Loading summary
Transcript1 lines
- [00:03]
A
In early 2026, a British hedge fund sent a letter to the Japanese company Toto. Yeah, the company making tubs, sinks and those high tech bidet toilets. Not because Toto was doing particularly bad in those businesses, but because Toto also has an advanced ceramics division that is the world's leading supplier of electrostatic chucks for the 3D NAND industry. Now I know what you guys are thinking. How did a Japanese toilet company get involved in semiconductors? The answer is that Toto doesn't see itself as just a toilet company. In today's video we travel with Toto, the Porcelain King. This video is brought to you by the Asianometry Patreon. In 1839, Ichizeemon Morimura was born to an Edo based merchant family that that once made leather harnesses. Morimura grew up in the turbulent 1850s and 1860s when Japan opened up to the world and modernized. His family lost everything after a devastating 1855 Tokyo earthquake. But Morimura worked hard, opened his own store and paid off his family debts. Throughout his career, Morimura invested in a wide variety of things. Copper mines, fishing, silk and military armaments. Most of these ventures failed or were shut down. In 1876, he came to the belief that the only way to earn back the gold and silver then flowing out of Japan was with exports. So he starts the Morimura Group, a trading company. They eventually open a retail shop in New York City to sell Japanese made goods to Americans. Murimura at the start sold simple tourist goods that sold because they were Japanese and exotic, but then pivoted to Western style tableware produced in Japan. At the time, hand painted porcelains from Europe were popular but too expensive for ordinary folks. Murimura's products, sugar bowls, milk bowls, water pitchers, etc. Cost far less. Such items were initially marketed in the United States as being from Nippon, creating the rather discombobulating phrase Nippon China. Anyway, Muramura's real ace in the hole was distribution. He could reach America's growing middle class through Sears and Roebucks famous mail order catalogs. Revenue and profits soared and and in 1882 the company switched entirely to wholesale. Ijizeemon Murimura also brought his relatives into his growing business. This included brother in law Magobe Okura. Magobe's son Kazuchika Okura would also join and quickly rise up the group's ranks. In the 1900s, American households increasingly desired tableware made from a porcelain that was thin, hard and durable. Such porcelain also had to be colored white so that elaborate floral and lifestyle Designs can be painted onto them. At the time, such porcelain tableware can only be had from Europe, examples being France's prestigious Sevres or Germany's Dresden style porcelains. Japan's porcelains on the other hand, were a grayish blue color, often with some Japanese landscape motif painted onto them. They were also thick walled and brittle, unsuitable for making full Western style dinnerware table sets. But producing a suitable hard white porcelain turned out to be an immense challenge. In 1904, the Morimura Group founded a separate company called the Nippon Tokigomei Gaisha or Noritake for short, to do it. The name means Japan pottery and they were led by the aforementioned Kazuchika Okura. They built a new factory with a large coal fired kiln imported from Germany and their goal was to use it along with a special clay mixture brought back from Europe to produce the white porcelain to make the desired dinner set. But the clay mixture fell short. To get the white color, Noritake added Amakusa pottery stone into the mixture. However, the stone's high silica content caused the ceramic to deform and sag when fired inside the kiln. So Noritake stirred in a highly plastic iron rich clay called Gairome, only to find the opposite effect. The ceramic held up structurally inside the kiln, but lost its pure white color. With the New York office relentlessly pushing for results, Noritake sent its chief engineer to a state owned pottery research institute in Berlin in 1910. The German scientists studied the clay mixtures and made suggestions to both the mix and the porcelain firing process node that fixed the problem. In 1913, Noritake unveiled its game changing hard white porcelain. A year later, in June 1914, Noritake finally unleashed their long awaited white porcelain dinner set. The timing was fortuitous. The outbreak of World War I halted shipments of European porcelain to the United States. Noritake's Exports surged from 2,000 sets in 1914 to 40,000 in 1918. Now let us rush for the toilet. In the 1850s and 1860s, Japan rapidly absorbed items brought over from the West. Such things included stuff that went into your mouth, foods and culinary styles. Another thing involved the other end of all that, the toilets. Traditional Japanese toilets were located outside of the main home. They were basically just wood shacks with holes in the ground that you squatted over. The night soil, so it was called, is then later sold off for use as agricultural fertilizer. Generally speaking, it wasn't too bad in terms of sanitation. The latrines were outside, separate from the living quarters and I'm aware that some people think that squat toilets are better for your health. Definitely works your thighs. The first Western style sit down toilets were brought into Japan in the early 1900s by contractors like Toyojiro Suga. Suga's firm, Suga Kyo, operates to this day as a well respected building services engineering firm. Recognizing the incongruities between Western flush toilets and Japanese customs, Suga produced hybrid toilets to bridge the gap. He also recognized though, that Western style toilets can potentially be very popular in Japan if made cheaply enough. Key to being able to produce a good flush style toilet was the porcelain. Unfortunately, artisanal Japanese kilns, often based in the region of Seto, which has long had a good reputation in ceramics, struggle to make a suitable industrial porcelain. What makes a good porcelain? To start, it cannot leak. It should have good strength and cannot warp very much lest it breaks its connection to the metal pipes in the floor. Furthermore, the porcelain surface must be easily washed of various chemicals and materials. The toilet porcelain should not absorb water. Imagine the smell if it did. And with regards to manufacturing, it must be easily moulded into a single piece. And finally, yes, the blindingly obvious it cannot cost a lot. Making this stuff is hard. For much of the 1800s, early British toilet manufacturers like Jennings and Twyford did not even make their toilets from porcelain. Instead they they used another ceramic called earthenware. It is also made with clays fired in a kiln, but at lower temperatures than porcelain. It is cheaper and easier to mold, but at the cost of being more prone to cracking. Earthenware is also porous and without a lot of glazing, can absorb lots of water. Japan only figured out how to make earthenware tableware in 1902, thanks to a technological transition from wood to coal fired kilns. Doing big single pieces of porcelain for toilets and other sanitary items like sinks was a major step beyond. In 1912, Kazuchika Okura of the Morimura Group proposed that the company make a toilet. This was a big ask. The company was still putting its finishing touches on its white porcelain production line. Nevertheless, Okura and his father both strongly believed that modern toilets are necessary for Japan's modernization. So they invested their own money to start a small R and D lab inside Noritake to study the techniques to make these large toilet bowls and sinks. The toilet bowl is a large single piece with many curves, and at the time it was not possible for the Japanese to cast them as one single piece. All those curves would cause the clay to get stuck or collapse under its own weight. So the Japanese turned to an older western technique called teokoshi, which means to press. It is also called the hand layup method. We take long flat strips of clay called tatara boards and press them by hand onto wooden moulds. A craftsman then pushes and pounds the clay into every nook and cranny of the mould to achieve a uniform thickness while squeezing out air bubbles. The lab does extensive experimentation to achieve the right mix of clay, glaze and mold, including 17,280 test firings. In the end, they had to import molds from the United States. But finally, in 1914, the company announced its first run of domestically produced toilet bowls. To produce these bowls, Okura afterward started building a new large porcelain factory on the Japanese main island of Kyushu, where land and coal can be found in plentiful amounts. This plant was completed in 1917. Soon afterwards, Muramura founds a new company to run it and sell its porcelain sanitary and tablewares. This company was named Toyotoki, which means Eastern porcelain. This shortened to Toto all caps, though this name was not officially adopted until the late 1960s. Toto's primary production facilities were sited in the city of Kokura in Kyushu. They retain a presence there today, like with their Toto museum. By the way, like many other Japanese companies, To Toto has done a great job of recording its own history. In 2018, they released an 822 page tome in celebration of their centennial. It is a strong source for these early years, though I triangulated it with others of the time. Anyway, the hand layup method required exceptional skill, strength and time from the laborer. It does not work for mass production. But there does exist a more suitable method called slip casting. Rather than using hard manual labor to press the clay into the mold, we rely on liquid dynamics and chemistry. We formulate a clay slurry called the slip and pour it into a porous mold until full. The mold then absorbs the liquid slip, leaving behind a solid clay shell. Humans are so smart. Slip casting has been done before in Japan for smaller items like tableware. But larger items are more challenging because the clay physically shrinks so much. 3% during drying and 10% during firing. So mold designers must calculate and intentionally build distortions into their mold designs to accommodate. It was not until 1920 that that Toto figures out a mix of clays that do not warp and break the molds during drying, paving the way for their first slip casted toilets. They continued to refine these processes throughout the decade. These Efforts were. Despite great economic challenges during the period after World War I, the global economy fell into a recession. The United States raised tariffs, the depriving Toto of its core tableware export market. They pivoted to the Japanese and Southeast Asian markets, but still had to cut costs and jobs. The company was saved in part by the reconstruction effort after the Great Kanto earthquake in 1923. The tragedy mandated reinforced concrete buildings with central plumbing. This gave Toto its first big break in Japan with with jobs like the Marinucci Building, then one of Asia's biggest office buildings. Then in the early 1930s, the creation of the Manchukuo puppet state kicked off a big construction boom that drove Toto's export sales of sanitary ware goods. Some years 60% of their sales went to there. Then came World War II. Over 10% of the male workforce was drafted. The government rationed raw materials and the war of course shut off export markets. Toto's factories were repurposed to produce armaments or metal replacements using ceramic. This nearly got them obliterated. As I mentioned, Toto's primary production facilities were in the city of Kokura and they had been converted to making weaponry. This and the nearby presence of an armory were why in August 1945, Kokura was designated the primary target for the second atomic bomb. But heavy cloud cover that day, as well as smoke from the firebombing of the nearby city of Yawata, obscured the view. I mentioned this before in the Japan Steel video. The plane circled Kokura three times and then diverted the bomb to Nagasaki. This saved both the city of Kokura and Toto from complete annihilation. At the war's end, Japan and its economy laid in ruins. Early on, the Allied occupation requisitioned various buildings and ordered them to be retrofitted with Western style flush toilets and plumbing. As Toto was the leading Japan based supplier, they got most of the orders. But to fulfill those orders, they first had to learn metalworking. When thinking about the toilet, most people obviously think about the porcelain bowl. But there is another set of critical components that I briefly mentioned earlier. The metal fittings. The metal fittings refer to the various small metal items that anchor the toilet, connect it to the plumbing and do the actual flush. Without them, the toilet is just a big porcelain bowl. Toto had a saying, if the toilet were a car, then the ceramic is the body and the metal fitting is the engine. By the way, this same applies to all sorts of porcelain bathroom goods like sinks, valves, faucets, tubs and more. The issue is that Metal and porcelain have different mechanical properties and getting a good fit between the two is very challenging. A bad fitting can easily crack the bowl. For this reason, American toilet makers often design and sell the sanitary ceramics and metal fittings together. But before the war, Toto didn't do this, sourcing their fittings from an outside supplier named Janssen. This was for several reasons. Janssen made cheaper, better fittings. Okura lacked confidence in his company's metalworking skills and most of the pre war demand was for non flush Japanese style toilets anyway. Metal fitting content in there was not that large. But now we have this order from the Allied occupation forces and they want it done like how things are done back home in the States. Janssen is no longer available as a supplier, so Toto must learn how to make the metal fittings themselves. With effort, they pull it off, requiring extensive metalwork modernization. However, this forced expansion puts all the pieces together. Toto is now a fully integrated sanitary ware maker, capable of selling the complete package at home in Japan and abroad. This also enables Toto to expand into new categories like water taps or drainage fittings. They grew rapidly, rising like Kilimanjaro above the Serengeti. By the mid-1960s, Toto held 47% market share in those areas and an estimated 60% of the market in toilets. Major domestic competitors included Inax, which ironically was one of their sibling companies for many years within the larger Muramura Group company. Later in 2001, Inax left the company, triggering lawsuits. In the 1960s, the larger Muramura Group reorganized itself to focus on sanitary wear, and in 1970, they finally exited their legacy tableware business. The 1960s saw a wave of new construction. Building investment nearly tripled in the five years leading up to 1964 as Tokyo urbanized for its first hosted Olympics. This helped Toto sell more sinks, toilets and other sanitary ware. To keep up, the company introduced these unit bathrooms where the whole room is prefabricated in the factory and shipped to the construction site for fast installation. To differentiate themselves from other commodity toilet makers, Toto branched into new products. Back in the 1950s, they tried porcelain tiles and plastics. Those didn't work. Same goes for one of my personal favorite products of all time, the Sunny Standard, a urinal for ladies. Then in 1964, they licensed and imported a bidet made by the American Bidet Company. This product was a detachable toilet seat with a nozzle that blew warm water and then air onto your dirty butt. It was primarily designed for those suffering hemorrhoids. Toto Dubbed it the Wash Air seat and marketed it as eliminating the need for toilet paper. Unfortunately, it did not sell very well. The price was high and the water temperature unstable. Cold, sometimes hot others. Shusaku Endo is the famed author of the book Silence, a very serious treatise on faith. He also published a very humorous negative review of this first Wash Air Seat. The most famous passage goes, A stream of unbelievably hot water suddenly assaulted my anus. Incredibly hot water. The kind of thing you use once and only once. Japanese magazines also refused to take adverts for something as profane as a toilet. By 1979, Toto was selling just 500 units a month, mostly to wealthy people, hospitals and welfare centers. And by the way, Toto was not the only Japanese company selling a bidet. In 1967, their corporate sibling in Porcelain's Inax released the Sanatorina 61. It was a pioneer, but didn't sell too well either. Maybe because it required a foot pedal to pump the nozzle. Thinking that the opportunity existed, but the existing products didn't quite fit, Toto re evaluated the product. Toto needed to determine a new product spec, but at the start lacked the anatomical data for it. To collect this data, the Toto development team crowdsourced it. Starting with themselves, they built a test seat with a wire stretched across its center. The subject sits down onto it and the staff marks the location of the anus on on the wire with a piece of paper. The result is a distribution of data points. The team started with themselves, but realized they needed more data, particularly of women. So with permission, they surveyed 300 Toto employees for measurements of their proctological anatomy, along with their temperature preferences. The final product specs were surprisingly specific. A toilet seat temperature of about 36 degrees Celsius, water temperature of about 38 degrees, a nozzle volume of 500cc per minute and an angle of 43 degrees and an air dry of 50 degrees Celsius. Now to build it, the water temperature proved to be especially challenging. The old wash air seat maintained water temperature using something called a bimetallic strip consisting of two metals bonded together. The metals shrink or expand at different rates when heated, allowing it to turn the heater on or off. They used it for toasters. The issue with the bimetallic switch is that it is crude and mechanical, only turning on or off at a certain heat level. Ergo why the nozzle water temperature fluctuated so much. And if the strip malfunctions, the switch might get stuck, leading to the scalding hot water rushing towards your anus. What is most ideal is to replace this unreliable mechanical switch with an electronic setup with integrated circuits and temperature sensors. The IC would sample the temperature and adjust the heater up or down accordingly. However, the ICs on the circuit board had to resist water and salt containing urine. How to protect them While pondering this problem, a Toto engineer came upon a traffic light operating outside in difficult wind and rains. It occurred to him that whatever protected the chips inside the traffic light can also be used for the warm water toilet. So they tracked down the light's manufacturer and set up a joint venture to encapsulate the chips with resin for protection. Toto introduced the first washlets in June 1980. The name comes from the phrase let's wash, derived from the larger slogan from now on, it is an era of washing your behind. Let's wash. To get people to buy the product, Toto initially tapped its existing strengths in toilet porcelains, in particular their network of plumbers, who they recognize as having a strong say in deciding what gets installed into people's homes. Toto adjusted the washlet design so that it needed a plumber to install and emphasized to them that they made more by selling the $1,000 washlet over a $400 traditional toilet. Despite some initial product hiccups, the Washlet grew steadily in its first two years. The through word of mouth. Then in 1982, Toto ran its first television adverts bought up by the same team behind the ad campaign for the Sony Walkman. The TV campaign was rather unconventional. One advert had a slogan that the Japanese had the cleanest private parts in the world. Their breakthrough ad featured the actress and avant garde singer Jun Tagawa saying that if you wash your hands with water, then you should wash your bottom the same way, all the while with some weird hollering music in the background. This ad ran at prime time 7pm and immediately blew up. Go watch it. It's on YouTube. To drive further awareness, Toto built custom demonstration bands so people can try the washlet themselves. Sounds like a plumb perfect experience. They also encouraged installations in fancy public venues like restaurants, stores and golf clubs, correctly betting that if people tried it there, then they would want it at home. It worked. Washlet sales boomed. By 1987, Toto had sold a million units, and over the subsequent years, Toto kept adding new features to the device. Such premium bidet toilets would drive the majority of of the industry sales growth. 25 years after its release, the washlet had sold 20 million units, cumulatively toto continues to develop new features for their premium toilets. My favorite is the tornado flush, which cleans the bowl by creating a huge whirlpool. I've been sitting on the toilet for so long. Where are the semiconductors? The 1970s energy crises ended the old economic growth model of heavy industry and cheap energy. Finding its core markets difficult, Toto began applying the vast material casting and firing knowledge that it originally developed for housing fixtures to new applications. Their initial contact with the semiconductor space came via three large scale high precision measuring instruments. The first of which was the Airslide. This device generates a thin layer of pressurized air to support a load. Second is a ceramic guide shaft for 3D coordinate measuring machines. Such machines are used to measure an item's physical characteristics. Third, and finally, the XY2 axis table. It helps move an item around. These three instruments sold well to research institutes and semiconductor makers, giving Toto their first taste of the industry's intense demands. Another semiconductor related space that the company tried were ceramic substrates. These refer to the physical platforms on which we place transistors, chips, resistors, whatever. Kind of like circuit boards, but not quite. Ceramics fit the bill because of their ideal physical properties. Stable and rigid. Their good thermal conductivity means they can spread heat, but they are electrical insulators. Toto used its substrates to produce hybrid integrated circuits, things with multiple ICs and devices placed onto a single substrate and interconnected, a type of pre advanced packaging. They use such hybrid ICs for their washlets, a nice move for synergies. They also produce standard semiconductor packages. Dyes need protection when in the wild, so we packaged them in resins or ceramics. The latter being what Toto tried to make, unfortunately did not compete well due to high cost and withdrew in 1990. Nonetheless, results in the other fields were encouraging enough that in December 1984 Toto formally established their fine ceramics division. Today it is called their advanced Ceramics division. Creator of the Chuck in 1984 I think it's 1984, NTT Laboratories, Japan's version of Bell Labs, approached Toto's research institute and asked if they can produce something for them. They needed a chuck. So far as I can tell, the word chuck is Scottish or Middle English. It started out as something on a lathe or drill that affixes the workpiece and holds it steady. In the semiconductor world, the chuck holds the wafer inside the tool as it undergoes some process step. NTT Laboratories was working on a tool using a new technology called electron beam lithography. This tool employs an electron beam to Write a chip design onto a wafer, like how a cathode ray tube draws a picture onto a TV screen. The tool needed a wafer chuck, but the existing options did not work well. Mechanical chucks physically clamp the wafer, but that can end up scratching it, generating particles that crush yields. An alternative would be a vacuum chuck, But E beam litho needs a high vacuum environment because you don't want stray particles scattering the electrons. No external atmospheric pressure means that a vacuum chuck cannot clamp the wafer. The most compelling alternative would be a chuck based on electrostatic forces. Something like what allows a charged balloon to pick up small pieces of paper. The printer industry already used such forces to hold down pieces of paper flat against the surface whilst printing. So NTT labs approached Toto to build an electrostatic chuck for them. With an electrostatic force of 1 kg per square centimeter. The job was assigned to a recently hired researcher named Toshiya watanabe. In a 2025 retrospective for the Japan Patent Attorneys association magazine, Watanabe recalled examining entities demands and initially finding them impossible to achieve. All electric static chucks are made out of a special dielectric ceramic plate, or puck, with a layer of electrodes beneath or embedded inside it. After the wafer is placed onto the chuck, we activate the electrodes with a DC voltage, creating an electrostatic force that securely clamps the wafer onto the chuck. Initially, Watanabe assumed that the chuck would operate using the principle of coulombic charge attraction. There is a parallel plate capacitor setup between the chuck's electrodes, the chuck's dielectric ceramic plate and the silicon wafer. This creates opposing charges on both sides of the dielectric. The strength of the force depends on the thickness of the dielectric and its material, a material that has to be a strong enough insulator that does not allow much current to leak through. Watanabe, however, did the math and recognized that achieving NTTs requested 1 kg per square centimeter strength via Coulombic principles would need a ridiculously thin dielectric far beyond what can be possibly made in practice. Befuddled, he went back to NTT scientists and asked what they were smoking. And to his surprise, the scientists explained that they were thinking of a quite different principle of electrostatic clamping. Note that I said that the dielectric material in the columbic chuck should not allow any current leakage. So the material has to have a very strong resistance. But what if we tweak the dielectric material. So to give it some resistance, but not too much, making it slightly leaky. That sounds like a recipe for failure, right? Well, if the dielectric is just leaky enough, then a small amount of current can migrate through the bulk of the dielectric to its surface, the part interfacing with the wafer. This turns that dielectric interface surface into a virtual electrode located just micrometers from the wafer. We now have two opposing charges very close together, creating a powerful clamping force that cannot be achieved with only coulombic based forces. We call this the Johnson Rabec or JR Effect. Discovered and named after two Danish engineers in the 1920s. IBM had worked on variants of this since at least the late 1950s. To make it work for a chuck, Watanabe and his team painstakingly doped the alumina based ceramics with various dopants like titanium oxide in various amounts until they discovered the perfect recipe. Unfortunately, E beam lithography technology did not outperform the incumbent photolithography for throughput reasons. So Toto switched courses and tried to produce a chuck suitable for plasma etch equipment. Etch is a step that goes after lithography. Lithography tells you where to etch. Etch processes use an acid or plasma to make the actual cut. Plasma etch systems use plasma for their etch. Obviously, of course, issues began to emerge. The inside of an etched chamber is a far harsher environment than that inside an E beam machine. Early versions of chucks failed to properly grasp and release the wafer. One of the flaws of the JR effect is that trapped forces on the chuck surfaces can create residual stickiness that makes it hard to declamp the wafer. Watanabe recalled responding to the complaints of a major customer in Osaka. The chuck had cracked inside the plasma chamber, stopping the entire line and earning him a serious reprimand. Finding and fixing the root cause for the crack would take six months. But Toto refused to give up. They worked closely with the customers to achieve their desired goal. This close working relationship, developed over years of trust and hardened by fire, explains how Toto's chucks keep getting specced into processed nodes. Toto debuted their chuck in 1988. Yields and sales steadily grew in the years afterwards. In the 1990s, it became one of their three flagship products within the advanced ceramics division. And of course, as the hedge fund said, it contributes anywhere from 30 to to 50% of the company's operating profits today, growing at a rapid pace. The hedge fund also notes that Toto is The dominant supplier of Chucks to Lam Research for their 3D NAND cryogenic etch processes. What is that? I'm thinking about doing a video about it later, so here I shall just give you the brief rundown. 3D NAND is special in that we have many memory cells produced all at once in large vertical stacks. To produce it, we deposit dozens of alternating layers of dielectric onto a substrate. Then we pattern the layers and etch deep channels or holes through them so that we can produce the memory cells inside them. The hard part with these etches is making sure that we can etch them deep and straight with the side walls smooth. This gets harder with more stacks. Modern 3D NAND might feature memory holes 100 nm wide and and 10 micrometers deep, a very high 100:1 aspect ratio. Existing methods do not work well, leaving issues like bowing incomplete etches, twisting or whatnot. So they developed cryogenic etching, a new type of etching done at cryogenic temperatures of -70 or lower. Companies have been researching this since the 1980s but it only recently reached the market due to the growing needs of 3D NAND. The need for very cold temperatures was one of the major issues for fabs to overcome. It must have been very hard for Toto to produce a ceramic chuck that does all the same things as the old chucks did. But in the cold, guess why it contributes so much of their operating profit? Thank you to friend of the Channel Gen for suggesting this topic. I enjoyed making this one. I got to study toilets and chips, two of my favorite items in the household. The story of Toto is so interesting because of its similarities to others like Fujifilm or in a prior video comparing them with Kodak, I mentioned that Kodak saw itself as a consumer brand with a technological base. Fujifilm instead saw itself as a technology company that happens to have a consumer brand. Toto does not see itself as a tableware toilet or even a sanitary ware company. They are a technology company with an expertise in porcelains honed over a hundred years. They find problems that they can solve and try to solve them with their technology. Right now, that is being rewarded. Alright everyone, that's it for tonight. Thanks for watching. Subscribe to the Channel, sign up for the Patreon and I'll see you guys next time.