ZnO with a wurtzite structure is naturally an n-type semiconductor because of a deviation from stoichiometry due to the presence of intrinsic defects such as O vacancies O(v) and Zn interstitials Zn(i) [1]. Undoped ZnO shows intrinsic n-type conductivity with very high electron densities of about 10^21 cm^−3. Although it is experimentally known that unintentionally doped ZnO is n type, whether the donors are Zn(i )and O(v) is still controversial. The first-principles study suggested that none of the native defects show high concentration shallow donor characteristics. However, Look et al.[2] suggested that Zn(i) rather than O(v) is the dominant native shallow donor in ZnO with an ionization energy of about 30– 50 meV. It has also been suggested that the n-type conductivity of unintentionally doped ZnO films is only due to hydrogen H, which, acts as a shallow donor with an ionization energy about 30 meV. This assumption is valid since hydrogen is always present in all growth methods and can easily diffuse into ZnO in large amounts due to its large mobility. First-principles calculations also suggested that unintentionally incorporated hydrogen acts as a source of conductivity and behaves as a shallow donor in ZnO. n-type doping of ZnO is relatively easy compared to p-type doping. Group-III elements Al, Ga, and In as substitutional elements for Zn and group-VII elements Cl and I as substitutional elements for O can be used as n-type dopants.

Ref:

[1]. Ozgur, U., Alivov, Y. I., Liu, C., et al. A comprehensive review of ZnO materials and devices. Journal of Applied Physics 98, 041301 (2005).

[2]. ;D. C. Look, D. C. Reynolds, J. R. Sizelove, R. L. Jones, C. W. Litton, G. Cantwell, and W. C. Harsch, Solid State Commun. 105, 399(1998).

Dear Gouri S. Tripathi,

Thank you very much for your scientific, exciting and convincing answer.

Dear Labidi Herissi and Gouri S. Tripathi :

This is a very interesting topic.

I support and believe the hypothesis that the H impurities are the cause of the intrinsic n-type conductivity in undoped ZnO material.

Is it true that in general scientific literature, Oxygen vacancies are pointed to be the cause of the n-type character of the ZnO. But the Oxygen vacancies are deep donors in the bands structure of ZnO, they are found deep from the CBM, so they cannot account for the n-type conductivity in ZnO.

This hypothesis has been supported well with the technique called Electron Paramagnetic Resonance (EPR). This technique is similar than NMR (Nuclear Magnetic Resonance), but instead of getting a spectrograph of the resonance frecuency of H1, or C13 (which are atomic nuclei) it does a similar thing but for electrons.

There are some candidates of a Shallow Donor in a naturally grown ZnO n-type, but Hydrogen has been postulated as the more likely candidate. One of the reason is its high Energy Formation, and its low Barrier to Diffusion. And additional supporting logic to this hypothesis is what Gouri S. Tripathi pointed out; Hydrogen is present in practically all grow methods and it has a large difussivity.

As Prof. Gouri pointed out too, First-principles calculations using some theories of DFT have correctly computed the formation energy and ionization potential for Hydrogen in the ZnO bands structure, and this direction also points to confirm that the most likely impurity to give the n-type conductivity to undoped ZnO is Hydrogen.

There are another defects which can be present in ZnO, which are present in lower amounts than the Oxygen and Zinc Vacancies, However these are Shallow Donor defects (as Hydrogen is a Shallow Donor too) : The Zinc Intersticials. However, these shallow donor defects are unestable and are not the cause for the n-type conductivity.

Let me leave a very nice review in ZnO as a semiconductor. Here you can find all above mentioned and more references about the subject.

Good Luck ! :)

I think that high resistivity is relate with n type semiconductors while low resistivity is related with p type semiconductors thus ZnO is n type as a result of domination

of hydrogen donors

Dear Bushra, I think is the opposite, here is the logic :

Resistivity is mainly dependent of the free carriers concentration (either 'holes' or 'electrons'), but is true that the kind of carrier affects the value of Electrical Resistivity, since it (Electrical Resistivity) also depends on the Mobility of the carrier. And in most semiconductor materials Holes have larger Effective Mass, hence lower Mobility than free Electrons, in the same material.

Hence, Resistivity is the Inverse of Conductivity (rho=1/sigma), and the ideal expression for Conductivity in a semiconductor in thermal equilibrium is: sigma =mi*n*q.

From this formula we can see that; if you vary mi (Mobility), from a high Mobility corresponding to free Electrons (n-type) to a lower Mobility (corresponding to free Holes=p-type). Then, considering the same material, doped with Donor Impurities in the first case and with Acceptor Impurities in latter, (both escenarios with a similar Free Carrier Concentration). the p-type will end with a higher Resistivity (translated into a lower Conductivity) than the n-type semiconductor.

Regards !

Dear Franklin Uriel Parás Hernández & Bushra A Hasan,

Thank you very much for your scientific and exciting discussion, But the most (not all) n-type semicondictor have a conductivity more than p-type semicondictor. It may be caused by the mobility or the number of charge carriers.

Greetings

Dear Labidi,

The Oxygen vacancies Vo are the origin of n-type conductivity in ZnO material. Also the Oxygen vacancy (VO) is a common native point defect that plays crucial roles in determining the physical and chemical properties of metal oxides such as ZnO, take this article :

- http://oxidesemi.iphy.ac.cn/result/paper/2016/2016-Lishu%20Liu-Oxygen%20vacancies.pdf.

Dear Idris Bouchama,

Thank you very much for your scientific, exciting and convincing answer.

Dear Labidi Herissi :

did you read the review article I sent you ?

What do you think about that theory ?

Regards !

Dear Franklin Uriel Parás Hernández,

I commented on your previous answer The article is completely dust-free and it contains very important information and raises several questions in order to delve into the research My greetings

A Zn+2 ion occupy an 'interstitial vacancy'[1] (and/or self-insertion[2] of Zn) in Zinc oxide(s) (Zn-O) lattice. These electron might be de-localized, and they are, mainly, contributing to the n-type conductivity.

1. A free copy from Bijit Choudhuri, answer in : https://www.researchgate.net/post/Why_Zinc_oxide_is_naturally_n_type_and_nickel_oxide_is_naturally_p_type_semiconductors

2. Zinc self-diffusion, electrical properties, and defect structure of undoped, single crystal zinc oxide https://aip.scitation.org/doi/abs/10.1063/1.371832

Without having read any of the papers mentioned so far, I still wonder about the very high electron density of about 1021 cm−3 (as mentioned above by Gouri Sankar Tripathi; very first answer in this thread).

Doesn't that mean that the density of "effective dopants" must be in the same order of magnitude? But how likely is it, then, that lattice defects play a major role as "effective dopants"?

Dzar Ioannis Samaras & Jan-Martin Wagner,

Thank you for the scientific and valuable contribution

Dear Franklin Uriel Parás Hernández & Ioannis Samaras & Jan-Martin Wagner,

The charge carriers in zinc oxide are always electrons regardless of their density, which varies according to the method of preparation which may reach great values ​​due to several effects (O vacancies O(v), Zn interstitials Zn(i), H impurities and also the mobility of charge carriers that are related to the grain size).

Dear Labidi Herissi :

If you have a piece of material of ZnO, you will always have both kind of free carriers in the material, always, a certain amount of free electrons and a certain amount of free holes, but depending on which of the two kinds end dominating the charge transport, the conductivity of the material will be p-type or n-type. So, even if you have a n-type ZnO material, there is both kind of free carriers (holes and electrons) inside the material, but as it is always in the case of ZnO, the free electrons end dominating the hole's charge transport

Now, in the case of ZnO, the observed conductivity type for this material have been (almost always) n-type. I say almost always because the very few studies were people have claimed to be obtained a p-type conductivity , other people have had problems to be able to replicate those results, but there is nothing that theoretically prevent the a particular way of shyntesis ZnO to end with a p-type conductivity.

The problem is that the ZnO, when it grows, the atoms (Zn and O) at the lattice (due to its particular structure, binding energies and Formation Energies) they tend to incorporate and bound to +(plus) charged impurities, as H+ and/or N. This fact boost the tendency to a n-type conductivity.

So, to externally make ZnO p-type conductive, first you have to compensate all the effects the large concentration of impurities gives to ZnO. So, one thing you can do is to dope ZnO with acceptor impurities (Al, Ga or In), but it continues to be very difficult to obtain a p-type conductivity. Claims on this have been made though, but with poor replicable results.

Maybe if you could synthesize (or grow) a ZnO with a technique in an ambient almost totally free of Hydrogen and Nitrogen (which would be a hard thing to do, since there is always Hydrogen and Nitrogen in the air, and a lot of Hydrogen content in any wet ambient, as wet solutions) it could be possible to end with a p-type ZnO conductive, likely with a small carrier concentration though.

So, this paper, and many other several works have realised that neither O vacacies nor Zn intersticials could be possibly the defects which are giving the ZnO its n-type conductivity, mainly due to two reasons:

1. O vacancies are not a shallow Donor in ZnO, but a deep Donor. This means that these defects are well below the Fermi Level and its lack of charge cannot contribuite to the electrical conductivity in a significan manner.

So the O vacancies could n't be the reason behind the n-type conductivity, and a shallow Donor is needed.

2. Even though the Zinc interstials are shallow donors. Due to its Formation Energy, they do not form in ZnO in a significant way (again) to be the reason of the n-type conductivity

This fact was revealed more importantly after theoretical results from ab initio simulations were carried out.

all of this is in the article I sent you.

However, with the use of two very sensitive characterization techniques: Electron Paramagnetic Resonance, and Muon Magnetic Resonance, these groups have been able to prove that either H and N impurities are present in everycase when ZnO were synthesize or grown. this is because, if you think on any of the techinques used to grow ZnO: ALD, CVD, MBE .. or the wet techniques which grow the ZnO starting from a solution, like Chemical Bath Deposition or Sol-Gel methods, all of these techniques have Hydrogen present in the its environment.

And Hydrogen (either interstitional or subtitutional) is a Shallow donor.

Again, the Formation Energies of all these defects were calculated from numerical simulations using first principles techniques.

Regards !

Dear Franklin Uriel Parás Hernández,

Yes, I agree with you "for Zinc oxide: the free electrons ends dominating the hole's charge transport or The density of charge carriers is electrons".

Most likely if you could synthesize or grow a p-type ZnO conductive, it will return to the n-type zinc oxide when it comes in contact with nature.

Regards

Yes, Labidi Herissi, such effects have indeed been observed: Have a look at the publication given below, or just search for "p-type ZnO stability" using any scientific search engine.

Article Stability of p-type conductivity in nitrogen-doped ZnO thin film

Because it contains free electrons.

Labidi Herissi

Dear Beddiar Mohammed Ismail,

I agree with you because it is the short and correct answer.

I hope not to be pedantic here , since this is not my objective, but for the sake of understanding,

I belive we have to be careful with the general description to not misunderstand the semiconductor physics background here

Even if the ZnO was to have a p-type conductivity, it as well would have free electrons.

Both types (-p & -n semiconductors) have both carriers; free electrons and free holes. The importance here is which one is the dominant carrier in charge transport.

Best Regards !

Dear Franklin Uriel Parás Hernández,

For ZnO, I think the electrons are the charge carriers that ones controlling because of their presence rate and mobility.

My greetings

ZnO is a n-type semiconductor due to Oxygen vacancies, interstitials or antisites of Zn ion.

Be aware, A M M Tanveer Karim, that above, Franklin Uriel Parás Hernández has pointed out that there are arguments against this common explanation. (I just wonder which first-principles paper he refers to; could you name it again?)

The attached file may be helpful to understand the real scenario.

Dear Jan-Martin Wagner :

The paper I refered is attached in this question , is on the third entry , at the beginning of the answers , the name is :

Fundamentals of Zinc Oxide as a Semiconductor

Indeed, is a really good paper, I absolutely recommend it

Regards :)

Thanks a lot for the clarification; interestingly, it is the same paper that also M N H Liton has recommended.

M N H Liton Excellent paper. Explains a lot.