JPH02216435A - Time division spectroscopic method - Google Patents

Abstract

PURPOSE:To make time division measurement of even the emitted light of the wavelength as long as >=1mum by making use of the linearity that the intensity of the light emitted from a light emitting material possesses with respect to the intensity of stimulating light. CONSTITUTION:An incident light pulse 1 is made into a stimulating light pulse 15 by a beam splitter 12 and mirrors 13, 14. A time delay 16 can be provided between the superposed two stimulating light pulses 15 by moving either of the mirror 13 or 14 in the incident direction of the light pulse at this time. The pulses 15 pass through a dichroic mirror 17 and are thereafter condensed and projected by a lens 18 to a sample 19. The emitted light from a sample 19 is reflected by a mirror 17 via a lens 18 and is made into signal light 20. The signal light 20 is converted by a photodetector 21 to an electric signal which is then time averaged and is measured as a voltage or current.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a method for performing spectroscopic measurements of light emission commands of various materials, particularly semiconductors.

[Conventional technology]

Measurement of high-speed light emission processes using light pulses with a time width of In5 (10"" seconds) or less is extremely useful for gaining knowledge about the quality evaluation of various device materials, and has been rapidly gaining popularity in recent years. It is becoming popular. In such measurements, it is usually necessary to use a high-speed weak light detection system with a time width of less than ins.A typical example is the ps generated from a mode periodic laser or a pulsed current excited semiconductor laser as an optical pulse. There is a method that uses a laser pulse having a time width of (10'' seconds) and uses a streak camera or a single photon counting system as a high-speed weak light detection system.

Regarding such a high-speed luminescence measurement system and a specific method of luminescence measurement, see, for example, Fouq.
et), Burnham, IEEE Journal of Quantum Electronics (IEEE, J, QuantumEle)
ctronics) magazine, QE-22, 1986, 9
A detailed description can be found in the paper published on pages 1799 to 1810 of the issue. In this paper, approximately 250
A GaAs/AlGaAs quantum well sample is excited using a ps light pulse, and time-resolved measurement of light emission from this sample is performed using a high-speed photon counter.

[Problem to be solved by the invention]

The major problem with the above measurement method is that there is no photodetection element that is sufficiently fast and sensitive at wavelengths of 1 μm or more, so the wavelength range of the emitted light to be measured is 300 to 90 Or+m.
It is limited to a certain extent.

An object of the present invention is to provide a luminescence measurement method that eliminates the drawbacks of the conventional measurement methods described above and allows time-resolved measurement of luminescence with a long wavelength of 1 μm or more.

[Means to solve the problem]

In order to solve the above-mentioned problems, the time-resolved spectroscopy method provided by the present invention irradiates a light-emitting substance with two light pulses at the same location with a time delay between the light pulses, and detects the light emission from the light-emitting substance. The present invention is characterized in that a signal detected by time-integrating is measured while changing the time delay.

[Effect]

The present invention utilizes the fact that the intensity of light emitted from a light-emitting substance has nonlinearity with respect to the intensity of excitation light.

Such characteristics are observed in many materials, but are particularly noticeable in undoped semiconductors.

For example, in many cases where light emission decays rapidly at room temperature, the rate at which photoexcited carriers in a semiconductor are annihilated by non-radiative recombination is extremely large.

At this time, if the doping concentration is sufficiently small compared to the optically excited carrier density, the emission intensity has a characteristic that it is proportional to the square of the excitation light intensity. The same thing can be said about the time-integrated emission intensity when the excitation light is a short optical pulse.

Utilizing this characteristic, for example, if a light pulse that is split into two is overlapped again after a time delay between them, and the luminescence is observed by time integration when it is irradiated onto a semiconductor sample, the intensity will change over time. It is maximum when the delay is zero and decreases as the time delay increases. If the time width of the optical pulse is sufficiently shorter than the luminescence lifetime, the luminescence lifetime can be directly determined from the characteristics of intensity changes due to time delay. In this case, the important point is that since what is observed is the time-integrated emission intensity, a high-speed photodetector is not required. Therefore, even if a high-speed photodetector is used, it is possible to measure the temporal change in luminescence at a substantially high speed even at long wavelengths in the infrared region by using a highly sensitive photodetector.
When the emission intensity is proportional to the square of the excitation light intensity, if the light pulse is divided into two halves each with 1/2 intensity and the beams are superimposed on the sample surface to observe the emission, the time-integrated emission intensity I(τ ) is calculated to have the following relationship with I (τ) (1+exP (−l rτl ) ) with a time delay. Here, γ represents the reciprocal of the lifetime of light emission. The luminescence lifetime can be immediately determined from the measurement result of the dependence of the scattered luminescence intensity on τ. When the sample is a semiconductor, 1/γ indicates the lifetime of excited carriers.

〔Example〕

Embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows a schematic configuration of an embodiment of time-resolved spectroscopy to which the present invention is applied. The incident light pulse 11 is split into two parts each having half intensity by the beam splitter 12, and the split light pulses are vertically reflected by the mirrors 13 and 14, and each half of the intensity is reflected downward by the beam splitter 12 again. , and are superimposed along the optical path to form an excitation light pulse 15. At this time, mirror 13
Alternatively, by moving one of the mirrors 14 back and forth along the incident direction of the light pulses, a time delay 16 can be provided between the two superimposed excitation light pulses 15. The excitation light pulse 15 passes through a dichroic mirror 17 and is focused onto a sample 19 by a lens 18 . The light emitted from one sample is collated by a lens 18, reflected by a dichroic mirror 17, and becomes a signal light 20. This is converted into a "C" electrical signal by a photodetector 21, and then the electrical signal is converted into a voltage or current on a time average. It is measured as.

In this example, as the incident light pulse 11, a light pulse with a pulse width of about Loops and a wavelength of 830 nm from an AJ2GaAs diode laser driven by a pulse current was used, collimated with a lens. In addition, as sample 19, for the purpose of measuring infrared light with a wavelength of 1.5 to 1.6 μm,
An InGaAs thin film epitaxially grown on P was used. Furthermore, the photodetector 21 has a response speed of Lop.
A Ge photodetector with a wavelength of approximately 1.5 s was used.

FIG. 2 shows an example of measurement results using the configuration shown in FIG. 1, and a nonlinear interlayer curve 31 of emission intensity shows the time delay dependence of emission intensity from an InGaAs thin film. From this measurement result, it was found that the luminescence lifetime of the sample was about 1.2 ns.

In this example, a light pulse light source, a sample to be measured, and a photodetector were selected for the purpose of measuring light emission with a wavelength of 1.5 to 1.6 μm, but the present invention is not limited to this combination. It is clear that the excitation light pulse wavelength, its light source, and photodetector can be selected depending on the emission wavelength of various luminescent materials.

Furthermore, the nonlinearity of the emission intensity with respect to the excitation light intensity of the material does not need to be square dependence, and the present invention is applicable to all cases where the nonlinearity deviates from linearity. In addition, in general, the two excitation light pulses with a time delay do not need to be two parts of a light pulse from the same light source, and may have different wavelengths and intensities.
What is important when making measurements under these various conditions is that
It is necessary to analytically predict the change in luminescence intensity with respect to time delay and calculate the luminescence lifetime from the measured data based on this prediction.

〔Effect of the invention〕

As described above, by using the present invention, it becomes possible to measure the luminescence lifetime of a luminescent substance at high speed without requiring a high-speed photodetector.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a schematic configuration of an embodiment of time-resolved spectrometry to which the present invention is applied. Also, Figure 2 shows the first
InGaAsfillK measured by the configuration shown in the figure
FIG. 3 is a diagram showing the dependence of the intensity of light emitted from the cell on time delay. DESCRIPTION OF SYMBOLS 11... Incident light pulse, 12... Beam splitter, 13... Mirror, 14... Mirror, 15... Excitation light pulse, 16... Time delay, 17... Guicroic mirror 18... Lens, 19... Sample, 2
0...Signal light, 21...Photodetector, 31...Emission intensity nonlinear correlation curve.

Claims (1)

[Claims] A light emitting substance is irradiated with two light pulses at the same place with a time delay between the light pulses, and a signal detected by time-integrating the light emission from the light emitting substance is measured while changing the time delay. Characteristic time-resolved spectroscopy method.