Quantum Maxwell's Demon Paradox: Trick? Or Treat?
Вставка
- Опубліковано 30 жов 2022
- Maxwell's Demon has haunted the foundations of physics for over 150 years - and recently it has been messing with quantum technologies. In our latest Paradoxes video, Oxford PhD researcher Maria Violaris explains the classical Maxwell's Demon thought experiment, and the microscopic version known as Szilard's engine. Then she tests out what happens when the demon is quantum, showing you how to code your own quantum Maxwell's Demon as a quantum circuit in Qiskit!
Find out more on the Qiskit blog: / quantum-maxwells-demon...
All the code is in this Jupyter Notebook on GitHub: github.com/maria-violaris/qua...
The code in this video uses Noise Models, covered in this One Minute Qiskit: • How can I make a noise...
This One Minute Qiskit will also be helpful:
• How can I create a cus...
The coding part of this video assumes familiarity with the basics of quantum computing and Qiskit, covered in the first two chapters of the Qiskit textbook and the Appendix on Linear Algebra: qiskit.org/textbook/preface.html.
#qiskit #quantumcomputing #learnquantum - Наука та технологія
Very very well explained. Moreover, what is this minimum entropy cost?
Thank you! The minimum entropy cost of erasing 1 bit of information is ln2, where ln is the natural logarithm (log to base e). In the classical case, this is equal to the Shannon entropy of a random coin; see the definition of Shannon entropy here: www.quantiki.org/wiki/classical-information)
And in the quantum case, this is equal to the von Neumann entropy of a maximally mixed qubit; see the definition of von Neumann entropy here: www.quantiki.org/wiki/von-neumann-entropy#:~:text=and%20if%20%CE%BBi%20are,of%201%20for%20a%20qubit.
The definitions of these entropies are also at the end of the Jupyter Notebook: github.com/maria-violaris/quantum-paradoxes/blob/main/quantum-maxwells-demon.ipynb.
Hope that helps!
@@qiskit Thankyou
Spooky
Nice video! Bigger font and more zoom next time please!
In the sub-atomic scale, the deamon wouldn't be able to measure speed AND position of the particle at the same time. Heisenberg forbids this.
So, there's no way to distinguish slow and fast particles, or left and right side of the box. This experiment makes lots of assumptions.
According Mr. Gibbs, the entropy even increases, if you put water into a cup filled with water. Gibbs-Paradox.
So, we also have to clearify the problem with indistinguishable particles. In other words, can we label these particles?
For me, this rises more questions than answers.
The rabbit hole you are pointing to is fascinating and has been the subject of quite a bit of research. Keep sitting with those questions! You might want to read about quantum feedback control (e.g. based on continuous measurement, in which the observer takes on a role analogous to Maxwell's demon), and/or look into any of the existing proposals for quantum engines driven by measurements; there are many in the recent literature beyond Szilard!
@Maria How do we know that whether it is a slow particle or fast moving particle if we are using a single particle?
In the single-particle version of the thought experiment, instead of using information about whether the particle is slow-moving or fast-moving to decrease the entropy of the box, we use a different piece of information: whether the particle is on the Left of the box, or on the Right. (We could do a similar thing with many particles, and use information about their positions to get all the particles on one side of the box and create a pressure difference instead of a temperature difference. The conclusions are the same!) Hope that makes sense.
@maria_violaris thanks for the reply.
Isn't what Maxwell's demon does just the same as evaporative cooling?
rad
💀
These papers which are independently claiming reversing entropy by mixing Raoult's law with osmosis principle and Extended Gibbs Donnan Equilibrium .
Title of the papers:
Experimental Demonstration of Energy Harvesting by Maxwell's Demon Device
DOI: 10.20944/preprints202403.1698.v1
....
An Autonomous Mechanical Maxwell's Demon
DOI: 10.14293/S2199-1006.1.SOR-.PP5S6NK.v1