@@BushyBrowsHD I was watching I think the 3rd episode of broken silicon and he said that it's worked out that graphene chips are some type of certainty just it's gonna take big companies and fabs putting money into it... I want to know where he gets his information.
@@isbestlizard I don't quite agree. We are simply investing way too little money into fusion. Any oil company spends more in exploration than we do in fusion, and there is nothing that should really be impossible about fusion. Graphene will eventually also hit the streets. It's simply too awesome a material to not use it eventually. Room temperature superconductors on the other hand seem more far fetched. But even cheap liquid nitrogen temperature ones could be extremely useful.
It's great to work with in the lab. I ended up publishing 2 papers on it. But it really is disappointing that it's just something to play with in the lab rather than in industry.
Having worked in both TMDs and Graphene a bit in past few years, pretty accurate description of material interaction here, barring one clarification. Device mobility of 25 is different from Intrinsic carrier mobility which is 2orders higher that clearly mentioned in video. However, the device mobility depends on the gate model and its interaction with gate material - Schottky barrier. Therefore, improving this barrier aka finding more suitable material can improve device mobility. From authors viewpoint, intention of the work seems to show the isolation of SiC-graphene, characterization of its bandgap without trade-off on intrinsic mobility and not building best in-class device.
Being honest, if I heard "square centimeters per volt second" in dialogue of a sci fi movie, I would have said that's obviously a nonsense unit that's just made up.
Ahh, one can make up compounded units any way one wishes. Whatever is practical. Now idiots will use dimensions twice (eg meter per square meter ) or use different units on different sides of an equation.
@@JasminUwU Aggree, I like to give the electrical resistivity ρ in units of ohm square meter per meter instead of Ohm meters, because it immediately shows, that a formula should include an area and a length, not just a length.
"The silicone carbide chips are placed inside a crucible, where they are subjected to high heat and McCarthyist accusations." Line delivered stone cold.😂
I published a paper on it being an electrochemically transparent material as well as a hydrophobic conductor back in 2021, and now I got an accepted manuscript on its ability to perform orientationally dependent catalysis. It's a truly amazing material that can do so much! The bottleneck in production is such a damn shame.
Does your research delve mainly into chemical science? im intrigued to know how it might be used in a photonic based system, i had an idea to perform combinatorial logic on the waves them selves using the modulators but whatever controls the modulators are limited by electricity still. Parallel modulators with a much better conductivity may be the key.
No, in fact Asianometry got that bit wrong. The p-orbitals projecting perpendicular to the basal plane overlap on alternate atoms, resulting in a weak chemical bond, as originally explained by J. D Bernal in his 1924 article about the crystal structure of graphite, and in several later articles by others. The elastic constants of graphite are consistent with this model. Van der Waals forces are a dynamical, electrostatic effect of itinerant electrons, which does not involve orbital overlap, and would give different elastic constants for graphite from the observed values if they were responsible for interlayer binding. The interlayer binding of graphite is often mistakenly attributed to Van de Waals forces because it is weak. Okay? :-)
There's an interesting compound called siloxene that's formed by dissolving calcium silicide in hydrochloric acid. A yellow solid is formed that when washed and dried is fairly stable. When treated with oxidizing agents it gives off a yellow orange chemiluminescence. It's one of a very few inorganic chemiluminescent agents. Apparently it's strange stuff - it has intercalated hydrogen, so it burns quietly when ignited, but apparently it's structure is similar to graphene in some respects, being made up of numerous sheets of silicon bonded to oxygen and OH groups. Not as exciting as graphene maybe, more like graphene oxide, but exotic nontheless....
I'm watching a video about another breakthrough on a 12-year-old Core i5. If I was using 12-year-old CPU in 2004 (back from 1992) I would have been committed (and moved to a psychiatric facility asap).
Just as a note on the physics, when electrons are excited to the conduction band, current flows via movement of electrons and by the holes left in the valence band as well (think of holes like air in a spirit level) where the absence of an electron allows the flow of other electrons and therefore conduction.
paragraf already has, this is maybe one of the most glossed over videos he has ever done. He missed a lot on it. 2 inch graphene wafers are already in mass production and so are working usable chips. GEIC in UK also has tons of research on it, he didn't mention at all. and he missed several 2d elements discoveries borophene, phosphorene, goldene etc.
It's pretty neat when you realise you can get one of the most intriguing materials ever manufactured by connecting scotch tape to a rock and pulling a few times. Not very sci-fi
Great presentation, I enjoy your precise, tight, staccato evocative of Rod Serling's vocal presentations of the Twillight Zone while you also casually throw in humor. Very effective way to communicate technical subjects.
Interesting advancement have been made towards single molecule transistor Where quantum wave effect have been used to create destructive or constructive wave, eliminating tunnelling effect
One way to make a fet out of graphene is to put a very precise gold layer so individual atoms sit in the c6 cells, next deposit another graphene later over that. If you apply an electric field the graphene will conduct to the second layer, if you apply a positive charge it will resist current flow. It behaves similar to a tunnel diode at zero charge.
@@ZoranRavic Yes, I know. However, they understood less of the fundamental physics, and vacuum tubes were well-established and highly manufacturable technology. The transistor was obviously superior for important niche (mainly military) applications...something the GFET is not. (If it were otherwise, we'd have commercial devices by now...or, at the very least, an announced licensing agreement.)
Great work - very detailed and honest overview of this work. It cuts through a lot of the hype you see around these papers. Super appreciate your work.
I’ve never been able to understand why it’s taken 20 years to get here. I would have thought graphene electronics would be in millions of devices by now. I appreciate this video for explaining some of the roadblocks, especially the missing band gap.
What about GaN HEMT devices? They rely on a 2-D "electron gas" that is spontaneously generated by piezoelectric stress from the lattice mismatch stress between an AlGaN - GaN interface. Does this count as a 2-D transistor?
Who cares how it's classified. GaNFETs are available at Digikey right now, and they are better than conventional Silicon MOSFETs is most ways that matter. People are already designing tiny but high powered switch-mode voltage converters with them. My next solid state relay is design going to use GaNFETs instead of conventional silicon ones.
Graphene’s bandgap issue is a fundamental limitation when it comes to its use in semiconductors. Here’s a detailed look at how the electronic structure of graphene compares to traditional semiconductor materials like silicon and germanium and why it’s challenging to open a usable bandgap in graphene. 1. Understanding Bandgaps in Semiconductors In semiconductor materials like silicon and germanium, a bandgap separates the valence band (filled with electrons) and the conduction band (where electrons can move freely). The size of this bandgap determines the semiconductor’s electrical properties. Silicon: Silicon has an indirect bandgap of ~1.1 eV, which allows it to conduct electricity under certain conditions while staying insulating at low voltages. This bandgap size strikes a balance, enabling silicon to switch on and off easily in digital electronics, which is crucial for transistor functionality. Germanium: Germanium has a smaller indirect bandgap of ~0.67 eV, making it more conductive than silicon but less ideal for creating a clear “off” state in digital circuits. Despite this, it’s still useful in some high-speed applications, though it's more sensitive to temperature variations than silicon. 2. Graphene’s Bandgap: A Zero-Gap Semiconductor Graphene is unique because it’s a zero-bandgap material. Its valence and conduction bands meet at the Dirac points (in a "cone" shape in energy-momentum space), meaning there is no energy gap to prevent electrons from jumping between these bands. This structure makes graphene act like a semimetal rather than a semiconductor, giving it remarkable electrical conductivity but eliminating the possibility of an “off” state without a bandgap. In traditional transistors, this inability to fully switch “off” would result in leakage currents that limit energy efficiency and overall functionality. 3. Attempts to Induce a Bandgap in Graphene Researchers have tried various methods to introduce a bandgap in graphene, but each approach has notable drawbacks due to counteracting effects: Strain Engineering: Applying mechanical strain to graphene can distort its lattice structure, creating a slight bandgap. However, strain-induced bandgaps are typically small (tens of meV) and can vary depending on the amount of strain, making them impractical for consistent semiconductor applications. Chemical Doping: By introducing atoms like nitrogen or boron, researchers attempt to disrupt graphene's symmetry and open a bandgap. However, doping often results in uncontrollable defects that degrade graphene’s remarkable properties, including its mobility, and introduces unwanted scattering. Bilayer Graphene with Electric Fields: Stacking two layers of graphene and applying an electric field can also induce a bandgap (up to ~250 meV). But this setup is complex, and even this bandgap is much smaller than silicon's, limiting the bilayer's usefulness in digital circuits. 4. Comparison with Silicon and Germanium Silicon and germanium have stable, tunable bandgaps that allow control over electron flow, enabling them to switch reliably between conductive and non-conductive states. In comparison: Graphene’s Conduction: Because it lacks a bandgap, graphene has a very high electron mobility and conducts almost like a metal, which is advantageous in some contexts but is unsuitable for digital switches. Thermal Sensitivity: Silicon’s bandgap provides a stable “off” state even under fluctuating temperatures, a feature that graphene lacks since it cannot isolate an “off” state. Induced Bandgap Limitations: Any forced bandgap in graphene is inherently unstable or too small to compete with silicon, as it’s either difficult to control or comes with trade-offs, such as reduced mobility and material integrity. 5. Why Graphene is Still Being Studied Despite its limitations in traditional semiconducting applications, graphene’s high carrier mobility, strength, and flexibility make it valuable in areas like high-frequency electronics, sensors, and photodetectors. In these fields, its lack of a bandgap isn’t as critical since it benefits from high conductivity and speed without needing the on-off functionality of digital transistors. Conclusion Graphene’s lack of a natural bandgap and the difficulty in introducing a stable one make it unsuitable for conventional semiconductor applications like transistors. Silicon and germanium, with their stable bandgaps, remain far better suited for digital electronics where controlled switching is essential.
Some day. Just not today lol. But fr though, they will find a way, mark my words. Whether it's 2, 4, 8, or 16 years idk. I mean, they already have graphene based thermal pads, so it's gotta be going somewhere ;) lol.
it is happening, paragraf has 2 inch wafers in mass production as well as chips being made. GEIC in UK has tons of other research on it. This video misses a lot
The electron mobility of HEMT devices (high electron mobility transistors) are in the ball park of these proposed GFETs (and usually better). III-V processes use a 2D electron gas at the interference between the semi insulator and the metal. The most common III-V are AlGaAs, InGaAs, GaN, and InP. Of course these are all usually used as amplifiers. Anyway, I point this out because these are decades old processes that give some grounding in reality. Graphene is cool, but it's also a buzz word. It's amazing how many papers come out with titles filled with buzz words but which don't do anything novel.. or useful.
“McCarthiest Accusations” Obviously you’re checking to see if anyone is listening. The problem with Graphene is not small size but connector stability. Graphene is composed of 4n+2 carbon atoms in resonance stabilized oribitals. Its properties are quantum in nature given the fact that electrons have many more orbitals than a molecule like butadiene, per carbon. There is an orbital that covers the whole sheet, theoretically. Practically no, the energy difference between an orbital covering a square cm and 1/4 cm is trivial. The near light speed capacity (and near zero resistance) for carrying electrons is dependent on the length of the orbital doing the carrying. Thus at some point adding length doesn’t accomplish anything. However, the problem is at the contacts of the edges of the graphene, these areas are subject to higher resistance and thus you can’t dump a lot of electrons on these without the contacts heating up. This problem is solved by making the contact wider. Presumably if you could engineer a low resistance contact then you could treat the graphene rectangle like a superconductor for any area that might be in a chip. That’s good, but. If you think about what making faster transistors in a CPU do is that in increases the clock cycle, electrons are fed down a circuit in a burst, the the circuit goes to zero and then there’s another burst. To go faster means the burst of electrons are closer together. Heating at the contacts goes up. That’s a problem.
somehow i find it funny that we have had graphite for a long time in pencils and also tapes for almost a hundred years and it took so long for someone to figure out you can pull the graphene sheets with a tape 😄
If graphene can reach hundreds of GHz or even THz then it doesn't matter if power consumption goes up by 50x. The reason the transistor count is so absurdly high in today's CPUs and GPUs is because of the lack of switching speed. You roughly speaking need to increase transistor count by 70% to increase IPC (instructions per clock) by 15%. So a Pentium sized chip at 1THz would run rings around a modern core while still using significantly less power. Of cause, if the transistors can also be made as small (or thereabout) as today's leading silicon nodes, then the area of a Pentium would shrink to a pin-prick.
Keep in mind that power scales linearly with frequency. So if your transistors are twice as power efficient, you can only double your frequency, (ignoring wire resistance). If graphene transistors are faster than silicon transistors, but take more power than silicon transistors, then your chip will not necessarily run faster However, the power dissipation in the wires is the significant bottleneck on frequency, not transistor switching speed
"increase transistor count by 70% to increase IPC by 15%" -- that's only true for single-threaded workloads.Scaling by adding more compute units (like GPUs, vector machines, multicore, etc) scales much closer to 1:1. So all these whole wafer AI designs are still going to be the same size, regardless of how fast you can switch.
i think the biggest breakthrough will be when electronics focus on trying to become more efficient under high temperatures. Reason graphene is so innovative is because it is made from coal the 4th most abundant element on earth and it is atoms thin. Coal has a temperature rating of 3 which is in the mid point of a conductor and an insulator. When graphene over heats which probably barely happens, graphene will burn away meaning that a product with intended disposability and low cost for electronics will create a more efficient product. The reason its more effective with temperature is because the hexagonal structures creates two paths for the electron bonds to share photons and energy. I think that graphene will create the new wires and connection points for motherboards in the future. Transistors themselves i think that replacing the gates with nanotubes will the best option for what is in the market today to prevent the leaks currently creating panic amongst so many EE. Moores Law. All Love Doe
Great video, as always! Lower band gap = faster switching + higher energy/power consumption. But there may come other factors affecting leakage currents. So the bottomline may look differently in the future. I also disagree with the conclusions. Graphene is hard to techologize, while TMDs allow more room to play with substrates, chemistry, available techniques and... band gap. Yes, graphene epitaxy in between SiC layers is cool and promising, yet we are 10-20 years away from the defining the winner. P.S. I am a PhD in single carbon nanotubes: CVD growth and electric/chemical manipulation. Grew and played wit the CVD graphene as well. The field hasn't moved far in 15 years, as I observe
we already have chips with graphene, already in mass production. They use way less power, almost 1000 times less. video is inaccurate as of 2 years ago.
@@Gaste11o GEIC in England does all the research they have multiple buildings dedicated to it. The company & website paragraf has them for sale and bulk order. Hall effect sensors and 1 or 2 others. The 2 inch wafers are on some documentary videos about their campus/company, much better than this video. paragraf follows the industry they are there to succeed and make profit so it hopefully won't cease to exist like most lab stuff does. I have been a many year subscriber of asianometry and this is probably his worst video.
One possible configuration of microtransistor is a vacuum lamp. At that scale it doesn't needs neither vacuum nor heating, but the operating voltages are high (close to 12V).
Maybe you should stop listening to msm sensationalist rage-baiting urinalists and look at actual scientific advancements directly. Urinalists don't care if they lie, they juat care about views. Besides, Graphene is already being used in batteries, thermal systems, composites, and other things.
What I like about this channel is the voice-over - it is either well researched human or the AI is way beyond anything that AI is capable of IME. The almost fluff, the (geddit?) not to mention the pun it carried. Can AI be that subtle and still carry genuine information? Yea, Yea one day, and I shudder. AND "Macarthyist accusations" is a phrase that doesn't appear on the web, so explain it applying to graphene deposition - someone.
Graphene transistors also generate less heat and require less W to power/work with for same performance. We,re talking about trully efficient chips, going further that can easily work with just passive cooling and last for weeks.
The current chip industry is so cost effective and optimized that large leaps in technology such as advanced materials or extreme-large-scale particle etching machines can only be commericalized in markets where western chips are banned from export.
Professor James Tour at Rice University is developing a means of mass-producing graphene from any organic material, including waste plastic (flash graphene). Please look at his RUclips channel for more information. He also has some very thoughtful commentary on religion and Christianity.
This seems like it’s not too dissimilar to GAN transistors, which conduct with an atom-thick “electron gas”. It’s such a good conductor that this atom-thick layer is enough for lots of current.
even if doesnt make into ultra fast (GHz) computer chips due to the mobility and complexity, it can be used in power electronics for renewable and motor drives, these circuits hovers in the KHz band and normally are monolitic structures
My favorite part about graphene is how it's speculated that it can do all these amazing things, and yet none of these applications seem to have actually been performed
Did my PhD in a related field and worked with 2D and 1D materials myself, including making transistors in the clean room. I can only agree with all the assessments.
Hi Im currently studying computer engineering and I want to specialize in chip design, what would you recommend for me to get in closer contact with that field?
I think even with these new materials, the practical limit for production processes for integrated circuits using ASML/TSMC process is maybe around 1 nanometer. We may be finally finally reaching the limit of Moore's Law. This is where stacked chips may start to become important.
I noticed a bunch of Dutch names and they were actually Dutch guys, not just people with Dutch names in for example the US. Some were even at Philips, where ASML comes from.
I think it might turn out that the graphene will actually never leave the lab. However, things might end up in such a way, that solutions and techniques for creating graphene, invented during development, will be used in the creation of some other metamaterials. So basically, Carbon in graphene, would be exchanged with some other atoms, but the development/manufacturing techniques (not necessarily the ones in this video) will be the same
Little to no band gap would mean it could function as a diode with very low forward voltage, but since the band gaps don't overlap, it would still function as a diode? That's amazing for high powered diodes, like in solar panel arrays and such.
You have one of the cleanest and least pretentious content development and delivery styles on the interwebs. I think I could listen to you describe moonlight raindrops on a puddle in a vacant alley. In what must be close to 100 of your videos I've not heard a verbal crutch from "um" or "ah" to those seemingly ubiquitous toss off phrases like "what's up guys?" or "[insert thirty seconds of drivel]...these bad boys...". Well done kiddo, I don't know what makes you do it but I hope those clouds of inspiration remain in your sky.
Perhaps those chips shall finally manifest using graphene quantum dots or other structures doped with some special element. Maybe the key to endowing it with a band gap is structuring it in a special way!
I’ve built high power, high efficiency class E radio transmitters using SiC fets (silicon carbide) because they have low gate capacitances and allow for higher frequencies for switching. I wonder if graphene gates would permit high power fets designed for PWM applications to be used in the hobby world for higher frequency switching radio amp designs. It all depends upon gate capacitance at those frequencies, I guess.
I watched a news report when I was a kid about 40 years ago about how Graphene was going to change the world in just a few short years..... I am still waiting for my personal jetpack.
Excuse my simple approach (I have done some semiconductor physics in my studies BTW) but transistors rely on SEMICONDUCTOR action (through doping). In a FET, the channel's conductivity is modulated from the relatively high intrinsic resistance of the pure(ish) semiconductor (usually silicon) to a lower level which results from doping. Graphene however is a CONDUCTOR. It has no ability to have its conductivity modulated ! Whatever these experimental transistors may be, there are NOT 'graphene transistors' since graphene is ALWAYS a CONDUCTOR. It can have NO transistor action by definition. Researchers here are making rather far-fetched claims. I suspect it's the silicon carbide doing the work here, not the graphene !
Because it is a 2D material, no consumer device will ever be viable if it will rely on notably large continuous sheets of graphene, as thermal expansion will constantly crack the pieces apart.
I wonder if some kind of bacteria or cellular automata could be used to create graphene. Imagine a computer seed you feed raw materials, and it grows the computer.
Funny enough, we have already had to deal with increased power consumption to get faster switching speeds, ask anyone that has bought into the last few generations of Intel processors, when a CPU itself is drawing over 250 watts to get work done, that is a problem. That said, I have been perfectly happy with my insanely fast for what it is AMD Ryzen 5 5600X, it brought down power consumption vs my outgoing AMD Rysen 5 1600X by 30 watts and yet I got a HUGE increase in performance. It would be awesome to see graphene become part of CPU tech, because I can't imagine what it would be like to have the performance I have now, with an even lower power draw.
The ability of Graphene to be produced from coffee grinds is one of its peculiarities. (microwave process). Although, it doesn't taste very nice to drink …
i gather graphene local interconnect would be even more important. Gate all around makes sillicon transistors far more scalable than what any current local interconnect tecnology allows, making that the bottleneck
YEARS AND YEARS ago, like almost 30 years ago, I realised that I could use a lone drawn on a piece of paper with a sketching pencil as a resistor, or at least that's what my multimeter showed.
Could graphene act as a transistor in a next-gen chip? Scientists come to overwhelming consensus of "maybe"
Thank you, you save 15 minutes of my time lool
Graphene "maybe" can revolutionize every part of electronics.
Consensus of "yes, but only in a lab"
Can I get my assignment done on time? "maybe" , maybe is so neutral, the perfect way to say yes AND no at the same time, its my go to
@@termiterasinSo can superconductors... Maybe
Let’s all give a huge applause to Asianometry for always putting immense time into research. Always superb work.
Graphene can basically do anything, except leave the lab. waste of time..
And some corny memes for a laugh. hehe
@@teemoammo they said the same thing when trying to make the blue LED, but look how revolutionary that was once they finally figured it out.
@@BushyBrowsHD I was watching I think the 3rd episode of broken silicon and he said that it's worked out that graphene chips are some type of certainty just it's gonna take big companies and fabs putting money into it...
I want to know where he gets his information.
"Graphene can basically do anything, except leave the lab"
Right?? Sick of hearing about it! Stick it in the same pile as 'fusion' and 'room temperature superconductor'
@@isbestlizard I don't quite agree. We are simply investing way too little money into fusion. Any oil company spends more in exploration than we do in fusion, and there is nothing that should really be impossible about fusion.
Graphene will eventually also hit the streets. It's simply too awesome a material to not use it eventually.
Room temperature superconductors on the other hand seem more far fetched. But even cheap liquid nitrogen temperature ones could be extremely useful.
It's great to work with in the lab. I ended up publishing 2 papers on it. But it really is disappointing that it's just something to play with in the lab rather than in industry.
People can do anything with graphene, except for getting a new joke.
my monitor has a graphene heatsink, doesnt need a fan :)
Having worked in both TMDs and Graphene a bit in past few years, pretty accurate description of material interaction here, barring one clarification.
Device mobility of 25 is different from Intrinsic carrier mobility which is 2orders higher that clearly mentioned in video. However, the device mobility depends on the gate model and its interaction with gate material - Schottky barrier. Therefore, improving this barrier aka finding more suitable material can improve device mobility. From authors viewpoint, intention of the work seems to show the isolation of SiC-graphene, characterization of its bandgap without trade-off on intrinsic mobility and not building best in-class device.
Being honest, if I heard "square centimeters per volt second" in dialogue of a sci fi movie, I would have said that's obviously a nonsense unit that's just made up.
Ahh, one can make up compounded units any way one wishes. Whatever is practical.
Now idiots will use dimensions twice (eg meter per square meter ) or use different units on different sides of an equation.
@@lrrrruleroftheplanetomicro6881meter per square meter makes sense sometimes
@@JasminUwU Aggree, I like to give the electrical resistivity ρ in units of ohm square meter per meter instead of Ohm meters, because it immediately shows, that a formula should include an area and a length, not just a length.
To be fair, it was just made up
@@lrrrruleroftheplanetomicro6881metres per second per second. That one is used for acceleration due to gravity.
Dude keeps dropping bangers after bangers.
You should check out his backlog
"The silicone carbide chips are placed inside a crucible, where they are subjected to high heat and McCarthyist accusations." Line delivered stone cold.😂
2 months ago
Must be patreon
I first thought that I heard that wrong. LOL
i dunno man mccarthy turned out to be right
@@manitoba-op4jx Lol no he didn't. The guy was just paranoid and prejudiced, nothing more.
I published a paper on it being an electrochemically transparent material as well as a hydrophobic conductor back in 2021, and now I got an accepted manuscript on its ability to perform orientationally dependent catalysis. It's a truly amazing material that can do so much! The bottleneck in production is such a damn shame.
Does your research delve mainly into chemical science? im intrigued to know how it might be used in a photonic based system, i had an idea to perform combinatorial logic on the waves them selves using the modulators but whatever controls the modulators are limited by electricity still. Parallel modulators with a much better conductivity may be the key.
@@devilsolution9781 yes, it's mostly electrochemistry. I'm afraid I'm far less educated on photonics.
"for fear of entering the quantum realm" That was a good line
Ok there ant man
Ya best start believin' in quantum realms. Yer in one!
And the accompanying slide stayed on for way too long lol. I laughed at that line twice.
Quantum Manifold Modulator is really throde off😆
2:41 One of the authors of this paper is Horst Stormer. My Dad worked with him at Bell Labs.
Subjected to McCarthyist accusations! Sounds brutal.
Graphite: "Anyway, here's Van der Waal"
No, in fact Asianometry got that bit wrong. The p-orbitals projecting perpendicular to the basal plane overlap on alternate atoms, resulting in a weak chemical bond, as originally explained by J. D Bernal in his 1924 article about the crystal structure of graphite, and in several later articles by others. The elastic constants of graphite are consistent with this model. Van der Waals forces are a dynamical, electrostatic effect of itinerant electrons, which does not involve orbital overlap, and would give different elastic constants for graphite from the observed values if they were responsible for interlayer binding. The interlayer binding of graphite is often mistakenly attributed to Van de Waals forces because it is weak. Okay? :-)
There's an interesting compound called siloxene that's formed by dissolving calcium silicide in hydrochloric acid. A yellow solid is formed that when washed and dried is fairly stable. When treated with oxidizing agents it gives off a yellow orange chemiluminescence. It's one of a very few inorganic chemiluminescent agents.
Apparently it's strange stuff - it has intercalated hydrogen, so it burns quietly when ignited, but apparently it's structure is similar to graphene in some respects, being made up of numerous sheets of silicon bonded to oxygen and OH groups. Not as exciting as graphene maybe, more like graphene oxide, but exotic nontheless....
I'm watching a video about another breakthrough on a 12-year-old Core i5. If I was using 12-year-old CPU in 2004 (back from 1992) I would have been committed (and moved to a psychiatric facility asap).
^ This. Hasswell i7 here. 4.8 Ghz single thread. I am a man of simple needs, and I will ride this CPU out until it's death.
If it does the job... It's good
@@Im_TheSaint chances are: that i7 will outlive you.
@@n00b247 Then burry me with it.
@@n00b247 Then bury me with it.
Just as a note on the physics, when electrons are excited to the conduction band, current flows via movement of electrons and by the holes left in the valence band as well (think of holes like air in a spirit level) where the absence of an electron allows the flow of other electrons and therefore conduction.
Wake me up when they can actually commercialize graphene
About the same time fusion power becomes viable.
Only commercially injectable according to facebook...
Graphene is not even that expensive anymore.
@@cjc363636Fission is better.
paragraf already has, this is maybe one of the most glossed over videos he has ever done. He missed a lot on it. 2 inch graphene wafers are already in mass production and so are working usable chips. GEIC in UK also has tons of research on it, he didn't mention at all.
and he missed several 2d elements discoveries borophene, phosphorene, goldene etc.
They won the Nobel Prize using scotch tape?
Yet another Condensed Matter Physicist win against the Particular Physicist needing decades and the entirety of CERN just to compete
that’s part of the reason why they won. They were able to discover a groundbreaking material using something as simple as scotch tape.
Yes, but they were in the UK, so probably called it Sellotape. :-)
They won the Nobel Prize for being physicists who *realized* that Scotch tape would work, and then *tried it.*
It's pretty neat when you realise you can get one of the most intriguing materials ever manufactured by connecting scotch tape to a rock and pulling a few times. Not very sci-fi
15:23 That was very easy to understand even with very basic knowledge about how transistors work like I do. Thank you.
McCarthyesque accusations? I’m not familiar with that physical effect.
Those carbon atoms will never be able to work in hollywood again!
@@puffman06 They will be Carbonlisted.
@@puffman06McCarthy only accused government members
Huge amount of pressure, applied somewhat randomly with the energy source being anxiety.
Great presentation, I enjoy your precise, tight, staccato evocative of Rod Serling's vocal presentations of the Twillight Zone while you also casually throw in humor. Very effective way to communicate technical subjects.
He who controls the graphene controls the universe.
The Graphene must flow!
Interesting advancement have been made towards single molecule transistor
Where quantum wave effect have been used to create destructive or constructive wave, eliminating tunnelling effect
Source?
@@electroflame6188 search: scaling beyond 1nm
@@electroflame6188 I'd swear I replied, youtube vid
Scaling Beyond 1nm Anastasi In Tech
10:48 I have always suspected that electricity is witchcraft.
Do you mean 7:10? How is anything at that time stamp at all which craft related?
@@vadepierce4542The Crucible is a play about the Salem Witchcraft Trials.
@@Gwanzan3325 ohhhhhhh I asked someone else about the joke. I clearly did not get it. Thank you for explaining it to me
One way to make a fet out of graphene is to put a very precise gold layer so individual atoms sit in the c6 cells, next deposit another graphene later over that. If you apply an electric field the graphene will conduct to the second layer, if you apply a positive charge it will resist current flow. It behaves similar to a tunnel diode at zero charge.
For the sake of comparison, the CK703 was being manufactured ONE YEAR after the discovery of the point contact transistor.
That was the very beginning of transistors, with nothing remotely comparable. Not to mention that it was "easy" to manufacture.
@@ZoranRavic Yes, I know. However, they understood less of the fundamental physics, and vacuum tubes were well-established and highly manufacturable technology. The transistor was obviously superior for important niche (mainly military) applications...something the GFET is not. (If it were otherwise, we'd have commercial devices by now...or, at the very least, an announced licensing agreement.)
Very nice presentation. Appreciate the final closing. This is not ready for production.
Nice to be able to put it on a timeline.
babe wake up new asianometry just dropped
Great work - very detailed and honest overview of this work. It cuts through a lot of the hype you see around these papers. Super appreciate your work.
"All I know is my gut says.....Maybe" - Zat Branigan (Futurama)
Fun Fact about TMDCs: A TMDC monolayer absorbs about ~4% of the light transmitted through it, which is a ton considering it's three atoms thick!
I’ve never been able to understand why it’s taken 20 years to get here. I would have thought graphene electronics would be in millions of devices by now. I appreciate this video for explaining some of the roadblocks, especially the missing band gap.
What about GaN HEMT devices? They rely on a 2-D "electron gas" that is spontaneously generated by piezoelectric stress from the lattice mismatch stress between an AlGaN - GaN interface. Does this count as a 2-D transistor?
Who cares how it's classified. GaNFETs are available at Digikey right now, and they are better than conventional Silicon MOSFETs is most ways that matter. People are already designing tiny but high powered switch-mode voltage converters with them. My next solid state relay is design going to use GaNFETs instead of conventional silicon ones.
@@SaltyPuglord Are you going to make a video about it?
@@ZoranRavic Possibly. But I have to build and test it first. And right now I'm building and testing an OBD-II cable tester.
Graphene’s bandgap issue is a fundamental limitation when it comes to its use in semiconductors. Here’s a detailed look at how the electronic structure of graphene compares to traditional semiconductor materials like silicon and germanium and why it’s challenging to open a usable bandgap in graphene.
1. Understanding Bandgaps in Semiconductors
In semiconductor materials like silicon and germanium, a bandgap separates the valence band (filled with electrons) and the conduction band (where electrons can move freely). The size of this bandgap determines the semiconductor’s electrical properties.
Silicon: Silicon has an indirect bandgap of ~1.1 eV, which allows it to conduct electricity under certain conditions while staying insulating at low voltages. This bandgap size strikes a balance, enabling silicon to switch on and off easily in digital electronics, which is crucial for transistor functionality.
Germanium: Germanium has a smaller indirect bandgap of ~0.67 eV, making it more conductive than silicon but less ideal for creating a clear “off” state in digital circuits. Despite this, it’s still useful in some high-speed applications, though it's more sensitive to temperature variations than silicon.
2. Graphene’s Bandgap: A Zero-Gap Semiconductor
Graphene is unique because it’s a zero-bandgap material. Its valence and conduction bands meet at the Dirac points (in a "cone" shape in energy-momentum space), meaning there is no energy gap to prevent electrons from jumping between these bands. This structure makes graphene act like a semimetal rather than a semiconductor, giving it remarkable electrical conductivity but eliminating the possibility of an “off” state without a bandgap.
In traditional transistors, this inability to fully switch “off” would result in leakage currents that limit energy efficiency and overall functionality.
3. Attempts to Induce a Bandgap in Graphene
Researchers have tried various methods to introduce a bandgap in graphene, but each approach has notable drawbacks due to counteracting effects:
Strain Engineering: Applying mechanical strain to graphene can distort its lattice structure, creating a slight bandgap. However, strain-induced bandgaps are typically small (tens of meV) and can vary depending on the amount of strain, making them impractical for consistent semiconductor applications.
Chemical Doping: By introducing atoms like nitrogen or boron, researchers attempt to disrupt graphene's symmetry and open a bandgap. However, doping often results in uncontrollable defects that degrade graphene’s remarkable properties, including its mobility, and introduces unwanted scattering.
Bilayer Graphene with Electric Fields: Stacking two layers of graphene and applying an electric field can also induce a bandgap (up to ~250 meV). But this setup is complex, and even this bandgap is much smaller than silicon's, limiting the bilayer's usefulness in digital circuits.
4. Comparison with Silicon and Germanium
Silicon and germanium have stable, tunable bandgaps that allow control over electron flow, enabling them to switch reliably between conductive and non-conductive states. In comparison:
Graphene’s Conduction: Because it lacks a bandgap, graphene has a very high electron mobility and conducts almost like a metal, which is advantageous in some contexts but is unsuitable for digital switches.
Thermal Sensitivity: Silicon’s bandgap provides a stable “off” state even under fluctuating temperatures, a feature that graphene lacks since it cannot isolate an “off” state.
Induced Bandgap Limitations: Any forced bandgap in graphene is inherently unstable or too small to compete with silicon, as it’s either difficult to control or comes with trade-offs, such as reduced mobility and material integrity.
5. Why Graphene is Still Being Studied
Despite its limitations in traditional semiconducting applications, graphene’s high carrier mobility, strength, and flexibility make it valuable in areas like high-frequency electronics, sensors, and photodetectors. In these fields, its lack of a bandgap isn’t as critical since it benefits from high conductivity and speed without needing the on-off functionality of digital transistors.
Conclusion
Graphene’s lack of a natural bandgap and the difficulty in introducing a stable one make it unsuitable for conventional semiconductor applications like transistors. Silicon and germanium, with their stable bandgaps, remain far better suited for digital electronics where controlled switching is essential.
".. subjected to high heat and McCarthyist accusations.." !! I enjoy how you drop these gems in at random, keeping us paying attention. 👏👏👏
I’ve been hearing about how graphene is almost there for 20 years. I don’t think it’s going to happen.
Bruh, it was only discovered years ago... unlike your pp which is so microscopic it still hasn't been discovered yet.
Just like with fusion energy and practical quantum computing, it's just 5 years away at all times.
Some day. Just not today lol. But fr though, they will find a way, mark my words. Whether it's 2, 4, 8, or 16 years idk. I mean, they already have graphene based thermal pads, so it's gotta be going somewhere ;) lol.
Electric car manufacturers still have their 2010 short sellers.
it is happening, paragraf has 2 inch wafers in mass production as well as chips being made. GEIC in UK has tons of other research on it. This video misses a lot
The electron mobility of HEMT devices (high electron mobility transistors) are in the ball park of these proposed GFETs (and usually better). III-V processes use a 2D electron gas at the interference between the semi insulator and the metal. The most common III-V are AlGaAs, InGaAs, GaN, and InP. Of course these are all usually used as amplifiers. Anyway, I point this out because these are decades old processes that give some grounding in reality. Graphene is cool, but it's also a buzz word. It's amazing how many papers come out with titles filled with buzz words but which don't do anything novel.. or useful.
“McCarthiest Accusations”
Obviously you’re checking to see if anyone is listening.
The problem with Graphene is not small size but connector stability. Graphene is composed of 4n+2 carbon atoms in resonance stabilized oribitals. Its properties are quantum in nature given the fact that electrons have many more orbitals than a molecule like butadiene, per carbon. There is an orbital that covers the whole sheet, theoretically. Practically no, the energy difference between an orbital covering a square cm and 1/4 cm is trivial. The near light speed capacity (and near zero resistance) for carrying electrons is dependent on the length of the orbital doing the carrying. Thus at some point adding length doesn’t accomplish anything.
However, the problem is at the contacts of the edges of the graphene, these areas are subject to higher resistance and thus you can’t dump a lot of electrons on these without the contacts heating up. This problem is solved by making the contact wider.
Presumably if you could engineer a low resistance contact then you could treat the graphene rectangle like a superconductor for any area that might be in a chip. That’s good, but.
If you think about what making faster transistors in a CPU do is that in increases the clock cycle, electrons are fed down a circuit in a burst, the the circuit goes to zero and then there’s another burst. To go faster means the burst of electrons are closer together. Heating at the contacts goes up. That’s a problem.
somehow i find it funny that we have had graphite for a long time in pencils and also tapes for almost a hundred years and it took so long for someone to figure out you can pull the graphene sheets with a tape 😄
If graphene can reach hundreds of GHz or even THz then it doesn't matter if power consumption goes up by 50x. The reason the transistor count is so absurdly high in today's CPUs and GPUs is because of the lack of switching speed. You roughly speaking need to increase transistor count by 70% to increase IPC (instructions per clock) by 15%. So a Pentium sized chip at 1THz would run rings around a modern core while still using significantly less power. Of cause, if the transistors can also be made as small (or thereabout) as today's leading silicon nodes, then the area of a Pentium would shrink to a pin-prick.
Would be great to have all 3 of them. More transistors, less ppwer consumption and higher frequency.
Keep in mind that power scales linearly with frequency. So if your transistors are twice as power efficient, you can only double your frequency, (ignoring wire resistance). If graphene transistors are faster than silicon transistors, but take more power than silicon transistors, then your chip will not necessarily run faster
However, the power dissipation in the wires is the significant bottleneck on frequency, not transistor switching speed
You still have to cool a processor. Even if electricity is cheap, a 50x increase in heat density would be very challenging.
@@E4tHamyeah, capacitance is difficult to reduce. I'm thinking more sensitive transistors like tunnel FETs might be better.
"increase transistor count by 70% to increase IPC by 15%" -- that's only true for single-threaded workloads.Scaling by adding more compute units (like GPUs, vector machines, multicore, etc) scales much closer to 1:1. So all these whole wafer AI designs are still going to be the same size, regardless of how fast you can switch.
i think the biggest breakthrough will be when electronics focus on trying to become more efficient under high temperatures. Reason graphene is so innovative is because it is made from coal the 4th most abundant element on earth and it is atoms thin. Coal has a temperature rating of 3 which is in the mid point of a conductor and an insulator. When graphene over heats which probably barely happens, graphene will burn away meaning that a product with intended disposability and low cost for electronics will create a more efficient product. The reason its more effective with temperature is because the hexagonal structures creates two paths for the electron bonds to share photons and energy. I think that graphene will create the new wires and connection points for motherboards in the future. Transistors themselves i think that replacing the gates with nanotubes will the best option for what is in the market today to prevent the leaks currently creating panic amongst so many EE. Moores Law. All Love Doe
Great video, as always!
Lower band gap = faster switching + higher energy/power consumption. But there may come other factors affecting leakage currents. So the bottomline may look differently in the future.
I also disagree with the conclusions. Graphene is hard to techologize, while TMDs allow more room to play with substrates, chemistry, available techniques and... band gap. Yes, graphene epitaxy in between SiC layers is cool and promising, yet we are 10-20 years away from the defining the winner.
P.S. I am a PhD in single carbon nanotubes: CVD growth and electric/chemical manipulation. Grew and played wit the CVD graphene as well. The field hasn't moved far in 15 years, as I observe
Why does lower band gap mean higher power consumption?
we already have chips with graphene, already in mass production. They use way less power, almost 1000 times less. video is inaccurate as of 2 years ago.
@@mqb3gofjzkko7nzx38 because of leak currents. With short BG there are more electrons to break away i.e. leak
@@gg-gn3re I would be delighted if you pointed me in the direction of mass-produced graphene chips. Thanks!
@@Gaste11o GEIC in England does all the research they have multiple buildings dedicated to it. The company & website paragraf has them for sale and bulk order. Hall effect sensors and 1 or 2 others. The 2 inch wafers are on some documentary videos about their campus/company, much better than this video. paragraf follows the industry they are there to succeed and make profit so it hopefully won't cease to exist like most lab stuff does.
I have been a many year subscriber of asianometry and this is probably his worst video.
One possible configuration of microtransistor is a vacuum lamp. At that scale it doesn't needs neither vacuum nor heating, but the operating voltages are high (close to 12V).
This channel is so good I can't believe it exists
Oooooph! I've been waiting for a this research to bare fruits.
Great video as usual, keep it up.
I heard graphene was about to change everything. That was a long time ago.
Maybe you should stop listening to msm sensationalist rage-baiting urinalists and look at actual scientific advancements directly. Urinalists don't care if they lie, they juat care about views. Besides, Graphene is already being used in batteries, thermal systems, composites, and other things.
Yeah, it turns out going from "the physics should work" to "we know how to make economically viable factories for this stuff" takes about a century.
@@bengoodwin2141 already 2 inch wafers in mass production so only 2 decades
Is similar with aluminum and plastic, they still didnt found a way to make them cheaper.
What I like about this channel is the voice-over - it is either well researched human or the AI is way beyond anything that AI is capable of IME. The almost fluff, the (geddit?) not to mention the pun it carried. Can AI be that subtle and still carry genuine information? Yea, Yea one day, and I shudder. AND "Macarthyist accusations" is a phrase that doesn't appear on the web, so explain it applying to graphene deposition - someone.
Graphene transistors also generate less heat and require less W to power/work with for same performance.
We,re talking about trully efficient chips, going further that can easily work with just passive cooling and last for weeks.
No a transistor that does not turn off is not efficient at all.
The current chip industry is so cost effective and optimized that large leaps in technology such as advanced materials or extreme-large-scale particle etching machines can only be commericalized in markets where western chips are banned from export.
Professor James Tour at Rice University is developing a means of mass-producing graphene from any organic material, including waste plastic (flash graphene). Please look at his RUclips channel for more information. He also has some very thoughtful commentary on religion and Christianity.
"2D materials" sounds like some futuristic shit from a 90s scifi anime.
It’s referred to as 2d because graphene is a sheet that is 1 carbon atom thick. This is as thin as it gets before you start breaking down atoms
This seems like it’s not too dissimilar to GAN transistors, which conduct with an atom-thick “electron gas”. It’s such a good conductor that this atom-thick layer is enough for lots of current.
graphene, graphyne, borophene, germanene, silicene, stanene, plumbene, phosphorene,antimonene, bismuthene, goldene, etc. we have found a lot
imagine its paper just thinner
@@megalonoobiacinc4863 it's only 2D so you can stack it infinitely high
Great Video! Relates to CubicWonder - QuadStep 3D Geometry spot on.
Graphene transistors, flying cars, fusion, ... where can I go looooooooong on their stock?
even if doesnt make into ultra fast (GHz) computer chips due to the mobility and complexity, it can be used in power electronics for renewable and motor drives, these circuits hovers in the KHz band and normally are monolitic structures
My favorite part about graphene is how it's speculated that it can do all these amazing things, and yet none of these applications seem to have actually been performed
Did my PhD in a related field and worked with 2D and 1D materials myself, including making transistors in the clean room. I can only agree with all the assessments.
Hi Im currently studying computer engineering and I want to specialize in chip design, what would you recommend for me to get in closer contact with that field?
I think even with these new materials, the practical limit for production processes for integrated circuits using ASML/TSMC process is maybe around 1 nanometer. We may be finally finally reaching the limit of Moore's Law. This is where stacked chips may start to become important.
Loved it when you threw in the McCarthy reference!
I'm still waiting on those SUPER solid state batteries and nuclear fusion! 🤷♂
Only have to wait an other 20 years ? 🙂
So you need a new power source for your lawnmower?😁
@@crazyedo9979 Why not?
@@autohmae Yeah...
@@Mr.SharkTooth-zc8rm Im only asking. I build a volcano in my backyard after my lawnmower broke down the third time.😁
I noticed a bunch of Dutch names and they were actually Dutch guys, not just people with Dutch names in for example the US. Some were even at Philips, where ASML comes from.
Can’t wait till I can hear how a Graphene Transistor sounds when it’s distorted.
I think it might turn out that the graphene will actually never leave the lab. However, things might end up in such a way, that solutions and techniques for creating graphene, invented during development, will be used in the creation of some other metamaterials. So basically, Carbon in graphene, would be exchanged with some other atoms, but the development/manufacturing techniques (not necessarily the ones in this video) will be the same
Little to no band gap would mean it could function as a diode with very low forward voltage, but since the band gaps don't overlap, it would still function as a diode? That's amazing for high powered diodes, like in solar panel arrays and such.
I live in Atlanta and have taken a tour of Georgia Tech's clean room before, but I never knew they were on the bleeding edge of graphene research 😮
I've seen dozens of these videos and papers over the last 5 years. Nothing changed, we still use the same materials
at 10:46 "Using McCarthyest accusations" woke me up
I’ve said for years, Carbon is a semi-conductor, thus can substitute Silicon.
Edge effects kill the mobility in nanoscale devices.
You have one of the cleanest and least pretentious content development and delivery styles on the interwebs. I think I could listen to you describe moonlight raindrops on a puddle in a vacant alley. In what must be close to 100 of your videos I've not heard a verbal crutch from "um" or "ah" to those seemingly ubiquitous toss off phrases like "what's up guys?" or "[insert thirty seconds of drivel]...these bad boys...". Well done kiddo, I don't know what makes you do it but I hope those clouds of inspiration remain in your sky.
Super convinient timing great vid👍
Perhaps those chips shall finally manifest using graphene quantum dots or other structures doped with some special element. Maybe the key to endowing it with a band gap is structuring it in a special way!
Love the quantum realm graphic 7:18 is that a scroll in it's left hand?
I’ve built high power, high efficiency class E radio transmitters using SiC fets (silicon carbide) because they have low gate capacitances and allow for higher frequencies for switching. I wonder if graphene gates would permit high power fets designed for PWM applications to be used in the hobby world for higher frequency switching radio amp designs. It all depends upon gate capacitance at those frequencies, I guess.
I belong to the research group that published the paper displayed at the ending of the video
OK you got me adding an ad where I have to go off and do research is allowed.
I watched a news report when I was a kid about 40 years ago about how Graphene was going to change the world in just a few short years.....
I am still waiting for my personal jetpack.
I just want a part from Mouser that enables new products to be designed.
ahhh graphene , the answer to tall our problems since 05 , eventually i hope it will make it into a product .
there are already 2 inch wafers in mass production and several chips being sold. So your hopes have been answered. There are products.
Smoothly including an Arthur Miller reference: a man after my heart.
Thanks for the graphene facts Squilliam!
This is excellent content! Thank you
A Very Great Presented video 👍👍👍😍😍😍 thank you very much.
Next please do a video on surface plasmon polariton and plasmonic transistor 😃
Been studying this nano material for some time now great summery for chip integration!
Ah yes, Graphene breakthroughs. Like Fusion, they’re just 15 years away.
The joke at 10:56 cracked me up👌
Excuse my simple approach (I have done some semiconductor physics in my studies BTW) but transistors rely on SEMICONDUCTOR action (through doping). In a FET, the channel's conductivity is modulated from the relatively high intrinsic resistance of the pure(ish) semiconductor (usually silicon) to a lower level which results from doping.
Graphene however is a CONDUCTOR. It has no ability to have its conductivity modulated !
Whatever these experimental transistors may be, there are NOT 'graphene transistors' since graphene is ALWAYS a CONDUCTOR. It can have NO transistor action by definition.
Researchers here are making rather far-fetched claims. I suspect it's the silicon carbide doing the work here, not the graphene !
Because it is a 2D material, no consumer device will ever be viable if it will rely on notably large continuous sheets of graphene, as thermal expansion will constantly crack the pieces apart.
Good to see The Gap Band get some more recognition.
Sounds like it's ready for any application today!
God dammit, it’s like trying to make a blue led!
You should not use God's name in vain.
I wonder if some kind of bacteria or cellular automata could be used to create graphene. Imagine a computer seed you feed raw materials, and it grows the computer.
Funny enough, we have already had to deal with increased power consumption to get faster switching speeds, ask anyone that has bought into the last few generations of Intel processors, when a CPU itself is drawing over 250 watts to get work done, that is a problem. That said, I have been perfectly happy with my insanely fast for what it is AMD Ryzen 5 5600X, it brought down power consumption vs my outgoing AMD Rysen 5 1600X by 30 watts and yet I got a HUGE increase in performance. It would be awesome to see graphene become part of CPU tech, because I can't imagine what it would be like to have the performance I have now, with an even lower power draw.
Tianjin China and Georgia USA? Immediately you see the problem.
The ability of Graphene to be produced from coffee grinds is one of its peculiarities. (microwave process).
Although, it doesn't taste very nice to drink …
Dude! I have a friend who works on that! Vertically, conduction behaves like a FET.
Sound to me like the scotch tape deserved the Nobel prize more than the 2 scientists
i gather graphene local interconnect would be even more important. Gate all around makes sillicon transistors far more scalable than what any current local interconnect tecnology allows, making that the bottleneck
I expect photonic/optical transistors to be "the next big thing" rather than gfets.
YEARS AND YEARS ago, like almost 30 years ago, I realised that I could use a lone drawn on a piece of paper with a sketching pencil as a resistor, or at least that's what my multimeter showed.
It was taught in the 1988 radio shack book.
Scotch tape is a brand, you don't need to use that particular brand.