Charge Trap Flash vs Floating Gate

Both charge trap flash and floating gate are non-volatile memory technologies used in flash storage devices. Basically, these two technologies indicate how the memory cell inside the flash memory stores the data in the form of charge. The charge trap is a more advanced method with better capabilities of data retention. The floating gate technology is

The basic difference between these storage technologies is how a single storage cell holds the charge. The way of keeping the charge at its plays an important role in the data retention. So, there is no serious advantage in terms of performance with any of these. However, charge trap flash is better at retaining the charge and less prone to leakage.

Use-cases of CTF and FG

Floating Gate (FG) Flash use cases

The major application of the FG Flash is in the 2D NAND flash and NOR Flash storage. Traditional Solid State Drives use the floating gate transistors as the storage medium. You will find most of the low-end and even some mid-range SSDs using the floating gates. Because the FG technology is cost-effective and is the dominant technology in the SSD market, it consumes most of the SSD market to date. Along with SSDs, smartphones, USB drives, and flash drives, use the FG Flash for permanent data storage purposes. Embedded systems and firmware storage using NOR flash also employ floating gates as their storage mediums.

Charge Trap Flash (CTF) use cases

CTF is used mainly when the storage density is critical. It is primarily employed in the 3D NAND technology in both consumer and enterprise storage devices. Also, it is used in data centers, cloud storage, AI processing, and mostly in every place where endurance is critical. In most high-end and flagship smartphones and tablets, you will find the CTF memory used as the primary storage. Also, in emerging technologies like AI and IoT, and also in automotive storage, the CTF is utilized. Although CTF has found its application in many areas, it is still found in high-end devices whether they are smartphones or NVMe SSDs. There is a lot of room for scalability and we may find this flash in cheap prices as well.

Difference in Working

The way the CTF and FG store the data is a little different. The difference is how a single memory cell holds the charge. CTF is better in terms of data retention as compared to the FG. This is because of the way the CTF stores the charge. Let’s discuss how.

Floating Gate Transistor Working

The floating gate transistor is an advanced version of the normal BJT or Transistor. A simple transistor can be used as a switch or an amplifier provided we are setting it up in a proper circuit. The normal transistor would allow the current to flow through its two terminals (collector to emitter) when we provide some voltage to its gate. It works like in the image.

The floating gate transistor comes with three extra layers between the gate and the main body. One layer is the conductive floating gate and two extra oxide layers cover the floating gate. The oxide layer works as an insulator to hold any charge inside the conductive floating gate. A floating gate is called floating because it is a conductor isolated from the main circuitry and is floating depending upon the control gate and transistor channel.

How the charge is pushed inside the floating gate is the process of writing the data in the FG-based memory. Depending on the number of bits stored inside a cell, a specific amount of charge is pushed by the controller in your SSDs or other storage device. Once the charge is pushed, the oxide layer would keep this charge from leaking out. In this way, the data is kept in its place even without the need for power.

But, over time, the floating gate leads the charge because it is stored in a conductive layer which is prone to leakage. The oxide layer around the floating gate degrades over time and this will result in much more leakage. All in all, the floating gate has no inherited properties to keep the charge at its place on its own. It just depends on the oxide layer for holding the charge.

Charge Trap Flash Working

In the charge trap flash, everything remains the same but the floating gate which was a conductor is replaced by a non-conductive or doped polycrystalline silicone. Now, we can’t call this a floating gate because the charge isn’t stored in a single continuous layer. The term-charge trap is used because the charge is now trapped in different locations inside a dielectric material (not floating in the conductive layer). The charge is now stored in isolated pockets in a non-conductive material, it is called a “trap” instead of a “float”.

This simple difference in how the charge is held at its place plays an important role in the overall reliability of the stored data. The total P/E cycles in the CTF is higher than the FG. Also, the error rates are less because the chances of data leakage are lower in the CTF.

Because the charge is stored in discrete and isolated trapping locations on a non-conductive material, the chances of leakages are very low. Even though the oxide layer is present here as well, the dielectric trap layer has the properties to keep the charge in place.

2D NAND vs 3D NAND

It isn’t that the 3D NAND only uses the charge trap flash. Although the current 3D NAND just uses the charge trap, the early deployments were done with the Floating gate NAND flash. I believe it is important to know the difference between the 2D NAND and 3D NAND in order to understand this concept better.

2D NAND is a simple deployment of the floating gate transistors on a single plane. Although 2D NAND is well-established because of being a mature technology, it has some drawbacks. Mainly, there are storage density limitations. Also, the scalability is tougher because the cell size can’t be reduced too much. That is why, companies started to invest in the 3D NAND and now the 4D NAND is also under development.

The manufacturers have tried to reduce the cell size to smaller process nodes (below 20nm). But, shrinking cells caused problems like cell-to-cell interference, charge retention issues, and reduced endurance. However, 2D NAND is still found in older or more cost-sensitive storage devices such as basic USB drives, SD cards, and some low-end SSDs.

In 3D NAND, memory cells are stacked vertically in multiple layers (often referred to as vertical stacking). This allows more storage capacity without shrinking the size of each cell. This makes the layers into a tower-like structure. 3D NAND offers better scalability because manufacturers can add more layers to increase density instead of shrinking the cells.

With the help of 3D NAND Flash, we get higher capacity, lower cost per gigabyte, better endurance, and improved performance over 2D NAND. 3D NAND can have 100+ layers. However, 3D NAND is complex and costly to manufacture, though advancements are continually improving yields and reducing costs.

3D NAND flash is found in modern SSDs, smartphones, high-capacity USB drives, and enterprise storage. Samsung has its marketing term for the 3D NAND i.e. V-NAND (Vertical NAND).

Do all modern SSDs and storage devices use CTF?

No, charge trap flash is used only in high-end storage devices like the modern NVMe SSDs, expensive smartphones, etc. Brands like Samsung, Crucial, WD, etc use the CTF in their 3D NAND products. Traditional 2D NAND is less common in the latest designs, but you can still find it in some drives mostly the slower and older ones. The SATA drives like Crucial BX500, and MX500 still use the planar floating gate technology for data storage.

But, it is indeed hard to find the modern drives coming up with the 2D Floating gate storage. Most of them will have multiple layers of NAND flash and will be using CTF instead of the floating gate.

Data Retention difference between CTF and FG

Is the floating gate still relevant?

The adoption of the CTF is huge. Charge Trap Flash (CTF) technology is increasingly used for its advantages in scalability and reliability. 3D NAND does not use a floating gate but a different charge-trapping mechanism. FG-based SSDs and other storage drives are easy and cost-effective to manufacture. But, in the upcoming days, I believe the trends of floating gate flash will go down and down. The CTF and 3D NAND and perhaps the 4D NAND would rule the markets.

I hope this helps!