WEBVTT
00:00:00.160 -->
00:00:01.360Actually, we didn't get the idea.
00:00:01.360 -->
00:00:05.759Sony came to us and and said, We're developing this battery and we need separators.
00:00:06.160 -->
00:00:09.919Before they launched, they came to you like before they launched late 80s or early 90s.
00:00:10.240 -->
00:00:11.119In the late 80s.
00:00:38.000 -->
00:00:40.399Welcome to the Battery Potential Podcast.
00:00:40.399 -->
00:00:41.840I'm Vincent Chevrier.
00:00:41.840 -->
00:00:43.439And I'm Kevin Eberman.
00:00:46.640 -->
00:00:47.439Kevin?
00:00:47.439 -->
00:00:48.399Vincent?
00:00:49.359 -->
00:00:55.520Today we are delighted to welcome Robert Spotnitz, or Bob.
00:00:55.520 -->
00:00:59.200Welcome Bob, and a huge thanks for joining us on the podcast.
00:00:59.200 -->
00:01:02.240Your career covers many interesting facets of the field.
00:01:02.240 -->
00:01:14.319After earning your PhD in chemical engineering from UW Madison, you spent over a decade in RD at WR Grace, where you notably co-invented the multilayer battery separator.
00:01:14.319 -->
00:01:30.959You then managed the Battery Applications Lab at Celanese, the parent company of Celgard, before joining PolyStor as director of advanced product development, leading early efforts to build lithium-ion batteries for hybrid electric vehicles back in 1998.
00:01:30.959 -->
00:01:43.840In 1999, you founded Battery Design LLC, which has had a long life as a cornerstone of cell design in the field, and then finally you co-founded Enovix and American Lithium Energy in 2007.
00:01:43.840 -->
00:01:46.959Your LinkedIn profile currently lists you as retired.
00:01:46.959 -->
00:01:54.239We crossed paths more than 10 years ago at a DOE grant review meeting, and then again a couple of times after that.
00:01:54.239 -->
00:01:55.840Welcome.
00:01:56.400 -->
00:01:57.040Thank you.
00:01:57.040 -->
00:01:58.879Thank you for having me.
00:01:58.879 -->
00:02:01.840I really think it's a great thing you guys are doing.
00:02:01.840 -->
00:02:03.200It's really a great podcast.
00:02:03.200 -->
00:02:04.159I really enjoy it.
00:02:04.159 -->
00:02:04.640So thank you.
00:02:04.640 -->
00:02:05.280Thanks, Bob.
00:02:05.280 -->
00:02:06.000We're excited to have you.
00:02:06.400 -->
00:02:06.799Thank you so much.
00:02:06.799 -->
00:02:12.400Well, you know, we've it one thing we like is getting people from really different parts of the of the world.
00:02:12.400 -->
00:02:17.680And we haven't had someone that was so much from the modeling side also here or the separator.
00:02:17.680 -->
00:02:22.719So they're gonna you're gonna bring a lot of um a lot of freshness to the conversation.
00:02:22.719 -->
00:02:23.840Thanks for joining us.
00:02:24.159 -->
00:02:27.280We often start with how did you get into batteries?
00:02:27.280 -->
00:02:28.319What was the spark?
00:02:28.639 -->
00:02:35.439Yeah, I mean it's take us back, like put your head in your high school self and how did you get from there to ending up in batteries?
00:02:36.000 -->
00:02:38.479I went to Arizona State my sophomore year.
00:02:38.479 -->
00:02:43.520I read this book called Electrochemistry of Cleaner Environments by J.O.
00:02:43.520 -->
00:02:43.680M.
00:02:43.680 -->
00:02:44.560Bakris.
00:02:44.560 -->
00:02:56.400And he explained how electrification was a centuries-long trend, and it was inevitable for everything to become electrified.
00:02:56.400 -->
00:02:59.120The uh transportation industry, the chemical industry.
00:02:59.120 -->
00:03:06.479I then after that I read his book, Modern Electrochemistry, got interested in electrochemistry, and decided to do that in in grad school.
00:03:06.800 -->
00:03:07.840So you were compelled.
00:03:07.840 -->
00:03:10.080You you saw the vision.
00:03:10.080 -->
00:03:11.280He convinced me.
00:03:11.360 -->
00:03:14.400Yeah, he just uh very convincing guy.
00:03:14.400 -->
00:03:18.400And he wrote so clear he writes so clearly, it really made every sense.
00:03:18.400 -->
00:03:19.759Uh, it just totally sold me.
00:03:19.759 -->
00:03:21.520What what's the book again?
00:03:21.520 -->
00:03:24.479Electrochemistry of Cleaner Environments.
00:03:24.879 -->
00:03:26.400Okay, I have to check that out.
00:03:26.639 -->
00:03:26.960Yeah.
00:03:26.960 -->
00:03:29.120What was your PhD on?
00:03:29.520 -->
00:03:31.599Electrooxidation of sulfur dioxide.
00:03:31.599 -->
00:03:37.599So the idea was to use a fuel cell to convert sulfur dioxide to sulfuric acid.
00:03:37.599 -->
00:03:40.479And so sulfur dioxide would be a waste product.
00:03:40.800 -->
00:03:41.120Yeah.
00:03:41.360 -->
00:03:47.759And what I what I found was that the SO2 electrochemically oxidized SO3.
00:03:47.759 -->
00:03:55.919But then the sulfur trioxide would react with water instantly, and then it was so irreversible you didn't get much potential at all.
00:03:55.919 -->
00:04:01.120That was one of the projects in the electrochemistry of cleaner environments that Bachrus had suggested.
00:04:01.120 -->
00:04:01.840Wow.
00:04:01.840 -->
00:04:02.080Yeah.
00:04:02.159 -->
00:04:05.120I think I've run into a startup doing that, something like that now.
00:04:05.439 -->
00:04:12.159So is this somewhat akin to the idea of electrolyzers, you know, taking CO2 and making fuels out of CO2?
00:04:12.159 -->
00:04:12.639Yeah.
00:04:12.639 -->
00:04:12.879Yeah.
00:04:12.879 -->
00:04:15.599I mean that's still a that's still a recurring theme of trying to make it.
00:04:15.599 -->
00:04:21.920It's still an active area industrial processes and making them like less power intensive through electrochemistry.
00:04:22.240 -->
00:04:24.959Yeah, it's really it's really compelling.
00:04:24.959 -->
00:04:29.600And then at WR Grace, though I started off in electroorganic synthesis.
00:04:29.600 -->
00:04:31.600And we did a lot of processes.
00:04:31.600 -->
00:04:38.399WR Grace set up uh they hired this guy from GM called Lewis Hagetis, who was a brilliant modeler.
00:04:38.639 -->
00:04:38.800Okay.
00:04:39.120 -->
00:04:45.519And he set up an electrochemistry department with Charles DePias as his advisor.
00:04:45.519 -->
00:04:54.560And he picked areas of batteries, electroorganic synthesis, and I was in uh electroorganic synthesis.
00:04:54.560 -->
00:04:59.920He hired this guy, uh Menahem Anderman, to be in charge of batteries.
00:04:59.920 -->
00:05:07.279Um but then the electroorganic synthesis I began to understand was not really a good fit for electrochemistry.
00:05:07.279 -->
00:05:18.560Oh because it it was capital intensive and chemical production, when you use a pot, it scales up by volume, whereas electrochemicals, it's by area.
00:05:18.560 -->
00:05:21.519So it's way more capital intensive.
00:05:21.519 -->
00:05:29.920And also electrochemistry fundamentally you generate free radicals, which is a very unselective when you're trying to do chemistry.
00:05:30.319 -->
00:05:30.560Right.
00:05:31.360 -->
00:05:39.439Um but what we did do, and I I worked with this inorganic chemist, Bob Cray, we developed this uh mediated electrosynthesis.
00:05:39.680 -->
00:05:39.920Okay.
00:05:40.319 -->
00:05:42.800And we using cerium as a mediator.
00:05:42.959 -->
00:05:43.199Okay.
00:05:43.360 -->
00:05:45.519And that had some very selective chemistry.
00:05:45.759 -->
00:05:46.079Interesting.
00:05:46.319 -->
00:05:54.399And that was a really nice process, um, but Grace couldn't find a fit for it, and we licensed it to Hydro Quebec.
00:05:54.800 -->
00:05:56.000What what time period is that?
00:05:56.000 -->
00:06:00.160So I think you started in 82, and from there until licensing to Hydro Quebec.
00:06:00.160 -->
00:06:02.959I don't remember the exact date, but I think it's like mid-80s.
00:06:02.959 -->
00:06:03.519Okay.
00:06:03.519 -->
00:06:12.720But that insight you gave, I thought was a powerful one that that the scaling by volume is is cheaper than scaling by surface area.
00:06:12.720 -->
00:06:22.160And if you're if your electrodes are what's needed, you know, big surface of an electrode, then you have to make a much bigger system in order to scale up.
00:06:22.160 -->
00:06:24.000Uh that's a it's a fascinating point.
00:06:24.480 -->
00:06:40.399One of the uh business people at Grace, uh Robert Spillman, we published a paper uh doing an economic analysis of electrochemical processes and showed that there's a a narrow window where they make sense, that that certain conditions are right.
00:06:40.399 -->
00:06:44.480It it's the value of the chemical, the cost of the electricity, things like that.
00:06:44.959 -->
00:06:52.639Yeah, I think I'd like to see that because that theme keeps coming to me about once a month with us as startup companies.
00:06:52.639 -->
00:06:54.639Like we should do all this electrochemically.
00:06:54.639 -->
00:06:56.079Oh, maybe we should.
00:06:56.079 -->
00:06:57.839I don't know, but I should read your paper.
00:06:58.800 -->
00:07:06.959But at at WR Grace in the research division, I got into supporting the battery uh team, and that was led by Menahem Anderman.
00:07:06.959 -->
00:07:12.639He started off using um, I think it was it was some sort of a conductive polymers.
00:07:12.639 -->
00:07:14.399That was a big rage back in the 80s.
00:07:14.399 -->
00:07:15.920Conductive polymers.
00:07:15.920 -->
00:07:22.319Yeah, and then he got into TIS2 based on uh Whittingham's work, um as a cathode material.
00:07:22.319 -->
00:07:24.079As a cathode material.
00:07:24.079 -->
00:07:32.160And then Menachem left and he he joined a NICAD company in Arizona, which he eventually became president of.
00:07:32.160 -->
00:07:35.920Uh David Zuckerbrot took over his place, and I still work with that group.
00:07:35.920 -->
00:07:38.399And we did some modeling of TIS-2 electrodes.
00:07:38.399 -->
00:07:41.040So Grace Grace had some technology.
00:07:41.040 -->
00:07:44.959Grace owned a company called Duramic, which made battery separators.
00:07:45.199 -->
00:07:45.519Okay.
00:07:45.920 -->
00:07:56.879And their technology was to blend ultra-high molecular weight polyethylene with silica, and you could make these tens of microns thick separators for lead acid batteries.
00:07:56.879 -->
00:07:57.360Yeah.
00:07:57.439 -->
00:07:57.759Yep.
00:07:58.000 -->
00:08:00.240And they had a huge business based on this technology.
00:08:00.240 -->
00:08:04.000They had the base patents in it, and it was a huge, huge business.
00:08:04.240 -->
00:08:04.560Okay.
00:08:04.879 -->
00:08:11.040And then they also had a division called Cryovac, which made food packaging.
00:08:11.040 -->
00:08:14.959And the the food packaging is really complicated technology.
00:08:14.959 -->
00:08:17.439There's like many layers involved in it.
00:08:17.439 -->
00:08:23.439And they had developed their own blown film extrusion dies for multi-layers.
00:08:23.439 -->
00:08:27.519And so, as part of the battery thing, we got the idea.
00:08:27.519 -->
00:08:28.959Actually, we didn't get the idea.
00:08:28.959 -->
00:08:34.720Sony came to us and said, We're gonna we're developing this battery and we need separators.
00:08:34.960 -->
00:08:38.879Well before they launched, they came to you like before they launched late 80s or early 90s.
00:08:39.200 -->
00:08:40.080In the late 80s.
00:08:40.080 -->
00:08:42.879We needed separators.
00:08:42.879 -->
00:08:47.440And we couldn't, we there's not an there's no mark, there's no supply of separators.
00:08:47.840 -->
00:08:56.960Okay, and then they came to GRACE because GRACE was already providing through a subsidiary separators to the lead acid industry.
00:08:57.039 -->
00:08:57.759Aaron Powell That's right.
00:08:57.759 -->
00:09:03.039So I I got put on a project to try to make separators for lithium-ion batteries.
00:09:03.519 -->
00:09:03.759Okay.
00:09:03.759 -->
00:09:07.600And what was the like right off the bat, how were the constraints different?
00:09:07.600 -->
00:09:13.759Like what what made the separator for lead acid not applicable to lithium ion?
00:09:13.759 -->
00:09:15.759And what were your sort of like design goals?
00:09:16.320 -->
00:09:22.879Well they had to be very thin, whereas lead acid's tens of microns, and they had to be extremely strong.
00:09:23.279 -->
00:09:23.519Right.
00:09:23.759 -->
00:09:29.679And the the the tensile strength requirement was just unbelievable how how how how strict it was.
00:09:29.679 -->
00:09:39.279We started making samples and we gave it to them, and they rejected them because they had defects, they called them defects.
00:09:39.440 -->
00:09:39.759Okay.
00:09:40.000 -->
00:09:44.159And they were gel particles from the polymer, you know, in the extrusion process.
00:09:44.159 -->
00:09:44.960We got gel.
00:09:44.960 -->
00:09:49.600So with these the lithium-ion battery separators, we were making them without a filler.
00:09:49.600 -->
00:09:58.559We originally started with the filler, but then it was we we started making just pure polymer separators and making them porous by extracting out on oil.
00:09:58.559 -->
00:10:02.000But the gel particles from the polymer was still left in them.
00:10:02.000 -->
00:10:09.919And we tried filtering, we we actually we never solved the problem of gels, and that that was unacceptable to Sony.
00:10:09.919 -->
00:10:12.720But also the tensile strength was unacceptable.
00:10:12.720 -->
00:10:18.559In our process, we didn't do orientation, and that's how you get strength with a membrane.
00:10:18.559 -->
00:10:20.799You by actually oriented it, oriented.
00:10:21.279 -->
00:10:21.519Right.
00:10:21.519 -->
00:10:29.519And the requirement of tensile strength, of course, came because the separator was driving the winding of the electrodes.
00:10:29.840 -->
00:10:30.240Yes.
00:10:30.639 -->
00:10:33.600Whereas in lead acid, that's not the case.
00:10:33.600 -->
00:10:35.200The separator sits between plates.
00:10:35.200 -->
00:10:35.919That's right.
00:10:35.919 -->
00:10:36.320That's right.
00:10:36.320 -->
00:10:36.720That's right.
00:10:36.720 -->
00:10:37.519Right, right, right.
00:10:37.519 -->
00:10:38.080Yeah, okay.
00:10:38.080 -->
00:10:44.960So that I mean that's really interesting that how the construction of the cell really guided the design and the tensile strength was the key aspect.
00:10:45.759 -->
00:10:47.679Grace decided that we just couldn't do it.
00:10:47.679 -->
00:10:55.840We didn't have the you know, we we did go, we did some trials at a company, we paid to do some trials of co-extrusion, and it was just too involved.
00:10:55.840 -->
00:11:01.279I mean, you just couldn't develop a co-extrusion process with the requirements that were so strict.
00:11:01.759 -->
00:11:02.399I see.
00:11:02.399 -->
00:11:07.919So they didn't have uh like uh a way to by actually stretch to get the orientation either?
00:11:08.320 -->
00:11:08.639No.
00:11:08.639 -->
00:11:09.679Not not in-house.
00:11:09.679 -->
00:11:15.840They had, you know, we we could rent a lab to do it, but we never figured out how to do it right.
00:11:16.399 -->
00:11:23.440Um And had you figured out that by actually stretching it it would solve the problem that you just couldn't do it, or you hadn't identified that as a solution?
00:11:23.840 -->
00:11:30.639We identified it as a solution, but we just couldn't get the right we it definitely improved the strength, but we couldn't get the the properties that we needed.
00:11:30.960 -->
00:11:38.879In the end, did they did the the the separator that was in the first batteries was it was it as strong as they were requesting, or was it not quite that strong?
00:11:38.879 -->
00:11:39.600It was.
00:11:39.600 -->
00:11:39.919Yeah.
00:11:39.919 -->
00:11:41.360Okay, so the goal was achieved.
00:11:41.360 -->
00:11:43.360I thought maybe they'd give you an unachievable goal.
00:11:43.759 -->
00:11:46.799No, so the thing like Celgard was strong enough.
00:11:47.120 -->
00:11:47.600I see.
00:11:47.759 -->
00:11:53.360Um people were doing things like uh Molly Energy.
00:11:53.360 -->
00:11:57.759You take a layer of polyethylene, a layer of polypropylene, put them together because you needed shutdown.
00:11:57.759 -->
00:12:00.399That was another thing you wanted in a lithium ion, is shut down.
00:12:00.399 -->
00:12:19.039Um and so with Grace, one neat thing we had through our connection with the the company that made food packaging, Cryovac, we got access to all these polymers that had very sharp melting points, you know, like 110, 120, right, 100.
00:12:19.039 -->
00:12:24.480And we made trilayer separators with those polymers in the in the middle.
00:12:24.480 -->
00:12:28.559And so we had shutdown temperature that was, you know, whatever you wanted.
00:12:28.559 -->
00:12:30.080We could make a shutdown temperature.
00:12:30.320 -->
00:12:30.559I see.
00:12:30.799 -->
00:12:35.600You know, the separators weren't strong enough or they had gels, but you knew how to do shutdown.
00:12:35.600 -->
00:12:36.879But we could do shutdown.
00:12:37.039 -->
00:12:37.360I see.
00:12:37.759 -->
00:12:47.360And we thought that would be a really compelling story, but it turned out that the polyethylene, which uh was used with 135 melting point, was good enough.
00:12:47.600 -->
00:12:48.559Uh I see.
00:12:48.720 -->
00:12:55.919And then they had a drying process where they would dry the separators for a long period of time at like you know, 100, 100, maybe above 100.
00:12:55.919 -->
00:13:00.480So having a lower shutdown temperature really caused problems in that process.
00:13:00.480 -->
00:13:01.679Shrink shrink.
00:13:02.159 -->
00:13:02.320Yeah.
00:13:02.320 -->
00:13:09.759Before sort of doing some background research uh before having you on the on the podcast, I hadn't heard of WR Grace.
00:13:09.759 -->
00:13:20.639It sort of on its own has a sort of this interesting history of initially it was uh uh built on the guano industry of harvesting uh bird droppings, apparently.
00:13:20.879 -->
00:13:21.600Oh wow.
00:13:22.240 -->
00:13:30.480That went, then they evolved into shipping and then finally it evolved into specialty chemicals, which is, you know, they were already that by the time you joined.
00:13:30.480 -->
00:13:41.360But the late 80s was sort of this interesting time where you had a few big US industrials and Canadian as well, kind of going after the battery space.
00:13:41.360 -->
00:13:46.559Can you give us like a little bit of a of a sense of who you thought of as your competitors?
00:13:46.559 -->
00:13:49.039You've already mentioned Celgard.
00:13:49.039 -->
00:13:53.600But yeah, could you give us a sense of like what was the scene like in the late 80s?
00:13:54.559 -->
00:14:03.039So for separators, we we thought of uh 3M would be a potential competitor, they were doing that.
00:14:03.039 -->
00:14:05.679Um and then Celgard.
00:14:05.679 -->
00:14:16.879There was another company that was made lead acid, we thought might be but the lead I tell you the lead acid companies didn't take lithium ion that seriously.
00:14:16.879 -->
00:14:19.279You know, lead ath lead acid was king.
00:14:19.279 -->
00:14:22.399It was so big, all these things were niche markets.
00:14:22.399 -->
00:14:24.799They're they yeah, we didn't want to mess with these niche things.
00:14:27.039 -->
00:14:32.799Yeah, 3M had uh um a TIPS process separator patent or two uh in those early days.
00:14:32.799 -->
00:14:34.320Maybe you saw all those.
00:14:34.720 -->
00:14:35.440Sure, sure.
00:14:35.679 -->
00:14:36.320Yeah.
00:14:36.799 -->
00:14:40.720Um do you know what separator ended up going into the Sony cell?
00:14:41.039 -->
00:14:52.879So Tonen invented the ultra-high uh molecular rate polyethylene separator by actually oriented by the wet process, and that that's that really made the lithium-ion battery possible.
00:14:53.200 -->
00:14:55.759And that's what ended up going to the first Sony release?
00:14:55.759 -->
00:14:56.399Yeah.
00:14:56.399 -->
00:14:56.720Okay.
00:14:56.720 -->
00:14:58.240And Tonen still exists.
00:14:58.559 -->
00:14:59.519Tonen, yeah, yeah.
00:14:59.759 -->
00:15:00.240Yeah, yeah.
00:15:00.480 -->
00:15:01.200Yeah, we had.
00:15:01.200 -->
00:15:06.399I remember we had rolls of Tonen on the shelf when I trained three on them and rolls of Celgard as well.
00:15:06.399 -->
00:15:06.799Yeah.
00:15:07.519 -->
00:15:20.240So WR Grace was a big industrial chemicals company, and what you were working on for batteries was just a small sliver of that big research apparatus.
00:15:20.240 -->
00:15:21.039Is that correct?
00:15:21.360 -->
00:15:22.320That's right.
00:15:22.320 -->
00:15:29.039But the the thing about WR Grace was that it was a corporate research center.
00:15:29.039 -->
00:15:35.519WR Grace was built up of a lot of companies that it owned, and each of the companies had their own research center.
00:15:35.519 -->
00:15:41.519And so they didn't really want their money to be going to this corporate research center.
00:15:41.519 -->
00:15:43.919So there was a lot of tension all the time.
00:15:43.919 -->
00:15:48.399Just that they didn't really want corporate corporate to be doing anything.
00:15:48.399 -->
00:15:57.679And the projects that corporate did were you know, the if it was overlapped with a division, the the division would say, we could do it ourselves.
00:15:57.679 -->
00:15:58.960We don't really need these guys to do it.
00:15:59.360 -->
00:15:59.519Okay.
00:15:59.519 -->
00:16:02.799So did did battery ever make any money for Grace?
00:16:03.759 -->
00:16:05.120Other than government contracts.
00:16:05.120 -->
00:16:07.679That that's where the battery ultimately getting funded.
00:16:07.679 -->
00:16:14.159None of the divisions wanted to pay for it, and it ended up it was successful in getting government contracts.
00:16:14.799 -->
00:16:16.080Can you explain a little bit more?
00:16:16.080 -->
00:16:21.759Like, was that making products for like defense or what what what does that mean?
00:16:23.120 -->
00:16:31.600Oh, I think it w uh the Department of Energy was giving out these huge multimillion dollar contracts to develop rechargeable lithium battery.
00:16:31.600 -->
00:16:35.919Not these was it wasn't lithium ion, this was lithium metal uh battery technology.
00:16:35.919 -->
00:16:38.240Because it was seen as an important technology.
00:16:38.240 -->
00:16:38.480Trevor Burrus, Jr.
00:16:38.639 -->
00:16:51.120This this completely dovetails with uh what was being done at 3M and at Oregon National Labs and Hydro Quebec, and Valence was also uh working on those systems in that context.
00:16:51.120 -->
00:16:55.759And so WR Grace also uh was part of that effort.
00:16:55.759 -->
00:16:56.720Aaron Powell That's right.
00:16:56.960 -->
00:17:01.039My my first introduction to lithium ion was in the late 80s.
00:17:01.039 -->
00:17:14.160WR Grace's industrial espionage group or whatever, they got word that this new technology was being evaluated being invented in Japan that was going to revolutionize battery technology.
00:17:14.160 -->
00:17:23.039And since we were working on rechargeable lithium TIS2, I I was given an assignment to check and see if this was real or not.
00:17:23.039 -->
00:17:27.519And the understanding was that it was using some sort of hard carbon, which I couldn't get.
00:17:27.519 -->
00:17:48.799Um, but I I tried these uh different uh organic chemicals that I would pyrolyze and grind up and make electrodes, and I never got anything that wasn't really good, but I I was able to get a small signature of a voltage window of really low capacity.
00:17:48.799 -->
00:17:54.799I don't know what you know, it's a few milliamp hours per gram, just you know, very, very low capacity.
00:17:54.799 -->
00:17:57.920But it was incredibly reversible.
00:17:57.920 -->
00:18:02.880Uh yeah, I could go right up and down that voltage curve hundreds and hundreds of times.
00:18:03.519 -->
00:18:12.559You were able to get a sort of hard carbon adjacent system working, not very high capacity, but extremely reversible.
00:18:13.200 -->
00:18:13.759Exactly.
00:18:13.759 -->
00:18:17.119I I was very, very impressed with it.
00:18:17.519 -->
00:18:18.319Oh, that's fascinating.
00:18:18.559 -->
00:18:25.519And so I I I was convinced this was really a a breakthrough, but Grace did not decided not to go with it.
00:18:25.519 -->
00:18:38.400And then the manager of our department, Joe Lundquist, and one of the engineers in the battery group, Mark Isaacson, they quit and went to work with uh Valence.
00:18:38.400 -->
00:18:48.640And then after that, people started to get more uneasy about the group and everything.
00:18:48.640 -->
00:18:52.559And the guy in charge of the group then was uh David Zuckerbroad.
00:18:52.559 -->
00:18:58.319He had gotten a uh invitation to visit Hoechst Celanese, uh, Celgard.
00:18:58.319 -->
00:19:00.720But he didn't want to leave.
00:19:00.720 -->
00:19:05.759He was in charge of the battery group, he liked living in Maryland, and he he didn't want to go.
00:19:05.759 -->
00:19:09.279And he asked me if I would be interested in doing it.
00:19:09.279 -->
00:19:11.759And I I said, yeah, I would.
00:19:11.759 -->
00:19:21.359And so I got an interview, and Celgard was getting it was the company was actually Hoechst Celanese.
00:19:21.359 -->
00:19:27.200They made Sell Guard, but it was a small, the Sell Guard division, it was a very small division.
00:19:27.200 -->
00:19:40.559And in Charlotte, North Carolina, Hoechst Celanese had a research, a big research department, and then further away they had the small division called Celgard.
00:19:40.559 -->
00:19:48.640And the research division had invented this process for making microporous membranes, and it was an incredible process.
00:19:48.640 -->
00:20:01.279It was so simple, and it gave you these very strong uh membranes, and everybody thought there should be a huge market somewhere for them, but they didn't know where it was, and that there were so all these niche applications.
00:20:01.279 -->
00:20:06.240But not you know, the but it was only a very small business, a few million dollars.
00:20:06.240 -->
00:20:13.599Um, but they were always trying to find applications, and they had a really strong research group.
00:20:13.599 -->
00:20:23.680Um so you imagine they had this uh factory for making sail guard membranes, and within a month or two they could make enough membrane for the entire market.
00:20:23.680 -->
00:20:30.319And so the rest of the time they just did research, you know, with with you know industrial scale equipment.
00:20:31.039 -->
00:20:35.359But it sounds like Sony didn't go to them when they were looking for a separator, or maybe they didn't know about this.
00:20:35.599 -->
00:20:36.079Oh, they did.
00:20:36.079 -->
00:20:47.279They did, they did go to they did go to uh Sony, but Celgard did not make a separator that was meeting Sony's they didn't make a one mil or uh something competitive with Tonen.
00:20:47.279 -->
00:20:54.720Celgard was getting was very connected in the battery industry because they made primary battery separators.
00:20:54.720 -->
00:20:59.200And so all the Japanese companies, a lot of the Japanese companies made primary batteries.
00:20:59.200 -->
00:21:06.640Celgard had very good connections with them, and plus all the you know the Duracell, everybody.
00:21:06.640 -->
00:21:10.400But Celgard was so connected with the battery industry.
00:21:10.400 -->
00:21:12.480This is why they wanted to hire somebody.
00:21:12.480 -->
00:21:14.000They wanted to hire a battery guy.
00:21:14.000 -->
00:21:18.480So that the the fact that I had done separators and had that that was not of interest to them.
00:21:18.480 -->
00:21:20.160They wanted a battery guy.
00:21:20.160 -->
00:21:25.839Because the the limiting factor in making lithium-ion batteries was getting separator.
00:21:25.839 -->
00:21:28.240Tonen could not supply the market.
00:21:28.240 -->
00:21:34.319They had to shut down the fact that the lithium ion batteries factories couldn't produce because they couldn't get separator.
00:21:34.319 -->
00:21:41.279And but and so they were so desperate to get separator, they were telling Celgard that anything you made was okay.
00:21:41.279 -->
00:21:49.519And you know, so Sony would Sony would say, Oh, you need the very strict, you need to, you know, have it can't have these gels, it can't have has to have this tensile strength.
00:21:49.519 -->
00:21:52.559And other companies would say, Oh no, this is fine, don't worry about this.
00:21:52.559 -->
00:22:06.319You know and and so they were the management at Sell Guard didn't was they wanted someone who had New batteries and could tell them you had to have this, you had to have that, and evaluate what the customers were telling them.
00:22:06.319 -->
00:22:08.720That's where where I came in.
00:22:08.720 -->
00:22:12.079And they had three major approaches, Celgard.
00:22:12.079 -->
00:22:15.440They had uh bilayer technology.
00:22:15.440 -->
00:22:20.960So this was using the MOLI approach where you just took a polyethylene, polypropylene, and put them together.
00:22:21.200 -->
00:22:21.759Right.
00:22:22.240 -->
00:22:28.559But they were making them thinner so that you you know it would meet the requirement, the thickness requirement.
00:22:28.559 -->
00:22:30.880Um then they had the trilayer separator.
00:22:31.039 -->
00:22:31.440Yeah.
00:22:31.440 -->
00:22:35.519Because this is going to inform what you're what you're talking about here.
00:22:35.519 -->
00:22:42.480Can you walk us through the patent that you filed in 91, I believe?
00:22:42.480 -->
00:22:49.920So you have a patent that you filed in in 91 that talks about multilayer separators.
00:22:49.920 -->
00:23:01.680And in the examples, you have bilayer, you have trilayer, and that gets filed and assigned to WR Grace, and later on gets assigned to Hoechst Celanese.
00:23:01.680 -->
00:23:08.559Can you walk us through that invention and then how that sort of informed the work at Celgard later on?
00:23:09.039 -->
00:23:17.759So we had access to all these different uh polymers, and the GRACE had a lot of good co-extrusion technology.
00:23:17.759 -->
00:23:20.960So we were able to make multilayered membranes very easily.
00:23:20.960 -->
00:23:26.400And we made this whole ri array of multilayer membranes using wet process.
00:23:26.400 -->
00:23:28.960But we had a really good patent attorney.
00:23:28.960 -->
00:23:32.160You know, he said, you know, why limit the patents to just the wet process?
00:23:32.160 -->
00:23:34.640You know, why don't we use the you know patent the dry process too?
00:23:34.640 -->
00:23:37.039You know, they could probably make tri-layer if they wanted to.
00:23:37.200 -->
00:23:37.440Right.
00:23:37.599 -->
00:23:41.519And so even though we didn't make those membranes, we went ahead and patented it.
00:23:41.920 -->
00:23:42.559Nice.
00:23:43.119 -->
00:23:51.119You know, the grace division that really was responsible for the patents was the battery separators, and they didn't really take the lithium ion thing seriously anyway.
00:23:51.599 -->
00:23:56.079Can you explain the difference between the wet process and the dry process for making separator?
00:23:56.480 -->
00:23:57.039Oh, yeah.
00:23:57.039 -->
00:24:12.480So the Celgard process, and this was like a there's this really brilliant researcher at the at Hoechst Celanese who had picked up a piece of extruded, I think it was polypropylene, and pulled it and noticed that it turned white.
00:24:12.480 -->
00:24:17.920And he had the brilliance to realize that it that was because it had to be introducing microporosity.
00:24:17.920 -->
00:24:33.759And it turns out that certain groups of polymers that have uh what's called a row nucleated crystal structure, if you orient them in the in the machine direction, that those crystals will separate and leave voids behind, and you end up with a microporous membrane.
00:24:33.759 -->
00:24:35.839So it's like incredibly simple, right?
00:24:35.839 -->
00:24:38.079You just extrude a film and stretch it.
00:24:38.079 -->
00:24:42.960And machine direction orientation is really easy, you know, just one role moving faster than another.
00:24:42.960 -->
00:24:56.319So it's very inexpensive, very simple process, and it makes these really good, strong membranes, but they're limited to polymers that have that row nucleated crystal structure.
00:24:56.640 -->
00:24:58.079And that's called the dry process, right?
00:24:58.400 -->
00:24:59.599That's the dry process.
00:24:59.599 -->
00:25:04.880So cheap, easy to do, and very well protected by Celanese uh patents.
00:25:04.880 -->
00:25:10.880But the wet process is there there's the tips that 3M used.
00:25:10.880 -->
00:25:13.119You get you get temperature-induced phase separation.
00:25:13.119 -->
00:25:29.839But basically the idea is that you take a membrane, you dissolve it in oil, extrude it, and now you have a film that's filled with oil, and you it you stretch that film in the machine direction and the cross machine direction.
00:25:29.839 -->
00:25:35.759And that gives you strength by orienting the polymer, and you extract out the oil.
00:25:35.759 -->
00:25:40.079So that process you have to do extraction, which is expensive.
00:25:40.079 -->
00:25:43.039You know, you have to you have a solvent that you have to recover, right?
00:25:43.039 -->
00:25:47.839And you have biaxial orientation, not just the machine direction, but the cross machine direction.
00:25:47.839 -->
00:25:51.279And cross-machine direction orientation is very expensive.
00:25:51.279 -->
00:25:57.039You know, you have to have these clamps that grab on the sides, and it's a very elaborate machine.
00:25:57.039 -->
00:25:59.440So it's a very expensive process.
00:25:59.440 -->
00:26:11.200But the advantage of the wet process is that you can use a polymer called ultra-high molecular weight polyethylene, and that polymer is stronger than steel when it's highly oriented.
00:26:11.200 -->
00:26:13.440So you get this incredible strength.
00:26:13.440 -->
00:26:20.960So Celgard couldn't compete with the wet process on strength, but the cost advantage was unbeatable, you know.
00:26:20.960 -->
00:26:26.079So you gotta trade-off, you could have a super strong or lower cost.
00:26:26.480 -->
00:26:26.880Right.
00:26:26.880 -->
00:26:35.359And you know, just from experience in the lab, if you have a roll of separator, you know, you pull on the machine direction uh quite strong.
00:26:35.359 -->
00:26:40.240Pull the transverse direction and it will just split like nothing.
00:26:40.240 -->
00:26:42.480That would be from the dry process, right?
00:26:42.799 -->
00:26:43.359That's right.
00:26:43.359 -->
00:26:44.240That's right.
00:26:44.559 -->
00:26:48.799And then if I have something from the wet process, it'll be strong in both directions.
00:26:49.119 -->
00:26:49.680That's right.
00:26:49.680 -->
00:26:55.599Yeah, what one neat thing at Celgard did was they built this battery applications lab.
00:26:55.599 -->
00:26:56.880So they built a dry room.
00:26:56.880 -->
00:27:04.960And and so we could make batteries, and I from great grace I knew how to make batteries, and so I could make I made these little double A cells.
00:27:04.960 -->
00:27:08.960And um the customers would give us electrodes.
00:27:08.960 -->
00:27:13.680And so we could make our own batteries in-house and do testing.
00:27:13.680 -->
00:27:19.119But after a little bit, the the companies said, you know, why even bother making your own?
00:27:19.119 -->
00:27:21.440We'll just give you cells if you want to test them with something.
00:27:21.440 -->
00:27:23.359We'll just we'll just give you the cells to test.
00:27:23.359 -->
00:27:34.480Um, but the main focus of the battery applications lab was to develop a suite of tests for characterizing separators, and then we could evaluate the different types of separators that Celgard was making.
00:27:34.480 -->
00:27:43.440So that they had three approaches the bilayer, the trilayer, and then a coated, a polypropylene coated with some sort of low melting wax.
00:27:43.440 -->
00:27:46.079And so they wanted to decide which was the best option.
00:27:46.079 -->
00:27:51.279And my testing came out that the trilayer was the best option.
00:27:51.279 -->
00:27:53.440It had so many advantages, the trilayer.
00:27:53.440 -->
00:28:02.799It was stronger in in terms of puncture strength on both directions, whereas the the bilayer was only strong in one direction, not the other direction.
00:28:02.799 -->
00:28:12.880There were other things too, but it at any case, one of the customers in Japan was had a very persuasive technical manager.
00:28:12.880 -->
00:28:20.400He was very charismatic and he really convinced the management team that the bilayer was the way to go, regardless of our internal testing.
00:28:20.400 -->
00:28:22.160He convinced them to do that.
00:28:22.160 -->
00:28:26.079And so and also they like they liked the idea of bilayer.
00:28:26.079 -->
00:28:28.720It just seemed simpler in terms of using the existing equipment.
00:28:28.720 -->
00:28:30.960They just extruded these films, laminated them.
00:28:30.960 -->
00:28:32.799It just seemed like an easier thing to make.
00:28:32.799 -->
00:28:36.640But then that customer ran into problems using it in production.
00:28:36.640 -->
00:28:39.839And so then we flipped back to trilayer.
00:28:39.839 -->
00:28:49.759The trilayer process was this this uh brilliant guy, Wei Cheng Yu, developed this whole process for making the trilayer separator with Celgard.
00:28:49.759 -->
00:28:54.880And we started making it, it was a fantastic success.
00:28:54.880 -->
00:29:02.160And the the problem though, with even Celgard had those defects that Sony didn't like.
00:29:02.160 -->
00:29:05.839But Celgard had nowhere near the data defects that we had at Grace.
00:29:05.839 -->
00:29:16.720They had what what they would do is at the end of the production line, they would have somebody look, and when they saw a defect, they would make a note that okay, at this point we have a defect.
00:29:16.720 -->
00:29:24.240And then when they went to slitting, they would stop the they would cut the roll where that defect was, and so that it wouldn't go to the customer.
00:29:24.240 -->
00:29:29.119And so Celgard could only make rolls a certain length be before they would hit a defect.
00:29:29.119 -->
00:29:31.680And then all the other stuff was just wasted.
00:29:31.680 -->
00:29:33.680Right, and you couldn't splice them together.
00:29:33.680 -->
00:29:35.839No, no, you couldn't couldn't splice them together.
00:29:35.839 -->
00:29:36.240Right.
00:29:36.240 -->
00:29:45.920Um But remember Celgard had this small production line, but even at the time it had only been running like a month or two a year and the rest of the time doing research.
00:29:45.920 -->
00:29:50.720But then when lithium ion came around, we started running it 24-7.
00:29:50.720 -->
00:29:58.559And we we had to get everybody in the company volunteered for certain shifts.
00:29:58.559 -->
00:30:02.240You know, I I would I would have you know like a 6 to 8 a.m.
00:30:02.240 -->
00:30:02.960shift or whatever.
00:30:02.960 -->
00:30:06.559So and then we would look it to see whether you got those defects or not.
00:30:06.559 -->
00:30:10.640That would be the job of the you know, the people like me who weren't experts at manufacturing.
00:30:10.640 -->
00:30:12.319You just say, okay, there's a defect.
00:30:12.559 -->
00:30:14.079Oh, I see, no vision system for this.
00:30:14.079 -->
00:30:15.599This is humans looking for the defects.
00:30:15.759 -->
00:30:16.240That's right.
00:30:16.240 -->
00:30:16.880Exactly.
00:30:16.880 -->
00:30:18.480What year are we here?
00:30:18.480 -->
00:30:21.519This is uh mid-90s.
00:30:21.519 -->
00:30:27.039Um, and then Celgard decided to invest in scaling up and in a new plant.
00:30:27.039 -->
00:30:29.279And then everything changed.
00:30:29.279 -->
00:30:36.640They got a vision system for identifying defects, they got bigger production lines, every it also they improved the process of the defects.
00:30:36.640 -->
00:30:48.559I think I forget the exact numbers, but it was like we started off with like 200 meter rolls, then 300 meter rolls, then 400-meter rolls, then 500, and then it got to be as long as your customer wanted, you know.
00:30:48.960 -->
00:30:49.279Right.
00:30:49.279 -->
00:30:55.119I've got one thing is I'm curious about, and it's just so basic, which is how thick were separators then?
00:30:55.119 -->
00:30:59.920Uh like when that Tonen separator that came out in the 1991 Sony battery, how thick was that?
00:30:59.920 -->
00:31:06.400And then how thick was it by you know the 94 to 98 era when when uh Celanese is there?
00:31:06.400 -->
00:31:10.000You know, kind of because I know they're they gotta get thinner over the years.
00:31:10.319 -->
00:31:13.11925 microns was the initial thickness target.
00:31:13.119 -->
00:31:14.960And then they they got thinner and thinner.
00:31:15.359 -->
00:31:16.480What were the defects?
00:31:16.480 -->
00:31:17.359What would you look for?
00:31:17.359 -->
00:31:20.799What what was the cause of the defect or what was that defect?
00:31:21.279 -->
00:31:29.119I think it's just a gel, the gelling of the polymer, but it's like you know when it goes under a light, you can see it's like a just a dis a discoloration.
00:31:29.119 -->
00:31:39.839And I think what happens in a lithium-ion battery, it's not a it's a poor it's a non-porous region, and so you get you can cause lithium metal to grow, you know, if you block off part of the uh the negative.
00:31:40.240 -->
00:31:40.480Right.
00:31:40.480 -->
00:31:44.079You suddenly have a region of really low conductivity.
00:31:44.720 -->
00:31:45.200Exactly.
00:31:45.599 -->
00:31:47.200I had a question about people.
00:31:47.200 -->
00:31:51.920So how big was the group at uh at Celgard or at Celanese?
00:31:52.319 -->
00:32:01.839So for the battery group, it was just me and a technician, Mark Farabee, who was really he was a brilliant guy, and he ended up becoming an engineer.
00:32:01.839 -->
00:32:06.319But we had we had these grad students come in that would help us out doing things.
00:32:06.319 -->
00:32:09.440And you know, they were brilliant PhD students that helped them.
00:32:09.440 -->
00:32:11.680So it really that really helped out.
00:32:12.160 -->
00:32:14.880Okay, so they didn't have a battery group for primaries.
00:32:14.880 -->
00:32:16.000Like that was it.
00:32:16.000 -->
00:32:18.319Like you were the battery group with one other person.
00:32:18.799 -->
00:32:25.759They had a guy, he was like a sort of a technical sales type guy, Deut Hoffman, who's a really good guy.
00:32:25.759 -->
00:32:30.240He was very knowledgeable about batteries, but uh just to digress, it's an interesting story.
00:32:30.240 -->
00:32:37.279Mashusta, uh Panasonic, there's they had were trying to find separators for their primary batteries.
00:32:37.279 -->
00:32:47.759And they the sales guy that there that they put in charge to go out and find a separator to buy for their primary batteries, he did this extensive technology review.
00:32:47.759 -->
00:32:54.400And he came to Celgard and told them, hey, this material you're making would be good in primary batteries.
00:32:54.400 -->
00:32:58.160And and that's how they got into that business was through this.
00:32:58.480 -->
00:32:59.359Oh wow.
00:32:59.359 -->
00:33:00.880Wow.
00:33:00.880 -->
00:33:03.279Talk about business development.
00:33:03.279 -->
00:33:13.279So, you know, there's a fairly prominent figure in in the battery world, and he was, I believe, CTO of Celgard, John Zhang.
00:33:13.279 -->
00:33:14.880Did you overlap with John Zhang?
00:33:15.119 -->
00:33:16.640No, he was he was my replacement.
00:33:16.640 -->
00:33:19.119I quit Sell Guard and they they hired John Zhang.
00:33:19.359 -->
00:33:19.839I see.
00:33:19.839 -->
00:33:20.160Okay.
00:33:20.160 -->
00:33:22.880So maybe you can tell us why did you quit Sell Guard?
00:33:22.880 -->
00:33:23.759What happened?
00:33:24.160 -->
00:33:28.079So in 1998, Sell Guard was a huge success.
00:33:28.079 -->
00:33:36.079And uh Menahem Anderman had was at PolyStor, and he was the VP of technology.
00:33:36.079 -->
00:33:42.319And he gave me this opportunity to go there and work on lithium ion batteries.
00:33:42.319 -->
00:33:53.359And I was I I really liked the modeling, understanding batteries through modeling, and I had modeled separators and stuff, and I there was not more too much I could do more with separators, I felt like.
00:33:53.359 -->
00:33:57.200I mean, it just it wasn't really very challenging at that point.
00:33:57.200 -->
00:34:06.079And I'd like the I the idea of working with a battery company, and especially Poly Store had a very good reputation in terms of the quality of their cells, and so it's just an opportunity.
00:34:06.079 -->
00:34:07.680And also, I like California.
00:34:07.680 -->
00:34:11.280I mean, I like the Bay Area, I really liked that I wanted to live here.
00:34:11.440 -->
00:34:14.480So Okay, so so there was a broadening of horizons.
00:34:14.480 -->
00:34:24.639Basically, it was like, okay, I'm gonna go from a company that's just making separator to a company that's actually gonna make cells, and I get to see the final product.
00:34:25.039 -->
00:34:26.000That's right.
00:34:26.000 -->
00:34:26.880Okay.
00:34:27.440 -->
00:34:30.159So so what was um PolyStor's story?
00:34:30.159 -->
00:34:33.519Like just like a very brief sort of the total arc of PolyStor.
00:34:34.239 -->
00:34:39.840Uh PolyStor, they were gonna make lithium-ion batteries to compete with Sony.
00:34:39.840 -->
00:34:41.119I see.
00:34:41.119 -->
00:34:43.679Which I didn't really understand until I got there.
00:34:43.679 -->
00:34:50.159I thought I was gonna be working on batteries for hybrid electric vehicles, which I was a super believer in.
00:34:50.159 -->
00:35:04.639I had read these uh reports you know from the uh I don't know, it was the National Engineering Academy or whatever, National Science Academy, or they'd been studying battery technology and thought it was a really good fit for this thing.
00:35:04.639 -->
00:35:09.280It was for there was just no question, it was hybrids made a lot of sense, a lot of sense.
00:35:09.280 -->
00:35:09.920Right.
00:35:09.920 -->
00:35:12.639And that's what I was interested interested in doing.
00:35:12.639 -->
00:35:32.320And then when I got to PolyStor, I was very, very disappointed because I realized that first of all, with Celgard, I had had the opportunity to go to Japan many times, and I had visited all the companies and had seen plants, and you know, the Sony plants are just awesome, you know, they're just the the robotics and everything.
00:35:32.320 -->
00:35:39.840And then you come to this little place that's like you know almost like a small warehouse type thing.
00:35:39.840 -->
00:35:43.519It's like you know, really primitive and felt more like a workshop, yeah.
00:35:43.519 -->
00:35:45.280And these guys are gonna beat Sony.
00:35:45.280 -->
00:35:48.159It's like what you know, it just wasn't serious.
00:35:48.159 -->
00:35:50.960But they had a lot of good people there.
00:35:50.960 -->
00:35:54.639They had uh Mark Juskow, you know, really early.
00:35:54.639 -->
00:35:55.360Oh, yeah.
00:35:55.360 -->
00:36:24.719Um they had uh Dave Fouchard, another uh Molly guy, they had uh Chris Cassadine from Electrovya, and then they had my uh two guys that that well at least that I knew from uh were working in my lab at uh at Celgard, and I had a really nice team, about five PhDs to work on rechargeable, uh work on hybrid electric vehicle batteries.
00:36:24.719 -->
00:36:36.159And we had a research uh program from USABC, and we had this guy, uh Harold Haskins from Ford, who was just a brilliant engineer.
00:36:36.159 -->
00:36:39.760He had done a lot of the specifications for what hybrid batteries should do.
00:36:39.760 -->
00:36:42.559I mean, I was the manager, but he was really sort of directing the work.
00:36:42.559 -->
00:36:49.920I mean, he was like really brilliant and knew what was good, knew what we needed to do, and I was pretty open with them what our capabilities were.
00:36:49.920 -->
00:36:57.599And so we were making a lot of progress on making these these larger batteries, and we had a but we had a very long way to go.
00:36:57.599 -->
00:37:00.239I mean, the the they were really, really dangerous.
00:37:00.400 -->
00:37:05.039Uh what size cells were you uh making then or trying to make prototyping?
00:37:05.360 -->
00:37:11.119I I think they were like uh maybe the size of two beer cans together or something like that.
00:37:11.440 -->
00:37:12.000Ah, okay.
00:37:12.239 -->
00:37:13.119They were big cells.
00:37:13.119 -->
00:37:14.079Um cylindrical.
00:37:14.079 -->
00:37:15.039Cylindrical, yeah.
00:37:15.039 -->
00:37:15.280Okay.
00:37:15.280 -->
00:37:20.320Chris Castledine did did some safety testing and they just they were really, really dangerous.
00:37:20.400 -->
00:37:25.840Uh so were they were they bigger than what would be um like a 4680?
00:37:26.400 -->
00:37:27.440Yeah, yeah, yeah.
00:37:27.440 -->
00:37:28.079They were bigger.
00:37:28.320 -->
00:37:29.039And wound.
00:37:29.360 -->
00:37:30.079And wound.
00:37:30.079 -->
00:37:30.880Okay.
00:37:30.880 -->
00:37:34.320So what do you mean when you say a really dangerous?
00:37:34.320 -->
00:37:36.480Do you need like did you ever see any of the safety tests?
00:37:36.639 -->
00:37:40.159Yeah, he yeah, he blew some up that he would put them together and blow them up.
00:37:40.159 -->
00:37:42.719It was really exciting.
00:37:45.199 -->
00:37:46.239I can't imagine.
00:37:46.239 -->
00:37:48.480I mean, I've I've seen a lot of these sort of tests.
00:37:48.480 -->
00:37:57.599How how did he I mean now now when we do that kind of thing, we have like a special facility set up for for but was uh can you tell how that tell us how that test was done?
00:37:57.599 -->
00:37:59.199Was it like done outside or was it done?
00:37:59.199 -->
00:38:00.880Oh, it was outside.
00:38:01.280 -->
00:38:11.280But it it it became clear I I was at PolyStor almost a year, and it was really clear that they were fixed fixated on beating Sony.
00:38:11.280 -->
00:38:26.320Okay and I it was just so crazy to me that you know that I just I argued with Menachem, my boss, I argued with the management, I talked to everybody I could, and right, and I said this is crazy, and I and I and I quit.
00:38:26.320 -->
00:38:27.599I I just couldn't take it.
00:38:27.599 -->
00:38:29.760It's just it was just so crazy.
00:38:29.760 -->
00:38:41.199Um but the a the USABC people they liked me for and they they gave me a consulting job, which was pretty good to keep working with them.
00:38:41.199 -->
00:39:00.639So and then also we had hired uh this from Celgard, I had met one of the professors at University of South Carolina, Professor uh John Van Z, and he had been a consultant for with PolyStor too, and we got to talking and we we he was really into modeling.
00:39:00.639 -->
00:39:06.719Okay, and he got he and I we had this idea of starting a company that did battery modeling.
00:39:06.719 -->
00:39:12.159You know, because it was clear battery modeling, you know, batteries were gonna take off and people needed models.
00:39:12.159 -->
00:39:12.880Yep.
00:39:12.880 -->
00:39:17.760And so we started this company uh battery design.
00:39:17.760 -->
00:39:20.079And he he really was great.
00:39:20.079 -->
00:39:28.159I mean, he set up an office, he hired a couple people, uh uh uh computer engineers to work on uh work on it.
00:39:28.400 -->
00:39:28.639Right.
00:39:28.800 -->
00:39:37.679And we got uh some NSF funding and we developed a product and we started selling it, and things were looking good.
00:39:37.679 -->
00:39:42.480Um, but then he got he was also really interested in fuel cells.
00:39:42.800 -->
00:39:43.119Okay.
00:39:43.360 -->
00:39:48.800And he got this very lucrative offer to develop a fuel cell product.
00:39:48.800 -->
00:39:49.440Huh.
00:39:49.440 -->
00:39:59.039And so he left the company and went to work on that, and it was just m me uh uh with this with the two uh engineers.
00:39:59.440 -->
00:40:06.480So at this point, you're basically self-funded with the addition of grants, um, but you don't have outside investors?
00:40:06.880 -->
00:40:08.159No outside investors.
00:40:08.480 -->
00:40:11.840And at this point, do you have a a product in the market already?
00:40:12.159 -->
00:40:14.800It took us about four years, and we got a product.
00:40:15.039 -->
00:40:15.360Oh wow.
00:40:15.360 -->
00:40:18.639So were you just uh not paying yourself for four years?
00:40:19.280 -->
00:40:20.079Yeah, just about.
00:40:20.079 -->
00:40:23.519I I was make making payroll was my challenge in life.
00:40:23.840 -->
00:40:24.400I see.
00:40:24.400 -->
00:40:29.119So you were you were you were probably personally funding those two guys for four years?
00:40:29.280 -->
00:40:29.599Yeah.
00:40:29.599 -->
00:40:30.159Yeah.
00:40:30.159 -->
00:40:31.840The true bootstrap method.
00:40:31.840 -->
00:40:32.159Yeah.
00:40:32.400 -->
00:40:36.239But I did a lot a lot of consulting jobs and things like that I would do.
00:40:36.239 -->
00:40:37.840So I was traveling a lot.
00:40:37.840 -->
00:40:41.280In the consulting about separators, or what would you consider?
00:40:41.280 -->
00:40:42.480Yeah, separators.
00:40:42.480 -->
00:40:47.039Every big name company you can think of was interested in the in this market now.
00:40:47.039 -->
00:40:53.119Just as a side note, the the Grace division that made lead acid battery separators dynamic.
00:40:53.199 -->
00:40:53.679Yep.
00:40:54.639 -->
00:40:59.039And they remember they were not interested in lithium-ion technology, they thought it was a joke.
00:40:59.039 -->
00:41:01.360They ended up acquiring Celgard.
00:41:02.320 -->
00:41:02.639Ah.
00:41:02.639 -->
00:41:05.760Yeah.
00:41:06.320 -->
00:41:10.079And they held it for a few years and ended up selling it again to Asahi.
00:41:11.039 -->
00:41:11.440Yeah.
00:41:11.840 -->
00:41:12.159Right.
00:41:12.159 -->
00:41:15.599And currently it's it's it resides with Asahi today.
00:41:16.719 -->
00:41:16.880Right.
00:41:16.880 -->
00:41:17.519Got it.
00:41:17.519 -->
00:41:24.000So anyway, you're up at Polly Store, you're there for a year, you decide there's no way we're we're gonna compete with with Sony.
00:41:24.000 -->
00:41:26.159Why what are they thinking here?
00:41:26.159 -->
00:41:35.280And uh and then you leave, and then you start the battery design LLC, and and that and then you're alone there again, uh, with but with the the two guys with you.
00:41:35.280 -->
00:41:42.159Uh, how does what's your outlook right there of like um okay, I'm I'm gonna make this work?
00:41:42.159 -->
00:41:45.199Or was there a period where you thought, I'm not sure I'm gonna make this work?
00:41:45.199 -->
00:41:47.280Uh, you know, kind of put us in your mindset.
00:41:47.280 -->
00:41:58.000Because I mean, this is another interesting thing about that that we can learn in this podcast is like from that entrepreneur's uh perspective, what's it like from the the hard position moving forward?
00:41:58.400 -->
00:42:01.199Oh, I I had I I had drank the Kool-Aid.
00:42:01.199 -->
00:42:02.880I mean, I was sure it was gonna work.
00:42:02.880 -->
00:42:04.719That it just people just had to see it.
00:42:04.719 -->
00:42:12.639You know, I I would go to these battery companies and they would have their each each company would have one guy who did modeling.
00:42:12.639 -->
00:42:21.440You know, Molly had Jan Reemers and uh uh Medtronics had a guy, and everybody, every every company had some guy that did modeling.
00:42:21.440 -->
00:42:28.480And right, so they would say, Well, you know, you you're just one guy, you know, we have a guy.
00:42:28.480 -->
00:42:29.760What's the you know?
00:42:29.760 -->
00:42:43.760And I would say but I have a product, and you know, if if you buy the product, then I'll be able to hire more guys and we'll be work on the product, and then having a product will get input from all these different companies, and it didn't sell.
00:42:43.760 -->
00:42:46.480It it you know it just did not sell.
00:42:46.480 -->
00:42:57.280Uh but occasionally uh periodically, these companies would hire me as a consultant to do stuff for them, and really they just wanted to update their maybe their internal stuff to make sure it was as good as what I was doing.
00:42:57.280 -->
00:43:01.679I got a best business to keep me going, but I wasn't going anywhere.
00:43:01.679 -->
00:43:03.199I was just staying afloat.
00:43:03.199 -->
00:43:03.760Right.
00:43:04.000 -->
00:43:07.280And Battery Design Studio did a number of things, right?
00:43:07.280 -->
00:43:14.480Like one thing that it did is that it allowed you to visualize what will my cell look like, like the jelly roll physically, what will it look like?
00:43:14.480 -->
00:43:16.320How will my electrodes line up?
00:43:16.320 -->
00:43:17.440What else did it do?
00:43:17.440 -->
00:43:19.440Like what was the breadth of the offering?
00:43:19.840 -->
00:43:27.519The main feature at the time was that it implemented Professor Newman's model, the the Doyle uh Fuller Newman model.
00:43:27.519 -->
00:43:27.920Yep.
00:43:27.920 -->
00:43:30.559And it had it was very powerful in that.
00:43:30.559 -->
00:43:40.800So I did so another backtrack a little bit, but story when I when I was at WR Grace, I I we did a model of the lithium TIS2 cell.
00:43:40.800 -->
00:43:46.159And we used the uh Atlung and West model for the TIS2 electrodes.
00:43:46.159 -->
00:43:59.760What was really fantastic was we've we've we tried to fit the data at different different discharge rates to the model, we found out that the only way we could get a good fit was by assuming that the surface area of the active particles.
00:43:59.760 -->
00:44:02.159Was severely blocked.
00:44:02.159 -->
00:44:11.519And it turned out that uh ultra-hymoleculate polyethylene is not a good binder for TIS2 because it does block the surface.
00:44:11.519 -->
00:44:12.320I see.
00:44:12.320 -->
00:44:17.119And when we went to a different binder, all the rate capability problem went away.
00:44:17.119 -->
00:44:18.639It solved all the problems.
00:44:18.639 -->
00:44:23.039And and that made me like such a true believer in cell modeling.
00:44:23.280 -->
00:44:23.840Ah, right.
00:44:23.840 -->
00:44:27.840Yeah, the modeling told you right there, like, hey, this must be what the issue is.
00:44:27.840 -->
00:44:28.639This is what we should go.
00:44:28.639 -->
00:44:36.800It's hard to know when you're just looking at overall cell data, but the model could look at the nuances of the cell data and say, oh, I know I can spot your problem.
00:44:36.800 -->
00:44:37.519It's the binder.
00:44:37.519 -->
00:44:38.800It's blocking the surface.
00:44:38.800 -->
00:44:39.679Exactly.
00:44:39.920 -->
00:44:42.639And when I got to PolyStor, I had uh Dr.
00:44:42.639 -->
00:44:50.639Nagarajan, who was also a brilliant modeler, and he was running models of the cells 24-7.
00:44:50.639 -->
00:45:05.599And we we we started to understand how the for hybrid vehicles, you had to get the uh the rate capability so high, you had you had to optimize both electrodes to get that good rate capability, and we were seeing a lot of effects.
00:45:05.599 -->
00:45:13.760So it's so I really just got more and more into uh into in the convinced of the value of the modeling that it that it was true.
00:45:13.760 -->
00:45:18.480And so we we took the Newman model, we added a lot of features to it.
00:45:18.480 -->
00:45:26.159So we before Newman did it, we put in different particle sizes and different blends of active materials.
00:45:26.159 -->
00:45:37.360But the the the real brilliant thing of Newman's was that he had non-ideal electrolyte theory in the model, whereas the earlier models did not.
00:45:37.360 -->
00:45:40.400They just had solid phase diffusion, basically, and then they use ideal.
00:45:40.880 -->
00:45:41.199I see.
00:45:41.599 -->
00:45:47.519And in lithium ion batteries, taking into account the non-ideality of the electrolytes is critical.
00:45:47.519 -->
00:45:53.599And Molly Energy did a fantastic paper on this, where they looked at what happened.
00:45:53.599 -->
00:46:00.719They they they had a, I guess, a research grant where they actually measured all the properties of electrolyte, lithium-ion battery electrolyte.
00:46:00.719 -->
00:46:12.400And they showed with their cells what the result was of making different assumptions that people made of, you know, and that you really had to take into account the non-ideality to get it right.
00:46:12.400 -->
00:46:22.079And I'm gonna I'm gonna blank on his name, but we met this one scientist from Idaho National Labs who had developed a model for developing properties of electrolytes.
00:46:22.400 -->
00:46:23.360Kevin Nearing?
00:46:23.599 -->
00:46:25.920Kevin Gehring, right, right, thank you.
00:46:25.920 -->
00:46:47.360And and Kevin provided us with all these electrolyte uh properties, sets of properties, and we put those into the Newman model, and so that you could get really good fidelity simulations of lithium-ion cells, because you had all the right physical properties there, and it really it really was impressive.
00:46:47.360 -->
00:46:53.519Um so so we had just about every capability you could do with the Newman model we had in there.
00:46:53.519 -->
00:47:00.639What what happened finally was in like uh in 2010.
00:47:00.639 -->
00:47:13.8402010, CD Adapco, this company, which was a a big uh they they came and asked me if I would be interested in having them take over the battery design studio software.
00:47:13.840 -->
00:47:17.679They they would have me as a consultant, but they would do uh d handle it.
00:47:17.679 -->
00:47:20.719And at that point, I I gave up.
00:47:20.719 -->
00:47:24.000I said, okay, yeah, let them do it.
00:47:24.400 -->
00:47:39.119And and was their pitch basically, look, we're an engineering software company, we have the whole apparatus for sales and distribution, you have the brains on this, you know the science the science, we can do the the business development.
00:47:39.119 -->
00:47:40.960Is it was that sort of the pitch?
00:47:42.000 -->
00:47:42.880It was interesting.
00:47:42.880 -->
00:47:47.039They they had this guy, this British guy named Richard Johns.
00:47:47.039 -->
00:47:52.719He call he called me up and said, Hey, I'm in downtown.
00:47:52.719 -->
00:47:56.960How would you like to meet for uh for uh lunch?
00:47:56.960 -->
00:48:00.239Okay, you know, downtown Pleasanton, where I live.
00:48:00.400 -->
00:48:01.920You know, yeah, okay, right on.
00:48:02.079 -->
00:48:03.840It's like he's called me out of the blue.
00:48:03.840 -->
00:48:08.239He said, I'm you know, I'm interested in battery modeling and I'd like to just meet you and talk to you.
00:48:08.239 -->
00:48:12.639And he was a uh one of the senior, uh more senior guy at CD Adapco.
00:48:12.639 -->
00:48:23.599And we I had lunch with him and talked to him about modeling, and you know, he did his modeling was in combustion, but just his the way of looking at it was like a kindred soul.
00:48:23.599 -->
00:48:34.960I mean, I I liked I just this we had we got a lot we ended up talking for like three, four hours just about about modeling just and the potential for software and stuff like that.
00:48:35.280 -->
00:48:36.079Awesome.
00:48:36.320 -->
00:48:43.760And then he said, Why don't you go to uh meet our CEO in New York?
00:48:43.760 -->
00:48:50.480And they said, and I said, Okay, and he I flew out there, I was expecting to spend some time.
00:48:50.480 -->
00:48:54.320I go into this office, the the CEO, the guy is Steve McDonald.
00:48:54.320 -->
00:48:56.159He started the company, CD Adapco.
00:48:56.159 -->
00:48:59.679Okay, and he said, Look, this is the way we do it.
00:48:59.679 -->
00:49:05.039It's sales is one third, development is one third, and ownership is one third.
00:49:05.039 -->
00:49:08.239So I'll give you one-third of everything.
00:49:08.239 -->
00:49:10.320Is that okay with you?
00:49:10.320 -->
00:49:12.239And I said, Yeah.
00:49:12.239 -->
00:49:13.119He said, Okay.
00:49:13.119 -->
00:49:14.239And that was it.
00:49:18.559 -->
00:49:20.559A really quick and simple conversation.
00:49:20.559 -->
00:49:21.599It was like two minutes.
00:49:21.599 -->
00:49:22.239I love that.
00:49:22.320 -->
00:49:23.760I know I flew all the way there.
00:49:23.760 -->
00:49:32.000It's like, and I got a contract in the mail and uh signed it, and that was that was it.
00:49:32.880 -->
00:49:33.679Awesome.
00:49:33.920 -->
00:49:34.239Okay.
00:49:34.239 -->
00:49:39.280And then were they were they able to grow the user base significantly?
00:49:39.679 -->
00:49:40.079Yeah.
00:49:40.079 -->
00:49:40.960Oh yeah.
00:49:40.960 -->
00:49:48.559Because, you know, all the people who said, well, you're just one guy, we have one guy, you know, now it was this big company.
00:49:48.719 -->
00:49:48.880Right.
00:49:48.880 -->
00:49:49.599I see.
00:49:49.599 -->
00:49:50.000Right.
00:49:50.000 -->
00:49:52.159Yeah, that kind of changes that conversation.
00:49:52.559 -->
00:49:56.960But the the real dynamic was the market was in Asia.
00:49:56.960 -->
00:49:59.119That's where all the lithium ion batteries were made.
00:49:59.119 -->
00:50:06.639And so they had a big office in Japan, China, Korea, all these Japanese uh pr companies bought it.
00:50:06.639 -->
00:50:07.760It just took off there.
00:50:07.760 -->
00:50:09.199And that's nice.
00:50:09.199 -->
00:50:10.320That was the big difference.
00:50:10.320 -->
00:50:14.800Is that the US really didn't have much of a m industry, you know, to really use it.
00:50:15.119 -->
00:50:15.760Right.
00:50:16.159 -->
00:50:16.400Right.
00:50:16.480 -->
00:50:23.760So you needed to be you needed to have a presence in Asia in order to make the in order to in order to grow sell, make sales.
00:50:24.000 -->
00:50:24.559Yes.
00:50:24.559 -->
00:50:29.599So then its footprint increases and grows with CD Adapco.
00:50:29.599 -->
00:50:33.920It sort of becomes a a reference in the field for for cell design.
00:50:33.920 -->
00:50:36.079And then and then what happens?
00:50:36.079 -->
00:50:37.920Siemens comes into play, right?
00:50:38.239 -->
00:50:38.480Yeah.
00:50:38.480 -->
00:50:38.880Okay.
00:50:38.880 -->
00:50:42.480So the contract with CD Adapco was was five years.
00:50:42.480 -->
00:50:46.480They had a right to either acquire it or give it back to me.
00:50:46.480 -->
00:50:48.880And then it happened exactly at that point.
00:50:48.880 -->
00:50:53.199Siemens offered to buy the CD Adapco for a billion dollars.
00:50:53.599 -->
00:50:53.920Wow.
00:50:53.920 -->
00:50:54.719Oh, jeez.
00:50:55.920 -->
00:51:01.920And Steve McDonald, you know, who is this incredible work, you know, CD Adapco was his life.
00:51:01.920 -->
00:51:06.719He started the company, you know, made it to the billion dollar company from nothing.
00:51:06.719 -->
00:51:08.719Um, he died.
00:51:08.719 -->
00:51:10.400Oh no.
00:51:10.400 -->
00:51:16.559And then the people left over didn't know what to do with the battery software.
00:51:16.559 -->
00:51:23.199Um, but finally at the end, they decided to go ahead and do it and renew the contract for another five years.
00:51:23.199 -->
00:51:26.159And so it was part of the Siemens sale.
00:51:26.960 -->
00:51:33.119And then when Siemens acquires CD Adapco, do you remain involved?
00:51:33.119 -->
00:51:34.800And I mean, are you working with Siemens?
00:51:35.119 -->
00:51:37.280Yeah, yeah, they kept me on, they kept me on as a consultant.
00:51:37.280 -->
00:51:40.079And then at the end of that five years, they they acquired it.
00:51:40.079 -->
00:51:42.239That the sales had grown enough and so on.
00:51:42.239 -->
00:51:45.199And then they added a lot of good capabilities to it.
00:51:45.199 -->
00:51:48.400So like the Newman model was always my interest.
00:51:48.400 -->
00:52:04.559But with their technical service people said, you know, that the circuit model was actually a lot more useful to people because the the circuit model you could automate the fitting of the parameters.
00:52:04.559 -->
00:52:16.239So the the battery engineers at the companies could do some tests, automate the fitting, you know, just put in those test results, get a model that then would that reproduce their experimental results exactly.
00:52:16.239 -->
00:52:21.920So there's a model called the RCR model, which was developed by USABC.
00:52:21.920 -->
00:52:31.199So Harold Haskins actually developed it and nobody could figure out he it's it's an analytical exp it's an analytical equation.
00:52:31.199 -->
00:52:36.880And nobody could figure out how he did it, because he didn't give any notes, he just gave the equation.
00:52:36.880 -->
00:52:43.519And finally, uh Mark Verbruggey at GM did a derivation of it.
00:52:43.519 -->
00:52:45.599He did he figured out how to derive it.
00:52:45.599 -->
00:52:55.519Um but it was it's so it's this analytical model that you can fit to uh uh automatically, and you know it's it's just incredible.
00:52:55.519 -->
00:53:05.840You know, you just you get really good fits, it you know, it reproduces the experimental results r really well, and people just loved it.
00:53:07.039 -->
00:53:31.599And is it is it a case where it's more of a kind of more of an empirical, I'm gonna fit the data, and then I mean well to reproduce the data in the field, and the evaluation of the model is pretty cheap compared to the new middle model, which is more of a physics-driven model where I'm trying to sort of model the actual properties of the of the components that are part of my system.
00:53:32.400 -->
00:53:37.440The Newman model is very attractive to the battery developers.
00:53:37.440 -->
00:53:41.280Because they want to understand what's going on inside the battery.
00:53:41.599 -->
00:53:41.920Right.
00:53:42.159 -->
00:53:46.320But the battery developers themselves are not that into modeling.
00:53:46.320 -->
00:53:50.320You know, they they sit there and say, Oh, I can build batteries and test them.
00:53:50.320 -->
00:53:51.039I don't need a model.
00:53:51.039 -->
00:53:55.679I can if I have a question about something, I just try it out experimentally, yeah, and I don't know the answer.
00:53:55.679 -->
00:54:03.599But automotive companies, they love this, they love the they they live by model, you know, uh you this cars are designed through models.
00:54:03.599 -->
00:54:07.679You know, it's modeling is their bread and is their lifeblood, whatever.
00:54:07.679 -->
00:54:09.840You know, their whole structure is modeling.
00:54:09.840 -->
00:54:11.920And so they love these models.
00:54:12.480 -->
00:54:21.679If the Newman model is more of a physics-based and the RCR model is more like you fit your data, are they equally good at extrapolation?
00:54:21.679 -->
00:54:27.199I would sort of tend to think that the Newman model would be end up being better at extrapolation, right?
00:54:27.519 -->
00:54:28.639Oh, yeah, exactly.
00:54:28.639 -->
00:54:32.320The RCR model is is terrible at extrapolation.
00:54:32.320 -->
00:54:33.679You can't believe anything it says.
00:54:33.679 -->
00:54:36.159Whereas in the Newman model, you have real insight.
00:54:36.159 -->
00:54:38.559I mean, you can extrapolate, you can you can look at things.
00:54:38.559 -->
00:54:44.400You know, if you if you're trying to develop a uh a charging algorithm, you're trying to see how fast I can charge it.
00:54:44.400 -->
00:54:49.119With it with the Newman model, you can sit there and say, Oh, I'm gonna start getting lithium deposition at this rate.
00:54:49.119 -->
00:54:50.800I'm gonna start having all these problems.
00:54:50.800 -->
00:54:53.039RCR model doesn't tell you anything.
00:54:53.199 -->
00:54:56.079You know, it just Right, right, yeah, exactly.
00:54:56.079 -->
00:54:59.360So you just fed the elephant, but you don't know what it means.
00:54:59.360 -->
00:54:59.599Yeah.
00:55:00.000 -->
00:55:00.800Exactly.
00:55:00.800 -->
00:55:11.360But the RCR turned out to be way just the ease of use and the fidelity that you got, it just made it incredibly attractive to people.
00:55:11.360 -->
00:55:14.400And then like I say, in the bigger market is really the automotive.
00:55:15.039 -->
00:55:19.039I I have one memory that maybe you can just shed insight on.
00:55:19.039 -->
00:55:22.960So when I I was I left 3M in 2006 and came back in 2010.
00:55:22.960 -->
00:55:33.920During that time I went to Medtronic and I was working in the battery research group, and part that Gomadon was working on uh modeling, and he used uh ComSol all the time and kind of just calculated from ground up.
00:55:33.920 -->
00:55:44.960And uh one of the things we were looking at was that Medtronics started making their own lithiumine cells for neurological devices and selling them maybe around 2001 or something like that.
00:55:44.960 -->
00:55:55.360And in some cases, we could see um age cells or highly cycled cells could have like uh lithium plating on the edges of the jelly roll, like a pr a small prismatic jelly roll.
00:55:55.360 -->
00:55:58.239And it was sort of like hard to figure out why that was happening.
00:55:58.239 -->
00:56:15.599But Newman came out with this paper that showed like, oh, if you if you have excess electrolyte and there's electrolyte on these on these uh edges, you get this edge effect where you can kind of get this concentration of the field that uh puts extra electro um extra lithium on the anode uh at that edge.
00:56:15.599 -->
00:56:22.159Uh was that was that kind of thing modelable with um with what you had in in better design studio?
00:56:22.159 -->
00:56:25.599No, that do you know is is what I'm describing familiar to you?
00:56:25.840 -->
00:56:39.039Oh, yeah, I know the sort of I saw at Medtronics, it wasn't Partha, it was another guy, I forgot I'm sorry, I forget his name, but he was Partha's boss, I think, gave a talk at the Electrochemical Society meeting about this edge effect.
00:56:39.039 -->
00:56:46.320And Newman had these weekly meetings where he would let people attend and they would review papers.
00:56:46.320 -->
00:56:47.599And so I would go to that.
00:56:47.599 -->
00:56:50.320I mean, that was one of the big big pluses of living here in the Bay Area.
00:56:50.320 -->
00:56:52.320I had that incredible experience.
00:56:52.320 -->
00:57:03.440Newman is right, I mean, it's just hard to uh overstate how what a a genius is, a genius he is, and what a gentleman he is, and just what just an incredible person.
00:57:03.440 -->
00:57:15.039One of his things was he solicited projects from people, and so I had heard this one this problem that Medtronics had said and suggested that as a project.
00:57:15.039 -->
00:57:20.639And then Maureen Tang took it up as her PhD project and developed the theory for it.
00:57:20.639 -->
00:57:22.559So I I knew that story.
00:57:22.559 -->
00:57:25.039So it is connected, you are connected to that story.
00:57:25.039 -->
00:57:26.079That's fascinating.
00:57:26.079 -->
00:57:28.239Oh wow, that's cool.
00:57:28.239 -->
00:57:43.920I have to say that uh what what I did not do, and I I really bad about that because I messed up in consulting on this, was that I was using the Newman model, and Molly Energy had developed a model that took into account the tab effects and things like that.
00:57:43.920 -->
00:57:44.800I see.
00:57:44.800 -->
00:57:48.639And so I did not appreciate the importance of tab effects.
00:57:48.639 -->
00:57:55.280And most of the batteries I had dealt with were low-energy applications, you know, low-rate applications, and were it wasn't that important.
00:57:55.280 -->
00:58:00.719And so I I did give bad advice to people who were starting to look at higher rates.
00:58:00.719 -->
00:58:13.679Um, when CD Adapco acquired the software, they gave me the project of extending the model to include Tab Effects, and we put that in the model, and then I saw all these effects that I really regret not knowing about later on.
00:58:13.679 -->
00:58:14.880Oh wow.
00:58:15.760 -->
00:58:16.320Wow.
00:58:16.320 -->
00:58:26.159One of the uh things that lithium ion is borrowing from the capacitor industry, really, more and more is the there's different names.
00:58:26.159 -->
00:58:28.880Some some call them tablets, some call them continuous tab.
00:58:28.880 -->
00:58:46.159But basically, instead of in a wound cell, instead of having sort of discrete little metallic strips where you end up connecting the current collector, you have a whole edge which is bare, and you connect the whole edge on the end of the jelly roll to the can.
00:58:46.159 -->
00:58:55.920Um what do you think about that that approach now that you know having now seen all these tab effects from like a modeling perspective, like what's your what's your take on that approach?
00:58:56.239 -->
00:58:57.519Well, it's fantastic.
00:58:57.519 -->
00:59:03.360Um I point out SAFT had those cells years and years ago.
00:59:03.360 -->
00:59:13.039Um so it and when we were when I was at Poly Store, we were competing against USABC was giving us contracts and staff contracts.
00:59:13.039 -->
00:59:20.079And so they were our direct competitors on these developing and they developed this tablets concept for their batteries way back then.
00:59:21.199 -->
00:59:21.519Wow.
00:59:21.760 -->
00:59:25.840Um But the thing was it was the cost, the complexity of doing it.
00:59:25.840 -->
00:59:28.079So that you know everything was low cost.
00:59:28.719 -->
00:59:34.559Yeah, it was always identified as a as a kind of uh an ideal, but the implementation was difficult.
00:59:34.960 -->
00:59:35.679Exactly.
00:59:36.079 -->
00:59:41.599So then you ended up uh being co-founder of of some companies as well.
00:59:42.000 -->
00:59:42.559Oh yeah.
00:59:42.559 -->
00:59:57.679So this really famous scientist, uh Vin Kat Srinavasin, gave me a call and asked me if I would like to have a meeting with some of these entrepreneurs that he knew that wanted somebody to help them with batteries.
00:59:57.679 -->
01:00:11.440So this the it was a small group of guys, Harold Rust, Ashurk Lahari, and Morali, Rama Supermanyan, and all these guys had been with a startup that was successful, and they all made some money.
01:00:11.440 -->
01:00:18.320And so they were young guys, they had made money, and they and they wanted to do something in the batteries.
01:00:18.320 -->
01:00:20.079They thought batteries was the future.
01:00:20.079 -->
01:00:33.519And they had they were experts in etching silicon, they could make any sort of pattern, and so they had on the micro batteries the idea of uh interdigitated electrodes, right?
01:00:33.519 -->
01:00:36.960And they looked at those structures and said, Oh, we can do that easy.
01:00:36.960 -->
01:00:38.320Let's make those structures.
01:00:38.559 -->
01:00:38.880Right.
01:00:39.119 -->
01:00:55.280I didn't I to be honest, I wasn't really sure how uh technically valid what they were doing was, but these guys were the quintessential Silicon Valley entrepreneurs, you know, they they just had the can-do attitude, the and I I just I loved it.
01:00:55.280 -->
01:00:56.559I I wanted to be part of it.
01:00:56.559 -->
01:01:02.559And so I started helping those guys uh one one day a week working with those guys.
01:01:02.559 -->
01:01:03.599Enovix.
01:01:03.599 -->
01:01:05.440That's Enovix, that's right.
01:01:05.679 -->
01:01:07.760And what about the other one, the American Lithium Energy?
01:01:07.760 -->
01:01:10.000Is there a story there as well on how that got started?
01:01:10.320 -->
01:01:26.719So Jong Fan, who had worked at PolyStor, he had gone to um oh, this Taiwanese battery maker, and he went and got this job in San Diego working for this company with a guy named Doug Magnuson, who went to 3M later.
01:01:26.719 -->
01:01:27.440Oh yeah.
01:01:27.679 -->
01:01:28.960Yep, and Doug.
01:01:29.599 -->
01:01:37.039And this Taiwanese company started making lithium ion batteries and became very, very successful in making lithium-ion batteries.
01:01:37.039 -->
01:01:41.199But they decided to shut down their San Diego laboratory.
01:01:41.199 -->
01:01:54.480And so Jong decided, and and I say Jong the he he is the has the most intuitive grasp of lithium ion batteries I've ever seen anybody in my in my life.
01:01:54.480 -->
01:02:04.079He at Pauli's door, he would do our electrolyte formulations and things, and I I never could understand his thought process, but he would come out with these things that just worked.
01:02:04.079 -->
01:02:06.079It was it was just incredible.
01:02:06.320 -->
01:02:06.719Wow.
01:02:06.960 -->
01:02:16.719And so he he had the idea of buying the assets for their lab, specifically, you know, like uh the dry room.
01:02:16.719 -->
01:02:25.920Uh we couldn't take the dry room, but we could take the the dryer and the test uh the equipment and stuff like that, and starting his own company to make specialty batteries.
01:02:25.920 -->
01:02:29.519Because he was also on board with me, you know, that that's the market, you know, specialty batteries.
01:02:29.519 -->
01:02:30.960You're not gonna compete against Sony.
01:02:30.960 -->
01:02:48.320And and we went through with John and I, we went through an economic analysis and showed that there was no way you could, you know, you the economy of scale was such, you know, you know, if you looked at like you need an engineer to monitor winding machines.
01:02:48.320 -->
01:02:57.599So if you only if you only have one winding machine, you're paying an engineer to monitor to monitor it, whereas that same engineer could be monitoring 50 winding machines.
01:02:57.599 -->
01:03:02.639And so Sony having 50 winding machines has the economy of scale.
01:03:02.639 -->
01:03:06.800They just have that one engineer, whereas you need one engineer, you know, it's right, right, right.
01:03:06.800 -->
01:03:11.360Just the labor costs kill you with uh with lithium ion.
01:03:11.360 -->
01:03:15.920He got the assets of that company and started his own lab.
01:03:15.920 -->
01:03:17.760And what was your involvement then?
01:03:17.760 -->
01:03:23.519So I I would fly down to San Diego one or two days a week, and I helped set up.
01:03:23.519 -->
01:03:31.840I mean, we bought a uh a freezer, a walk-in freezer, and we hooked up the the dry room, uh, helped write proposals.
01:03:31.840 -->
01:03:36.800So his focus was military, built cells, did everything, whatever it took, we we did.
01:03:37.199 -->
01:03:38.880What eventually happened to that company?
01:03:39.119 -->
01:03:40.239American Lithium Energy?
01:03:40.239 -->
01:03:40.559Right.
01:03:40.559 -->
01:03:44.719So when CD Dapper came around came along, I couldn't go to San Diego anymore.
01:03:44.719 -->
01:03:46.000It's just too much time.
01:03:46.000 -->
01:03:48.719So I had to give up working with him.
01:03:48.719 -->
01:03:50.559But he kept the company going.
01:03:50.880 -->
01:03:55.280I I want to go back and ask a question about um modeling.
01:03:55.280 -->
01:04:02.159Um we're kind of we're getting towards the end of the podcast where we can sort of explore things a little bit more wildly.
01:04:02.159 -->
01:04:04.320So I'm gonna take a stab here.
01:04:04.320 -->
01:04:14.320Um so uh, you know, I'm like everybody else, just trying to make sure that I don't fall behind on uh using uh AI and so on.
01:04:14.320 -->
01:04:18.559And in the world of modeling, there's uh certainly a a big, a big hit there.
01:04:18.559 -->
01:04:42.320And so I had a conversation just recently with uh professor friend of mine in Germany, and and he showed me what he was able to do when he just uh without knowing anything about how to code and how he worked with Cloud Code and could sort of immediately make a uh really good cell model complete with um beautiful outputs of um dynamic graphs that he could move around and play with and and so on.
01:04:42.320 -->
01:05:04.719And it kind of makes me wonder like, have you played at that and what do you you know, do you think that what's gonna happen here with with uh battery modeling specifically, but it but really you know this kind of modeling in general, is it since you come from this whole world of modeling and um you know you're always on the cutting edge of things, you probably have have seen what's cap what's possible now.
01:05:04.719 -->
01:05:13.280You know, what what's your view of what's gonna happen to to modeling uh with the the the incredible capability that is kind of in everyone's hands?
01:05:13.599 -->
01:05:14.960Oh Kevin, that's really interesting.
01:05:14.960 -->
01:05:17.039That that is a really interesting question.
01:05:17.039 -->
01:05:21.199I I've been trying to wrap my head around AI.
01:05:21.199 -->
01:05:39.119You know, I've read a lot of books about it, and um I've played with with Claude and Gemini co-pilot, and I I still am not clear what's gonna happen because it's like I I think with AI, you you're not gonna need software programs like Battery Design Studio anymore.
01:05:39.119 -->
01:05:40.559It's like you're saying, I mean, you can do stuff.
01:05:40.559 -->
01:05:42.320I mean, you're not gonna need these type of programs.
01:05:42.320 -->
01:05:49.119You're just and I just don't I don't I don't really have not myself been able to wrap my head around what's gonna happen yet.
01:05:49.519 -->
01:05:57.840But you do see that that difference of like, oh okay, I can today I can I can create a something that's kind of like Battery Design Studio like in a day.
01:05:58.159 -->
01:05:59.280But but yeah, but why bother?
01:05:59.280 -->
01:06:03.360Why not just Just ask it for the results you want to get the answer to my question.
01:06:03.360 -->
01:06:03.840Right.
01:06:03.840 -->
01:06:04.559Exactly.
01:06:04.719 -->
01:06:05.039Yeah.
01:06:05.280 -->
01:06:05.760Exactly.
01:06:08.400 -->
01:06:08.960Good point.
01:06:08.960 -->
01:06:10.320Why would I even need a model?
01:06:10.320 -->
01:06:11.199All I need is the answer.
01:06:11.199 -->
01:06:11.840Exactly.
01:06:13.199 -->
01:06:18.400Just as a developer, what bothers me is that I can't train the model myself.
01:06:18.400 -->
01:06:20.960It's like uh 50 first dates.
01:06:20.960 -->
01:06:27.760Every day the model wakes up and it, you know, it remembers what's a nice buffer, but you gotta, you know, it doesn't know anything.
01:06:28.079 -->
01:06:29.199You gotta remind it every time.
01:06:29.519 -->
01:06:29.920Every time.
01:06:29.920 -->
01:06:34.000And whereas the developers, they can they do the training.
01:06:34.000 -->
01:06:46.239So they're gonna be able to take all the know-how and train the models, and then they have the battery engineer that knows everything and can answer all the questions.
01:06:46.239 -->
01:06:49.920And so I can't compete with that, you know, because I don't have that capability.
01:06:49.920 -->
01:06:51.440They're the only ones who have that capability.
01:06:51.440 -->
01:06:51.920Right.
01:06:51.920 -->
01:07:04.320And so it's uh it's really gonna change the competitive landscape about, you know, instead of having a whole bunch of different companies, you're gonna have just these small few AI companies that are in charge of everything.
01:07:04.320 -->
01:07:05.199Uh I don't know.
01:07:05.679 -->
01:07:06.239Right.
01:07:06.239 -->
01:07:13.199Yeah, yeah, I guess you're saying the end game is kind of uh it sounds very concentrated.
01:07:13.199 -->
01:07:16.239Yeah, interesting.
01:07:16.559 -->
01:07:18.559Yeah, I have a a question.
01:07:18.559 -->
01:07:31.119Back in '98, when you were thinking of when you joined PolyStor, you said, Well, I had seen what Sony had in their factories, their automation, their economies of scale.
01:07:31.119 -->
01:07:35.440I get to PolyStor, and I realize we can't compete with Sony.
01:07:35.440 -->
01:07:39.840You know, this was 30 years ago, 28 years ago.
01:07:39.840 -->
01:07:45.119Um My impression is that dynamic still occurs today.
01:07:45.119 -->
01:07:50.239We have companies that start up in the US and think, oh, I'm gonna compete.
01:07:50.239 -->
01:07:53.360And the same video gets played over again.
01:07:53.360 -->
01:07:55.039What do you think of that assessment?
01:07:55.039 -->
01:07:56.639What do you think the solution is?
01:07:57.199 -->
01:08:04.639So th this I'm gonna be pretty maybe far out there, but it took me a long time to come to this realization.
01:08:04.639 -->
01:08:07.679But the problem with the US is market.
01:08:07.679 -->
01:08:09.280It's demand.
01:08:09.599 -->
01:08:09.920Okay.
01:08:10.800 -->
01:08:19.199You know, the US companies, auto companies, and even the European ones, they don't want batteries.
01:08:19.199 -->
01:08:22.079I mean, they have battery groups and stuff like that.
01:08:22.079 -->
01:08:29.920But every chance they get, they lobby the government to stop the battery research, to just, you know, do it, bring it to dead ends and things like that.
01:08:29.920 -->
01:08:31.840They they don't want to go there.
01:08:31.840 -->
01:08:33.359They just there's no market.
01:08:33.359 -->
01:08:40.640And what really opened my eyes was the dieselgate thing, the Volkswagen scandal in 2015.
01:08:40.960 -->
01:08:41.520Yes.
01:08:41.520 -->
01:08:42.239Yeah.
01:08:42.560 -->
01:08:50.800When you you realize that all a lot of the other companies, I think, in my opinion, must have known what was going on, right?
01:08:50.800 -->
01:08:58.479That they were gaming this, and then a lot of it turned out that a number of other companies were doing similar type things and got caught.
01:08:58.479 -->
01:09:01.600And then a lot of companies were just gaming the system.
01:09:01.600 -->
01:09:12.159So these guys, and over the when I started thinking back of my interactions and stuff, and then I started to realize that these guys really never believed in battery technology.
01:09:12.159 -->
01:09:28.239They just and you know, this thing about oh, the there's not a there's not a domestic, you know, they they could they lobbied the government to build these domestic plants, you know, like the Dow Kokam or uh the LG, but there's no sales, there's no market for it.
01:09:28.239 -->
01:09:35.840They just they don't if you look at like uh Tesla, Panasonic build a plant.
01:09:35.840 -->
01:09:38.640They build a plant because they got a market, they build a plant.
01:09:38.640 -->
01:09:42.960If there's a market, you'll the people will build a plant for it.
01:09:42.960 -->
01:09:44.640It doesn't make sense to build a plant.
01:09:44.640 -->
01:09:49.840It's not like a field of dreams where you you know you build it, they'll come, it's the other way around.
01:09:49.840 -->
01:09:52.560You get a market and then you build you supply that market.
01:09:52.560 -->
01:09:57.760And the the Western companies are just not serious about that market.
01:09:57.760 -->
01:10:05.359They just don't they don't believe in battery technology, they just they just don't believe it really can happen, that it can really replace gas.
01:10:05.760 -->
01:10:08.960And do you think that they they influence the consumers as well?
01:10:08.960 -->
01:10:09.760Oh, absolutely.
01:10:09.760 -->
01:10:11.359Okay, got it.
01:10:11.439 -->
01:10:13.199Yeah, look at the advertising, you know.
01:10:13.199 -->
01:10:13.520Right.
01:10:13.520 -->
01:10:20.399When I see the advertising for electric vehicles on TV, they're not anywhere as persuasive as they could be.
01:10:20.399 -->
01:10:21.520I think.
01:10:21.840 -->
01:10:22.479Right, exactly.
01:10:22.800 -->
01:10:28.640This is really interesting because mo most of the guests up to now have not answered this answer.
01:10:28.640 -->
01:10:31.920They've said, well, look at policy.
01:10:31.920 -->
01:10:43.520You know, in Asia you have typically you have policy that really supports the market, you have long-term policy that sort of extends beyond just the the uh you know a four-year cycle of an administration.
01:10:43.520 -->
01:10:55.520But you're bringing a different perspective, which is like just the internal pull, the the customer pull is either tepid or inconsistent or just not there.
01:10:55.840 -->
01:10:56.239That's right.
01:10:56.239 -->
01:11:01.359And and the companies in China started off for the most part as electric vehicle companies.
01:11:01.359 -->
01:11:04.800They weren't, you know, legacy uh ice companies.
01:11:04.800 -->
01:11:06.399They're they were EV companies.
01:11:06.399 -->
01:11:07.920And so that was their business.
01:11:08.239 -->
01:11:11.119So they have to make it work and they have to build the demand as well.
01:11:11.119 -->
01:11:11.680Yeah.
01:11:11.680 -->
01:11:13.680They're really trying to sell EVs.
01:11:14.319 -->
01:11:15.840They're not Yeah.
01:11:16.000 -->
01:11:20.319So it's a recipe for kind of inevitable disruption, I suppose.
01:11:20.319 -->
01:11:26.880Um, I guess I say that because I owned three EVs and I it's like obvious to me that the value proposition is better.
01:11:26.880 -->
01:11:49.199But um, but I guess you know, uh most people don't see it the way I do, and it'll and so it'll just be it's just gonna eventually be like painfully obvious that uh you can get a better car from uh from somewhere else that's an electric vehicle, and um and then and then we'll fall over and start uh having electric vehicles in the US, maybe.
01:11:49.199 -->
01:11:52.319What's your crystal ball look like?
01:11:52.560 -->
01:12:18.479Oh I my understanding is that uh Canada has recently agreed to let the uh EV factory be built there, and Mexico already has uh deals, and so it's just gonna take a you know a change of administration or something, and it's gonna be you're gonna have uh access cheap electric vehicles available from other countries, and it's just so what a wasted opportunity for the US.
01:12:18.479 -->
01:12:38.479You know, we we could have led this country country, but the you know it I I I I I I don't hate to be in a soapbox, but it's like that that we allowed the fossil fuel industry to spend billions of dollars to sway our political system is just uh uh unexcusable.
01:12:38.479 -->
01:12:44.960You know, it's just caused so much damage to our society that it's just it's really right, and puts us behind technologically significantly.
01:12:45.359 -->
01:12:45.840Yes.
01:12:46.000 -->
01:12:46.640Yeah.
01:12:46.960 -->
01:12:47.600Yeah.
01:12:47.600 -->
01:12:49.199Yeah.
01:12:49.199 -->
01:12:50.159Wow.
01:12:50.159 -->
01:12:51.520Wow.
01:12:51.520 -->
01:12:54.079Well said, Bob.
01:12:54.079 -->
01:12:55.359Yeah, thank you.
01:12:55.680 -->
01:13:00.159I mean, I think this is a good place to to conclude.
01:13:00.479 -->
01:13:02.960Yeah, thanks a lot, Bob, for coming on.
01:13:02.960 -->
01:13:06.319That is so powerful, powerful finish.
01:13:07.199 -->
01:13:07.680Thank you.
01:13:07.680 -->
01:13:09.119Thank you for the opportunity.
01:13:09.119 -->
01:13:10.239I appreciate it.
01:13:10.560 -->
01:13:12.239Yeah, thank you so much.
01:13:12.239 -->
01:13:21.680Battery potential is produced by cyclical music by big flavor.
