At its core silicon is a great picks and shovels play to get exposure to multiple mega trends, especially Quantum Computing. Mega trends often seem more fiction than fact to the casual observer, but as part of the article, I will cover concrete examples where this technology is likely applied during the next couple of years. After giving an intro to silicon and Quantum Computing, I revisit Silex Systems (ASX:SLX) and explain my investment thesis from a point of view quite different to that of my last Silex article. Moreover, I am uncovering potential most investors have not realized exists and is likely bigger than the laser uranium enrichment business unit I covered last time.
Structure
When I wrote my first article about Silex Systems in March 2022 my focus was on uranium enrichment, I saw the Zero-Spin Silicon (ZS-Si) technology as "optionality which de-risks the investment case" only. However, I wasn't wrong to focus on laser uranium enrichment, as it turned out to be a very timely topic, but based on the news flow that followed my article, I was wrong to classify ZS-Si purely as optionality. ZS-Si is expected to be in commercial production in 2023 and has successfully completed all stages of the joint pilot project with partner Silicon Quantum Computing (SQC) so far. The mistake I (and most of the other Silex investors/followers) made is to underestimate SQC and the offtake agreement Silex has with this company. Not many are aware of the fact that SQC is one of the few companies that was able to raise a significant amount of money during the market turmoil of the last 12 months. On June 14th 2022 SQC announced the launch of its A$130M Series A capital raising to fund the company’s technical development, operations and strategic activities from 2023 to 2028.
This Series A round followed SQC’s successful A$83 million seed capital raising in 2017.
At a time where even startups that pursue more traditional business ideas have trouble raising any money, raising A$130M is significant and a tell that SQC is not just a science project. Another tell is the fact that SQC announced the successful fabrication of the world’s first integrated circuit manufactured at the atomic scale, two years ahead of schedule on June 23rd 2022.
Based on the collaboration between Silex Systems and SQC, Australia is on track to become leaders in (Silicon) Quantum Computing and the production of highly enriched silicon. This insight connected many dots for me and also made me look into another company again, which leverages silicon for innovative technologies: Twist Bioscience.
What I realized by connecting the dots is that at its core silicon is a picks and shovels play to get exposure to mega trends like Quantum Computing, synthetic DNA & DNA data storage. I am of the impression that this is a puzzle not many investors (especially not the broader stock market) have yet put together and thus I am happy to share my thoughts with you.
Let's dive into the topic and why there are currently exciting developments involving silicon that offer progress in Quantum Computing, Synthetic DNA & DNA data storage.
What excites me the most as a retail investor, is that there is a way to invest in these themes via publicly traded companies. The opportunity this investment theme offers is akin to riding a time machine to the 1969s and be an early stage investor in companies involved in classical microchips.
Today I will focus on Quantum Computing and the company Silex Systems (ASX:SLX). At a later time, I might also write an article about synthetic DNA, DNA data storage and the company Twist Bioscience.
What is silicon?
Even though we use the term "Silicon Valley" as everyday language nowadays, we rarely think about the origins of the name and that it hints at the major role silicon has been playing in technology breakthroughs. The origins of the Silicon Valley we know today reach back to 1956 when William Shockley, the co-inventor of the first working transistor (with John Bardeen and Walter Houser Brattain), moved from New Jersey to Mountain View, California, to start Shockley Semiconductor Laboratory. In contrast to leading researchers of that time who favored Germanium as semiconductor material, Shockley focused on silicon as the material for making transistors. It turned out to be a challenging endeavor to make silicon work as semiconductor material and in 1957 Shockley decided to end research on the silicon transistor. Two of the original employees of Fairchild Semiconductor, Robert Noyce and Gordon Moore, left the company and founded Intel.
This kick started the Silicon Valley we know today, home to a large number of innovators and manufacturers, specializing in silicon-based transistors and integrated circuit chips.
Today, silicon still is the leading element that is the basis for and at the center of modern microchips. In my article about CyberOptics, I explained where silicon comes from and how it is transformed to wafers that are used for microchips. Sand is the simple raw material that is the basis for all (micro)chips, as sand is primarily made up of silicon dioxide. In order to produce an extremely pure mono crystalline silicon ingot called a blank (only one impurity atom for every 10 million silicon atoms), complex chemical and physical processes are required to convert silica sand to silicon. Silicon blanks are fabricated in a range of different diameters the most common sizes are 150, 200 and 300 millimeters wafers. The extremely thin wafers (one millimeter thick) are then cut from the silicon blanks using a special sawing technique. In microelectronics, photovoltaics and microsystems technology these wafers are usually used as a substrate (base plate) for electronic components, including integrated circuits (IC, "chip"). Silicon is a semiconductor. This means it can conduct electricity and also act as an insulator. These wafers are the basic building blocks for subsequent microchip production.
From the 60s to today the global semiconductor industry has reached massive scale and the ability to inexpensively manufacture high-precision silicon computer chips.
"By finding a way to create quantum computing processors in silicon, you can take advantage of all of the years of development, knowledge and infrastructure used to manufacture conventional computers, rather than creating a whole new industry for quantum manufacturing”.
When I first heard about Quantum Computing a couple of years ago, my initial thoughts were "this is a topic that is more science fiction than reality". However, through my research regarding Silex Systems (ASX:SLX), I also learnt more about Quantum Computing and that it is actually closer to being commercially relevant than most people expect. Let's go down the rabbit hole and explore what Quantum Computing is and what practical applications it will offer very soon. In turn, these applications offer the value propositions that make me bullish on holding SLX. Silex Systems aims to produce the raw material that is needed to produce the microchips needed for Quantum Computers.
Quantum Computing
“Quantum Physics makes me so happy. It’s like looking at the universe naked” Sheldon, The Big Bang Theory
Source: Qashi YouTube
Quantum physics is a branch of physics that deals with physical phenomena at a very small length scale, being very strong at the nano to atomic length scales. Quantum Computing is the development of computer technology based on the principles of quantum mechanics & quantum physics.
In this insightful 2018 lecture Prof Andrea Morello from the UNSW Sydney explains the concepts of Quantum physics. He explains the differences between Quantum- and regular physics and gives practical examples of how Quantum Computing differs from classical computing.
Source: SibosTV YouTube
This video is definitely worth a watch for everybody who wants to better understand Quantum Computing. I will not go into every detail of the topic, but mention the talking points that are relevant for investors on a high level.
In Andrea's opinion, silicon transistors, which are the building blocks of classical computers, are the most amazing achievement of humankind, even more impressive than flying to the moon. The reason he is so impressed by transistors is that in your smartphone there are a billion man-made objects, on a size scale of a few hundreds of atoms. On average, only a hundred transistors out of a billion transistors on a chip are broken. However, the manufacturing of transistors is possible at mass scale, meaning that they can be bought for only a few dollars in a store. Quantum Computers (QC) can be built with technology that is very similar to that used in classical computers. With advances in technology, things that his professor told him 25 years ago cannot be done are now achievable. This is not because Quantum theory has changed, but rather the technology available to bring Quantum theory into practice has changed.
Digital Computer vs Quantum Computer
The below images shows an Intel 286 which is a 16-bit microprocessor that was introduced in 1982.
Source: SibosTV YouTube
Classical computers are digital computers. In order to do a calculation like "three times five" by using 4 bits, the number three would be represented by 0011 and five would be 0101. The computer will calculate that 3 times 5 equals 15, which is 1 1 1 1 in digital code.
When dealing with quantum objects we are entering the world of qubits instead of bits, as quantum objects have two natural states to encode quantum information. Take two protons and one electron as an example:
Electron on the Left = zero
Electron is on the right = 1
It can also be 0 and 1 at the same time (superposition), that's a legitimate code on this quantum bit
Source: SibosTV YouTube
To fully understand the power of quantum computing, we have to use examples with multiple bits. With three classical bits you can write eight numbers (see image below). You have 8 choices but all the information in this example still is 3 bits of information.
Source: SibosTV YouTube
Looking at the same example but using quantum bits: 8 basis states, but because we are dealing with quantum objects, we are allowed to make superpositions of all these 8 states. If I want to describe what the quantum state of those 3 qubits is, I need eight complex numbers, where 8 is 2 to the power 3.
In general, if you have n quantum bits you need 2 to the power n numbers to describe the full quantum state.
Why do I need 8 here and only 3 in the classical example? When doing this kind
of superpositions, most of the states you get are entangled States. You get a lot of those situations where the bits exist only in relation to each other and not in their own individuality. This is something you cannot do on a classical computer, but it is a perfectly legitimate digital coordinate in a QC.
Source: SibosTV YouTube
If I had 300 quantum bits and they are all perfectly fully entangled, to describe their quantum state, I would need as many numbers as there are particles in the universe.
Source: SibosTV YouTube
Quantum supremacy
Source: SibosTV YouTube
When dealing with quantum computing the numbers can get big pretty fast. It also begs the question at what level the amount of data that can be processed by QC exceeds the capabilities of what could ever be achieved by using a classical computer?
Quantum supremacy starts at 70 qubits. At that level you would need about 10 to the power 21 numbers. This would be 1 zettabyte of data and equal to the amount of digital data that is in the world at the moment.
If you could make a QC with 70 qubits that are all fully talking to each other, entangled with each other and you could run a calculation on this computer that uses the entire 2 to the power 70 information content of the computer, you could legitimately say that you've done something that cannot under any circumstance be done on a classical computer.
Practical applications of Quantum Computers
Mass media coverage of QC so far has not done a great job of bringing across the potential for real life applications of these super computers. The two biggest misconceptions (I also held) are that QC are "just" good for doing complex calculations and that GC are decades away from being relevant to our everyday life. Many readers of this article will be surprised by a.) how far along QC already are and b.) what practical applications QC will be used for.
QC are good at solving problems where the computational complexity explodes exponentially. A basic example of such a problem would be a combinatoric problem, e.g. solving matrices. To simulate a molecule like Beryllium hydride you would have to solve a 40,000 X 40,000 matrix for example. This could still be done with a classical computer, but is a very heavy calculation. In 2017 IBM successfully calculated this molecule on a purpose-built 7 qubit quantum processor. Since 2017 there has been rapid development in QC. Recently IBM announced "IBM Osprey" which has the largest qubit count of any IBM quantum processor currently. IBM Osprey is a 433 qubit processor with the stated goal of reaching 4,000+ qubits by 2025.
Source: SibosTV YouTube
The value of simulating molecules that one can find in nature, can be best explained by another concrete example. The Haber-Bosch process combines nitrogen & hydrogen to form ammonia in industrial quantities for the production of fertilizer. The challenge currently is, that the production of ammonia consumes about two percent of the world's energy. In addition, the process is inefficient and requires very high temperatures. Moreover, this problem is worsening, as we are in an energy crisis and energy prices are rising rapidly worldwide.
QC can offer a solution to this problem. There are bacteria (Cyanobacteria) in nature that spontaneously produce ammonia without having to heat up to 600 degrees. The chemical reaction that goes on inside of Cyanobacteria is a series of quantum evolutions and we just can't make sense of them on a classical computer.
Source: SibosTV YouTube
As the global population is growing and raising the demand for fertilizers, figuring out how Cyanobacteria produces ammonia would have a positive impact on the environment and help to decrease energy consumption. "The global ammonia market garnered an approximate revenue figure of USD 74 billion in the year 2021 and is estimated to grow at a CAGR of ~8% over the forecast period, i.e., 2022 – 2031". The company solving this problem would also disrupt the multi billion USD ammonia market.
One company that has actually come closer to bringing real life, scalable QC applications to the world is Silicon Quantum Computing, a private company from Australia which is collaborating with UNSW & Prof Morello.
Currently research groups and companies around the world are working on multiple qubit technologies, with the major ones being superconducting qubits, photonic qubits and quantum dot qubits.
IBM's approach leverages a super conducting qubit, they already offer high 3 digit qubits, but beyond 10,000 qubits will likely run into engineering challenges. In addition, super conducting quantum computers need to operate at very low temperatures in large, sophisticated coolers. This is the case because the temperature they have to run at is near absolute zero (0 Kelvin), which is equal to −273,150 °C. Similar to the first classical computers, their whole system would fill a number of rooms, making it not really applicable to real world usage.
Silicon quantum computing, also referred to as quantum dot qubits, has multiple benefits over superconducting quantum computers. First, quantum dot qubits can be run in environments that are 15 times warmer (1.5 Kelvin) than those the superconducting qubit technology has to run in. Second, quantum dot qubit technology can run in tandem with conventional chips (conventional chips would run at a temperature of 4.0 Kelvin). Third, the silicon semiconductor supply chain is very mature and trillions of dollars have been invested to build it up. Apart from small tweaks that would have to be done, a foundry for classical chips could be used to build quantum dot chips, offering faster time to market than building a supply chain for the other qubit technologies from scratch.
Even though the quantum dot qubit approach has been perceived as the laggard in the QC landscape, it is an approach that offers the possibility to scale better than superconducting qubits.
Silicon Quantum Computing (SQC) and their breakthrough
“if you want to understand how nature works, you must be able to control matter at the same length scales from which matter is constructed” Prof R. Feynman in 1959
This inspired Michelle Simmons (Director of SQC) and her team to build an integrated circuit using atomic components in silicon. Silicon Quantum Computing (SQC) is Australia’s first Quantum Computing company pioneering a globally unique atom-based manufacturing technology, to build a commercial-scale Quantum Computer.
The company has a bold vision and three main objectives:
10 Qubit Prototype: Demonstrating the capability required to reliably produce a ten (10) qubit prototype quantum integrated processor by 2023.
100 Qubit Quantum Processor: Delivering a programmable device based on a one hundred (100) qubit quantum processor embodying error correction before 2030.
Universal Quantum Computer: Enabling access to useful Quantum Computing solutions for a broad audience of users and multiple uses by the mid-2030s.
In June 2022 SQC announced the world’s first quantum integrated circuit manufactured at the atomic scale. The fabrication of the world’s first single atom transistor has been achieved 2 years ahead of schedule. This was a real breakthrough, as they not only fabricated the quantum integrated circuit, but also successfully modeled the quantum states of a small, organic polyacetylene molecule. Polyacetylene usually refers to an organic polymer. The simulation of molecules by QC that Prof Morello described as a potential application in 2018 is now a reality in 2022. SQC has proven the validity of the company’s technology for modelling quantum systems.
"This ability to simulate materials at the atomic level will revolutionize the way people look at solving problems. Soon it will be able to simulate materials that already exist or to simulate new materials that have never existed before. This will allow SQC and our customers to construct quantum models for a range of new materials, whether they be pharmaceuticals, materials for batteries, or catalysts.” Michelle Simmons
This video explains the breakthrough in more detail.
Source: Vimeo
Once this technology is developed further, it will have widespread implications in many industries. Note that the big deal here is not only replacing processes like Haber-Bosch, but also to develop new materials that have never existed before. It isn't even impossible right now to come up with a figure for the total addressable market of this technology, as this technology will disrupt many industries and build new ones. It is a shame that SQC is a private company and with the exception of venture capital, it would be very tough to invest in SQC currently. However, retail investors have the option to indirectly invest in SQC via their partner Silex Systems (ASX:SLX).
Enriched Silicon: Silex Systems' connection with SQC
In order to fabricate their silicon QC processor chips, SQC has a need for enriched silicon.
Here Silex Systems comes into play. Silex is an Australian company that developed a laser isotope separation technology which can be used to enrich silicon to the level needed for QC. Silex is listed on the ASX under the ticker SLX.
The project to develop a process for the commercial production of high-purity ‘Zero-Spin Silicon’ (ZS-Si) was launched in 2019. "Natural silicon (Si) consists of 3 isotopes: 92.2% Si-28, 3.1% Si-30 (each with zero electron spin state) and 4.7% Si-29 (with a spin state of ½). The presence of Si-29 in concentrations above 500 parts per million (ppm) (0.05%) prevents effective QC performance, so ZS-Si must be produced by elimination of the Si-29 isotope. The lower the concentration of Si-29, the better a silicon quantum processor will perform in terms of computational power, accuracy and reliability."
Supply side
When assessing the attractiveness to invest in a certain commodity or raw material, I usually put more emphasis on the supply side, as demand holds more uncertainty in terms of forecasting accuracy. Simply put, is is easier to evaluate, if a material is in under supply and if/how that deficit can be fixed. Enriched silicon currently is produced using outdated gas centrifuge technology and the produced quantities are as low as a few kilograms per year. In addition, most of the supply comes from Russia, making supply even more constrained due to geopolitical tensions.
The concentration level of Si-29 mentioned above also plays a key role in the supply/demand equation. ZS-Si's purity target is 99.995% or higher and SLX is aiming for commercial production of ZS-Si already in 2023, being first to market with such a high purity product. Details regarding the off take prices agreed on with SQC are unknown. However, as SLX is the only producer outside of Russia, offering this high purity product, the price for this material will be at a premium.
Demand side
The start of commercial production of ZS-Si will involve producing enriched silicon for SQC. However, SLX has already stated in their filings, that as the ZS-Si project progresses, SLX "will engage with other potential customers, possibly including someglobal computer chip manufacturerswho are also developing silicon quantum computing technology."
There are already some silicon QC companies I could find online, more will likely follow in the coming years. They all need enriched silicon and SLX would be the likely producer that can sell them the highest purity product out of a country that is seen as an ally by most global nations. Should the ZS-Si project be successful, it could potentially enable Australia to establish itself as a world-leader in enriched silicon production. If the market for ZS-Si evolves, this could create a new value-added export market for Australia
Among the list of silicon QC companies I could find are the following.
Quantum Motion
Equal1 Laboratories
Photonic Inc
Intel
The biggest name involved in silicon QC currently is Intel and they seem to be very bullish on the topic:
"The silicon quantum computer will be transformative for every aspect of society, creating the potential to solve highly complex problems in greatly reduced timeframes and to address problems that are currently beyond the capacity of conventional computers."
Conclusion
My first article about SLX focused on their laser uranium enrichment project. Today's article featured an overview of their silicon enrichment project. In conclusion, SLX is a great picks and shovels play, offering exposure to multiple mega trends. An investor looking to add the mega trend QC in his or her portfolio should do some due diligence on SLX. SLX already has a contract with SQC, but in contrast to a direct investment in SQC, offers diversification. SLX is not dependent on SQC in the future. They can win other customers and thus if SQC doesn't work out in the end, it will have very little impact on SLX.
SLX offers enriched silicon at high purity which they can likely sell at a premium price point. Taking into account the more efficient production method due to laser isotope separation, SLX will likely benefit from higher margins than the competitors that are using centrifuges. This is something the market is overlooking currently. The potential of ZS-Si is not factored into SLX current valuation. Imagine Intel (or another global chip manufacturer) signs an offtake agreement with SLX worth billions of dollars? Intel clearly has a focus on QC and will continue to invest R&D dollars into the supply chain. The current market cap of SLX is just $375M. Progressing towards multiple off-take agreements could easily lead to a multi billion dollar valuation of SLX over the years to come.
Owning SLX right now could be seen like owning Intel right before they launched the Intel 286 in 1982. Investing in INTC shares in 1982 would have delivered a +26,000% return (if one had sold at the peak in 2000).
Source: INTC monthly chart (1982- 2000) tradingview.com
For more information on SLX and my complete investment thesis, please read my article from March 2022.
Please subscribe to my blog and follow me on twitter (@ShovelStocks) to be notified of new content.
Silex Systems is part of my private portfolio and also part of the "Picks and shovels plays" wikifolio, which is traded on the German Stock Exchange. Sign up here for free to learn more about wikifolio:
Sources:
http://sqc.com.au/2022/06/14/130-million-series-a-capital-raising/
https://cosmosmagazine.com/technology/quantum-internet-silicon/
https://en.wikipedia.org/wiki/Polyacetylene#:~:text=Polyacetylene%20has%20no%20commercial%20applications,for%20film%2Dforming%20conductive%20polymers.
https://www.silex.com.au/silex-technology/silex-zs-si-production-for-quantum-computing/
