Silicon carbide has a variety of crystal types, but the silicon crystal structure the market needed is mainly 4H-SiC. So the silicon carbide crystal growth in crystal types is a defect. To a certain extent, it can be distinguished by the naked eye. A more accurate measurement method for testing the silicon carbide crystal distribution is Raman spectroscopy: Raman spectroscopy has characteristics for crystals, and the peak positions of the light emitted by different crystals are different.

1. What is Raman Spectroscopy?

I 1928 opdagede den indiske videnskabsmand Raman Raman-spredning i eksperimentet med at studere spredningsspektret af flydende benzen. Raman-spredningsspektroskopi er i enkle vendinger, at der bruger en lysstråle, der falder ind på et stof, frekvensen af ​​det indfaldende lys er v, og frekvensen af ​​det resulterende spredte lys vil være v, v + Av1, v-Av1, v + Δv2, v-Δv2 og så videre. Disse Av1, Av2 osv. Av har karakteristika. Med andre ord har hvert stof en bestemt forskel (topposition, peakintensitet), der kaldes Raman shift (det udsendte lys minus det indfaldende lys).

For eksempel er Raman-spektret af 4H-SiC:

The table above shown here is the Raman shift. During the silicon carbide crystal distribution measurement, the computer will help to calculate it and process it as a Raman shift spectrum.  

2. How to Test the Silicon Carbide Crystal Distrubution?

Typiske testbetingelser er: ved hjælp af 532 nm laser af Ar + laser fra LabRAM HR Raman-spektrometeret hælder den lodret, exciteringseffekten er 200 mW, og tilstanden til opsamling af spredt lys er backscatter-tilstand. Det indfaldende lys med forskellige bølgelængder har forskellige gennemtrængningsdybder. Generelt er 266nm-laser 0,2um, 325nm-laser er 2um, og 514nm-laser er 30um, hvilket betyder, at ultraviolet lys kun kan bruges til at måle tynde prøver.

Because the silicon carbide wafer has different positions, multiple measurements will be taken to obtain the silicon carbide crystal distribution:

The data has three indicators: the position of the peak, the height of the peak (light intensity), and the width of the peak. Only when the peak position is completely matched, can it be den kvalificerede 4H-SiC. Som med XRD, når der er andre toppe, er de stoffer i andre faser, hvilket er en defekt.

The difference in the position of the peak is due to the difference in the energy of the phonons brought about by the different silicon carbide crystal lattices, that is, the different frequencies. Each phonon has its corresponding energy level. The virtual energy level theory can be used to explain Raman (non-linear process):

Partiklen absorberer indfaldende lys til det virtuelle energiniveau (orange) og overgår derefter tilbage til et vibrationsenerginiveau (rødt), der adskiller sig fra det oprindelige energiniveau. Da det øverste energiniveau er et virtuelt energiniveau, kan frekvensen af ​​det indfaldende lys varieres, så længe det ikke er i konflikt med det oprindelige virkelige energiniveau.

Det er værd at bemærke, at LOPC-tilstanden (964,769 cm-1) kan bruges til at analysere bærerens koncentration:

n = 1.25 * 1017cm-2 *(964.769cm-1-VLOPC measurement)

As the carrier concentration increases, the interaction between atoms and the lattice increases, which makes the Raman peak blue shift (smaller), the intensity decreases, and the width increases. This method is not as accurate as other methods and can only be used as an aid to analyze the silicon carbide crystal distribution.

3. Why not Use XRD to Measure Silicon Carbide Crystal Distribution?

X-rays are optical radiation generated by the transition of electrons in the inner layer of atoms under the bombardment of high-speed moving electrons, including continuous X-rays and characteristic X-rays. Silicon carbide single crystal can be used as X-ray gratings, and the coherent scattering produced by these large numbers of particles (atoms, ions, or molecules) will cause light interference, increasing or decreasing the intensity of scattered X-rays. Due to the superposition of scattered waves from a large number of particles, the beams that interfere with each other to produce the highest intensity are called X-ray diffraction lines.

For at opfylde diffraktionsbetingelserne kan Bragg-formlen anvendes: 2dsinθ = nλ.

Den indfaldende stråle får hver spredning til at udstråle en lille del af dens intensitet igen som en sfærisk bølge. Hvis spredere er arrangeret symmetrisk med intervallet d, vil disse sfæriske bølger kun blive synkroniseret i den retning, hvor deres kurslængdeforskel 2dsinθ er lig med et heltal multiple af bølgelængden λ. I dette tilfælde afbøjes en del af den indfaldende stråle med en vinkel på 2θ, som vil frembringe refleksionspunkter i diffraktionsmønstret.

Use X-rays of known wavelengths to measure the θ angle to calculate the interplanar spacing d, which is used for X-ray structure analysis; the other is to use a silicon carbide seed crystal with a known d to measure the θ angle to calculate the characteristic X-ray wavelength, and then the elements contained in the sample can be found in the existing data.

The measurement formula is 2dSinθ=λ. While the d value among the cubic silicon carbide crystal is close, and the characteristic is not obvious enough, the accurate silicon carbide crystal distribution cannot be precisely determined. For these reasons, it is not suitable to use the XRD to measure the distribution of silicon carbide crystals.

For mere information, kontakt os venligst e-mail på victorchan@powerwaywafer.com og powerwaymaterial@gmail.com.