CN104833450A - Raman test system and method of in-situ LED stress test - Google Patents

Abstract

The invention relates to a Raman test system and method of in-situ LED stress test. The test system is provided with a laser, an optical path system, a long wave pass or band pass filter, a Raman filter, a microscope and a CCD, wherein the long wave pass or band pass filter is arranged at the front/rear end of an optical path of the microscope, or the front/rear end of the Raman filter or the front end of the CCD, and combined with the laser whose wavelength is far from the light spectrum of the LED. The test method comprises that the microscope is used to adjust the position of a sample, so that the sample is processed into a focusing state; and a light source of the microscope is closed, the laser is started to emit laser rays required by Raman test, the optical path system transmits input and output signals of laser, signal processing and filtering are carried out on reflected light via the Raman filter, and final signals are detected by the CCD. The signal-to-noise ratio of Raman signals is improved, and in-situ LED stress test can be realized under different current injection conditions.

Description

A kind of Raman test macro of in-situ test LED stress and method of testing thereof

Technical field

The present invention relates to semiconductor optoelectronic and semiconductor test, particularly relate to a kind of Raman test macro and method of testing thereof of in-situ test LED stress.

Background technology

Light emitting diode (LED) generally adopts heteroepitaxy, in pulse current injectingt situation, because the grating constant of storeroom, thermal expansivity, current/voltage are on differences such as the impacts of lattice ^{[1], [2], [3]}, the stress of LED can change, thus causes the problems such as the luminous intensity of LED declines, the skew of efficiency droop, emission wavelength, encapsulation mismatch ^{[4], [5]}.The stress test of conventional semiconductors generally adopts Raman test macro, but because Raman scattering signal intensity is more weak, 2 ~ 3 orders of magnitude low compared with the fluorescence intensity of LED under pulse current injectingt condition, cause the Raman signal of traditional Raman test macro can flood by the fluorescence spectrum of LED, the Raman spectrum that signal to noise ratio (S/N ratio) is high cannot be obtained, therefore, traditional Raman test macro is adopted cannot to carry out stress test to the LED under pulse current injectingt.

The LED stress test under pulse current injectingt condition cannot be solved in view of existing Raman test macro, be difficult to the impact of determination and analysis stress on LED properties, therefore be necessary a kind of method proposing new in-situ test LED stress.

List of references:

[1]Wu J 2009When group-III nitrides go infrared:New properties andperspectives Journal of Applied Physics106.

[2]Piprek J 2010Efficiency droop in nitride-based light-emitting diodesPhysica Status Solidi(A)Applications and Materials Science207 2217–25.

[3]Aumer M E,LeBoeuf S F,Bedair S M,Smith M,Lin J Y and Jiang H X 2000Effects of tensile and compressive strain on the luminescence properties ofAlInGaN/InGaN quantum well structures Applied Physics Letters77 821–3.

[4]Kisielowski C,Krüger J,Ruvimov S,Suski T,Ager J,Jones E,Liliental-Weber Z,Rubin M,Weber E,Bremser M and Davis R 1996Strain-relatedphenomena in GaN thin films Physical Review B54 17745–53.

[5]Yan Q,Rinke P,Janotti A,Scheffler M and Walle C G Van De 2014Effectsof strain on the band structure of group-III nitrides Physical Review B125118 125118.

Summary of the invention

The object of this invention is to provide a kind of Raman test macro of in-situ test LED stress.

Another object of the present invention is to provide a kind of Raman method of testing of in-situ test LED stress.

A Raman test macro for in-situ test LED stress, is provided with laser instrument, light path system, long-pass or band pass filter, Raman filter, microscope, CCD; Described long-pass or band pass filter are located at microscopes optical path front end in Raman light path, microscopes optical path rear end, Raman filter front end, Raman filter rear end or CCD front end etc., preferred microscopes optical path rear end, and be combined with the laser instrument of wavelength away from LED luminescent spectrum.

Described long-pass or band pass filter allow to be greater than 600nm by the wavelength of signal, and the signal that wavelength is less than 600nm is filtered, the long-pass filter plate of long-pass or band pass filter optimal wavelength >600nm.

The quantity of described long-pass or band pass filter is N sheet, N >=1, preferably 1.

The wavelength of described laser instrument must away from the electroluminescent spectrum region of LED (wavelength is greater than the laser instrument of 600nm), and to reduce the interference of electroluminescence spectra to Raman signal, optimal wavelength is the He-Ne laser instrument of 633nm.

A Raman method of testing for in-situ test LED stress, adopts the Raman test macro of described in-situ test LED stress, comprises the following steps:

First adopt microscope to carry out regulating the position of sample, make sample preparation focus state; Then, close microscopical light source, open laser instrument, laser instrument is used for launching Raman and tests the laser beam needed, light path system transmission laser input and output signal, reflection ray carries out signal transacting and filtration through Raman filter, final signal detect by CCD.

The present invention introduces long-pass or band pass filter in the light path system of Raman test macro, and select optical maser wavelength away from the laser instrument of LED electrical photoluminescence spectrum, with the interference avoiding Raman signal to be subject to the electroluminescent fluorescent spectrum of the LED under pulse current injectingt condition, and then promote the signal to noise ratio (S/N ratio) of Raman signal, realize the in-situ test of the LED stress under different pulse current injectingt condition.

Present invention can be implemented in the in situ detection of the LED stress under different pulse current injectingt condition, described in-situ Raman test macro can carry out the in-situ Raman test of stress at the LED light emitting diode of 100 ~ 600nm to wavelength.

Accompanying drawing explanation

Fig. 1 is the composition schematic diagram of the Raman test macro embodiment of in-situ test LED stress of the present invention.

Fig. 2 is the principle schematic of the Raman method of testing of the in-situ test LED stress of the embodiment of the present invention.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.

See Fig. 1, the Raman test macro embodiment of in-situ test LED stress of the present invention is provided with laser instrument 1, light path system a/b/c/d, long-pass or band pass filter 2, Raman filter 3, microscope 4, CCD 5; Described long-pass or band pass filter 2 are located at microscopes optical path front end in Raman light path, microscopes optical path rear end, Raman filter front end, Raman filter rear end or CCD front end etc., preferred microscopes optical path rear end, and be combined with the laser instrument of wavelength away from LED luminescent spectrum.

Described long-pass or band pass filter 2 allow to be greater than 600nm by the wavelength of signal, and the signal that wavelength is less than 600nm is filtered, the long-pass filter plate of long-pass or band pass filter optimal wavelength >600nm.

The quantity of described long-pass or band pass filter 2 is N sheet, N >=1, preferably 1.

The wavelength of described laser instrument must away from the electroluminescent spectrum region of LED (wavelength is greater than the laser instrument of 600nm), and to reduce the interference of electroluminescence spectra to Raman signal, optimal wavelength is the He-Ne laser instrument of 633nm.

In FIG, mark A is lens, and B is reflective mirror, and P is sample.

The Raman method of testing of in-situ test LED stress is as follows:

First adopt microscope to carry out regulating the position of sample, make sample preparation focus state.Then, close microscopical light source, open laser instrument, laser instrument is used for launching Raman and tests the laser beam needed, light path system transmission laser input and output signal, reflection ray carries out signal transacting and filtration through Raman filter, final signal detect by CCD.

Described Raman test macro selects wavelength away from the optical maser wavelength of LED luminescent spectrum, the laser instrument of preferred 633nm.The filter plate inserted in Raman light path, the preferably long-pass filter plate of 1 wavelength >600nm, introduce position in the rear end of microscopes optical path b, by the filtration of filter plate to the photoluminescence under different pulse current injectingt, the Raman signal that signal to noise ratio (S/N ratio) is higher can be obtained, be converted to stress by formula, thus the STRESS VARIATION of the LED under different pulse current injectingt can be monitored in real time.

The Raman method of testing of a kind of in-situ test LED stress proposed by the invention is shown in accompanying drawing 2 to the principle schematic that blue light-emitting diode is tested.As shown in Figure 2, the main glow peak 455nm of the blue light-emitting diode of GaN base, yellow luminescence peak is 500 ~ 600nm, and the band-edge luminescence peaks of GaN material is 365nm.Traditional Raman test macro generally adopts Ar+488nm laser instrument, Nd:YAG 532nm laser instrument and He-Ne 633nm laser instrument to be used as the light source of laser, the E that the above three kinds of LASER Light Source is corresponding ₂the wavelength location of mould lays respectively at 501.9nm, 548.6nm and 656.6nm.Although the E of all LASER Light Source ₂mould all can away from the main glow peak of blue-ray LED, but the E that 488nm and 532nm laser is corresponding ₂mould can be subject to the impact at 500 ~ 600nm defect peak, yellowish leukorrhea peak of blue-ray LED, and especially in Bulk current injection condition, the intensity at yellowish leukorrhea peak can sharply strengthen, and severe jamming Raman signal causes declining to a great extent of signal to noise ratio (S/N ratio).Therefore, for blue-ray LED, 633nm laser instrument is relatively suitable lasing light emitter, because of its E ₂mould not only can away from main glow peak but also can away from yellow luminescence peak.

In addition, the mode of seed collecting backward scattering carries out Raman test, can carry out the change of calculated stress according to Raman frequency shift.The E of GaN ₂the frequency shift (FS) of mould calculates STRESS VARIATION (the Zhang L of GaN base LED according to following formula (1), Yu J, Hao X, Wu Y, Dai Y, Shao Y, Zhang H and Tian Y 2014Influence of stress in GaN crystals grownby HVPE on MOCVD-GaN/6H-SiC substrate.Scientific reports4 4179):

${\mathrm{\σ}}_a=\frac{\mathrm{\Δ\ω}}{4.3}\left({\mathrm{cm}}^{-1}G{\mathrm{Pa}}^{-1}\right)---\left(1\right)$

for frequency shift (FS), ω ₁for the wave number (or frequency) of Raman recorded, ω ₀for the Raman wave number (or frequency) under stress-free conditions, σ _afor stress intensity.The Raman wave number position ω of GaN gallium nitride under stress-free conditions ₀in in 567.1cm ^{-1 [6]}, in conjunction with the Raman wave number ω recorded ₁, substituting into formula (1) can calculated stress size.

Meanwhile, under pulse current injectingt condition, fluorescence signal intensity comparatively large 2 ~ 3 orders of magnitude of Raman signal of LED, cause the Raman signal of traditional Raman test macro can flood by fluorescence signal, be difficult to obtain the Raman spectrum of high s/n ratio.For obtaining the Raman signal of high s/n ratio, the present invention creatively adopts and insert long-pass or band pass filter in Raman light path, be preferably long-pass filter plate, its schematic diagram as shown in Figure 1, the signal that wavelength is shorter than 600nm is all filtered, the electroluminescent fluorescent spectrum of LED all can be shielded, the signal only allowing wavelength to be greater than 600nm passes through.

Claims (9)

1. a Raman test macro for in-situ test LED stress, is provided with laser instrument, light path system, Raman filter, microscope, CCD; It is characterized in that also being provided with long-pass or band pass filter, described long-pass or band pass filter are located at microscopes optical path front end in Raman light path, microscopes optical path rear end, Raman filter front end, Raman filter rear end or CCD front end.

2. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1, is characterized in that described long-pass or band pass filter are located at microscopes optical path rear end.

3. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1 or 2, it is characterized in that described long-pass or band pass filter allow to be greater than 600nm by the wavelength of signal, the signal that wavelength is less than 600nm is filtered.

4. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1 or 2, is characterized in that described long-pass or band pass filter adopt the long-pass filter plate of wavelength >600nm.

5. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1, is characterized in that the quantity of described long-pass or band pass filter is N sheet, N >=1.

6. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 5, is characterized in that the quantity of described long-pass or band pass filter is 1.

7. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1, it is characterized in that the wavelength of described laser instrument must away from the electroluminescent spectrum region of LED, laser instrument adopts wavelength to be greater than the laser instrument of 600nm.

8. the Raman test macro of a kind of in-situ test LED stress as claimed in claim 7, is characterized in that described laser instrument adopts wavelength to be the He-Ne laser instrument of 633nm.

9. a Raman method of testing for in-situ test LED stress, is characterized in that adopting the Raman test macro of a kind of in-situ test LED stress as claimed in claim 1, said method comprising the steps of:

First adopt microscope to carry out regulating the position of sample, make sample preparation focus state; Then, close microscopical light source, open laser instrument, laser instrument is used for launching Raman and tests the laser beam needed, light path system transmission laser input and output signal, reflection ray carries out signal transacting and filtration through Raman filter, final signal detect by CCD.