Positron creation in semiconductors?

[FieldIntensity]

I've posted a similar question to the high energy forum.

As I understand it, LEDs emit photons when electrons and holes have sufficient energy to cross a particular "well". I'm sure this explanation is lacking in many key ways. Why can't we "tune" the gap to create electron-positron pairs and shape the crystalline structure to draw electrons to one side and positrons to the other? IOW, can a PED (positron emitting diode) be physically possible?

 

[ZapperZ]

I've posted a similar question to the high energy forum.

As I understand it, LEDs emit photons when electrons and holes have sufficient energy to cross a particular "well". I'm sure this explanation is lacking in many key ways. Why can't we "tune" the gap to create electron-positron pairs and shape the crystalline structure to draw electrons to one side and positrons to the other? IOW, can a PED (positron emitting diode) be physically possible?

Er... what "positron"? And what does producing positron has anything to do with the band gap?

The band gap is the result of the collective interactions of all the ions of the material. The overlap of the local states of each ions with its nearest neighbor, next-nearest neighbor, next-next nearest neighbor, etc.. produces the band structure that we know of for almost all material. You "tune" the gap by varying the contituents of the material.

In any case, that explanation is moot considering that it has nothing to do with producing positrons. Can you please cite a reference where you got the connection between the band gap and production of positrons?

Zz.

P.S. Please re-read the PF Guidelines before you proceed any further. Multiple posting is clearly not allowed.

 

[Astronuc]

As I understand it, LEDs emit photons when electrons and holes have sufficient energy to cross a particular "well". I'm sure this explanation is lacking in many key ways. Why can't we "tune" the gap to create electron-positron pairs and shape the crystalline structure to draw electrons to one side and positrons to the other? IOW, can a PED (positron emitting diode) be physically possible?

Visible light is around 2 - 3 eV, while the rest mass of an electron or positron is 0.511 MeV. A big difference!

Positron-electron pairs are produced when a gamma-ray of at least 1.022 MeV interacts with the nucleus of an atom.

 

[FieldIntensity]

>>

Er... what "positron"? And what does producing positron has anything to do with the band gap?
<<snip>>
Can you please cite a reference where you got the connection between the band gap and production of positrons?

Zz. <<

I am asking if it is possible to make a semiconductor that produces positrons instead of photons by appropriately "tuning the gap". I need no reference since I am not citing previous work or theory, I am ASKING.

IF electrons can cross a gap of the right "size" to spontaneously generate PHOTONS, could a material be engineered that spontaneously generates positron-electron pairs? <see the question mark>

 

[ZapperZ]

>>

Er... what "positron"? And what does producing positron has anything to do with the band gap?
<<snip>>
Can you please cite a reference where you got the connection between the band gap and production of positrons?

Zz. <<

I am asking if it is possible to make a semiconductor that produces positrons instead of photons by appropriately "tuning the gap". I need no reference since I am not citing previous work or theory, I am ASKING.

IF electrons can cross a gap of the right "size" to spontaneously generate PHOTONS, could a material be engineered that spontaneously generates positron-electron pairs? <see the question mark>

No.

So now that the answer is clear, I am STILL curious as to the connection between "band gap" and "positron production". IF you are trying to say that the photon being produced is THEN used to generate electron-positron pair, then Astronuc has already answered that. A material with THAT magnitude of band gap simply doesn't exist and may not even be possible.

Zz.

 

[FieldIntensity]

>>

No.

So now that the answer is clear, I am STILL curious as to the connection between "band gap" and "positron production". IF you are trying to say that the photon being produced is THEN used to generate electron-positron pair, then Astronuc has already answered that. A material with THAT magnitude of band gap simply doesn't exist and may not even be possible.

Zz.

OK, maybe I am not using the right language. I am asking if a material can be manufactured that has the properties such that when sufficient energy is introduced (electricity in the case of semiconductors), instead of producing photons of visible light, you get electron-positron pairs?

I did not mean to imply a connection between "band gap" and "positron production", I am simply asking if there is a connection; even theoretical?

 

[FieldIntensity]

Visible light is around 2 - 3 eV, while the rest mass of an electron or positron is 0.511 MeV. A big difference!

Positron-electron pairs are produced when a gamma-ray of at least 1.022 MeV interacts with the nucleus of an atom.

OK, now we are in the ball park of the question. What would it take to cause spontaneous positron emission (with an electron mate of course)? We currently do it by slamming electrons into targets, but that is rather nasty and ineffcient. Surely there is a configuration of fields of sufficient intensity as to excite the local vacuum into pair emission when energy (a relatively large amount, granted) is pumped into the system?

I used semiconductors as a basis because it is a very elegant use of quantum properties to stimulate real effects. In that case, the excitation of the semiconductor's "local vacuum" into spontaneous photon creation.

If something similar, then what?

 

[ZapperZ]

OK, maybe I am not using the right language. I am asking if a material can be manufactured that has the properties such that when sufficient energy is introduced (electricity in the case of semiconductors), instead of producing photons of visible light, you get electron-positron pairs?

I did not mean to imply a connection between "band gap" and "positron production", I am simply asking if there is a connection; even theoretical?

No, because there's no established theory on the connection between what you are asking. That is why this is a rather puzzling question.

You need to know how electron-positrons are produced in the first place. Only THEN can you figure out of there are other ways to achieve the same thing. In a semiconductor (and in solid state physics in general), the "holes" ARE the "antiparticle" of the electrons! There is a complete analogue of particle-antiparticle creation by light with electron-hole creation in matter. When particle-antiparticle anhilate, light is created. When electron-hole anhilate, light is created.

This is why your replacement of these well-established holes with positrons is rather puzzling. We just don't create positrons this way. There are zero examples of electron-positron pairs being created using "semiconductors". Gamma passing through Tungsten, Beryllium, yes... but semiconductors using E-field? No theoretical foundation exists for that.

Zz.

 

[FieldIntensity]

There are zero examples of electron-positron pairs being created using "semiconductors". Gamma passing through Tungsten, Beryllium, yes... but semiconductors using E-field? No theoretical foundation exists for that.

Zz.

So, what you're saying is that no electric field, in any configuration or interaction, can cause pair production? Even if the field energy or particles in the field is greater than the rest mass of the pair? OK, maybe another way of asking is this: Can we set up the conditions conducive to creating virtual electron-positron pairs, and then, by adding eneough energy, make them real?

I started with semiconductors because they are precisely enegineered configurations of matter that allow certain quantum properties to be taken advantage of for the creation of low energy subatomic particles in the form of photons.

Can an analogous arrangement of matter be engineered to take advantage of certain quantum properties that would be advantageous to the spontaneous (that is, not involving nuclear collision) creation of e-p pairs?

 

[ZapperZ]

So, what you're saying is that no electric field, in any configuration or interaction, can cause pair production? Even if the field energy or particles in the field is greater than the rest mass of the pair? OK, maybe another way of asking is this: Can we set up the conditions conducive to creating virtual electron-positron pairs, and then, by adding eneough energy, make them real?

I started with semiconductors because they are precisely enegineered configurations of matter that allow certain quantum properties to be taken advantage of for the creation of low energy subatomic particles in the form of photons.

Can an analogous arrangement of matter be engineered to take advantage of certain quantum properties that would be advantageous to the spontaneous (that is, not involving nuclear collision) creation of e-p pairs?

Do you know how pair production is created?

You have high energy photons (gamma) passing by a very massive particle (nucleus). Why does it need to pass near a massive particle? Momentum conservation.

Also, "photons" are not subatomic particles.

I would also like to know what "quantum properties" in a semiconductor that you would like to take advantage of to produce such e-p pairs.

Zz.

 

[FieldIntensity]

Do you know how pair production is created?

You have high energy photons (gamma) passing by a very massive particle (nucleus). Why does it need to pass near a massive particle? Momentum conservation.

Also, "photons" are not subatomic particles.

I would also like to know what "quantum properties" in a semiconductor that you would like to take advantage of to produce such e-p pairs.

Zz.

Momentum is conserved when particle tragectories are opposed. Net momentum 0.

And what, pray tell, are photons if not quanta of the electromagnetic field?

 

[Rade]

...Can an analogous arrangement of matter be engineered to take advantage of certain quantum properties that would be advantageous to the spontaneous (that is, not involving nuclear collision) creation of e-p pairs?

Not sure if this is what you are looking for, but positronium (electron + positron) formation has been reported for semiconductors (see:
Dannefaer S., Kerr D., Craigen D. (1996b): J. Appl. Phys. 79, 9110
Itoh Y., Murakami H. (1994): Appl. Phys. A 58, 59)
See also these:
http://positron.physik.uni-halle.de/panet/text/intro/positronium.html
http://www.springerlink.com/content/j3875jw28u516x11/
http://www.springerlink.com/content/g740m38w7t2rlt76/
And this:
FRACTION OF POSITRONIUM FORMATION AT SEMICONDUCTOR SURFACE
S.B. Shrivastava and A. Upadhyay
School of Studies in Physics, Vikram University, Ujjain (M.P.), 456010 India
Received: May 4, 1998; revised version December 8, 1998; in final form March 30, 1999
ACTA Physica Polonica A

The fraction of positronium formation (fps) has been calculated in Ge(110), Ge(111), Si(110) and Si(111) surfaces by solving the diffusion equation for positrons in semiconductors and by setting up the rate equation to describe the processes that are supposed to occur when a thermalised positron encounters the surface including the trapping of positrons in neutral and negative vacancies. Certain parameters used in the evaluation of fps, e.g., the bulk annihilation rate (\lambdas), the positron diffusion length (L+), the diffusion coefficient (D+) and the implantation profile parameter (A), have been taken from the experiments. The calculated values of fps as a function of incident positron energy and temperature in Ge(110) and Si(111) have been compared with the experimental results. It has been found that in general the calculated results are in good agreement with the experimental ones. The calculation also confirms that the trapping rate of positrons into negative vacancy has a T-1/2 dependence with respect to the temperature.
PACS numbers: 78.70.Bj, 71.60.+z, 68.35.Fx

 

[FieldIntensity]

I would also like to know what "quantum properties" in a semiconductor that you would like to take advantage of to produce such e-p pairs.

Zz.

You obviously haven't figured out that that is more or less what I'm asking YOU. I do not suppose to have the answer, I have asked the question. Please quit beating about the head neck and chest as if I have made some radical presumption. I have made no statement alleging fact.

I used semiconductors, listen carefully now, as a starting point because they are familiar instances of engineered materials being used transform energy in the form of electricity into another kind in the form of electromagnetic radiation, WITHOUT heating the stuff to incandecent temps. Emission of light from an LED is purely a quantum mechanical process made possible by the specific arrangement of matter in the diode.

OK, maybe a semiconductor as we make them now cannot make anything other than electromagnetic radiation. Fine. IS THERE A WAY TO CONFIGURE A MATERIAL'S ATOMS AND MOLECULES TO COAX e-p PAIRS INTO EXISTENCE WHEN THE RIGHT AMOUNT OF ENEGY IS ADDED? Can we make a device that does for e-p pairs what a LED does for photons? Don't call it a semiconductor, it that isn't what it is.

 

[jtbell]

Not sure if this is what you are looking for, but positronium (electron + positron) formation has been reported for semiconductors [...]

These experiments let positrons from an external source strike a semiconductor, where they meet up with electrons to produce positronium. Definitely not what FieldIntensity is looking for.

 

[ZapperZ]

You obviously haven't figured out that that is more or less what I'm asking YOU. I do not suppose to have the answer, I have asked the question. Please quit beating about the head neck and chest as if I have made some radical presumption. I have made no statement alleging fact.

I used semiconductors, listen carefully now, as a starting point because they are familiar instances of engineered materials being used transform energy in the form of electricity into another kind in the form of electromagnetic radiation, WITHOUT heating the stuff to incandecent temps. Emission of light from an LED is purely a quantum mechanical process made possible by the specific arrangement of matter in the diode.

OK, maybe a semiconductor as we make them now cannot make anything other than electromagnetic radiation. Fine. IS THERE A WAY TO CONFIGURE A MATERIAL'S ATOMS AND MOLECULES TO COAX e-p PAIRS INTO EXISTENCE WHEN THE RIGHT AMOUNT OF ENEGY IS ADDED? Can we make a device that does for e-p pairs what a LED does for photons? Don't call it a semiconductor, it that isn't what it is.

But I have ALREADY given you an direct answer : NO!

Somehow, you continue to follow this line even after being given that answer. That is why I continue to INQUIRE on the reasoning of that question, because obviously, you wouldn't take "NO" for an answer. I'm not the one "beating" on it. You are!

You also have to remember that an irrational question will result in an irrational answer. To me, your question is irrational because you are using phrases such as "quantum properties" without bothering to define what properties you are invoking. You are using various things thrown into the soup pot hoping something might turn out. It makes it VERY confusing, and that is why I continue to ask for CLARIFICATION. Since you do not accept a simple answer, I then seek the origin of such a question in order to figure out the logic in such a scenario.

Again, you have ignored my explanation on why we already have "matter-antimatter" creation in a matterial as far as the analogous description is concerned, and that is the electron-hole pair. I do not see ANY mechanism for the creation of electron-positron pair in a semiconductor other than the STANDARD technique of passing gamma through a material.

That is my final answer. Take it or leave it.

Zz.

 

[FieldIntensity]

But I have ALREADY given you an direct answer : NO!


Again, you have ignored my explanation on why we already have "matter-antimatter" creation in a matterial as far as the analogous description is concerned, and that is the electron-hole pair. I do not see ANY mechanism for the creation of electron-positron pair in a semiconductor other than the STANDARD technique of passing gamma through a material.

That is my final answer. Take it or leave it.

Zz.

You keep focusing on semiconductors. I'm sorry that I used that term to begin with. The STANDARD technique is not the only way to create e-p pairs. The "quantum properties" I am referring to are the more general field dynamics brought about inside a material by a specific arrangement of matter.

When enough energy is present, and the conditions are right, e-p pairs can be created from the vacuum. You are acting as if the ONLY way to create them is from slamming a gamma photon into a nucleus. Can matter be made to excite the vacuum into a state favorable to e-p pair production? And if not, why not? It is energetically possible, but can it be physically possible?

 

[ZapperZ]

You keep focusing on semiconductors. I'm sorry that I used that term to begin with.

I'm a condensed matter physicist. A "semiconductor" is something that is right in my backyard and so, I assumed that you know what it is when you use that term, especially when you invoked the band gap. This explains why your description simply just makes no sense.

The STANDARD technique is not the only way to create e-p pairs.

Can you point out where other techniques have been used to generate e-p pairs?

The "quantum properties" I am referring to are the more general field dynamics brought about inside a material by a specific arrangement of matter.

I have no clue on what you just said here. What is a "general field dynamics"? If I give you a crystal lattice structure (i.e. what you referred to, I think, as "specific arrangement of matter"), then can you point out to me an example of what you mean by such field dynamics?

When enough energy is present, and the conditions are right, e-p pairs can be created from the vacuum. You are acting as if the ONLY way to create them is from slamming a gamma photon into a nucleus.

Again, I'll ask you to maybe look at CERN's LEP, SLAC, etc.. that all produced their e-p pair using the "standard" way. Can you point out to me an example of e-p production using your way? I can point to different technique of producing just positron via nuclear decay, such as that used in PET scans, or via bombardment of high energy particles to create unstable nuclei. But e-p pair?

Or are you confusing virtual particles and zero-point fluctuation as the formation of e-p pair?

This is getting confusing.

Zz.

 

[FieldIntensity]

OK, I've snipped the entire thing to make a general statement. First, thank you Zapper for all your responses to my question, and also thank you for pushing me to clarify the language of my question. That helps me understand what I am asking a little better, I think.

Second, on semiconductors and band gaps: I started with this because it is a very familiar instance of a material that is engineered to a precision that allows the direct control of the quantum level interaction of electrons and holes. In LEDs, we can manipulate the material to generate photons from electron-hole annihilation.

Using this as a jump off point, I envisioned that the energy of electrons "falling" into the "well" of a semiconductor might be converted (if enough energy is present) into e-p pairs instead of photons. Now I understand that semiconductors just can't do that.

Third, on standard positron creation: Energetically speaking, if enough energy is present in a system, and the right conditions exist (more on that in a moment), any particle-antiparticle pair can be generated. The usual way is to slam electrons into a dense metal target, creating e-p pairs which are then separated by magnetic fields into individual beams of positrons and electrons. The reason why this is the standard way is because it is just brute force and easy, though horribly ineffcient.

Other energetic systems should be able to do this more elegantly and efficiently, if enough energy is present. The reason why we don't do it now is because we lack the precision to accurately manipulate matter on the scale necessary to create an environment conducive to pair creation.

Physicists know that the quantum vacuum is affected in different ways by different arrangements of matter and fields. Virtual particles are constantly popping in and out of existence even in empty space. The presence of a single electron causes the vacuum to boil with fleeting e-p pairs. Sometimes virtual particles become real if a sufficient amount of energy is added to account for the rest masses of the particles; 1.022 MeV for an e-p pair.

We should be able, given enough precision, to create a vacuum condition in which pairs spontaneously appear when energy is pumped into the system. With the right setup, electrons will go in on direction and positrons in the other.

I guess it boils down to precision. Will we ever achieve that level of manipulation?

 

[Astronuc]

We should be able, given enough precision, to create a vacuum condition in which pairs spontaneously appear when energy is pumped into the system. With the right setup, electrons will go in on direction and positrons in the other.

A vacuum is simply the absence of matter in a particular volume. There will be no spontaneous generation of e-p pairs in a vacuum, even if there are 1.022 MeV photons passing through.

In matter, photons scatter off electrons, whereas in nuclei they can create e-p pairs, or photons could eject neutrons from the nucleus - so-called photo-neutrons, but the energy has to exceed some threshold like 1.6 MeV or so, i.e. the binding energy of a neutron.

In the case of LED's, the band widths are low energy - on the order of eV. If the band energy was on the order of 100's of keV, this is a huge energy and would require huge pressures to hold the matter together.

Some perspective - the binding energy of the K-electrons in U is 115.606 keV -
ref: http://xdb.lbl.gov/Section1/Table_1-1c.htm - which is way short of 511 keV, the rest energy of an electron. The band electrons in LED or any semiconductor are the 'outermost' electrons which have very low energies (~ eV).

 

[FieldIntensity]

A vacuum is simply the absence of matter in a particular volume. There will be no spontaneous generation of e-p pairs in a vacuum, even if there are 1.022 MeV photons passing through.

That is the classical view of the vacuum. Quantum mechanics puts a whole new spin on it. The quantum vacuum is very dynamic and fluid.

 

[ZapperZ]

That is the classical view of the vacuum. Quantum mechanics puts a whole new spin on it. The quantum vacuum is very dynamic and fluid.

Not for REAL particles, it isn't!

Again, if you look at the end of my previous message, I believe I got it right in terms of guessing where you were heading. You somehow promoted virtual particles into having the same footing as the real particles that are created out of pair production (note that you somehow missed the "conservation of momentum" part that is needed in pair production). So my guess being that you are using vacuum fluctuation as your mechanism for generating e-p pair appears to be right on the money.

Since that is the case, then I would ask you to actually figure out why we call these things "virtual particles", and how they differ from "real particles" that we deal with often.

Zz.

 

[FieldIntensity]

Not for REAL particles, it isn't!

Again, if you look at the end of my previous message, I believe I got it right in terms of guessing where you were heading. You somehow promoted virtual particles into having the same footing as the real particles that are created out of pair production (note that you somehow missed the "conservation of momentum" part that is needed in pair production).

Is not the net momentum of two particles moving in opposite directions zero? Doesn't this satisfy the conservation of momentum?

As for virtual particles, are there not instances where they can become real? For example, when an electric field is strong enough, one half of a virtual pair may fall toward the field thereby granting real mass to the other half that escapes. I will research this, but I thought this was a well established principle? That is confering mass to virtual particles?

[/QUOTE]Since that is the case, then I would ask you to actually figure out why we call these things "virtual particles", and how they differ from "real particles" that we deal with often.

Zz.[/QUOTE]

I understand the physics as it relates to nature. Here, we are talking about mediating forces between real fermions with virtual bosons.

Just as not all force carriers are virtual, it holds that not all virtual particles are force carriers. The difference between a virtual fermion and a real fermion is energy.

 

[dpak]

Not for REAL particles, it isn't!

Again, if you look at the end of my previous message, I believe I got it right in terms of guessing where you were heading. You somehow promoted virtual particles into having the same footing as the real particles that are created out of pair production (note that you somehow missed the "conservation of momentum" part that is needed in pair production). So my guess being that you are using vacuum fluctuation as your mechanism for generating e-p pair appears to be right on the money.

Since that is the case, then I would ask you to actually figure out why we call these things "virtual particles", and how they differ from "real particles" that we deal with often.

Zz.

I found this thread in an internet search, so I'm sorry to come back to it years later. But why DO we call them "virtual particles"? I thought it was because their very existence is too short-lived and they mutually annihilate before their existence can be measured. But is this true of all of them or only most of them? In other words is it possible to devise an experiment that will detect a small fraction of the total number of virtual e-p pairs that are believed to be produced in a system?

Re the conservation of momentum issue, could that come from the extremely small (10^-9 joules per cubic meter) net positive energy density of the vacuum that also accounts for the accelerated expansion of the universe?

Excuse my ignorance and don't jump down my throat. I'm here to learn.

 

[jtbell]

Virtual particles are not real in the same sense that "real" particles are. They are mathematical tools, artifacts of the perturbative methods that we use for calculating the approximate probabilities of processes in quantum electrodynamics.

This has been discussed here many times, for example in this thread:

https://www.physicsforums.com/showthread.php?t=302923

which as you can see quickly degenerates into basically an argument about what it means for something to be "real."