The proximity to the physical limit of silicon reminds us that this equation tells us how far we can go

If everything works out As expected by IBM and TSMCIn 2023, large-scale semiconductor production will begin 2 nm photolithography. This integration technology is actually around the corner. And it probably won’t take much longer for other large semiconductor manufacturers to introduce this photolithographic process into some of their plants.

When we talk about semiconductor production, it is imperative that we all think of TSMC, Samsung, Intel and GlobalFoundries. At the end of the day, these are the companies that account for a significant portion of the world’s chip production. However, they all use technology developed by a lesser known European company, which, without a doubt, He has a lot to say In this market: ASML.

As we told you in the article link hereThis Dutch company is owned by Philips, and designs and manufactures photolithography equipment Used by most semiconductor manufacturers in their facilities. In fact, the four companies you mentioned in the previous paragraph are your customers.

The proximity of 2nm lithography inevitably places us One step closer to the physical limit imposed by silicon technology, so we are at an ideal time to review the equation that reflects what parameters you minification condition of the elements that make up an integrated circuit.

Explanation of the Rayleigh criterion

The equation we are looking at It is the bible of ASML. I am not exaggerating at all. In fact, it’s something the researchers at this company recognize without hesitation. That is why it is worth reviewing for the sole purpose of knowing, as I mentioned a few lines above, the parameters that determine the development of an integration technology. This is the equation known as Riley Standard:

It soon seems like a complicated formula, but it’s actually not very complicated if we know what each term in the equation represents. I suggest we review them one by one from left to right. The first of them, “CD”, comes from the English expression monetary dimensiondetermine to what extent It can be reduced Components that make up an integrated circuit.

As we can guess, this is the parameter that semiconductor manufacturers want to reduce at all costs. In fact, all of them, and especially ASML, devote a huge amount of resources to developing technologies that allow this Critical dimension refinementwhich invites us to look at the expression on the right-hand side of the mathematical equality.

Semiconductor manufacturers dedicate huge amounts of resources to developing technologies to improve critical dimension (“CD”)

The “k₁” factor is a parameter determined by the physical parameters that define the semiconductor manufacturing process. What we are interested in keeping in mind is that the physical limit imposed by silicon photolithography is the maximum ‘k₁ = 0.25’Therefore, as we can guess, manufacturers are doing everything possible to improve their technology and bring this coefficient as close as possible to this limit value.

The next parameter, defined by the Greek letter lambda (‘λ’), tells us what . is The wavelength of light Used in the semiconductor manufacturing process. One of the most important challenges faced by the companies we are talking about, specifically, is to reduce the wavelength of light in order to increase the accuracy of the photolithography process.

However, every step forward in this path requires the development of new lithographic equipment, new light sources (usually ultraviolet light is used), new optical elements, new photoresist materials, and also a new fabrication process. In short, every time a plant reduces the wavelength of the light it projects on its flakes, it is forced to change more of its equipment and manufacturing process.

The last element in the recipe that we are interested in looking at is the parameter “NA” (numerical aperture), which determines Aperture value for optics It is used by the lithographic team. In this context, this parameter basically reflects the same opening value when we talk about it photo camera opticsso it determines the amount of light determined by the optical elements able to collect. As we can guess, the more light they collect, the better.

In conclusion, the conclusion we can reach after analyzing the information provided by the Rayleigh criterion is that in order to increase the accuracy of the lithographic process, semiconductor producers have to Refine the three parameters which coexist in the expression on the right side of the equation.

When TSMC, Intel, Samsung, or GlobalFoundries, among other semiconductor manufacturers, announce that they have a new integration technology ready, what they tell us between the lines is that they’ve been able to reduce the wavelength of light they use in their lithography processes, refining their elements photolithography to increase their light-gathering ability, and may also have been able to get a little closer to the physical limit imposed by silicon photolithography. At that ‘k₁ = 0.25’. No more, no less.

Cover Photo: ASML