Hydrogen Sub-task Presentation II
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 10 | |
| Autor | ||
| Lizenz | CC-Namensnennung - keine kommerzielle Nutzung - Weitergabe unter gleichen Bedingungen 4.0 International: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen und nicht-kommerziellen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben. | |
| Identifikatoren | 10.5446/42657 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
9
00:00
GummifederWeinfassSchwingungsmembranTrenntechnikProof <Graphische Technik>DruckluftanlageZylinderblockWalken <Textilveredelung>MaterialprüfungsamtVerdichterPatrone <Munition>SchraubstockSatz <Drucktechnik>SiebdruckFörderleistungHydraulikleitungMotorWarmumformenLeistenErdölgewinnungPostkutscheGreiffingerHochdruckpumpeSchlauchkupplungGleitsichtglasMaterialEntwicklung <Photographie>SeitenleitwerkBehälterKeramikRückezugFlussdiagramm
09:57
ModellbauerBohrmaschineBeschusszeichenSeeschiffBehälterProof <Graphische Technik>SchlauchkupplungProfilwalzenNassdampfturbineOrdner <Schreibware>GreiffingerSchwingungsmembranStoff <Textilien>TrenntechnikSchlichte <Textiltechnik>MaterialPatrone <Munition>LeistenVerdichterZylinderblockFörderleistungWarmumformenPostkutscheDruckluftanlageAbwrackwerftGelenk <Technik>TischlerAnker <Uhrentechnik>FahrzeugUnterseebootComputeranimation
19:55
SeeschiffLunkerStücklisteSchiffsrumpfKlopfenWebmaschineTiger <Panzer 2>Körner <Metallbearbeitung>GleitlagerElektrolokomotive Baureihe 181KissenErdöl- und ErdgastechnikPuma <Panzer>SurfbrettAmbossFlugkörperPapierstaubungNassdampfturbineBlatt <Papier>Drehung <Textiltechnik>Satz <Drucktechnik>Proof <Graphische Technik>VerdichterZylinderblockKopfstützeHochdruckpumpeKeramikUnwuchtFörderleistungLinienschiffDruckluftanlageMaterialHandwagenNähmaschineBiegenTürglockePostkutscheGreiffingerComputeranimationFlussdiagramm
29:52
LeitplankeSeeschiffFeinkohlePorzellanPanzerungPuma <Panzer>EinschienenbahnHydraulische Maschine
Transkript: Englisch(automatisch erzeugt)
00:00
The following content is provided under a Creative Commons license. Your support will help MIT OpenCourseWare continue to offer high quality educational resources for free. To make a donation or view additional materials from hundreds of MIT courses, visit MIT OpenCourseWare at ocw.mit.edu.
00:29
So good morning, everyone. Rebecca, Lauren, Ben, and I are going to be giving an update on the progress we've made in designing the hydrogen production plant. So first, we're going to give an overview of the UT3 process,
00:42
which is the hydrogen production process we chose for our plant. Then we'll go through the block diagram of our plant and the iterations we underwent to try to optimize for our desired hydrogen output for the biofuels group. Then we'll go through our individual contributions over the last month or so that we researched and were primary
01:01
concerns in the development of our UT3 plant. So the hydrogen oxygen separator membranes, the calcium reagent structures, and the hydrogen storage system. And last, we'll go into the next steps that we're going to take over the next month or so to complete our design. So the UT3 process is a process with four reactions.
01:24
And each reaction occurs in a separate reactor unit. There are two so-called larger reactor units that consist of calcium reactions and the iron reactions. And just for convenience, we normally denote these in their own boxes just to show the characteristics of the
01:43
reactions as a whole. The reason why I say that is because, so in the first reaction where calcium bromide reacts with steam, it makes calcium oxide and hydrobromic acid. Seeming, basically backwards from this reaction, calcium oxide reacts with bromine and makes
02:02
calcium bromide and oxygen. And so as you can see, the reactants of one reaction become the products of the other and vice versa. And this enables a cyclic, continuous hydrogen production process by switching the flow periodically in our UT3 plant. The iron reactions are analogous in that some of the
02:21
reactants become the products and vice versa. So I'm going to hand it out to Rebecca with the first block diagram. So this is the block diagram of the UT3 plant. So the four reactions which Derek mentioned take place in these four, I guess they're yellow on the screen,
02:42
these yellow blocks, which are the four different chemical reactors. And so basically steam comes in and it goes through the first calcium reactor and the first iron reactor and that produces hydrogen, which this red block is the hydrogen separator. And so hydrogen comes out here and either goes to our
03:01
storage area or will go off to the biofuels plant. And then there's a compressor here, which is what is used to switch the flow. I'll show you the backwards flow next. And then so the products from here continue flowing here, flow through the second iron reactor and the second
03:22
calcium reactor. And then we'll have oxygen at the end of that. And that's that blue block over on the left is our oxygen separator. And so each of these reactions takes place at a specific temperature. And so that's what these orange heat exchangers are to make sure that the reactants coming from one reactor
03:43
are going to the next one at the right temperature. And so, as Eric mentioned, since the products of one, the products of like the first calcium reactor are the reactants of the second one, what this does is we use this first compressor here
04:01
to change the pressure and that will change the flow so that when everything has gone to completion, we switch it. So instead of flowing around this direction, it then flows this direction. And so therefore we can reuse the products that we have in there.
04:22
So Ben will talk about the specific mass flows. Right. So as Rebecca mentioned, we get steam from the process heat group, and that provides both our fuel for the hydrogen extraction and also the majority of our heat
04:42
that we use for these processes. At each of the reactor units, we make products, gaseous products, which are then propagated through the system by the steam aided by pumps and compressors throughout the process.
05:00
Then at various stages, we extract some of the valuable products like hydrogen, or just keep propagating things along. Now this is what realistically happens. The assumptions we made are in an idealized case, everything would go to completion. So we'd use up the steam and then just propagate HBR
05:24
along. This is what we use for our calculations. Of course, in the actual case, we'll have steam going through the whole thing.
05:40
All right. And here we go with some of the qualities that we've actually calculated for our mass flow in our first iteration. Again, the assumptions we went off of were that each of the reactors goes to completion. We also know that they have to operate at a specific temperature. So that also grounded some of our nodes. And the other is the other assumption we made,
06:02
or requirement, was that we needed 0.1 kilograms per second to send to biofuels at 0.4 MPa and 25 degrees Celsius. So from that, we've been able to quantitatively identify the different mass flow rates if you're paying attention. So we have the hydrogen, the hydrobromine, the steam.
06:26
Basically, we just, in each of these pipes, you would sum up whatever's in these colored blocks to get the full mass flow rate. The other important thing to note is that our entire system runs at two megapascals, except for right here,
06:40
where this compressor creates the pressure differential, which creates the forward progression of flow. So we go from two megapascals to 2.04 megapascals, and that's what creates the precession. Another thing to note is that at the separators, the products that come out, the hydrogen and the oxygen,
07:01
come out at a much lower pressure and temperature. So we'll have to, eventually we will quantify this compressor to figure out what work needs to go into it to make sure that we get the 0.4 MPa and the 25 degrees Celsius. Oh, and we've also quantitatively identified
07:21
what these heat exchangers need. So the important thing to note is that in heat exchanger one and three, we are putting energy into the system because we have to increase the temperature of the products throwing through those heat exchangers. However, at heat exchanger two, because we're going from 452 degrees Celsius
07:43
to 220 degrees Celsius, we actually take heat out. Okay, moving on to some smaller topics that our group has to address to make our hydrogen plant work. One is the hydrogen separator membrane.
08:00
If you research membranes, you can realize that there are a ton of different types. You have organic, ceramic, metallic membranes. Because we are operating at two megapascals and a temperature of approximately 500 degrees Celsius, we've identified ceramic membranes as the most optimal
08:23
as far as getting a good permeance value through. We looked at two different ceramics. We looked at CVD, this TEOS membrane and the zirconium silica membrane. The CVD is well established, it's in use. There's a stated value of 4.0 for the permeance
08:44
times 10 to the negative eighth. However, the zirconium silica is more experimental. It's not used as frequently. They're still developing it. However, one of the figures suggested that we could get a permeance of 10.0 times 10 to the negative eighth for permeance. When you run through the calculations,
09:02
that means we need an area of 1,240 meters squared for the CVD membrane and we'd only need 496 meters squared. So obviously we've decided to go with the zirconium silica. The stability issues are the same for both. So we'll have a phase transition issue.
09:23
However, we're not expecting to have a phase transition. And then there are a couple of poisoning issues, but again, these shouldn't be a problem if we make sure to purify the steam. And the last thing that should be noted is we need 496 meters squared of membrane area for it to diffuse across.
09:41
This is gonna be a pretty significant engineering concern just because we're dealing with 0.1 kilograms per second. So that's a very small volume of hydrogen. But it should be fun moving forward.
10:00
So similarly with hydrogen separation, we also need to separate out oxygen because the whole point is we're breaking H2O and hydrogen and oxygen. So how these membranes work is you've got a mixed gas that comes in and then it diffuses across a sort of porous membrane.
10:22
This is aided by electrodes, which ionize the oxygen. And then it just diffuses across from differences in partial pressure and temperature. So what we're using for the oxygen separator
10:41
is doped CEO2. This has pretty good conductance at our temperature ranges and it's basically just a combination of an oxygen ion conducting material. That's the CEO2 and then it's doped with an electronic conducting material which helps out the ionization of the gas
11:02
and the conductance across the membrane. Okay, so the calcium reactor system, the UT3 process actually poses a significant materials concern
11:20
primarily because there's a 76% volumetric difference in the structures between calcium oxide and calcium bromide. So during the typical UT3 process with the switching of flow, you're gonna have a expansion and contraction of these solid reactants and products depending on the stage of the process, which can cause fines to form and could eventually lead to product centering
11:42
after substantial cycling. And so I researched a couple designs to mitigate these problems because we don't want our products to turn into powder. And so we looked at a couple of designs that basically form a stable calcium oxide pellet that can be cycled multiple times and last on the same order as the iron pellets
12:01
which are much more stable structurally. And so I'm gonna actually show you the design and then talk about it since it's more obvious. So the global calcium pellet has a couple of substructures nested in each other. And so the first substructure, you have the calcium titanium oxide which is just used as a structural binder for the pellet
12:20
which does not undergo substantial expansion or contraction because it doesn't actually react with any of the gaseous products. It's just there for structural integrity. And then the calcium oxide pellets, as you can see, have, they're not solid pellets like the calcium titanium oxide. And so there's actually voids presented in the calcium pellet
12:41
which you can see in this even smaller substructure. And basically the calcium oxide as it's brominated into calcium bromide expands due to the volumetric difference. And the extra void between all the pellets initially allows these pellets to expand without exerting substantial stresses on each adjacent particle.
13:00
The pore plugging, so the separation is designed such that when all of the calcium oxide is eventually transformed into calcium bromide is just at the point before they're exerting substantial structural stresses on each other as to prevent fines from forming and other bad things for the metals.
13:21
And so as a whole, if you can just cascade this back, the calcium pellet globally will not expand too much due to bromination. And even if it's going to expand and contract some because just inherent in this design. However, with the calcium titanium oxide coupled with a substantial space for calcium oxide expansion,
13:41
we can have a calcium pellet that is structurally stable and exists on the same order as the iron pellet. So now Rebecca is gonna talk about hydrogen storage. So as mentioned, we're going to store hydrogen so that if there's a case where we would need
14:02
to shut down, the biofuels could continue going. And due to a combination of storage size and safety concerns, we've decided that we're going to store one day's worth of hydrogen, which is about 9,000 kilograms. And so I looked at the different forms of hydrogen storage because it can be stored in solid, liquid or gas.
14:22
So as a solid, it's that you have hydrogen in the interstitial spaces between metals or covalently bonded. However, that hasn't been fully developed yet. And so we don't really know the thermodynamics or kinetics for solid forms. In addition, they have a very small percentage
14:43
of hydrogen by weight or mass. So that's obviously not a great form to start in for us. And so going to gas and liquid, the thing about gas is that it's at the very high pressure 35 to 70 MPa. And so in order to deal with the stresses
15:02
and that you need a very strong outer wall and it also has to be very thick. And so that's best for small amounts of hydrogen, such as that's the one that's used in vehicles, if you want to use hydrogen for that. But for large amounts, it's actually better to have liquid hydrogen, which is sort of a low pressure.
15:22
It is at negative 253 degrees Celsius. And that needs to be kept like that because if heat enters the system, the hydrogen will evaporate. And so that's a concern, but it actually turns out that because what you need is a good surface to volume ratio, it gets better as you get larger amounts
15:41
of liquid hydrogen. And so that means for the amount of hydrogen we want to store, because 9,000 kilograms is a lot, that means liquid hydrogen is the best. And so that's going to be 130 meters cubed of liquid hydrogen that we'll have stored. And so this means if something goes wrong within the hydrogen plant,
16:01
the biofuels team can run off the stored hydrogen for one day. Obviously, if we need to be shut down for longer than that, biofuels also need to shut down. And if anything else shuts down, we will also shut down. So our first analysis we did, that we showed you just now was done by hand.
16:22
And going forward, we're going to be using Ease and hopefully Aspen in order to be more rigorous. We will also look at the individual components, which we haven't discussed yet, such as the heat exchangers and compressors. We'll look at the physical size of the chemical reactors and the consumption rate of the solid reactants,
16:41
as well as the amount of solid reactants determines the timing of the flow switch. So when it switches from being the forward flow, when that is completed, to when we do the backwards flow, as well as recovery of bromine and hydrobromic acid. So are there any questions?
17:02
Yeah, you have two different flow rates for water that's going into the same heat exchanger. You had 0.97 kilograms coming into your... Yeah, if you wanna go back to the... You have the 0.97 kilograms per second going into that. Which is exchanger?
17:20
And you have 6.13 coming, I'm assuming that's where our heat exchanger is gonna go for processing on the far left. Yeah, for here? You have two very different mass flow rates. How are you gonna correct that? Because it's supposed to be on a circuit, right? So the idea is, yeah, it's hard that we were...
17:42
So in my creating these block diagrams, we were trying to explain how what happens with the 6.13 is that it's gonna have to go out because it will have to be reheated. It has to be at 760 degrees Celsius when it goes in. This 9.7 or 0.97, imagine that it's further out there.
18:01
So that's actually entering from, or rather... So the way it works is that we do have the 6.13 kilograms per second. That needs to be reheated and then that's gonna join in with the 0.97 coming from process heat specifically. So this is 6.13 kilograms that we can recover from our system
18:22
that then joins the 0.97 coming from process heat alone. Again, that needs to be reheated so that it can join processing. And then... You need to have some sort of feed water coming into the heat. That's the 0.9. Yeah, so that is there's feed water coming out, but once we have the cycle going in,
18:40
and there's steam being recovered, we don't need as much feed water coming out. And so those are joined before your heat exchange. So you do have a feed water system somewhere in there to get to the heat. I think the plan was just to have water coming in from outside, so yeah. So continuing on that,
19:01
are you going to be giving us feed water to heat or are we going to get the steam at 589 to heat to 760? Or some mixture of the two? It's looking like some mixture of the two. When can we have those numbers? What? When can we have those numbers?
19:24
These... Are we good? I mean, look at it. We have one. Like... Yeah. So, I mean, the thing is, yes, the heat water is going to create that 0.97 kilograms per second, and then this is going to be... After it's all flowed through,
19:41
that's our 6.13 that joins in. I don't know. We're still like optimized. We still need to put it in east. So these are not final numbers, but they're pretty close to what we're actually gonna need. I think the confusion is like, which stream were you responsible for heating and what temperature are we coming in at?
20:01
So if there's feed water coming in from the environment and it's coming in at room temperature versus this steam getting diverted in from the rest of the circuit and coming in at 589, I guess we're just wondering what the... All right. Yeah, we can definitely make that clear for you.
20:20
I'm sorry it hasn't been like that so far. So yeah, you will be... What we would like from you guys is for you to be responsible for this 0.97 kilograms per second coming in from feed water, heating it up from the, whatever, heating up to 760. And then in addition, you will also be responsible
20:41
for heating up our 6.13 kilograms per second from 589 degrees Celsius to... Well, here's a question. If you're taking in a small amount of feed water and you're giving it to process heat to heat up, would this be a good place to put in an intercooler? Something where some of the heat on the iron reaction
21:02
could be used to boil the water, make it as steam and then give the process heat a single phase to heat up. Where the 6.13 steam at 589 could be mixed with 0.97 at whatever the saturation temperature is at two megapascals. And that way they'll only have to worry about one phase.
21:21
So, okay, so you're saying take the 0.97 that we need from wherever that we heated up ourselves and then we send it to them as steam at some temperature that's already steamed. Yeah, heat it up using some of the heat from that iron reaction, which it seems like it doesn't have to be that hot in order to work unless I'm mistaken.
21:41
All of our reactors, like there have been papers upon papers among papers that have optimized the values at which these reactions run. So they're kind of like the reactors, so the 760, the 560, the 754 and the 220 are all fixed. Then how about, see the top heat exchanger
22:02
where you're dumping out 2.9 megawatts? Could you use some of that energy to boil the feed water and recover it? Yeah, sure. Okay, yeah, anywhere where you're dumping heat, if you can use that heat to make your life and process heat's life easier, that'd be a good use of energy. Sure.
22:21
Just ask one more thing, sorry to belabor this. So you want 7.1 kilograms per second totally going into calcium? Yes. Okay. Just to be sure. And that, I mean, the number may change slightly because we did assume that the reactions run to completion, so we might need a little more, but that's order of magnitude and pretty good ballpark.
22:45
And just a quick question, what's the sign convention for heat going in and out of heat exchangers? Because it looks. Yes, I mean, it's a little. Because if you take the bottom two as the example where positive is in and negative is out, then it shouldn't. I think that should have a negative.
23:01
No. The top one should be negative. What should be negative? Just checking. I apologize, that was just a confession. No problem. So does anyone else have any other questions? Tyrell, anything? Yeah, I have a few.
23:22
Just looking at this diagram, it looks like there's a lot of room for optimization with the, you know, the taking the hot water and re-teaching it as input to somewhere else because right now it's one's through cycle. You know, I think you can manage it a lot better.
23:42
But we can talk about that more. The other, one question I had is, you said you're running it through one way. So I guess what I'm gathering from that is you run it until all the products from your first calcium reactor
24:01
then become your reactants for your second and then switch you around. So how does, have you looked at all at the efficiency of your reaction as do you get the imbalance of products versus reactants? Because I imagine it's gonna asymptote, you know, and then you're gonna have to find some optimal time to switch it over.
24:24
No, I was gonna say, yeah, it's on our list. Okay, okay. Yeah, I mean, all the papers we've read and have done it with asymptote and it doesn't look like something that, so we're gonna try to do it. But right now we're making, as bad as it sounds, the best guess we can,
24:42
because they seem to be quite nonlinear according to the papers. So we're going on the best information we have right now. Yeah, and I think that's something that, you know, just being able to talk about, you know, the final presentation. I was also wondering, where does,
25:00
you have a, going from two megapascal to 2.04, like where did you come up with the 0.04? Is that just an arbitrary delta or? It's from another one of those papers that they determined that was the optimal pressure differential for that compressor. We're basically choosing papers that have quite similar designs to ours to try to.
25:25
We don't need like an aggressive, like speed, mass flow rate flowing through. So I think the 0.04 isn't just small enough where you're not requiring too much energy in that compressor to change the mass flow rate, but you do get that forward position.
25:41
Sorry, we've got a list here. How big is your calcium oxide pellet? The type that you're talking about with all the structures and structures. They're quite small. I mean, they vary depending on the actual structure when you decide on. So there's three competing structures. I was a little nervous because I felt like we did enough material science.
26:02
So what, go back to the calcium pellet. And so basically the calcium pellet in general are normally spheres, and these spheres are very small, comparatively, like on the order of centimeters. But the choice of, I mean they're porous too,
26:20
because you have these original voids and stuff. You don't want to make it too big so that you can't, nothing can permeate the mass flow rate of the wire. But the choice of how to load up these calcium pellets, you can either load them up, think of it as like a high-temperature gas reactor, like spheres on spheres,
26:40
or you can make it into a honeycomb structure and have them with a predetermined lattice, which you load up, or you can actually even make them so small that you can mix them and make it kind of like a nanoparticle fluid, if you will. So we're gonna also try to look into which design would be best for our mass flow rates of interest. Yeah, I imagine if you're flowing through a pebble bed,
27:02
your pressure drop's gonna be pretty high. So definitely take that into account. And one, I have several more, but I'll just limit it to one more. One big question I have is if you have this huge hydrogen tank,
27:21
what do you do if it starts leaking? That is something which I have not figured quite yet. Okay, I mean, that's the biggest handicap I see in the minimizing the surface area versus compartmentalizing it for ease of maintenance
27:44
and safety concerns and stuff like that. So I would at least have something that you could discuss for the final presentation on whether or not it's a final solution. So I'll let you go first. Oh, yeah, right.
28:00
I was gonna ask, I think I've got to limit it down. What are you guys using the oxygen for? We're not using the block. But we have to separate it out. Could you combine it with the hydrogen to get yourself a whole? What'd you say? Could you burn it with the... We'll burn it. The oxygen and then you get your feed water,
28:21
which is really hot. I'll be using it on a metric pump. Sure. Whether it's lifted as well, right? Yeah, I mean, we're optimizing this plant for a specific hydrogen rate for the biofuels. We have to make it even bigger. All of our things will change. So we have to redo all of them in the last moment.
28:41
I'm gonna cut the questions off there so I have one very quick one for you guys in terms of the materials. Since you're dealing with so much hydrogen at various temperatures, have you looked into hydrating in materials failure or into hydrogen cracking in terms of things that would be under high stress and would have a lot of hydrogen in it?
29:00
You don't have to yet, but I'd expect something to look at a little bit of like various sort of corrosion things and things along that line. So we're looking sort of like maybe some sort of ceramic material or like special elements. Okay, yeah, well, you may want to look into at what temperature and under what conditions does hydrating of things,
29:21
especially like metals become a serious problem because hydrogen diffusion at these temperatures, like the wind just goes right through. But wouldn't expect anyone to know all that stuff until the very end of the presentation. Good job, so nice job, hydrogen. I especially like some of the stuff you did like color-coding your block diagrams and your products.
29:42
That made it really, really easy to follow what was going on. So everyone take a hint from that. That made it extra clear, so thanks a lot.