Self-calibration apparatus and method for real-time temperature measurement system of MOCVD device
A self-calibration apparatus and method for a real-time temperature measurement system of a MOCVD device belong to the technical field of semiconductor manufacturing. The apparatus comprises a MOCVD reactor chamber ( 1 ) and an optical detector ( 6 ). The MOCVD reactor chamber ( 1 ) comprises an epitaxial wafer ( 4 ). A detection window ( 5 ) is provided on the top of the MOCVD reactor chamber ( 1 ). The optical detector ( 6 ) emits detection light beams whose wavelengths are respectively λ 1 and λ 2 toward the epitaxial wafer ( 4 ) through the detection window ( 5 ). The detection light beams are reflected by the epitaxial wafer ( 4 ) to form reflected light beams which are detected by the optical detector ( 6 ). In the method, points corresponding to the actual thermal radiation ratios are depicted on the theoretical thermal radiation ratio-temperature curve according to actual thermal radiation ratios, and values of the temperatures T corresponding to the points are substituted into formulas to obtain m 1 and m 2 respectively. With the method and apparatus, self-calibration of the real-time temperature measurement system of a MOCVD device is realized, thus ensuring consistent and accurate measurements of a growth temperature of the epitaxial wafer.
1. A self-calibration apparatus for a real-time temperature measurement system of a MOCVD device, comprising a MOCVD reaction chamber ( 1 ) and an optical detector ( 6 ), wherein the MOCVD reaction chamber ( 1 ) includes an epitaxial wafer ( 4 ), and a detection window ( 5 ) is provided on the top of the MOVCD reaction chamber ( 1 ), and the light detector ( 6 ) emits detection light beams whose wavelengths are respectively λ 1 and λ 2 toward the epitaxial wafer ( 4 ) through the detection window ( 5 ), and the detection light beams are reflected by the epitaxial wafer ( 4 ) to form reflected light beams which are detected by the optical detector,
wherein the optical detector ( 6 ) comprises a first light source, a second light source, a beam splitter, a first dichroic mirror ( 10 ), a first filter ( 11 ), and a first detector, a second dichroic mirror ( 8 ), a second filter ( 9 ), a second detector, a reference light detector, and a data acquisition unit;
the first light source emits a light beam of wavelength λ 1 , and the second light source emits a light beam of wavelength λ 2 ; the light beam of wavelength λ 1 and the light beam of wavelength λ 2 , after passing through the beam splitter ( 12 ), are split into two parts, one part being a reference light, the other part being a detection light beam of wavelength λ 1 and a detection light beam of wavelength λ 2 ; the reference light enters a reference light detector to form an electrical signal I refe ,
the detection light beam of the wavelength λ 1 and the detection light beam of the wavelength λ 2 are reflected by the epitaxial wafer ( 4 ) to form reflected lights, and the reflected lights, after passing through the beam splitter ( 12 ), are separated into two parts by the first dichroic mirror and the second dichroic mirror, wherein one part has a wavelength λ 1 , passing through the first filter and then entering the first detector to form an electrical signal I refl1 ; the other part has a wavelength 2 , passing through the second filter and then entering the second detector to form an electrical signal I refl2 ; and
the electrical signals I refe , I refl1 , and I refl2 are respectively acquired by the data acquisition unit.
2. The apparatus of claim 1 , wherein the MOCVD reaction chamber ( 1 ) further comprises a heating chamber ( 2 ) and a graphite susceptor ( 3 ), the graphite susceptor ( 3 ) being used to bear the epitaxial wafer ( 4 ), the heating chamber ( 2 ) being used to heat the graphite susceptor ( 3 ) and in turn heat the epitaxial wafer ( 4 ).
3. The apparatus of claim 1 , wherein a frequency of a light emitted by the first light source and the second light source is modulatable.
4. The apparatus of claim 1 , further comprising a light source control circuit for controlling the light emitted by the first light source and the second light source.
5. The apparatus of claim 1 , further comprising a processing unit, and the processing unit is configured to process the light source control circuit and the data acquisition unit.
6. A self-calibration method based on the self-calibration apparatus for a real-time temperature measurement system of a MOCVD device of claim 1 , comprising:
measuring response spectrums P(λ, T) of a black-body furnace at different temperatures;
calculating a ratio r 0 (T) of theoretical thermal radiation powers respectively corresponding to a first wavelength λ 1 and a second wavelength λ 2
r
0
(
T
)
=
P
0
(
λ
1
,
T
)
P
0
(
λ
2
,
T
)
=
∫
λ
1
-
Δλ
1
λ
1
+
Δλ
1
f
1
(
λ
)
g
1
(
λ
)
P
(
λ
,
T
)
/
τ
(
T
)
d
λ
∫
λ
2
-
Δλ
2
λ
2
+
Δλ
2
f
2
(
λ
)
g
2
(
λ
)
P
(
λ
,
T
)
/
τ
(
T
)
d
λ
according to the following formulas:
P 0 (λ 1 ,T )=∫ λ 1 −Δλ 1 λ 1 +Δλ 1 f 1 (λ) g 1 (λ) P (λ, T )/τ( T ) dλ
P 0 (λ 2 ,T )=∫ λ 2 −Δλ 2 λ 2 +Δλ 2 f 2 (λ) g 2 (λ) P (λ, T )/τ( T ) dλ
where,
P 0 (λ 1 , T) indicates a thermal radiation power corresponding to the first wavelength λ 1 ,
λ 1 indicates the first wavelength,
Δλ 1 indicates a bandwidth corresponding to the first wavelength λ 1 ,
f 1 (λ) indicates a response function of the optical detector 6 at the first wavelength λ 1 ,
g 1 (λ) indicates a transmittance of a radiation light corresponding to the first wavelength λ 1 in an optical device,
P(λ, T) indicates a response spectrum of the black-body furnace,
τ (T) indicates an expression of a spectral transmission curve,
P 0 (λ 2 , T) indicates a thermal radiation power corresponding to the second wavelength λ 2 ,
λ 2 indicates the second wavelength,
Δλ 2 indicates a bandwidth corresponding to the second wavelength λ 2 ,
f 2 (λ) indicates a response function of the optical detector ( 6 ) at the second wavelength λ 2 ,
g 2 (λ) indicates a transmittance of the radiation light corresponding to the second wavelength λ 2 in an optical device,
T indicates a temperature,
r 0 (T) indicates a ratio of theoretical thermal radiation powers respectively corresponding to the first wavelength λ 1 and the second wavelength λ 2 ;
performing a least square fitting according to the temperatures and the ratios r 0 (T) of corresponding theoretical thermal radiation powers to obtain a theoretical thermal radiation ratio-temperature curve;
measuring actual thermal radiation powers corresponding to the first wavelength λ 1 and actual thermal radiation powers corresponding to the second wavelength λ 2 at different temperatures, and obtaining actual thermal radiation ratios;
depicting points corresponding to the actual thermal radiation ratios on the theoretical thermal radiation ratio-temperature curve according to the actual thermal radiation ratios;
substituting values of the temperatures T corresponding to the points into the following formulas to obtain m 1 and m 2 respectively:
L
(
λ
1
,
T
)
=
m
1
×
∫
λ
1
-
Δλ
1
λ
1
+
Δλ
1
f
1
(
λ
)
g
1
(
λ
)
ɛ
(
λ
)
×
2
π
hc
2
/
λ
5
exp
(
hc
kT
λ
)
-
1
d
λ
L
(
λ
2
,
T
)
=
m
2
×
∫
λ
2
-
Δλ
2
λ
2
+
Δλ
2
f
2
(
λ
)
g
2
(
λ
)
ɛ
(
λ
)
×
2
π
hc
2
/
λ
5
exp
(
hc
kT
λ
)
-
1
d
λ
where,
L(λ 1 , T) indicates an actual thermal radiation power corresponding to the first wavelength λ 1 ,
L(λ 2 , T) indicates an actual thermal radiation power corresponding to the second wavelength λ 2 ,
m 1 indicates a calibration coefficient corresponding to the first wavelength λ 1 ,
m 2 indicates a calibration coefficient corresponding to the second wavelength λ 2 ,
f 1 (λ) indicates a response function of the optical detector ( 6 ) at the first wavelength λ 1 ,
g 1 (λ) indicates a transmittance of a radiation light corresponding to the first wavelength λ 1 in an optical device,
f 2 (λ) indicates a response function of the optical detector ( 6 ) at the second wavelength λ 1 ,
g 2 (λ) indicates a transmittance of the radiation light corresponding to the second wavelength λ 1 in an optical device,
ε(λ) indicates an emissivity of a surface of the epitaxial wafer ( 4 ),
T indicates temperature;
λ 1 indicates the first wavelength,
Δλ 1 indicates a bandwidth corresponding to the first wavelength λ 1 ,
λ 2 indicates the second wavelength,
Δλ 2 indicates a bandwidth corresponding to the second wavelength λ 2 ,
k indicates Boltzmann constant, k=1.3806×10 −23 J/K,
h indicates a Planck constant, h=6.626×10 −34 J·s, and
c indicates a speed of light in vacuum, c=3×10 8 m/s.
7. The method of claim 6 , wherein when the thermal radiation ratio-temperature curve is obtained by least square method, there are a plurality of thermal radiation ratios and corresponding temperatures T participating in the fitting, obtained respectively when the temperature of the reaction chamber is stable at T 1 , T 2 , . . . , T n .
8. The method of claim 7 , wherein T 1 , T 2 , . . . , T n are respectively obtained by heating by a black-body furnace heating system.
9. The method of claim 7 , wherein a temperature measurement range (T min , T max ) is (400° C., 1500° C.); the first wavelength λ 1 corresponds to a high temperature interval (T up , T max ), and the second wavelength λ 2 corresponds to a low temperature interval (T min , T down ).
10. The method of claim 9 , wherein (T min , T max ) is (450° C., 1200° C.), T up =750° C., T down =800° C., λ 1 =940 nm, λ 2 =1050 nm.
11. The method of claim 6 , wherein the actual thermal radiation ratio r(T) is calculated as follows:
r
(
T
)
=
L
(
λ
1
,
T
)
/
ɛ
1
L
(
λ
2
,
T
)
/
ɛ
2
where,
L(λ 1 , T) indicates an actual thermal radiation power corresponding to the first wavelength λ 1 ,
L(λ 2 , T) indicates an actual thermal radiation power corresponding to the second wavelength λ 2 ,
λ 1 indicates the first wavelength,
λ 2 indicates the second wavelength,
ε 1 indicates an emissivity of a surface of the epitaxial wafer 4 corresponding to the first wavelength λ 1 ,
ε 2 indicates an emissivity of the surface of the epitaxial wafer 4 corresponding to the second wavelength λ 2 , and
T indicates temperature.
12. The method of claim 11 , wherein,
when the epitaxial wafer ( 4 ) is an ideal opaque, smooth, flat surface,
ε=1− R/ΔT R
where,
ε indicates an emissivity of the surface of the epitaxial wafer ( 4 ),
R indicates a reflectivity of the epitaxial wafer ( 4 ), and
ΔT R indicates a reflectivity attenuation factor;
when the epitaxial wafer ( 4 ) is a transparent, single-side polished sapphire substrate,
ε=ε carr (1− R/ΔT R )(1− R diff ){1+ R/ΔT R *R diff +(1−ε carr )[( R diff +R/ΔT R (1− R diff ) 2 )]}
where,
ε indicates an emissivity of the surface of the epitaxial wafer ( 4 ),
R diff indicates a scattering rate of a non-smooth substrate,
ε carr indicates the thermal emissivity of the graphite susceptor ( 3 ), and
ΔT R indicates a reflectivity attenuation factor.
13. The method of claim 11 , wherein,
when the actual thermal radiation ratios are calculated, the temperatures T can be obtained by heating by the MOCVD reaction chamber ( 1 ).
14. The method of claim 8 , wherein a temperature measurement range (T min , T max ) is (400° C., 1500° C.); the first wavelength λ 1 corresponds to a high temperature interval (T up , T max ), and the second wavelength λ 2 corresponds to a low temperature interval (T min , T down ).
15. The apparatus of claim 3 , further comprising a processing unit, and the processing unit is configured to process the light source control circuit and the data acquisition unit.
16. The apparatus of claim 4 , further comprising a processing unit, and the processing unit is configured to process the light source control circuit and the data acquisition unit.