Flexible electronics has been done before, but not on this scale. Wearable electronic devices, such as watches and fitness trackers, represent the next logical step in computing. They have sparked interest in developing flexible electronics, which could include wearable devices such as clothes and backpacks. p>
There is a problem with flexible electronics: our processors are not flexible. Most efforts to counter this limitation involve breaking the processors into smaller units, connecting them with flexible wires, and then fusing all the components into a flexible polymer. To some extent, this is a return to the early days of computing, when a floating unit could be deployed on a separate chip.
But a group at ARM has now managed to implement one of the company's smaller embedded designs using flexible silicon. This design uses and implements all the instructions you would expect. But it also shows the compromises we need to make for truly flexible electronics.
Not fully CMOS
The basic idea behind flexible electronics is very simple: start with a flexible layer (such as plastic or paper) and use it as a substrate to create a thin layer using a flexible semiconductor. A variety of semiconductors are suitable, from thin atomic materials to semiconductor polymers. But most options are not mature technologies because they are used to build logic gates, so working with them involves two layers of experimentation - with the materials themselves and their flexibility. Ad
One of the more popular options is amorphous silicon. The silicon used to make existing treatments is crystalline, which means that it forms a regular group of atoms. Silicon is not amorphous, and therefore flexible. Plus, we know how to work with amorphous silicon, because we use it for things like solar panels and LCD displays. It is also inexpensive, in part because it can be converted into transistors with techniques simpler than crystalline silicon. The downside is that amorphous silicon, with a variety of measures including performance, power efficiency, and circuit density, has, as mentioned, many potential applications for flexible electronics that don't require much performance.
Based on the idea of minimal performance, the team at ARM partnered with PragmatIC Semiconductor to implement a version of the company's Cortex M0+ processor called PlasticARM. M0+ is a 32-bit processor that can execute a simple subset of "thumb" ARM instructions. It is optimized for small size and low power consumption and is generally used as an embedded processor.
Even by the standards of a very simple processor, PlasticARM has distinct features that set it apart from the rest. For one thing, the small bits of memory that processors use to store the data they're working on (called "registers") are usually in the processor itself, because the function of going to external RAM to read that memory gets corrupted. . To simplify the PlasticARM processor, the CPU registers are located in a RAM cache - and the system was built using only 128 bytes of RAM.Ads
Systems and applications running on PlasticARM on a 456-byte ROM chip that is also separated from the processing hardware. Currently, it is not possible to update the ROM (read-only), but the team hopes to change it in the next iteration.
All major components - CPU, RAM, ROM and connectors - are made of amorphous silicon and made of flexible polymer. The system also has pins for off-chip connection.
In general, the performance is not good. It has a maximum speed of 29 kHz and consumes about 20 MW at that speed. That might sound like a small amount, but the M0+, which runs on standard silicon, only needs a little over 10 microwatts to reach megahertz. On the plus side, it has over 18,000 separate gates, and it's more flexible than any previous processor. The researchers also managed to run all the programs in the ROM, though the researchers ran the experiment effortlessly with the processor's defining feature—they never dissuaded it.
Yes, the team is planning the next steps. This includes reduced power consumption, which is convenient given the gap between its performance and standard silicon. The researchers also hope that the number of gates will exceed 100,000, although they believe it will eventually rise to less than a million.
Why are they after that? The processor description article sums it up by talking about the potential of the Internet of Things, where things like clothing and food packaging can come up with a flexible processor. This paragraph is vague about what we will achieve, it simply indicates that “without unlocking innovations” “unlocking innovations.”
It has been said that several researchers are working to embed sensors and small power sources into items such as clothing, with the goal of monitoring everything from activity to exposure. Some of these applications require items to manage their behavior and data, and a flexible processor would make sense.
Nature, 2021. DOI: 10.1038/s41586-021-03625-w (about DOI). p>
Researchers are building a flexible ARM processor but they can hardly bend it
It is sometimes difficult to write objectively about the progress of SpaceX. The i...
Last Thursday, the Russian space station's new large unit, Navka, was finally conn...
We are now - often horrifyingly - watching what happens to the virus and ...