Slashdot Mirror
Scientists Discover a New Kind of Magnet (ieee.org)
Wave723 shares a report from IEEE Spectrum: A new kind of magnet, theorized for decades, may now have been experimentally proven to exist. And it could eventually lead to better data storage devices. In a normal magnet, the magnetic moments of individual grains align with each other to generate a magnetic field. In contrast, in the new "singlet-based" magnet, magnetic moments are temporary in nature, popping in and out of existence. Although a singlet-based magnet's field is unstable, the fact that such magnets can more easily transition between magnetic and non-magnetic states can make them well-suited for data storage application. Specifically, they could operate more quickly and with less power than conventional devices, says Andrew Wray, a materials physicist at New York University who led the research. Now, Wray and his colleagues have discovered the first example of a singlet-based magnet that is robust -- one made from uranium antimonide (USb2). "It ends up taking very little energy to create spin excitons for uranium antimonide," Wray says. "This is essential for the singlet-based magnet, because if it took a lot of energy, then there wouldn't be enough spin excitons to condense, stabilize one another, and give you a magnet." The research has been published in the journal Nature Communications.
How do they work?
TL;DR - Not a monopole (Sadly)
So our hard drives will be coated with uranium antimonide? That sounds like fun.
We have finally found a place to dump all the nuclear waste! It's going to our next gen data centers!
I see a glowing future ahead of us.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
I can't wait for USB sticks to come in 12" thick lead-lined concrete cases with free shipping from Alibaba.
Let's see the delivery guy try throwing that package across the front yard.
maybe this tech could be used in some thing like bubble memory but at a size, speed, and density that makes it equal or better to existing nonvolatile memory storage tech.
High temperature singlet-based magnetism from Hund’s rule correlations
Not paywalled. Thank you very much, Nature.
When all you have is a hammer, every problem starts to look like a thumb.
It sounds like this is some pretty basic science. If the only material this is somewhat stable in is Uranium Antimonide, we're still a long way from applications. Basically they found a new mechanism by which electrons in this rare material create magnetism. It looks like cool quantum mechanics, but not the path to near term applications.
Uh... no. Current gen under heavy use, when they exhaust the replacement blocks, fail dead.
It's really quite nasty in industrial high performance appliances.
First gen went read only unexpectedly... "locked". I still see this in SD and micro SD. The Raspberry Pi community is just now coming to grips with the issue.
True... If the utility is run. And when 100 percent of the spare storage cells are used, the next to fail means fail dead.
I speak of I have seen. Not marketing theory or wishful thinking.
There's a number of different kinds of magnetic memory systems that don't involve spinning discs. One I saw is just a micro version of the old core memory from way back when. Instead of a bunch of ladies weaving core memory like beaded jewelry the wires and magnets are printed out like an integrated circuit. There's also magneto-resistive RAM. These are not the spinning platters we've known.
I am armed because I am free. I am free because I am armed.
Uranium is very toxic, even if it’s not a radioactive isotope. Not sure I’d like to have that in my house.
-- Cheers!
Yeah, sure. USB-C solved the 'plugs in wrong way' problem.
In return, any given USB-C system may or may not support fast charging. Audio may or may not be supported. Cables are frequently proprietary to specific devices. You say USB-C is faster, but that faster speed is also optional. Low quality USB-C cables abound. Even the name has gone off the rails, with USB 3.2 Gen 2 Revision 5 Update 7 Amendment 9 Subchapter 16. Real consumer friendly there!
Fixing the problem of which way is up is worth a lot of the other problems. I hate USB-A and the difficulty to tell which way is up.
The theory of USB-C is great, but the reality is far less grand. USB is supposed to be Plug and Play, but it has turned into Plug and Pray with USB-C. No thanks, not until they get their sh*t together. And it has been long enough to do that yet it still hasn't happened.
It helps if you don't buy cheap shit cables. So long as they comply with the spec there should be little confusion. Part of complying with the spec is proper labeling on the connectors. The symbols on the cable should show what the cable is capable of doing, which includes both max rated data speed and amperage.
The alternative to these different cable types is requiring all cables to support every feature. That would mean the cables would be very expensive. A USB-C Thunderbolt cable (which supports 40Gbps and 5 amp power) is about $25, even though it's less than a meter long. A 2 meter long USB-C cable that supports 3 amp power and only USB 2.0 speed is about $8. Getting a cable that is both 2 meters long and supports Thunderbolt 3 would cost about $60.
Maybe you can argue about my prices a bit but the general relationship among them will be about right. A fast and short cable will cost about three times a slow and long one. A fast and long cable will be about three times the cost of the fast and short one. A 5 amp rated cable could cost as much as double that of a 3 amp cable. I'll put up with some of this confusion so I can get some quality "charge only" cables (which might really mean they support USB 2.0 for the power delivery negotiation) for only $8, and not have to spend $80 for a cable that does it all.
Oh, and I'm with you on the stupid speed naming conventions. They should have just labeled them by max data rate, such as 1.5M, 12M, 480M, 5G, 10G, and 20G.
I am armed because I am free. I am free because I am armed.
Possible but the same is said for spinning HD too. I've rarely had a spinning HD last more than 5 years. Most of the die after about 2 to 3 years. As for SSD dying, the rate of death of SSD deaths are greatly exaggerated. Even under heavy use a SSD will usually out last the point where its obeisance. One of the best things about SSD is being solid state devices they have a better chance to tell you when they are going to die than a HDD does. I've only had one or two HDD tell me they are going to croak with the SMART report. Most of them just fall over and die.
I read at +2. If your post doesn't reach that level I will not see or respond to it.