Vapor Pressure Effects on Material Growth

[marie2010]

hi,
I am confused about how the vapor pressure affects the materials growth. Can someone comment on the following : If the vapor pressure of a precursor, let's say N, is high then it will be hard to grow a material containing N because it will be hard to incorporate it onto the substrate, right? Or is the opposite true? How does the growth temperature affects the vapor pressure?
Thank you for any comment in advance.

 

[shawl]

What materials you asked? What growth method you are talking about?

In generally, crystal growth is required under a high enough temperature to melt the materials from solid phase into a liquid phase (or a vapor phase). And then the melted materials will form nucleation and growth.

According to the thermodynamics, if the vapor pressure is reaching the highest value, i.e. the saturated vapor pressure. Then the phase transition from solid to liquid will be depressed.

Therefore, one can adjust the vapor pressure to control the growth of crystals.

 

[marie2010]

hi,
I was thinking about GaN MOCVD growth with N2 as a N source. Thanks for replying.
So, to summarize, you need high temperatures for the precursors (here N2) to melt (that is, to dissociate from its molecular form), but if you go to high temp. the vapor pressure may become too high for the N incorporation. Is that right?
Thank you in advance.

 

[shawl]

Yes.

And sometimes you need a vacuum pump to reduce the pressure so as to control the growth rate and the axis orientation.

 

[marie2010]

Thanks for the reply. It helps.

 

[caffenta]

hi,
I was thinking about GaN MOCVD growth with N2 as a N source. Thanks for replying.
So, to summarize, you need high temperatures for the precursors (here N2) to melt (that is, to dissociate from its molecular form), but if you go to high temp. the vapor pressure may become too high for the N incorporation. Is that right?
Thank you in advance.

Not exactly. I believe Shawl is referring more to bulk growth, rather than epitaxial growth on a substrate like MOCVD, MBE, etc.

Heating the substrate is not only needed to provide enough energy to break the precursors (and the subsequent reaction to form GaN) as you mention, but also increase the mobility of the adsorbates on the surface. They need to move around in order to lead to anything resembling 2D growth. If they are not mobile enough, you get 3D island growth and bad (worse) material quality.

There is a constant dynamic adsorption and desorption process happening between the gas phase and the substrate's surface. As the substrate temperature increases, the desorption rate increases and the adsorption rate decreases, leading to an increase in vapor pressure. The relative adosrption/desorption rate is not the same for group III and group V species (group V desorbs faster), so there is less nitrogen on the surface. To compensate, you have to raise the relative nitrogen pressure, either by increasing the gas flow of the nitrogen precursor, and/or by making the ambient pressure more nitrogen-rich.

So the vapor pressure does increases with substrate temperature, but you have to increase it further to compensate. Maybe that was the source of the confusion.

 

[marie2010]

Thank you for the nice and thorough explanation. It really helped!
One quick question on the following line "To compensate, you have to raise the relative nitrogen pressure, either by increasing the gas flow of the nitrogen precursor, and/or by making the ambient pressure more nitrogen-rich." How do I make the ambient pressure more N-rich (other than increasing the gas flow of the N precursor as you mentioned as well)?
Thanks a lot in advance. I really appreciate it.

 

[caffenta]

Thank you for the nice and thorough explanation. It really helped!
One quick question on the following line "To compensate, you have to raise the relative nitrogen pressure, either by increasing the gas flow of the nitrogen precursor, and/or by making the ambient pressure more nitrogen-rich." How do I make the ambient pressure more N-rich (other than increasing the gas flow of the N precursor as you mentioned as well)?
Thanks a lot in advance. I really appreciate it.

The ambient gas is not necessarily the same as the precursor gas. You might use NH3 as a nitrogen precursor and a mixture of N2 and H2 for ambient, for example. Everything can affect growth quality: ambient pressure, ratio of the ambient mixture, precursor flow, growth temperature, and so on. All the parameters have to be optimized for a particular material. It gets complicated when you are growing multiple layers of various materials (heterostructures). Sometimes you must compromize.

 

[marie2010]

Thanks you for your last message. I think I finally start to understand the MOCVD growth process. One question (perhaps or hopefully the last one) that I would like to ask you is why we need the ambient gas. What is its role? How you decide what the ambient gas should be? I greatly appreciate your help.

 

[caffenta]

Thanks you for your last message. I think I finally start to understand the MOCVD growth process. One question (perhaps or hopefully the last one) that I would like to ask you is why we need the ambient gas. What is its role? How you decide what the ambient gas should be? I greatly appreciate your help.

The main role is pressure. At the temperatures needed for growth, the material is thermodynamically unstable in vacuum. Depending on the conditions, you might not be able to use the precursor gas at a pressure high enough, so you need a separate ambient gas. And sometimes, the ambient gas also participates in the reaction in various ways, which can help improve the material quality.

How do you choose the ambient gas? If the material you are growing has been around for a while and there is a mature process already, you go with that. If not, you check the literature to see what has been done, use your own experience, and a lot of trial and error until the process is optimized.

 

[marie2010]

hi,
Thank you for the reply. So, in other words, if the precursor gas cannot be used at pressures high enough, then you use ambient gas to suppress (or to control the desorption of N from the surface).
Thanks.

 

[caffenta]

So, in other words, if the precursor gas cannot be used at pressures high enough, then you use ambient gas to suppress (or to control the desorption of N from the surface).
Thanks.

Hmm. I wouldn't exactly say that. The precursor flow rates should be set for the correct growth settings (like growth rate and composition), and the ambient pressure/composition should be set for optimum growth conditions. There are so many processes going on inside the reactor that it's hard to know what affects what. A lot of it is trial and error. Especially with nitrides. There's still a lot of development going on. Very often, you'll find papers that contradict each other.

Essentially, the N/III ratio has to be high enough in order to get good nitrogen incorporation. The primary way to do this is to raise the nitrogen precursor flow.

 

[marie2010]

Hi,
when doing MOCVD experiments one needs to control precursors' flow rates. What are the units of the flow (flux) and what are typical values for MOCVD growth of nitrides?
Thanks for your help.