we know the polyethylene becomes a conjugated structure during the stabilization around 400°C. The other indication is the X-ray, because the polyethylene is the semi-crystalline so we see the crystalline peak. The farther one, the green curve, which you can see is very sharp, is a typical polyethylene crystalline peak, and when we hit up to more than 400°C, 440°C, these peaks completely disappear and the carbon, the graphene peak, starts to develop with the aromatic carbon and hit 1700, 1900, you can see very sharp peaks, that's a typical graphene peak, which we show. And so the polyethylene sort of convert to conjugate and become graphene in structure. The next one …
So we, because the XRD gives us the crystalline structures, so we can use the peaks to determine the size of the crystal. So in this slide I sort of like to compare. There are several samples, the temperature with the different samples, and the second column is the PE-Pitch, the results, based on the x-ray results. And the right side is commercial polyacrylonitrile carbon fiber, which you know, also under the similar carbonization temperature. Because the table is quite difficult to explain in a short time, I'd just like you to see the bottom one, the 1900°C (Sample 1). In fact the crystalline, graphene structure is very similar to the polyacrylonitrile carbon fiber.
The D spacing that's between the layers space, you know 0.348 and 0.351, is very similar. And the thickness of the crystal you can see you know in the PA is 3.1, in ours is 3.27, and you know the difference is we also can get near the A, nanometer, so if the crystal structure is quite similar, we can stretch. We can stretch the fiber, just the state of the steady heating, and the you know the material will be still wet from the crystalline phase, which is very similar to polyacrylonitrile, which very much comes from the polymer nature. Okay, and so then the next one …
Okay the other one is a … because the order/disorder ratio in the carbon fiber is very important for the heightened extremes and no one can tell after that because they are two peaks, one is D band and one is G band. D band is the carbon, we present that aromatic, the graph, the ordered structure, even if it's aromatic carbon in the disordered structure. So from the peak intensity ratio we can get to the ideal between the order/disorder phase and in the regular period based carbon fiber is about 1, that ratio, about 1, but you can see, even without stretching, with heating, the higher temperature you can see, that ratio becomes—approaches 1 but is still higher than 1. It's about 1.2, 1.3, but we believe with stretching this ratio will approach to 1 because stretching can alter the crystal orientation and the crystallization as well, okay, next one …
So I'd just like to tell you now at this stage of our research, you know, we find … today I showed you the PE-Pitch precursor, which uses some free Pitch as a plasticizer and these are low-cost material because polyethylene is very inexpensive, Pitch is very inexpensive, and we can do the melt processing, which will certainly save a lot of money, and the thermal converging, one step, we don't need oxygen, certainly also could save a lot of money. In addition if we don't need oxygen, we can fully convert the precursor into the carbon fiber and this is just simply based on thermal convergence, just heat, so heat can be homogenous, heat the fiber, so potentially we can get the more uniform structure and certainly we can get higher carbon yield and at this point in without even stretching, the morphology of the carbon fiber, very similar, you know, these nano-polycrystalline structures are very similar to polyacrylonitrile fiber.
So at this stage our research … you know what we are doing is certain … we have to do the carbonization of precursor fiber under the stretching, mechanical stretching, so we can orient it, the crystal, along the pro fiber direction as well as all the carbon structure along the fiber direction. But more importantly is that we have to remove all of the defects because during the stretching, during the carbonization, there are some [inaudible] that come out and create a void structure, a void which has to be removed, a stretcher also removes that, so we can achieve the high tensile strength. And that's our current research activities. Okay that's all my presentation, thank you.
Eric Parker:
Thank you, Mike. Okay, Sheng?
[some content removed due to audio difficulties]
Sheng Dai:
So without further, I think it's already explained quite well that the carbon fiber, the PAN fiber is one of the most—right at the heart of the modern composites materials and, Eric, could you go to the next slide?
And so Cassidy already gave a very nice introduction that the cost of the carbon fibers is really right at … the hydrogen storage tanks and for 700-bar … it's right at the carbon fiber production, composites productions. And so the key is that for managing the cost of the hydrogen storage is really developing economical, effective—a cost-effective process to do the production of carbon fibers, PAN-based carbon fibers. Next slide, please?
And traditionally, this has been covered by Matt and by Mike also, but traditionally the PAN fiber precursors mainly through the wet spinning process and this intimately related to the structure of the PAN fiber and as is shown here in the structures of the PAN fiber or well [inaudible] is that they are extremely polar polymers and therefore very little solvent can sustain or dissolves these in high constitutions. And the constitution of these precursor fibers, ranging on the web spinning, ranging from 10 to 20 percent, and there're quite some issues associated with these tedious preparations and that's the key things which drop the prices, and one of the key components. So go to the next slide, please.
So the objective of this project is to develop a novel plasticized melt-spinning process to replace current solution-spinning process based on nonvolatile task-specific ionic liquids. Specifically in this project we want to address four issues related to this new process being proposed. And first we want to investigate how we can use molecular structures of ionic liquids to dictate the interactions of the PAN fibers so we can control the glass transition temperature and rheological properties of the PAN fiber and that's allowing you to do precision to some extent, your melt-spinning process. And secondly to study how the chemical interaction of ionic liquids with PAN can be used to control the cyclization, the cross-lane interaction or formation for these precursor led polymers, which were shown in Matt's talk and also in Mike's talk. And thirdly is to integrate information gained above two tasks to develop ionic liquids based melt-spinning system and demonstrate the effectiveness of the [inaudible].
And the relevance of this project is very clear from Cassidy's talk that this is really the ability to melt-spinning the PAN fiber and identify significant cost drivers for high-strength carbon-fiber productions. And this potentially, which can cut the cost significantly of the fiber productions and also replace the volatile solution spinning process, with a nonvolatile ionic liquids process, this is a very—making the process more safe and less toxic also. Next slide … next slide, please.
Yes, so why ionic liquids? I want to very quickly go through "Why Ionic Liquids" and liquids consist of salt, basically organic salts have cation and anion, [inaudible] give a tremendous degree of freedom to tune the chemistries and this leads, these are very polar assisting so the PAN fiber have extremely high solubility, just nonvolatile, and another thing that's very important to emphasize is these are very highly recyclable and these are stable temperatures to make [inaudible] cycle assistance. Next slide, please … next slide, please … yeah, next slide, please …
So in these two … a little bit over two years period we at Oak Ridge and together with 525 we have successfully demonstrated that, we down-selected a number of ionic liquids for [inaudible] and we have shown that with these ionic liquids, which are shown on the right, and they can lead to, easily leads to more than 100-degree depressing of melting point, the PAN fibers, and as you can see that this also does something [inaudible] a very interesting trend is that the depression also highly relies and depends on what are the ionic liquids cation are we choosing and also relies on what ionic liquids' anion which we're choosing and this, again, demonstrated that our original approach works—that these ionic liquids really brought a new opportunity in terms of polymer melt-spinning process. Next slide, please …
And not only we can change the melting points, we can also change this process with PAN-IL composites, they can have melt rheological properties that can be highly variable and as shown over here they can easily change the rheological properties based on the—both the concentration and also the ionic liquids, for which we use the structure ionic liquids. This is really a [inaudible] also for this ionic liquids-based melting process. Next slide, please …
Yeah so this just as you can see that the right side that they have a spinning video shows that using this ionic liquids PAN fiber precursor, it can gradually spinning high-quality smooth fibers as shown on the right and, as you can very clearly that the loads temperature and the header's temperature itself are really the optimum temperature and it's really good for the production of these PAN fibers through this ionic liquids process, and next slide please?
So after spinning these fibers and we carry out some of the thermal analysis characterizations and the structural characterizations, as you can see that after this melt, the as-spun fiber exhibits a smaller and wider exotherm as compared to the composite and also it's very interesting because after the fiber had been watered or just washed, the carbon yield it's significantly enhanced compared to original PAN fibers. So the ionic liquids process itself actually helped in terms of producing some carbon that the fiber precursor at a much higher carbon yield than the original polymer so … and … next slide, please?
And so these are showing some of the SEM on the right and this is spun fiber and washed fibers, their diameters and their mechanical property actually is better than the traditional wet spinning PAN fibers, and due to time constraints I cannot talk about it. Next slide, please …
And so this is a further analysis of SEM, etc. Next slide, please …
And so these are the FTIR, the mechanisms actually to show that the observation process, based on this PAN fiber, based on ionic liquids melt-spinning is also a plus. Next slide, please …
And so these are the tensile strengths and modular strengths and as you can see over here it's actually quite good and the result actually is a little bit better than a traditional process like the same PAN fiber precursors. Go to next slide, please?
And so these are further mechanical performance testing. I will not go... so next slide, please? And next slide?
So let me just to give a very short summary so that I have some question left for the... Number 1 is the melting temperature of the PAN fiber has been demonstrated to be suppressed by over 100°C by addition of ionic liquids and which prove that the ionic liquids-based melt-spinning is a viable process towards the PAN fiber productions. Secondly, they have demonstrated the ability to successfully melt-spinning uniform and homogenous PAN fibers with this ionic liquids-based process. And the preliminary experiments showed that the PAN fibers can be stabilized at a low temperature with carbon yield greater than 50 percent … and the next slide, please?
I want to finally thank EERE for financial support and also the program manager for this project is Bahman and he is actively engaged with it and I want to thank him. I'm done.
Ned Stetson:
All right, thank you Sheng.
Sheng Dai:
Sorry for rushing through so very …
Ned Stetson:
Thank you, Sheng. Thank you, Mike. Thank you, Matt and Cassidy and as Eric mentioned to you at the beginning of the talk, I mean, you can submit questions through the chat function at the bottom of the right, right-hand side of your screen. So far we've only got a couple of questions and we do have a couple of minutes so we can take these questions. I will start with one of the easy ones. "Do we get access to the slides afterwards?" Yes, as was mentioned earlier, we will be posting both copies of the slides and a recording of the presentation to the FCTO website. It usually takes a couple of weeks so with the holidays I'm assuming it'd be early January when the slides would be available and then the recording's usually available a week or two after that.
So one question we have is actually a two-part question and I'm not sure who would like to take this but I think Cassidy or Matt may want to respond. So the first part of the question is, "Can carbon fibers be recycled at the end of component life?" And then the second part is, "If not, what can be done with it and can it be used to make new carbon fiber without major environmental cost?" So I would say that currently with the thermoset resins used in pressure vessels it's probably close to impossible to recover the continuous tow fiber to use it again as a continuous tow fiber so any recycle use afterwards it would probably have to be chopped fiber or some other form of fiber, but Cassidy or Matt or Mike or Sheng, if any of you want to respond?
Sheng Dai:
This is Sheng. I agree with Ned. I agree with you. Okay so almost … currently it's almost very impossible.
Matt Weisenberger:
This is Matt Weisenberger. I also would agree with Ned on the aspect of recycling towards recovering a continuous tow intact. It will be very difficult but there are ways to recycle the composites to recover the fiber in forms of discontinuous fiber for use in secondary composites.
Mike Chung:
Yes, I also agree with that. I think you know it's difficult to recycle as a carbon fiber form but can be chopped to a small fiber for other composites, yeah.
Ned Stetson:
Okay thank you. So another question: Is the overall design concept of the carbon fiber storage-tank system consistent with stainless-steel vessels? So for light-duty vehicles or even the medium/heavy-duty vehicles, there're typically two types of vessels used. One is known as a Type 3 and the other is Type 4. Type 3 is a metal liner, so it could be stainless steel or aluminum, with a carbon fiber composite overwrap, and the Type 4 are polymeric liners with a carbon fiber composite overwrap. For the lower pressure of 250-bar or even 350, for natural gas, so 350-bar for hydrogen, both Type 3 and Type 4 are in common usage. For the higher pressure, 700-bar, Type 4 is probably the more dominant one but however there are some Type 3s used for there as well.
And then actually, Sheng, we have a direct question for you.
Sheng Dai:
Yeah.
Ned Stetson:
How and when can you recover the ionic liquid?
Sheng Dai:
Okay so the ionic liquid is … and this is a very good question and the ionic liquids, which we cover is that after you spin the fiber and you can very easily that, using water to basically you know, wash the ionic liquids and after you evaporate the water, either naturally, it can … ionic liquids have no vapor pressure so you don't need any solvents to look at that and these ionic liquids can …
Ned Stetson:
Okay, thank you. Another one: So who are the industrial partners focusing on these processes' conventional and novel approaches? So I'd open that up to any of these, the four speakers?
Sheng Dai:
Okay the one we have been interacting with is 525 Solutions, in Alabama, and they are already the leading industrial partner in terms of developing ionic liquids process, scale up, etc.
Ned Stetson:
And Matt, you have an industrial partner providing the TechPAN, correct?
Matt Weisenberger:
Yes, we have a provider of the TechPAN. I would also mention that the process that we're developing we feel would be a drop-in spinning process at rather any existing precursor production facility.
Mike Chung:
Yeah and in the polyethylene pitch precursor I think the polymer, the melt processing is a very conventional method to form the fiber, so I don't worry about the industrial cooperation if we'd really be able to it and you know in the meantime I certainly talk with several carbon fiber producers and you know I sort of explained the potential, the way to do the melt processing, and certainly there are many, many people are interested in this idea.
Ned Stetson:
Okay, thank you. Cassidy, this one's actually probably good for you, considering some of the analysis that you recently did. "Are high-temperature thermoplastic resins being considered for COPVs and will this allow for recycling?"
Cassidy Houchins:
Sure, so I'm not certain about the recycling aspect other than what you pointed out earlier that the possibility for recovering the continuous tow fiber is a selling point of the … shoot. I'm sorry. I'm drawing a blank on the term.
Ned Stetson:
Thermoplastic.
Cassidy Houchins:
Thank you, but yeah, so we looked at a thermoplastic process that DuPont and Steelhead Composites was developing for under the IACMI umbrella and there's some promise there, but the manufacturing costs appeared to be high using the current approaches.
Eric Parker:
Sorry. Sorry, Ned. I need to wrap this up. I know we have way more questions than we'll ever get to today so thanks everyone who submitted them. I am going to make sure we capture all of them and if you want to contact the presenters directly, they are onscreen but that does conclude our time for today's webinar so thanks everyone for joining. A big thanks to our presenters for all that content. I know we had to get through it all very quickly and thanks to Ned for hosting today. And with that, thanks for joining and we'll see you in the next webinar.
[End of Audio]