Why Are Chips So Small? Building and Refining the Circuit City

Why Are Chips So Small? Building and Refining the Circuit City
Source: Institute of Physics, Chinese Academy of Sciences
Original Author: Xiao Xian
This article introduces the structure of chips from the perspectives of manufacturing and technological limitations.
Chips are hidden in the electronic devices that can be seen everywhere in the city. Smartphones, computers, and home appliances all rely on its control.
Why Are Chips So Small? Building and Refining the Circuit City
Image source: pixabay
The tiny chip integrates a large scale of circuits.
When magnified, it can be seen that there are densely arranged circuits inside, resembling a tightly woven highway, as if a well-organized circuit city has been built within an extremely small size.
Why Are Chips So Small? Building and Refining the Circuit City
Chip structure diagram (naked eye and microscopic) | Image source: pixabay
How small is the chip inside? The smallest process of chips we currently use in industry, which is the smallest size humans can create, has reached 3nm, allowing for the integration of hundreds of billions of transistors inside the chip.
Why Are Chips So Small? Building and Refining the Circuit City

The “multi-layer” approach to chip manufacturing

Are countless nano-level electronic components arranged on the chip pre-made and then placed one by one?
Why Are Chips So Small? Building and Refining the Circuit City
Image source: pixabay (above); Searchmedia – Wikimedia Commons (below)
No! We can look at this question from a different angle. Upon careful observation vertically, we can find that chips are made up of layers of sheet structures with different patterns stacked vertically. If we prepare each layer in advance and then stack them vertically, the two-dimensional structures can combine to form three-dimensional devices, ultimately creating functional chips.
Why Are Chips So Small? Building and Refining the Circuit City
Vertical observation of the chip’s internal structure | Image source: Searchmedia – Wikimedia Commons
Now our goal has become how to create sheet structures with specific patterns. First, we need a sheet material that can be used to print circuit diagrams, which is the silicon wafer. This is a highly pure silicon, processed and cut into smooth, extremely thin discs.
Why Are Chips So Small? Building and Refining the Circuit City
Silicon wafer | Image source: pixabay (left); Searchmedia – Wikimedia Commons (right)
Next, we, like carpenters, need to find suitable tools to carve the patterns. To create chips with complex and extremely small internal structures, the size of the processing tools is extremely demanding.
Why Are Chips So Small? Building and Refining the Circuit City
Smartly, we found light as our carving knife. Because light has rich wavelengths, we can use short wavelengths to achieve extremely fine processing.
Why Are Chips So Small? Building and Refining the Circuit City
The rich wavelengths of visible light (invisible light wavelengths are even richer) | Image source: Searchmedia – Wikimedia Commons
We hope to transfer the designed circuit patterns onto the silicon wafer using optical exposure, but light cannot affect silicon materials. Therefore, we need an intermediate material that can interact directly with light, which is photoresist.
Why Are Chips So Small? Building and Refining the Circuit City
Photoresist spun on silicon wafers (evenly covered by centrifugal force) | Image source: Searchmedia – Wikimedia Commons
To allow light to transmit the pattern information, we can create a pattern of light and dark by completely blocking or allowing light to pass through. Light passes through a mask with circuit patterns (mask plate), which can replicate the pattern information of the mask. Finally, after interacting with the uniformly covered photoresist on the silicon wafer, the required pattern information appears on the silicon wafer.
Why Are Chips So Small? Building and Refining the Circuit City
Photoresist imaging exposure process | Image source: Searchmedia – Wikimedia Commons
Photoresist is the main carrier medium for photoresist imaging, divided into positive and negative resists. The area exposed to light is more easily dissolved in developer solution for positive photoresist, while the area less likely to dissolve in developer solution is negative photoresist.
Why Are Chips So Small? Building and Refining the Circuit City
Two results of the exposure process (positive and negative resists) | Image source: Searchmedia – Wikimedia Commons
Assuming we use positive photoresist, after the exposure process, the developer solution can dissolve the exposed photoresist. Then, we use chemical substances to dissolve the exposed silicon wafer, and the remaining photoresist on the silicon wafer can protect it. This is the etching process.
Why Are Chips So Small? Building and Refining the Circuit City
Now we have achieved our goal, obtaining a silicon wafer with specific circuit patterns. Throughout this process, the general idea is relatively smooth, but the precision engineering of chip manufacturing, representing the pinnacle of human wisdom, involves countless stringent requirements.

What limits the internal size of chips?

The main component of chips is transistors. A large chip can have hundreds of billions of transistors. The smaller we can manufacture the transistors, the more components the chip can hold, and the power consumption of the transistors will also be lower.
Why Are Chips So Small? Building and Refining the Circuit City
In chip manufacturing, we hope to use light to create circuit patterns on a small scale. So why can light achieve this effect? What are the limits of light’s sculpting?

Diffraction

The main reason affecting the level of light sculpting is the diffraction effect of light. Light is an electromagnetic wave, and diffraction is inevitable during the propagation of light in photolithography, leading to a minimum feature size in the exposure range. The resolution of light, which is the ability of photoresist to reconstruct patterns based on light irradiation, has its limits.
Why Are Chips So Small? Building and Refining the Circuit City
Diffraction during the exposure process | Image source: Searchmedia – Wikimedia Commons
As shown in the figure, when a beam of parallel light passes through a narrow slit, light will interfere with countless sub-waves during propagation, forming a pattern of alternating light and dark.
Why Are Chips So Small? Building and Refining the Circuit City
Single-slit diffraction | Image source: Searchmedia – Wikimedia Commons
This means that when considering the propagation of light at a small scale, the areas of light and no light are no longer distinct, but rather blurry zones appear. An ideal point of light emitted from an object will deviate from the geometric optical straight line after passing the edge of an obstacle and will no longer form an ideal point image.
Why Are Chips So Small? Building and Refining the Circuit City
This is precisely because when the slit width is comparable to the wavelength of light, the wave effect of light takes center stage. Light can use the wave effect to bypass obstacles and disperse in space, creating the diffraction effect of light, resulting in the exposure range being no longer precise, and the resolution of light has its limits.
Why Are Chips So Small? Building and Refining the Circuit City
Wave effect of light (comparison of straight-line propagation and wave effect) | Image source: Searchmedia – Wikimedia Commons

Resolution

In the field of optical imaging, resolution is the ability to distinguish between two adjacent object points. Ideally, we hope that each object point can produce sharp image points, but due to diffraction, the actual result is light spots of a certain size. If the overlapping degree of two light spots (diffraction patterns) is too great, the image points are difficult to distinguish.
Rayleigh proposed an effective criterion, with the resolution calculation formula as follows:
Why Are Chips So Small? Building and Refining the Circuit City
This resolution expression describes the limit position at which two light spots can just be distinguished—when the maximum position of one light spot coincides with the first zero point of the other light spot. Here, λ represents the wavelength of the illuminating light.
Why Are Chips So Small? Building and Refining the Circuit City
The limit cases of indistinguishable and just distinguishable light spots | Image source: Searchmedia – Wikimedia Commons
NA is the numerical aperture, which describes the light convergence ability of the lens, specifically reflected in the degree of deflection of parallel light after incidence (converging to a focal point), with the calculation expression as follows:
Why Are Chips So Small? Building and Refining the Circuit City
Why Are Chips So Small? Building and Refining the Circuit City
Numerical aperture (n is the refractive index) | Image source: Searchmedia – Wikimedia Commons
Rayleigh criterion is commonly used to evaluate imaging quality, while photolithography systems are imaging in photoresist. Photoresist is a high-contrast imaging medium. Under certain exposure conditions, although optical resolution has reached below the limit of Rayleigh criterion, photoresist can still present good imaging results, achieving processing goals.
The resolution of photolithography is:
Why Are Chips So Small? Building and Refining the Circuit City
Rlitho is the resolvable pattern period of the lithography system; k1 is the process factor.

Photolithography

Photolithography is the most complex, expensive, and critical process in chip manufacturing, usually using projection photolithography systems to project the circuit structure diagram of the mask onto the surface of the silicon wafer.
Optical lenses can converge diffraction light to improve imaging quality. In photolithography technology, to achieve the smallest possible patterns, a projection imaging lens with a reduction ratio is used between the mask and the photoresist.
Why Are Chips So Small? Building and Refining the Circuit City
Projection photolithography system | Image source: network

How to sharpen the light as a carving knife?

We now know that the minimum processing scale of light (resolution) determines how small chips can be. How can we make chips smaller? We need to enhance the resolution ability and refine the functions of the circuit city on the chip.
Based on the three items in the photolithography resolution formula, we have three options to sharpen the light as a carving knife.

Increase the numerical aperture of the photolithography system

The larger the numerical aperture of the projection lens in the photolithography imaging system, the better the resolution capability. The specific operation is to design immersion lithography machines, that is, to fill a high refractive index medium between the last lens of the wafer and the projection lens.

Shorten the wavelength

The wavelength of light in the photolithography process has undergone the development of G line (432nm), I line (365nm), KrF(248nm), and ArF(193nm) deep ultraviolet bands. Currently, extreme ultraviolet lithography (EUV) with a wavelength of 13.5nm has been put into use.

Reduce the process factor

By optimizing photolithography process parameters, we can also improve the photolithography resolution, such as improving illumination conditions, photoresist processes, and mask design. These methods can reduce the process factor k1, known as resolution enhancement technology (RET).
Why Are Chips So Small? Building and Refining the Circuit City
Electromagnetic waves | Image source: network
Light is electromagnetic waves, thus containing information such as amplitude, phase, polarization state, and propagation direction. The resolution enhancement technology of photolithography is to control the above four pieces of information of light to obtain smaller graphic structures on the photoresist. For example, off-axis illumination technology can change amplitude and phase, optical proximity effect correction technology can change the amplitude of light waves, and source-mask co-optimization can change the propagation direction, amplitude, and phase of light waves.
Why Are Chips So Small? Building and Refining the Circuit City
Table of relationships between various process nodes and photolithography technology | Source: Sako Microelectronics official website, ASML, Zhongtai Securities Research Institute
Looking at the development history of photolithography machines, we are indeed running along the path of continuously shortening wavelengths. Observing the data in the table, when the wavelength of the light source remains the same, we are still continuously reducing the process, thanks to the numerical aperture, process factors, and other complex technologies.
References
[1] Wei Yayi. Calculating lithography and layout optimization [M]. 1. Electronics Industry Press, 2021.
[2] Stephen A. Campbell. Micro-nanoscale manufacturing engineering [M]. 3. Electronics Industry Press, 2010.
END
Reprinted content only represents the author’s views
Does not represent the position of the Institute of Semiconductor, Chinese Academy of Sciences
Editor: Schrödinger’s Cat
Editor-in-chief: Mu Xin
Submission Email: [email protected]
Previous Recommendations
1.The Institute of Semiconductor has made progress in the research of bionic covering-style neuron models and learning methods.
2.The Institute of Semiconductor has made significant progress in inverted structure perovskite solar cells.
3. Why is copper used as the interconnect metal in chips?
4.What exactly is the 7nm in chips?
5.Silicon-based integrated optical quantum chip technology.
6.How unusual is the quantum anomalous Hall effect? It may bring about the next information technology revolution!
Why Are Chips So Small? Building and Refining the Circuit City

Leave a Comment