3 ms·

Layman questions: - What are the potential benefits of spintronics to technology? Would it make electronics cheaper, smaller, more efficient, ...? Would it a

Layman questions: - What are the potential benefits of spintronics to technology? Would it make electronics cheaper, smaller, more efficient, ...? Would it allow something that we currently couldn't do? - "For example, whether placing a semiconductor in contact with another material would impede spin transport across the interface is far from well-understood." ...This seems like an eminently testable thing- just take a semiconductor, put it in contact with something else, put a, uh, spin current (layman!) through them and check for impedance. What hilariously ignorant mistake am I making? : )

tagrun 10y ago
> Would it make electronics cheaper, smaller, more efficient As the name implies, it uses transport of spin rather than electrons themselves. The electrons may not move at all (you can have a spin current with no electric current at all). So it'd no longer be electronics. The current premise is more dense memory devices with lower power consumption. > This seems like an eminently testable thing- just take a semiconductor, put it in contact with something else, put a, uh, spin current (layman!) through them and check for impedance. What hilariously ignorant mistake am I making? Well, where do I start? Think about these: how do you generate a spin current? How much is it? How do you measure it? (hint: no, there is no such thing as spinmeter) How do you store/switch/manipulate bits? How does it effect magnetization, temperature, electric current, chemical potential? (these are all intricately coupled things) What kind of spin waves is it going to generate? Can you treat them as quasi particles? (magnons) Will there be a flow of magnons? What are the transport properties and how does an interface affect it (it's not easy to isolate and measure only a single transport coefficient in practice)? What is your order parameter? Are we talking about a ferromagnet, antiferromagnet or ferrimagnet? Are they interfaced? What are the transport properties of your interface? What is your Landau free energy (from which you can calculate effective field or ground state)? Which phase of matter are you in? Helical phase? Skyrmion crystal phase? Canted/polarized phase? What is your geometry? What kind of stray field does it generate? What are your anisotropies? Does your system have a strong spin-orbit copuling? What is your overall spin transfer torque? How do all these depend of temperature / thermal fluctuations? etc etc etc. And keep in mind that almost all the physical observables we're talking about are really really really tiny. We're talking about quantum mechanical effects. I hope by now it is clear that you have to come up with an "impedance" on your own tailored to your system, taking all the relevant physical aspects of your system into account (which requires a good background in condensed matter physics). There is no generic "impedance" of "spinmeter" that will work in any situation. And generally, the issue is not measuring spin current or transport properties in an existing system. And you certain don't want to make devices randomly and measure its properties until something nice comes out miraculously: you have to "engineer" the desired properties (which typically requires very specific conditions), from ground up. So it's a little different from connecting a lamp and a resistor to a battery and see whether it lights up or not. Suffice it to say that it's more complicated than your average high school science experiments.
- agumonkey 10y ago
  So you'd get AC "differential behavior" where DC is the norm ?
- maxander 10y ago
  Your paragraph after "where do I start" is so incomprehensible that I'm not entirely sure you're not simply spouting random technobabble at me. But I get the idea. :D
s_p_lee 10y ago
Here's an introductory lecture on some aspects of spintronics from 2013 by Stuart Parkin of IBM: https://www.youtube.com/watch?v=cID4fKraWkE https://www.youtube.com/watch?v=cID4fKraWkE One application that he discusses is IBM's use of spintronics in the read heads of hard drives (i.e., spinning rust), which allowed them to increase sensitivity and to increase the storage density of the disks. Also, spin transfer torque (https://en.wikipedia.org/wiki/Spin-transfer_torque https://en.wikipedia.org/wiki/Spin-transfer_torque) is one technique for implementing magnetoresistive memory (MRAM) ( https://en.wikipedia.org/wiki/Magnetoresistive_random-access_memory https://en.wikipedia.org/wiki/Magnetoresistive_random-access...), which some people find exciting for its promise of DRAM-like read/write/erase speeds with data that persists after powered off.

tagrun 10y ago

> Would it make electronics cheaper, smaller, more efficient As the name implies, it uses transport of spin rather than electrons themselves. The electrons may not move at all (you can have a spin current with no electric current at all). So it'd no longer be electronics. The current premise is more dense memory devices with lower power consumption. > This seems like an eminently testable thing- just take a semiconductor, put it in contact with something else, put a, uh, spin current (layman!) through them and check for impedance. What hilariously ignorant mistake am I making? Well, where do I start? Think about these: how do you generate a spin current? How much is it? How do you measure it? (hint: no, there is no such thing as spinmeter) How do you store/switch/manipulate bits? How does it effect magnetization, temperature, electric current, chemical potential? (these are all intricately coupled things) What kind of spin waves is it going to generate? Can you treat them as quasi particles? (magnons) Will there be a flow of magnons? What are the transport properties and how does an interface affect it (it's not easy to isolate and measure only a single transport coefficient in practice)? What is your order parameter? Are we talking about a ferromagnet, antiferromagnet or ferrimagnet? Are they interfaced? What are the transport properties of your interface? What is your Landau free energy (from which you can calculate effective field or ground state)? Which phase of matter are you in? Helical phase? Skyrmion crystal phase? Canted/polarized phase? What is your geometry? What kind of stray field does it generate? What are your anisotropies? Does your system have a strong spin-orbit copuling? What is your overall spin transfer torque? How do all these depend of temperature / thermal fluctuations? etc etc etc. And keep in mind that almost all the physical observables we're talking about are really really really tiny. We're talking about quantum mechanical effects. I hope by now it is clear that you have to come up with an "impedance" on your own tailored to your system, taking all the relevant physical aspects of your system into account (which requires a good background in condensed matter physics). There is no generic "impedance" of "spinmeter" that will work in any situation. And generally, the issue is not measuring spin current or transport properties in an existing system. And you certain don't want to make devices randomly and measure its properties until something nice comes out miraculously: you have to "engineer" the desired properties (which typically requires very specific conditions), from ground up. So it's a little different from connecting a lamp and a resistor to a battery and see whether it lights up or not. Suffice it to say that it's more complicated than your average high school science experiments.

agumonkey 10y ago
So you'd get AC "differential behavior" where DC is the norm ?
maxander 10y ago
Your paragraph after "where do I start" is so incomprehensible that I'm not entirely sure you're not simply spouting random technobabble at me. But I get the idea. :D

s_p_lee 10y ago

Here's an introductory lecture on some aspects of spintronics from 2013 by Stuart Parkin of IBM: https://www.youtube.com/watch?v=cID4fKraWkE https://www.youtube.com/watch?v=cID4fKraWkE One application that he discusses is IBM's use of spintronics in the read heads of hard drives (i.e., spinning rust), which allowed them to increase sensitivity and to increase the storage density of the disks. Also, spin transfer torque (https://en.wikipedia.org/wiki/Spin-transfer_torque https://en.wikipedia.org/wiki/Spin-transfer_torque) is one technique for implementing magnetoresistive memory (MRAM) ( https://en.wikipedia.org/wiki/Magnetoresistive_random-access_memory https://en.wikipedia.org/wiki/Magnetoresistive_random-access...), which some people find exciting for its promise of DRAM-like read/write/erase speeds with data that persists after powered off.