Laser Cooling in Semiconductors
Chengao Wang
Optical Science and Engineering,
Department of Physics & Astronomy, University of New Mexico
Albuquerque, NM 87131-0001, USA
I. Introduction
The concept of laser cooling (optical refrigeration) by luminescence up-conversion in solids dates back to around 1930, when Pringsheim recognized that thermal vibration energy can be removed by anti-Stokes fluorescence if a material is excited with photons having energy below the mean fluorescence energy. With the laser invented in 1960, this kind of cooling machine seems promising. However, material purity problems prevented observation of this type of laser cooling until 1995, when it was first demonstrated in ytterbium-doped glass.
The next step is to achieve laser cooling in semiconductor materials. Heating is a major problem in semiconductor devices and current cooling devices( fans, liquid nitrogen) are inefficient and complicated. However, since optical refrigerators are solid-state devices without any moving parts, they would be free of vibration, mechanically robust and compact, and thus become an efficient cooling machine for semiconductors. If successful, the project can have far reaching implications in the area of optical detection systems and optoelectronic devices. Specifically, optical refrigerators might be used to cool infrared or gamma ray sensors to cryogenic temperatures to increase sensitivity. Down the road, the technique might be used to cool superconductor electronics.
The purpose of our research is to invent a practical all-solid-state optical refrigerator to cool semiconductors using laser cooling. Our hypothesis is that laser cooling in semiconductors can achieve temperatures ~10K and below. After solving all fundamental physics and engineering issues, a practical all-solid-state optical refrigerator in semiconductors can be made.
II. Overview of methodology
Research objective: invent a practical all-solid-state optical refrigerator to cool semiconductors using laser cooling.
Our research will include four major aims:
1. Develop a comprehensive theoretical model of laser cooling in semiconductors.
Deliverable: understanding of which materials are optimum for laser cooling in semiconductors.
2. Grow new semiconductor materials optimized for laser cooling using MOCVD
Deliverable: optimal materials that have a good chance of achieving net cooling.
3. Demonstrate experimentally the theory of laser cooling in semiconductor devices.
Deliverable: proof of net cooling in semiconductors
4. Build prototype optical refrigerator in semiconductors.
Deliverable: a prototype machine for practical applications.
Here, I will describe the four major steps in detail and mainly focus on the first step.
Major step 1: Develop a comprehensive theoretical model of laser cooling in semiconductors
The basic concepts about laser cooling in semiconductors have been established. We should first review these concepts and theories. In 1997, the theory of semiconductor cooling was also addressed. The basic cooling criteria in semiconductors are known.
However, attaining net cooling in semiconductors has still remained elusive. Two processes, luminescence trapping and red-shifting, have the potential to frustrate attempts to achieve semiconductor net cooling, while these two issues have not been addressed in the current preliminary theory. So, the next step is to develop a more sophisticated and comprehensive theoretical model, taking into account these processes.
The more important problem, luminescence trapping, occurs when total internal reflection prevents luminescence from efficiently escaping from a semiconductor. Fig. 1 shows how luminescence trapping occurs. In this figure, only the light marked with yellow lines can escape from the material, carrying energy away, while the light marked with dark lines bound back and forth inside the material due to total internal reflection, leaving the energy inside the material. Therefore, only a portion of the energy can be taken away from the material, which reduced the efficiency of laser cooling.
Fig. 1 Luminescence trapping. It’s the process when total internal reflection prevents luminescence from efficiently escaping from a semiconductor.
The process of red-shifting will also decrease the efficiency of cooling. It’s the process in which the emitted photons are red-shifted to low frequency and thus carrying less energy away.
We will include these two processes into our model describing laser cooling in semiconductors. Basically speaking, we will add two new terms into the electron-hole carrier density equation. By doing this, we expect to make a comprehensive theory describing the whole process.
We can use this comprehensive theory to determine which method to grow semiconductor materials and what specific characteristics are required for the materials. We may also predict the possible designs of the future optical refrigerators.
Major step 2: Use MOCVD( The metal organic chemical vapor deposition) to grow new semiconductor materials
First, we will use MOCVD to grow InGaP/GaAs Heterostructures, because we have predicted the InGaP/GaAs Heterostructures are good candidates for laser cooling in semiconductors. Then we may grow GaAs/InGaAs Devices and quantum wells since they have better characteristics in some aspects.
Next, in order to determine whether the material has the desired characteristic, we will test the effect of gas purity and growth temperature and perform microscopic analysis of each material. Based on these data, we may also optimize the materials for laser cooling.
We expect to have optimal materials that have a good chance of achieving net cooling.
Major step 3: Demonstrate experimentally the theory of laser cooling in semiconductor devices
First, we will try to fabricate devices to cool GaAs.
Second, we will enhance cooling GaAs using advanced methods. Since the starting temperature will influence the cooling efficiency greatly, we will investigate the role of starting temperature in laser cooling, trying to have the most efficiency.
Third, we will fabricate devices to cool InGaAs and quantum wells.
Based on these experimental data, we may reevaluate our theory about laser cooling in semiconductors and further optimize the materials.
Major step 4: Build prototype optical refrigerator in semiconductors.
In order to build a practical devise, we need to first solve some engineering issues.
After making the preliminary optical refrigerator, we may try to make it more compact and efficient.
We expect to make a prototype machine for practical applications at the end.
1