Intel’s 10 nm fabrication process was initially planned to enter mass production in the second half of 2016 but it has still not reached that stage and is barely used by the company. Currently, Intel uses its 10 nm process to manufacture just a few of its CPUs. The 10 nm process is expected to reach high-volume manufacturing (HVM) in the later stages of 2019.
This fabrication process was delayed by several years and it obviously affected the company’s product lineup and its business. Now, we have come to know that Intel is right on track with the development of its 7 nm technology and is expected to roll it out according to its original schedule.
Intel has admitted multiple times that it was too aggressive with its scaling/transistor density targets and deadlines for the 10 nm process and that is why they ran into problems during the developmental process. Intel’s 10 nm manufacturing technology uses Deep Ultraviolet Lithography (DUVL) with lasers having a wavelength of 193 nm. To achieve the fine feature sizes Intel planned for the 10 nm, a lot of multi-patterning was required and according to Intel, this was the exact problem in the process.
Now, with its 7 nm fabrication technology, Intel will be using Extreme Ultraviolet Lithography (EUVL) with lasers having a wavelength of 13.5 nm for select layers. This will reduce the use of multi-patterning for certain metal layers and will simplify the production while decreasing the cycle times. The company says that the 7 nm process was being developed separately from the 10 nm technology by a different team. Because of this, its development is right on track and is expected to enter the HVM stage in accordance with Intel’s unannounced roadmap.
Intel’s Chief Engineering Officer and President of Technology, System architecture and Client group, Murthy Renduchintala, said at the NASDAQ’s 39th Investor Conference, “7 nm for us is a separate team and largely a separate effort. We are quite pleased with our progress on 7 nm. In fact, very pleased with our progress on 7 nm. I think that we have taken a lot of lessons out of the 10 nm experience as we defined that and defined a different optimization point between transistor density, power and performance, and schedule predictability. […] So, we are very, very focused on getting 7 nm out according to our original internal plans.”
Renduchintala revealed Intel’s plans to start HVM production of client CPUs on its 10 nm process node in 2019, with data center products following on a bit later. Intel is not going to skip any of it’s already announced 10 nm products, but we can expect the 7 nm products to hit the market earlier than we might expect today. (i.e., four years after the 10 nm).
“One thing I will say is that as you look at 7 nm, for us this is really now a point in time where we will get EUV back into the manufacturing matrix, and therefore, I think, that will give us a degree of back to the traditional Moore’s Law cadence that we were really talking about,”
“[With 7 nm] we are going back to more like a 2X scaling factor […] and then really moving forward with that goal.”
Intel has not yet revealed any specifications of its 7 nm process, but a reduction in the usage of multi-patterning usage along with a conventional 2X scaling goal compared to 10 nm process mean a more substantial usage of EUVL.
According to ASML, each EUV layer needs one EUV step-and-scan system for every 45,000 wafer starts per month. This means that if Intel is planning to use EUVL for 10 to 20 layers, it will be needing around 20 to 40 EUVL scanner for a fab with a 100,000 wafer starts per month capacity. Since Intel is not the only company planning to use EUVL in the 2020s, procuring the required number of EUVL scanners for HVM at multiple fabs might be tough.
So far, Intel has only revealed its plans for just one 7 nm fab: the Fab 42 in Arizona. Along with this, it will be having some 7 nm-capable features at its D1 facility used for the development and trials.