The development of DRAM technology (2)
The Beginning of DRAM
In the 1960s, roughly 20 years after the invention of the IBM calculator by means of vacuum tubes, BJT (Bipolar Transistor) was the mainstream of the semiconductor industry, and MOSFETs had just begun to be noticed. Computer development was the top priority plan of most research institutions, and every developed country was competing for computer technology. Then, the memory required by computers was mainly composed of magnetic cores memory (as shown below).
Although the production of magnetic memory was quite industrialized and matured at that time. However, due to the limitations of volume, low density, and poor performance, it turned out the performance of the computer was limited. Thus further improvement and breakthrough in memory technology is required.
It was the situation background when the basic DRAM device concept was proposed by IBM's electronics engineers.
In 1966, IBM’s electrical engineers and the inventors R.H.DENNARD (photo quoted from IBM website) invented Dynamic Random Access Memory(DRAM) Cell – A transistor memory cell that stores every bit of information in an electronic circuit in the form of an electrical charge. He found it was capable of building capacitors, and that storing more charge or less charge in the MOS capacitor could represent the 1 and 0 of a bit, while the MOS transistor could control writing the charge to the capacitor This single transistor MOS DRAM cell technology makes a significant increase in memory density and memory access performance.
In 1967, Dennard and IBM filed the patent for his single transistor Dynamic Random Access Memory (DRAM) USPTO: 3387268, which was granted in 1968.
However, the first whole semiconductor MOS memory was commercialized in 1969 by sunny vale inc. That is a 1K bits DRAM was sold to Honeywell and others. Later Mostek inc. successfully developed and produced 4K and 16K bits before mid-1970. They dominated 75% memory share because of cost and performance advantages. And later MOSTEK and other US manufacturers were overtaken by Japanese manufacturers.
Recommended by LinkedIn
The USPTO: 3,387,286 patent is the one that enables today’s DRAM industry, driving the computers, the Internet, mobile phones, and even the ongoing AI artificial intelligence, Internet of Things, etc. Dennard himself said that when he invented Dynamic Random Access Memory, he didn't foresee it becoming such important: "I knew it was going to be a big thing, but I didn't know it would have a wide impact today."
It was an astonishing 50 years. In 1970, Intel ® revealed the first DRAM chip that the capacity was one-kilo bits. And today in 2022, DRAM single chip capacity is up to 16-billion bits (16Gbs has been mass production). The capacity of DRAM single-chip has increased by 16 million times, which means that the price and cost of single bit has decreased by 16 million times (1/16000000).
The development of the DRAM process perfectly explained the LAW OF MOORES. Leading the technological breakthrough of advanced semiconductors.
The next paragraph will be “Trench capacitor or Stacked capacitor”
It has been a confusing debate between two technologies, “stacked or trench capacitor”, and end up with all enterprises adopting trench capacitor fell down. If you were one of those CTO, would you choose the other side?
Pony, 2022.06.12