JPWO2014208349A1 - Optical measuring apparatus and optical measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a signal having a very small intensity with high sensitivity by effectively removing noise. An optical measurement apparatus includes a light source, a first optical element that divides light from the light source into probe light and reference light, a first path that guides the probe light to a sample, and the reference on a time axis. A second path of the reference light whose optical path length is adjusted so as to give a relative delay between the light and the probe light, the probe light that irradiates the sample, and the reference light whose optical path length is adjusted Detecting a common detection element and outputting a detection signal, and applying a reverse sign to the signal derived from the probe light and the signal derived from the reference light among the detection signals. A balancer that performs balancing and outputs a difference between a signal derived from the probe light and a signal derived from the reference light as a measurement result of the sample.

Description

  The present invention relates to an optical measurement device and an optical measurement method with low noise and high sensitivity.

  Optical measurement is performed in which a sample is irradiated with a light beam and the state of a substance or biomolecule is measured without contact. Depending on the object to be measured, the obtained signal strength is very small and may be buried in noise.

  As a technique for detecting a minute signal buried in noise, there is a method called a lock-in detection method. In this method, a product of a trigonometric function of a set frequency (lock-in frequency) and a signal is taken, and after converting the signal of that frequency to a low frequency, a signal component is taken out by a low-pass filter. In this respect, lock-in detection effectively works as a narrow-band bandpass filter. In this method, noise observed at a frequency different from the lock-in frequency is removed, and a signal in the vicinity of the lock-in frequency is detected, thereby improving the signal / noise ratio. In addition, by increasing the time constant of the low-pass filter, the bandwidth can be narrowed and noise can be removed more effectively.

  The lock-in detection method is applied to detection of low-intensity probe light and detection of minute modulation applied to the probe light. This modulation may be applied directly to the probe light source or may be induced by a stimulus applied to the sample. However, in any case, noise observed near the lock-in frequency cannot be removed in principle. In addition, if the time constant is increased in order to reduce the bandwidth and improve the signal-to-noise ratio by the integration effect, there is a problem that the response to changes in the signal is delayed and high-speed detection cannot be performed.

  A dual beam method is known as a method for canceling long-time scale fluctuations included in probe light by using lock-in detection. In this method, light from a light source is divided into probe light used for measuring the sample and reference light that does not pass through the sample, and is modulated at different frequencies. The probe light after passing through the sample and the reference light are superimposed and incident on the same detector. The signal from the detector is obtained by separately obtaining the signal of the probe light and the reference light by lock-in detection at each modulation frequency, and taking the ratio. By using the dual beam method, it is possible to remove the influence due to the fluctuation of the long-time scale commonly included in the probe light and the reference light and the influence on the fluctuation of the sensitivity of the long-time scale of the detector. However, since the frequencies at which the probe light and the reference light are detected are different, high-speed fluctuation noise observed at the same frequency as the lock-in frequency cannot be removed. Further, when the transmittance of the sample changes during measurement, correction may be necessary each time. Therefore, this method is not suitable for an object whose transmittance changes at the time of measurement because high-speed measurement is required at a large number of measurement points.

  There is a balance detection method as a method for removing noise observed at the same frequency as the detection included in the probe light (see, for example, Patent Document 1 and Non-Patent Documents 1 and 2). In the balance detection method, light from a light source is divided into probe light and reference light. After the probe light is incident on a sample, each light is detected by two detectors and the difference is obtained. Since the noise is included in the probe light and the reference light in phase, the noise observed at the same frequency can be removed by taking the difference.

  However, the known balance detection methods have some difficulties. First, every time the transmittance of the sample or the sensitivity of the detection system changes, calibration is required. Secondly, the characteristics of the two detection systems, such as the frequency characteristics of the detector and circuit, need to be the same. Third, it cannot be applied to spectroscopic measurement using a white light source. This is because the intensity of white light fluctuates differently for each wavelength, and optical noise cannot be canceled if the spectral conditions such as the observation center wavelength and the observation wavelength width for the probe light and the reference light are different. Furthermore, it is necessary to prepare two spectrometers having completely the same characteristics for the probe light and the reference light, but this is technically difficult.

  A method of detecting a minute signal over multiple wavelengths by detecting white probe light transmitted through a sample with a multi-channel lock-in amplifier and observing a Raman spectrum is known (for example, Non-Patent Document 3). reference).

U.S. Pat.No. 5,134,276

Alessio Gambetta, Vikas Kumar, Giulia Grancini, Dario Polli, Roberta Ramponi, Giulio Cerullo, Marco Marangoni, Opt. Lett., 35 (2) (2010) 226-228 Philip C. D. Hobbs, Appl. Opt., 36 (4) (1997) 903-920 N. Ishii, et al, “Optical frequency- and vibrational time-resolved two-dimensional spectroscopy by real-time impulsive resonant coherent Raman scattering in polydiacetylene”, Physical Review, A 70 (2004) 023811.

  As described above, in the dual beam method using a single detector, the frequency at which noise is observed is different between the reference light and the probe light, and noise that fluctuates at high speed near the lock-in frequency cannot be removed. Further, in the balance detection method using two detectors, it is difficult to perform calibration and characteristic adjustment between two optical paths and two detection systems.

  Therefore, it is an object to provide an optical measurement technique and configuration capable of detecting a signal with low noise and high sensitivity using a single detector.

  In order to solve the above-described problem, a new balanced circuit is used to efficiently remove noise near the observation frequency and detect a desired signal with low intensity with high sensitivity.

Specifically, the optical measuring device according to the first aspect of the present invention is:
A light source;
A first optical element that splits light from the light source into probe light and reference light;
A first path for guiding the probe light to the sample;
A second path of the reference light in which an optical path length is adjusted so as to give a relative delay between the reference light and the probe light on a time axis;
A detector that outputs the detection signal by detecting the probe light that irradiates the sample, and the reference light, the optical path length of which has been adjusted, with a common detection element;
Among the detection signals, a signal derived from the probe light and a signal derived from the reference light are balanced by applying opposite signs to the signal derived from the probe light and the signal derived from the reference light, respectively. A balancer that outputs the difference between the two as the measurement result of the sample.

  Here, “equilibrium” means to balance the absolute value of the intensity of the signal derived from the probe light and the intensity of the signal derived from the reference light.

As a second aspect of the present invention, an optical measurement method is provided. The optical measurement method is
Split the light from the light source into probe light and reference light,
Irradiate the sample with the probe light,
Giving a relative delay between the reference light and the probe light on the time axis;
Detecting the probe light irradiated on the sample and the reference light given the delay with the same detection element,
Among the detected signals, the signal derived from the probe light and the signal derived from the reference light are balanced by applying opposite signs to the signal derived from the probe light, and the signal derived from the probe light A difference between signals derived from the reference light is acquired as a measurement result of the sample.

  With one detector and a balanced circuit, noise near the observation frequency can be efficiently removed, and a minute desired signal can be detected with high sensitivity. Combining this technique with multi-channel lock-in detection using white probe light enables low-noise, high-speed, high-accuracy spectrum detection.

It is a schematic block diagram of the optical measuring device of 1st Embodiment. It is a figure which shows the structural example of the balance device used with the optical measuring device of FIG. It is a flowchart of the optical measuring method of 1st Embodiment. It is a schematic block diagram of the optical measuring device of 2nd Embodiment. It is a flowchart of the optical measuring method of 2nd Embodiment. It is a schematic block diagram of the optical measuring device of 3rd Embodiment. It is a figure which shows the spectrum of the white probe light used in 3rd Embodiment. It is a schematic block diagram of the optical measuring device of 4th Embodiment. It is a figure which shows the structural example of the measurement system and control system of the optical measuring device of FIG. It is a figure which shows the image by the sum total of the signal of several channels. It is a figure of the induced Raman spectrum in a specific coordinate. It is a figure explaining the effect by the optical measurement of embodiment. It is a schematic block diagram of the optical measuring device of 5th Embodiment. It is a figure which shows the modification of FIG. It is a schematic block diagram of the optical measuring device of 6th Embodiment. FIG. 16 is a diagram illustrating an application example of the optical measurement device of FIG. 15 to guided Raman measurement. It is a figure which shows the effect of the noise cancellation using the reference light delayed by 1/4 period of intensity modulation light, and the effect of feedback control. It is a schematic diagram of the sample used by measurement. It is a figure which shows the effect of the induced Raman measurement to which the method of embodiment is applied. It is a figure which shows the effect of the feedback control of embodiment.

The optical measurement technique and configuration of the embodiment will be described below with reference to the drawings.
<First Embodiment>
FIG. 1 is a schematic configuration diagram of an optical measurement apparatus 10A according to the first embodiment. The optical measurement apparatus 10 </ b> A includes a light source 11, optical elements 12 and 15, a detector 18, a transimpedance amplifier (TIA) 31, and a balancer 30. In this example, the light source 11 is a pulse light source that emits single-wavelength pulse light, and the optical elements 12 and 15 are polarization beam splitters.

Pulse light L emitted from the light source 11, the polarizing beam splitter 12, and the probe pulse light L _P, it is divided into the reference pulse light L _R. Probe pulse light L _P that has passed through the polarizing beam splitter 12 passes through the sample 20 along the path P1, and is incident on the polarization beam splitter 15. On the other hand, the reference pulsed light _{L R} reflected by the polarization beam splitter 12, along path P2, and is reflected by the mirrors 13 and 14, enters the polarization beam splitter 15. The polarization beam splitter 15, and the probe pulse light L _P passed through the path P1, the reference pulse light L _R passing through the path P2 is incident on the detector 18 are superimposed spatially.

Path P2 of the reference pulse light L _R is about half of the delay of the pulse period is set to a length given. The reference pulse light _LR is incident on the polarization beam splitter 15 with a phase delayed by 180 ° with respect to the pulse repetition period. In the output light of the polarization beam splitter 15, the reference pulse light _LR is a probe on the time axis. appearing in between pulses intermediate position of the pulse light L _P. In this example, the pulse repetition frequency of the light source 11 is 76 MHz.

The probe pulse light L _P transmitted through the sample, such as molecular vibrations of the sample, and information I minute strength. Reference pulse light L _R and the detected probe pulse light L _P by the detector 18 is input balancer (the balance circuit) 30. An electric pulse synchronized with the pulsed light L is input to the balancer 30 and is used for canceling noise. The balancer 30 outputs the information I after the noise is canceled.

  FIG. 2 shows a configuration example of the balancer 30. The balancer 30 receives a pulse signal (including a probe pulse signal and a reference pulse signal) detected by the detector 18 and converted into a voltage by the TIA, and an electric pulse synchronized with the pulsed light.

  The balancer 30 includes a bandpass filter 32, a phase adjuster 33, a comparator 34, a multiplier 38, and an averaging circuit 41. The balancer 30 may optionally include a bias power source 36, a capacitor 35, a resistor 37, an active low pass filter 42, and an active high pass filter 43.

  The current output from the detector 18 such as a photodiode is converted into a voltage signal of an appropriate level by the TIA 31 and input to the multiplier 38. This voltage signal includes a probe pulse component and a reference pulse component. On the other hand, the electric pulse signal synchronized with the pulse light from the pulse light source 11 is converted into a 76 MHz sine wave synchronized with the pulse light by the 76 MHz band pass filter 32 and the phase adjuster 33, for example. This sine wave is further converted into a square wave (rectangular wave) by the comparator 34, and square waves of +1 and −1 are input to the multiplier 38.

The bias power source 36, the capacitor 35, and the resistor 37 are used when a bias is applied to the square wave generated by the comparator 34. In this case, the capacitor 35 and the resistor 37 decouple the bias voltage B generated by the bias power source 36 and the square wave. Bias, applied in order to adjust so that the intensity of the probe pulse signal and the reference pulse signal determined by the optical division ratio of the reference pulse light L _R and the probe pulse light L _P (intensity ratio) is equal (for equilibration) Is done. As shown in FIG. 1, when the pulsed light L whose polarization plane is inclined by 45 ° is incident on the polarization beam splitter 12, the intensity of the horizontal polarization component (probe pulse light L _P ) and the vertical polarization component (reference pulse light L _R ). Are equal, it is not necessary to apply a bias. As will be described later in the embodiments, it is effective to adjust the signal intensity ratio between the probe light and the reference light by applying a bias when using multi-wavelength detection light.

  The multiplier 38 multiplies the probe pulse signal and the reference pulse signal by a square wave. The probe pulse signal is multiplied by 1 ((1 + B) times when a bias voltage is applied), and the reference pulse signal is multiplied by -1 ((-1 + B) times when a bias voltage is applied) and output from the multiplier 38. . The probe pulse signal and the reference pulse signal whose signs are reversed are averaged by the averaging circuit 41 and canceled out. The averaging circuit 41 is a passive low-pass filter 41 composed of, for example, a capacitor and a resistor, and outputs a difference by averaging the probe pulse signal and the reference pulse signal that have opposite signs. The output signal is the information component I derived from the sample after the noise is canceled out. The output Vout of the averaging circuit 41 is expressed as follows.

Vout = (probe pulse signal) × 1 + (reference pulse signal) × (−1)
The output of the averaging circuit 41 is further output through an active low-pass filter 42 having a cutoff frequency of 10 kHz using an operational amplifier and an active high-pass filter 43 having a cut-on frequency of 400 Hz. The combination of the active filters 42 and 43 functions as a band pass filter with a gain of 100 times.

The optical measuring device 10A shown in FIGS. 1 and 2 has the following characteristics.
(a) dividing the pulsed light generated at the same time from the same light source 11 by dividing the reference pulse light L _R and the probe pulse light L _P, contained in the reference pulse light L _R and the probe pulse light L _P The nature of noise is the same.
(b) detection system because of the same probe pulse light L _P and the reference pulse light L _R, the frequency characteristics of the detection system are identical. Therefore, noise can be canceled including the frequency at which the signal is observed.
The use of (c) a square wave, even if the timing of the reference pulse light L _R and the probe pulse light L _P is slightly deviated from one period of 2 minutes of repetitive pulses, the maximum equilibrium is obtained. That is, the degree of cancellation is not sensitive to the optical path difference, and the optical system can be easily adjusted.

FIG. 3 is a flowchart of the optical measurement method of the first embodiment. Irradiating the pulse light (S101), it divides the pulsed light to the reference pulse light _{L R} and the probe pulse light _{L P} (S102). The probe pulse light _{L P} is incident on the sample to measure the sample (S103). The reference pulsed light L _R giving half of the delay of the pulse period (S104), spatially overlap the reference pulse light and probe pulse light (S105). The superimposed optical signal is detected and converted into an electrical signal (S106).

On the other hand, an electric signal synchronized with the pulse light source is generated (S111), and the phase is adjusted (S112) to be converted to ± 1 logic (for example, ± 1 square wave function) (S113). The detected and voltage-converted electrical signal is multiplied by the logic (S115). The multiplication results are averaged by time averaging or the like and output (S116). Thereby, the probe pulse light and the reference pulse light are canceled, and a minute information component (measurement result) included in the probe pulse light can be extracted with a good signal-to-noise ratio.
Second Embodiment
FIG. 4 is a schematic configuration diagram of the optical measurement apparatus 10B of the second embodiment. The optical measuring device 10B splits the light from the light source 11 into probe pulse light and reference pulse light to give a delay to the reference pulse light, and the probe pulse signal detected by the same detector 18 in the balancer 30. It is the same as the optical measurement apparatus 10A of the first embodiment in that a reference pulse signal is input and an electric pulse synchronized with pulse light is input.

Optical measuring device 10B is arranged a polarization beam splitter 15 in front of the sample 20, and that the combined probe pulse light L _P and the reference pulse light L _R is incident on the sample 20, the probe pulse from the stimulation unit (not shown) in that the stimulus S to synchronize with light L _P is applied to the sample 20, different from the optical measuring device 10A of the first embodiment.

  By applying the stimulus S to the sample 20, a physical / chemical reaction or molecular vibration occurs in the sample molecule. Since the stimulus S is synchronized only with the probe pulse light, the influence of the stimulus S is reflected only on the probe pulse light. By obtaining the difference between the probe pulse signal light and the reference pulse signal using ± 1 logic in the balanced circuit 30, it is possible to cancel out the noise of the light from the light source and extract only the information component due to the influence of the stimulus S. .

The optical path from the sample 20 to the detector 18, so identical reference pulsed light L _R and the probe pulse light L _P, it is possible to cancel the fluctuation factors other than stimulation almost completely. For example, even if the transmittance of the sample 20 changes during measurement, it is not affected. This is particularly effective when the present invention is applied to imaging (image formation) for analyzing the spatial distribution of a measurement target. Usually, the sample is optically non-uniform and the transmittance of the sample changes during imaging. When the optical path of the probe pulse light L _P and the reference pulse light L _R from the specimen 20 to the detector 18 is different from the intensity ratio of the reference pulse light L _R and the probe pulse light L _P is changed during measurement, a possibility that the equilibrium is broken There is. On the other hand, by using the same optical path as shown in FIG. 4, it is possible to cancel out noise components in a balanced manner and extract necessary information.

FIG. 5 is a flowchart of the optical measurement method of the second embodiment. Irradiating the pulse light (S201), it divides the pulsed light to the reference pulse light _{L R} and the probe pulse light _{L P} (S202). The reference pulsed light _{L R} giving half of the delay of the pulse period (S204), spatially overlap the reference pulse light _{L R} and the probe pulse light _{L P} (S205). With the superimposed pulsed light incident on the specimen, providing a synchronized stimulus probe pulse light L _P to the sample (S206). In composite pulse light of the probe pulse light L _P and the reference pulse light L _R, measuring the sample under the influence of the stimulus (S207). The pulsed light transmitted through the sample is detected and converted into an electric signal (S208).

On the other hand, an electric signal synchronized with the pulse light source is generated (S211), and the phase is adjusted (S212) to be converted into ± 1 logic (for example, ± 1 square wave function) (S213). The detected and voltage-converted electrical signal is multiplied by the logic (S215). The multiplication results are averaged using a time average or the like and output (S216).
<Third Embodiment>
FIG. 6 is a schematic configuration diagram of an optical measuring device 50 according to the third embodiment. In the third embodiment, balanced detection is applied to multi-channel lock-in detection in which multiple signals are locked in in parallel.

  The optical measuring device 50 corresponds to a white light source 51, optical elements 52 and 55, a spectroscope 58, a plurality of detectors 18-1 to 18-n (collectively referred to as “detector 18” as appropriate), and each detector. TIA31-1 to 31-n (collectively referred to as “TIA31” as appropriate), a plurality of balancers 30-1 to 30-n (referred to as “balancer 30” as appropriate), and a multi-channel lock-in amplifier 59. . The optical element 52 is a polarization beam splitter as in the first and second embodiments. The detector 18 is, for example, an avalanche photodiode (APD). The white light source 51 generates white pulse light including a plurality of continuous wavelengths using, for example, a pulse laser having a pulse repetition frequency of 76 MHz and a photonic crystal fiber (PCF). The center wavelength of the pulse laser is, for example, 800 nm.

  FIG. 7 shows a spectrum of white pulse light emitted from the white light source 51. The white pulse light has an intensity from 575 nm to 780 nm.

Returning to FIG. 6, the white pulse light and probe pulse light L _P at the polarization beam splitter 52 is split into the reference pulse light L _R. Light component transmitted through the polarization beam splitter 52, a probe pulse light L _P, incident on the polarization beam splitter 55. Light component reflected by the polarizing beam splitter 52, as the reference pulsed light L _R, is incident on the polarization beam splitter 55 via the mirror 53 and 54. The optical path length of the reference pulse light L _R is set to a length delayed a half period (phase 180 °) relative to the pulse repetition period of the white light source 51.

Reference pulse light _{L R} and multiplexed probe pulse light _{L P} by the polarization beam splitter 55, on the dichroic mirror DM, superimposed and 800nm pumping pulse light _{L S.} The pump pulse light L _S is branched from the light emitted from the pulse laser used in the white light source 51, for example, and is intensity-modulated by a chopper (not shown). Pump pulse light L _S is synchronized with the probe pulse light L _P, but the reference pulse light L _R of phase 180 °.

The probe pulse light L _P , the reference pulse light L _R , and the pump pulse light L _S are incident on the sample 20. By irradiation with the pump pulse light L _S , molecular vibrations according to the wavelength are generated and scattering occurs. The influence of the scattering intensity of the probe pulse light L _P to synchronize the pump pulse light L _S is modulated. On the other hand, the reference pulse light L _R of different pump pulse light L _S and the timing is not affected. Of the light transmitted through the sample 20, the pump pulse light L _S is removed by short-pass filter SPF, Puroparusu light L _P having received the reference pulse light L _R and the intensity modulation, via the lens 57 and the optical fiber 61 Spectroscopy Guided to instrument 58.

Spectrometer 58 disperses reference pulse light _{L R} and the probe pulse light _{L P,} respectively. The separated light of each wavelength is detected by detectors 18-1 to 18-n. The detection results are input to the corresponding balancers 30-1 to 30-n.

Each of the balancers 30-1 to 30-n has the configuration shown in FIG. 2, and receives a voltage signal of the TIA 31 and an electric pulse synchronized with white pulse light. The ± 1 logic at each balancer 30 probe pulse light L _P and the reference pulse light L _R is canceled, only the portion which receives the intensity modulation is extracted as a sample information by scattering. The output of each balancer 30 is lock-in detected by the multi-channel lock-in amplifier 59 at the modulation frequency of the pump pulse light.

  By using white probe light, balanced detection and multi-channel lock-in detection, the measurement speed is remarkably increased compared with a method of obtaining a spectrum by normal one-channel lock-in detection. If the multi-channel lock-in amplifier 59 is, for example, a 128-channel lock-in amplifier, measurement can be performed at a speed 128 times that of normal one-channel lock-in detection. White light is usually noisy, but by using the balancers 30-1 to 30-n, noise can be canceled at each wavelength and minute information can be detected with high sensitivity and high speed.

  As in the first and second embodiments, the pulse light generated from the same light source at the same time is divided into probe pulse light and reference pulse light, which are included in the probe pulse light and reference pulse light. Noise becomes the same, and the effect of balanced detection is maximized. Similarly to the first embodiment and the second embodiment, by making the optical path after the probe pulse light and the reference pulse light overlap each other, the single spectroscope 58 can split the light into each wavelength. .

  Furthermore, each balancer 30 can adjust ± 1 logic (sign-inverted square wave) by applying a bias (see FIG. 2), which is particularly effective for sample measurement using white light. When the bias level is B, the signal output Vout obtained by each balancer 30 for each wavelength is expressed as follows.

Vout = (probe pulse signal) × (1 + B) + (reference pulse signal) × (−1 + B)
When observing white light by spectroscopic observation, in order to obtain the maximum balance, optical splitting is required in which the intensity of the probe pulse light and the reference pulse light are the same at all wavelengths. This is difficult to achieve at 52. In the third embodiment, even when the optical division ratio varies depending on the wavelength included in the white light, the signal intensity of the probe pulse light and the reference pulse light can be corrected evenly by applying an appropriate bias by the balancer 30 at the subsequent stage. Therefore, the maximum balance can be obtained at all wavelengths.
<Fourth embodiment>
8 and 9 are schematic configuration diagrams of the optical measurement apparatus 100 according to the fourth embodiment. In the fourth embodiment, the equilibrium detection and multi-channel lock-in detection of the third embodiment are applied to a guided Raman microscope.

  There is a stimulated Raman scattering method as a method of observing a Raman scattering spectrum and intensity reflecting molecular vibration information. In the stimulated Raman scattering method, a sample is irradiated with synchronized pulse laser beams of at least two wavelengths having different wavelengths. When the difference in wavelength corresponds to the energy of molecular vibration, the intensity of light having a short wavelength decreases and the intensity of light having a long wavelength increases. A decrease in the intensity of light having a short wavelength is referred to as induced Raman loss, and an increase in the intensity of light having a long wavelength is referred to as induced Raman gain. Since molecular vibration differs depending on the substance, qualitative analysis can be performed using stimulated Raman scattering. Furthermore, since the intensity of induced Raman loss and induced Raman gain is proportional to the concentration of molecules, quantitative analysis is also possible. By introducing these lights into a microscope, sweeping the sample or incident light, and acquiring the stimulated Raman scattering intensity at each point of the sample, the spatial distribution of the stimulated Raman scattering intensity and the stimulated Raman spectrum can be obtained. This is called guided Raman imaging.

  As in the third embodiment, the probe pulse light detected for each wavelength is subjected to multichannel lock-in detection, so that the stimulated Raman scattering intensity at each wavelength is obtained, and the sample is made relatively to the incident light. Sweep to obtain an image by stimulated Raman scattering spectrum.

  In FIG. 8, the optical measurement apparatus 100 includes a white light source 110, polarization beam splitters 119 and 122, a stage 127 that drives the sample 20 relative to the light pulse, and a measurement system that measures light from the sample 20. 150.

  The white light source 110 includes a pulse laser light source 101 and a photonic crystal fiber (PCF) 116. The pulse laser light source 101 is composed of, for example, a titanium sapphire pulse laser. The titanium sapphire pulse laser outputs pulse light at a center wavelength of 800 nm, a pulse width of 2.5 ps, a pulse repetition frequency of 76 MHz, and an average output of 450 mW by a passive mode-locking operation using titanium sapphire as a gain medium, and is synchronized with the pulse light. A pulse electric signal is output.

  The output light of the pulse laser light source 101 is guided to the beam splitter 104 via the isolator 103. The isolator 103 blocks the return light from the subsequent optical element to the pulse laser light source 101 and stabilizes the operation of the pulse laser light source 101.

  The beam splitter (BS) 104 splits the pulsed light output from the pulse laser light source 101, reflects 40%, and transmits 60%. The pulsed light transmitted through the BS 104 is guided to the half-wave plate 114 through a concave lens 112 having a focal length of 60 mm and a convex lens 113 having a focal length of 120 mm. The lenses 112 and 113 adjust the beam diameter so as to fill the pupil plane of the objective lens 115. The plane of polarization is adjusted by rotating the half-wave plate 114 around the optical axis.

  The pulsed light whose polarization plane has been adjusted is incident on a microscope objective lens 115 having a magnification of 40 and a numerical aperture of 0.65. The pulsed light collected by the objective lens 115 is introduced into a polarization-maintaining PCF 116 having a length of 30 cm. The PCF 116 is made of silica, and the cross section has a structure in which regularly arranged voids surround the core. A region having a void has a refractive index substantially smaller than that of the core and functions as a cladding. In such a fiber, the difference in refractive index between the core and the clad is large, the incident light is confined in a narrow space region, the energy density of the incident light is increased, and a large nonlinear optical effect is obtained. Due to this large nonlinear effect, the wavelength of the incident light is converted into white light having an intensity in a wide range of wavelengths by the power of the pulse laser 101. If the direction of the polarization plane of the incident light is appropriate due to the polarization maintaining function of the PCF 116, linearly polarized light is output.

  White light generated by the PCF 116 is collimated by a microscope objective lens 117 having a magnification of 40 × and a numerical aperture of 0.65. At this time, the direction of the polarization plane of white light is adjusted by rotating the PCF 116. The collimated white pulse light L becomes the output of the white light source 110.

White pulse light L is split into the reference pulse light _{L R} and the probe pulse light _{L P} by the polarization beam splitter (PBS) 119. Reference pulse light L _R is the pulsed laser light source 101 pulse repetition minute delayed half cycle with respect to the period, i.e. after a long optical path in phase is delayed 180 ° min, enters the PBS122, spatial and probe pulse light L _P Is superimposed on. The PBSs 119 and 122 both transmit horizontally polarized light and reflect vertically polarized light. When polarized light whose plane of polarization is inclined by 45 ° is incident on PBS 119, the intensity of transmitted horizontal polarized light and reflected vertical polarized light are substantially equal. In PBS122, transmits the probe pulse light L _P is a horizontally polarized light, by reference pulsed light L _R is vertically polarized light is reflected, the reference pulse light L _R and the probe pulse light L _P is overlapped with a high throughput .

Reference pulse light _{L R} and the probe pulse light _{L P} passes through the short-pass interference filter 111. The short-path interference filter 111 blocks white light having a wavelength component of 780 nm or more and superimposes the modulated pump pulse light L _S on the same axis.

The intensity of the 40% pulse light reflected by the BS 104 is modulated (on / off) by the optical chopper 105 and used as the pump pulse light L _S. The pump pulse light L _S passes through a convex lens 107 having a focal length of 170 mm and a concave lens 108 having a focal length of 150 mm. Lenses 107 and 108, the beam diameter of the pumping pulse light L _S equal to the beam diameter of the white pulse light L.

The pump pulse light L _S is guided to the delay stage 109. Delay stage 109 includes two mirrors M1, M2 in parallel to slide by a micrometer (not shown), synchronized with the probe pulse light L _P by adjusting the optical path length of the pump pulse light L _S. The power of the pump pulse light L _S is adjusted by the neutral density filter 106. The signal-to-noise ratio increases as the power of the pump pulse light L _S increases, but the large power causes the sample 20 to be destroyed. Therefore, the neutral density filter 106 is used to adjust the power of the pump pulse light L _S to a level that does not destroy the sample 20 while maintaining a large signal-to-noise ratio.

Pump pulse light _{L S} is on a short-pass interference filter 111, it is superimposed on the probe pulse light _{L P} and the reference pulse light _{L R.} The short path interference filter 111 functions as a mirror for 800 nm light. Pump pulse light L _S and the probe pulse light L _P is the timing matches, stimulated Raman signal is obtained as the intensity modulation of the probe pulse light L _P. On the other hand, the reference pulse light L _R do not interact since the pump pulse light L _S and timing is deviated.

And the probe pulse light _{L P} superimposed, the reference pulse light _{L R,} and the pump pulse light _{L S} is transmitted through the convex lens 124 of the concave lens 123 and the focal length 200mm focal length 100 mm. The lenses 123 and 124 expand the beam diameter of the combined light so as to fill the pupil plane of the objective lens 126 at the subsequent stage.

  The combined light passes through the beam splitter 125 having a reflectance of 8% and a transmittance of 92%, and then enters the objective lens 126. The beam condensed by the objective lens 126 irradiates the sample 20 fixed to the stage 127. The stage 127 is, for example, a piezo-driven stage, and sweeps the sample 20 with respect to incident light during measurement.

The beam reflected by the beam splitter 125 passes through the neutral density filter 129 and then forms an image on the CCD camera 131 by the imaging lens 130. The CCD camera 131, and the pump pulse light _{L S} of on the sample 20, to confirm the overlap of the probe pulse light _{L P} and the reference pulse light _{L R.} The neutral density filter 129 adjusts the intensity of the image formed on the CCD camera 131.

The light transmitted through the sample 20 is collimated by the condenser lens 132. The collimated light is guided to the short pass filter 133, and the pump pulse light L _S among the light included in the collimated light is blocked. Collimated light pumping pulse light _{L S} is removed is condensed by the convex lens 134 of focal length 100 mm, it is input to the measurement system 150 via a multimode optical fiber 135.

FIG. 9 shows a connection relationship between units of the optical measurement apparatus 100 with the measurement system 150 as the center. For convenience of illustration, the optical system up to the lens 134 in FIG. Optical signal obtained by the stimulated Raman microscope 160 (including a reference pulse light L _R and the probe pulse light L _P) is introduced into the spectrometer 136 by the multi-mode optical fiber 135. In the spectroscope 136, the light from the multimode optical fiber 135 is irradiated onto the diffraction grating (not shown) in parallel by a collimating mirror (not shown), and the light dispersed by the diffraction grating is bundled by the camera mirror (not shown). An image is formed on the end face of the fiber 137. The spectroscope 136 has a focal length of 300 mm and a diffraction grating engraving density of 1200 g / mm.

  The bundle fiber 137 is a bundle of 16 vertical fibers and 128 horizontal optical fibers, and divides the split light beam into 128. Every 16 bundle fibers 137 are bundled into one optical fiber 138.

  128 optical fibers 138 corresponding to 128 wavelengths (only one is shown for convenience of illustration), avalanche photodiodes (APD) 139-1 to 139-n each biased to 150V (this example) Then, n = 128). The APD 139 has a function (action) for amplifying a weak detection signal by itself by avalanche breakdown.

The probe pulse signal and the reference pulse signal detected by each APD 139 and converted into a voltage by the TIA 141 are input to the corresponding balancer 140. Each balancer 140 has the same configuration as the balancer 30 of FIG. The balancer 140 multiplies the probe pulse signal and the reference pulse signal by a square wave whose bias is adjusted, and the signs of the probe pulse signal and the reference pulse signal are opposite to each other. The probe pulse signal with the opposite sign and the reference pulse signal are added by a low-pass filter (see FIG. 2), and a difference is obtained. Signal output from the balanced device 140, the reference pulse light L _S and the probe pulse light L _P is canceled each other, it is placed only information representing the effect of stimulated Raman scattering. The timing at which the probe pulse signal and the reference pulse signal are reversed is controlled by an electric pulse signal synchronized with the pulse laser beam supplied from the pulse laser light source 101.

  The signals output from each of the balancers 140-1 to 140-n are connected to the input of the multi-channel lock-in amplifier 142. The multi-channel lock-in amplifier 142 is formed by connecting four 32-channel lock-in amplifiers to 128 channels, for example.

  The optical measurement apparatus 100 includes a chopper controller 144, a piezo stage controller 148, and a computer 146 as a control system. A reference signal for detecting lock-in is supplied from the chopper controller 144 to the multi-channel lock-in amplifier 142. The chopper controller 144 controls power supply to the optical chopper 105 (see FIG. 8). Further, upon receiving a signal of the rotation speed of the optical chopper 105, the rotation speed is feedback-controlled.

  The computer 146 controls the multi-channel lock-in amplifier 142, collects data from the multi-channel lock-in amplifier 142, and outputs a control signal to the piezo stage controller 148.

  The piezo stage controller 148 receives a control signal from the computer 146 and controls an actuator (not shown) of the piezo stage 127. While the output signal from the multi-channel lock-in amplifier 142 is collected by the computer 146, the sample 20 is swept by the piezo stage 127 to obtain a stimulated Raman spectrum at each measurement point. The multi-channel lock-in amplifier 142 and the piezo stage 127 are controlled by software installed in the computer 146.

  FIG. 10 is a stimulated Raman loss image obtained by the optical measurement apparatus 100. FIG. 10A shows an image obtained by summing the signals of the 40th to 50th channels, and FIG. 10B shows an image obtained by summing the signals of the 55th to 65th channels.

  As the sample 20, a film prepared by dissolving polystyrene (PS) and polymethyl methacrylate (PMMA) in toluene at a weight ratio of 1: 1 and dropping and drying the solution on a slide glass was used. The lock-in detection time constant is 1000 ms, the pixel size is 0.5 × 0.5 μm, and the image size is 20 × 20 μm. The lighter the shade, the higher the signal intensity, and the darker, the smaller the signal intensity.

  The signals of the 40th to 50th channels in FIG. 10A mainly show PS contribution. The signals of the 55th to 65th channels in FIG. 10B mainly show the contribution of PMMA. The images in FIGS. 10A and 10B are created from the same data obtained by one imaging. 10A and 10B show different contrasts, and it is possible to analyze that the concentration ratio of PS and PMMA varies depending on the position in the film.

  FIG. 11 is a diagram showing stimulated Raman spectra at different pixel positions. By using the optical measurement apparatus 100 shown in FIGS. 8 and 9, it is possible to acquire a stimulated Raman loss spectrum at each measurement point without sweeping any wavelength of the pump pulse light and the probe pulse light.

  The solid line in FIG. 11 is the spectrum at (X, Y) = (9, 6) pixels (the positions indicated by white triangles in FIGS. 10A and 10B), and the broken line is (X, Y). = Spectrum at (27,33) pixels (positions indicated by black triangles in FIGS. 10A and 10B). Each spectrum is a spectrum in which the spectrum of PS and the spectrum of PMMA overlap, but the intensity ratio of the Raman band derived from PS and the Raman band derived from PMMA is different. In the solid line, the intensity ratio of the Raman band derived from PS is large, and in the broken line, the intensity ratio of the Raman band derived from PMMA is large. Thus, by irradiating the sample with white pulse light, the states of different substances can be measured at once.

  FIG. 12 is a diagram showing the effect of the present invention. Through the first to fourth embodiments, the light from the same light source is divided into probe pulse light and reference pulse light, the reference pulse light is delayed by half the pulse repetition period of the light source, and the probe pulse light transmitted through the sample. The reference pulse light to which the delay is given is detected simultaneously, the sign of the signal of the probe light or the reference light is inverted by multiplication with ± 1 logic by the balancer 30, and the balance process is performed by the bias. In addition, noise can be reduced and a desired signal can be obtained with high sensitivity.

In FIG. 12, the balance 30 is connected to the lock-in amplifier, the signal intensity obtained by directly chopping only the probe pulse light (left column), and the signal obtained by simultaneously chopping the probe pulse light and the reference pulse light. Intensity is shown (right column). The intensity in the case of only the probe pulse light is 7.31 × 10 ⁻¹ V, whereas it becomes 3.01 × 10 ⁻³ V when the reference pulse light and the probe pulse light are incident simultaneously. The signal intensity after equilibrium is reduced to 1/243 compared to the case of probe pulse light alone. This means that noise included in the probe pulse light is also detected as 1/243. When the stimulus S is given to the sample as in the second embodiment, or when the pump pulse light is used as in the third and fourth embodiments, the stimulus and the pump pulse light are synchronized only with the probe pulse light. Intensity modulation by stimulation or induced Raman effect occurs only in the probe pulse light, and does not occur in the reference pulse light. By taking the difference between the probe pulse light and the reference pulse light by the balancer 30, only the influence of the stimulus S and the induced Raman signal can be detected with high sensitivity.
<Fifth Embodiment>
In the first to fourth embodiments, the pulsed light from the light source is divided into the probe light and the reference light, and the reference light is delayed by a half cycle of the pulse repetition, and then the optical signal is transmitted by the common photodetector. The detected signal was multiplied by a square wave of ± 1 to cancel the noise.

  In the fifth embodiment, a sine wave is used for the balancing process for noise cancellation. After splitting the intensity-modulated light of the light emitted from the same light source at the same time into the probe light and the reference light, and adding a time difference corresponding to about 1/4 of the intensity modulation period to the probe light and the reference light, It detects with a detection element. The detected signal is multiplied by a synchronization signal that is synchronized with the light from the light source and phase-adjusted. By adjusting the phase of the synchronization signal, the difference between the detection signal of the probe light and the detection signal of the reference light is minimized (equilibrium) to cancel the optical noise, and only the desired signal component is extracted.

FIG. 13 is a schematic diagram of an optical measurement apparatus 200 according to the fifth embodiment. Periodic intensity modulation is given to the light from the light source 201 by the intensity modulator 202 driven by the synchronization signal source 207. Light given intensity-modulated is divided into reference light L _R and the probe light L _P to be used in the sample measured by the beam splitter 203. The reference light L _R is given a delay corresponding to a quarter period of intensity modulation by an additional optical path after the beam splitter 203. Delay provided to the reference light L _R, since the reference light L _R and the probe light L _P need be distinguished, not be a quarter period strictly intensity modulation may be in the vicinity thereof. The probe light L _P and the reference light L _R are overlapped using the beam splitter 206 and are incident on the sample 210.

The probe light L _P and the reference light L _R after passing through the sample 210 are converted into an electric signal by the common light detection element 211 and amplified to an appropriate size by the preamplifier 212. For convenience, if the detection signal of the probe light L _P is A cos (wt), the detection signal of the reference light L _R is B sin (wt). Here, A is an amount proportional to the intensity of the probe light L _P , B is an amount proportional to the intensity of the reference light L _R , and w is an angular frequency of the synchronization signal. The optical noise is represented by fluctuations of A and B, but the probe light L _P and the reference light L _R divide the light at the same time, and therefore fluctuate in proportion to each other. Alternatively, the ratio of A and B is determined by the characteristics of the optical element and is constant. The electrical signal detected by the light detection element 211 is
A cos (wt) + B sin (wt)
It is expressed as This detection signal is input to the balancer 230 after amplification. The balancer 230 includes a multiplier 231, a phase shifter 232, and a low pass filter 233. The output of the preamplifier 212 is connected to the input of the multiplier 231. On the other hand, the phase of the synchronization signal output from the synchronization signal source 207 is adjusted by the phase shifter 232, and is input to the multiplier 231 and multiplied by the detected signal. For convenience, the synchronization signal input to the multiplier 231 is
C cos (wt + f)
It expresses. Here, f is the phase of the synchronization signal and is adjusted by the phase shifter 232. Therefore, the output from the multiplier 231 is
(1/2) AC {cos (2wt + f) + cos f} + (1/2) BC {sin (2wt + f) &#8211; sin f}
It becomes. This signal is input to the low pass filter 233, and only the low frequency component is output. As a result, the output signal of the balancer 230 is proportional to A cos f &#8211; B sin f. That is, when the phase f is appropriately selected, it can be seen that the difference between the detection signal of the probe light L _{P and} the detection signal of the reference light L _R is output.

Here, when f is adjusted so that cos f: sin f = B: A, the output of the balancer 230 becomes zero. Such f is tan ^-1 (A / B) + 2np (n is an integer). That is, even when the ratio of A and B is not equal due to the characteristics of the optical element, the output of the balancer 230 can be made zero by adjusting f, and can be balanced. This adjustment can cancel the optical noise even when the ratio of A and B varies.

On the other hand, a periodic stimulus is given to the sample 210 by the stimulus source 208 driven by the synchronization signal source 207. The phase of the stimulus is controlled to be in phase with the probe light L _P so that only the probe light L _P is subjected to intensity modulation induced by the stimulus. Since the effect of the stimulus modulates only the intensity A of the probe light L _P , the low-pass filter 233 outputs only the signal induced by the stimulus in which the optical noise has been canceled.

In this way, even if the detection signal does not contain a direct current component, it is possible to realize correction of the intensity ratio of the detection signal of the probe light and the reference light. By adjusting the phase of the synchronization signal with the light source 201 and multiplying the light detection signal, it is possible to cancel the optical noise of the probe light L _P and the reference light L _R and extract only the desired stimulus signal.
<Modification>
FIG. 14 is a schematic configuration diagram of an optical measuring device 300 which is a modification of the fifth embodiment. In FIG. 14, in addition to the configuration of FIG. 13, a spectroscope 311 is inserted immediately before the light detection element 211, and a resonator 312 that resonates with intensity modulation of the light source is inserted immediately after the light detection element 211. The configuration, arrangement, and functions of the other components are the same as those in FIG.

  As described in connection with FIG. 13, the configuration of FIG. 13 can make the intensity ratio of the detection signal of the probe light and the reference light constant even if the detection signal does not contain a DC component. With this feature, the resonator 312 can be used (a DC component is not included in the signal from the resonator).

  Since the resonator 312 does not include a load resistor that causes thermal noise, the thermal noise (electrical noise) is eliminated, and only a specific frequency component of the light detection signal is extracted and supplied to the balancer 230. It is possible to prevent mixing of frequencies that are not related to measurement, such as external noise, and to improve the signal to noise ratio.

Also in FIG. 14, after the probe light L _P and the reference light L _R are overlapped, they pass through a common optical path and are detected by a single photodetector 211, so the spectrometer is common and can be easily applied to spectroscopic measurement. It is.

The configuration of FIG. 14 cancels out optical noise due to fluctuations in the intensity ratio of the probe light L _P and the reference light L _R while reducing electrical noise using a resonator, so that the signal-to-noise ratio is maximized. can do.
<Sixth Embodiment>
FIG. 15 is a schematic diagram of an optical measurement apparatus 400 according to the sixth embodiment. Also in FIG. 15, the spectroscope 311 is disposed immediately before the light detection element 211, and the resonator 312 is disposed immediately after.

In the sixth embodiment, only the probe light L _P is guided to the sample 210, and the reference light L _R bypasses the sample 210. The reference light L _R that does not pass through the sample 210 and the probe light L _P that passes through the sample 210 are spatially overlapped and detected by a single light detection element 211, and a detection signal is input to the balancer 430. Further, the sample is irradiated with pump light (stimulation), the intensity of the pump light is modulated by the frequency of the reference signal, and the output of the balancer 430 is detected by lock-in at the frequency of the reference signal.

  As a feature of the sixth embodiment, the phase shifter 232 is feedback-controlled based on a direct current component included in the output of the multiplier 231. With this feedback control, even if the detection intensity ratio of the probe light and the reference light changes due to the sample state change due to sample scanning etc., the unnecessary state change is eliminated and only the intensity ratio change induced by the stimulus is taken out. Can do.

  The phase feedback control will be described more specifically. The balancer 430 includes a multiplier 231, a first low-pass filter 433, a second low-pass filter 434, an integrator 435, and a phase shifter 232. The second low-pass filter 434 extracts a DC component from the output signal of the multiplier 231, and the phase shifter 232 is driven by the integrated signal.

The output of the reference signal source 451 is input to the intensity modulator 452 and the lock-in detector 450. The stimulation signal from the stimulation source 208 is synchronized with the probe light L _P , is subjected to intensity modulation by the intensity modulator 452, and then enters the sample 210.

The frequency of the reference signal is set sufficiently higher than the frequency of the state change of the sample 210 due to imaging or the like. The cutoff frequency of the first low-pass filter 233 is set to be larger than the frequency of the reference signal. The cutoff frequency of the integrator 435 is set smaller than the frequency of the reference signal and larger than the frequency of the state change of the sample 210. In other words, the magnitude relationship of each frequency is
(Intensity modulation frequency of light source)> (cutoff frequency of first low-pass filter 433)
> (Reference signal frequency)> (Integrator 435 cutoff frequency)> (Sample state change frequency)
It becomes.

  Since the cut-off frequency of the first low-pass filter 433 is higher than the frequency of the reference signal (that is, the stimulus signal) and higher than the frequency of the state change of the sample 210, information on the state change induced by the stimulus of the sample 210 is included. The signal passes through the first low-pass filter 433 and is obtained as the output of the balancer 430. By detecting lock-in of this output at the frequency of the reference signal, a signal induced by stimulation can be obtained.

Since the cutoff frequency of the integrator 435 is smaller than the frequency of the reference signal, the output from the integrator 435 includes almost no signal induced by the stimulus. However, since it is higher than the frequency of the state change of the sample 210, information on the state change of the sample is included. That is, the signal from the integrator 435 does not include a change in the intensity A of the probe light L _P induced by the stimulus, but A and B (intensities of the reference light L _{R) that} change due to a change in the state of the sample 210 and other disturbances. ) Difference information is included. Based on the output signal of the integrator 435, the phase shifter 232 is feedback controlled so that the difference between A and B becomes zero.

Even if the ratio changes of A and B in such a state change of the disturbance or the sample, the slow feedback control does not negate until the signal changes of the stimulation by the state change, the detection signal intensity of the reference light L _R and the probe light L _P The ratio can be made constant. As a result, it is possible to observe the state change of the sample based on the signal induced by the stimulus while canceling the optical noise.

In the fifth embodiment (FIGS. 13 and 14), the sample 210 is irradiated with both the probe light L _P and the reference light L _{R in} order to prevent a change in the ratio of A and B due to a change in the state of the sample 210. in the sixth embodiment (FIG. 16), is corrected by feedback control of the change in the ratio of a and B according to the state change of the sample 210, it is not necessary to irradiate the reference light L _R on the sample 210.

Since the reference light L _R need not be applied to the sample 210, thereby preventing damage to the sample 210 by the reference light L _R. Therefore, the intensity of the light source 201 can be further increased. Further, optical loss due to irradiation of the sample 210 can be reduced. As a result, it is possible to increase the detection signal intensity and improve the signal-to-noise ratio of thermal noise, external noise, and photocurrent shot noise.
<Development examples>
FIG. 16 shows an example in which the configuration of the sixth embodiment is applied to stimulated Raman measurement.

In FIG. 16, a pulse laser is used as the light source 501. In this example, the pulse repetition frequency is 76.3 MHz, and the optical path length corresponding to a quarter period is about 1 m. The pulsed light is split by a beam splitter 505, and one light component is used as a stimulus source called pump light L _S. The intensity of the pump (pulse) light L _S is modulated by the chopper 552 and irradiated onto the sample 210. The pump light L _S is adjusted by an appropriate optical path so that the sample 210 is irradiated simultaneously with the probe light (pulse light) L _P.

The other pulse light component is incident on a photonic crystal fiber (PCF) 502. The PCF 502 converts the monochromatic pulse laser into white pulse light. The white pulse light passes through the polarizer 503 and is then divided into the probe light L _P and the reference light L _R by the polarization beam splitter 504.

The polarizer 503 is inserted to fix and adjust the polarization direction of white light. In general, since the split ratio of the beam splitter 504 changes depending on the polarization direction of incident light, if the polarization direction fluctuates for each pulse, the ratio of A to B fluctuates for each pulse, and optical noise cannot be canceled. Therefore, the polarization direction is fixed by the polarizer 503 immediately before the beam splitter 504 in order to suppress the variation in the division ratio for each pulse. The polarization beam splitter 504 transmits horizontally polarized light and reflects vertically polarized light. For example, if the polarization direction of the polarizer 503 is 45 °, the split ratio of the probe light L _P and the reference light L _R is 1: 1. The probe light L _P is incident on the sample 210, and information on molecular vibration is placed as intensity modulation by the induced Raman effect of the pump light L _S and the probe light L _P. The reference light L _R is spatially overlapped with the probe light L _P on the polarization beam splitter 507 after being given a delay corresponding to a quarter period of pulse repetition by an additional optical path. When the polarization beam splitters 504 and 507 are used for dividing and superposing white light, since the probe light L _P is horizontal polarization, it passes through the polarization beam splitters 504 and 507 with high transmittance. Since the reference light L _R is vertically polarized light, it is reflected with a high reflectance. Therefore, the optical loss can be suppressed smaller than that of a normal beam splitter.

The probe light L _P and the reference light L _R are spectrally separated by the common spectrometer 311, and an appropriate wavelength is introduced into the common light detection element 211. In FIG. 16, for the sake of illustration, only a single light detection element 211 is depicted. However, if a plurality of light detection elements 211 are prepared for each wavelength (see FIG. 9), a spectrum obtained by simultaneous multi-wavelength measurement can be obtained. it can.

Figure 17 shows the effect (feedback control of the balancer 430) of optical noise canceling effect and feedback of the probe light L _P measured in the system of FIG. 16. The horizontal axis is obtained by standardizing the difference between the intensity A of the probe light L _{P and} the intensity B of the reference light L _R with A, and represents an optical balance shift. The vertical axis is a signal obtained by normalizing the signal obtained without chopping light with A, and represents the noise in terms of modulation factor. Without noise, the signal strength is zero when there is no signal and no modulation is applied. It is due to optical noise that the signal is observed even though it is not chopped or subjected to stimulation-induced modulation.

  An asterisk indicates a measurement result of detecting only the probe light without noise cancellation. Square marks indicate measurement results when noise cancellation is performed by the balancer 430 but feedback control is not performed. Circles indicate measurement results when noise cancellation and feedback control are performed by the balancer 430.

As parameters for data acquisition, the frequency of the reference signal was 4.5 kHz, the cutoff frequency of the first low-pass filter 433 was 10 kHz, and the cutoff frequency of the integrator 435 and the second low-pass filter 435 was 1 kHz. The sample 210 was not inserted, and the intensity ratio between the probe light L _P and the reference light L _R was adjusted by the angle of the polarizer 503. When the reference light _LR is cut off and the probe light L _P is directly chopped with the feedback function disabled, the intensity A of the probe light L _P can be obtained. When the probe light L _P is blocked and the reference light L _R is directly chopped, the intensity B of the reference light L _R is obtained.

The star in the figure is a signal obtained by blocking the reference light without passing through the sample 210, and indicates the optical noise of the probe light L _P. Modulation translated at about ^{1.5 × 10 -4 (Hz) -} 1/2 of the optical noise is observed. At the measurement points indicated by the square marks, the feedback function is disabled, but by adjusting the polarizer 503 so that the value on the horizontal axis is zero, that is, A = B, 6.4 × 10 ⁻⁶ (Hz) ^-1 It can be seen that the noise is canceled up to ^{/ 2} . The lower limit of the noise is determined by the thermal noise of resistors in the circuit, noise due to active elements, and shot noise of photocurrent. The signal-to-noise ratio due to these noises is improved by increasing the incident light intensity of the light source 201.

  When the ratio of A and B is destroyed by adjusting the polarizer 503, the optical noise cannot be sufficiently canceled regardless of whether A is larger than B or B is larger than A. Noise is observed.

  At the measurement points indicated by circles, a feedback function is used, and even when the intensity ratio between the probe light and the reference light is broken, the optical noise is sufficiently canceled. Disabling the feedback function increases optical noise as the intensity ratio collapses (squares), but enabling the feedback function compensates for changes in the intensity ratio even if the intensity ratio of A and B collapses. , The optical noise is canceled out to the maximum.

  18 to 20 show measurement results when the optical measurements of the fifth and sixth embodiments are applied to a multi-wavelength simultaneous measurement guided Raman microscope. 18A is a top view of the sample 600 used, and FIG. 18B is a side view. A sample 600 in which polystyrene (PS) spheres 602 having a diameter of 4 μm are embedded in a polyvinyl alcohol (PVA) film 601 is produced.

The solid line in FIG. 19A indicates the spontaneous Raman spectrum of polystyrene (PS), and the broken line indicates the spontaneous Raman spectrum of PVA. PS has a Raman band with a peak at 3054 cm ⁻¹ and PVA has a Raman band with a peak at 2914 cm ⁻¹ . When the wave number difference between the pump light and the probe light is set near 3050 cm ⁻¹ , a signal mainly derived from PS is obtained, and when the wave number difference is set near 2915 cm ⁻¹ , a signal mainly derived from PVA is obtained.

FIG. 19B and FIG. 19C show imaging results when noise cancellation is performed and feedback is not performed, respectively. At the start of measurement, the intensity A of the probe light and the intensity B of the reference light are set to A = B, and the sample 600 is swept to obtain an image. The signal intensity is smaller as the color is black, and the signal intensity is greater as the color is white. FIG. 19B shows an image obtained at a wavelength of probe light resonating with a Raman band of 3050 cm ⁻¹ , and a strong signal is obtained in a disk shape where the PS sphere 602 exists. FIG. 19C is an image obtained at the wavelength of the probe light resonating with the 2915 cm ⁻¹ Raman band, and the signal intensity of the portion where the PS sphere 602 does not exist is almost uniformly strong. Since the PVA 601 is excluded where the PS sphere 602 exists, the signal intensity decreases in a disk shape.

For comparison, FIG. 19D shows an imaging result when the noise cancellation of the embodiment is not performed. Since the reference light is blocked at a wavelength that resonates with the Raman band of 3050 cm ⁻¹ , the optical noise of the probe light is not canceled out. In this case, the signal is buried in the optical noise of the probe light, and an image of the PS sphere 602 cannot be obtained. Further, although not shown in the figure, the same applies to a wavelength resonating with a Raman band of 2915 cm ⁻¹ . In the embodiment, it is understood that optical noise cancellation is achieved at the same multi-wavelength while passing through the spectroscope by performing the balancing process using the reference light.

FIGS. 19 (e) and 19 (f) show imaging results when feedback control is performed in addition to noise cancellation. FIG. 19 (e) is an image obtained with the wavelength of the probe light resonating with the Raman band of 3050 cm ⁻¹ , and FIG. 19 (f) is obtained with the wavelength of the probe light resonating with the Raman band of 2915 cm ⁻¹ . It is an image. The same contrast as in FIGS. 19B and 19C is obtained. By performing the feedback control, it is understood that the optical balance is automatically adjusted and the optical noise is canceled out. Furthermore, the principle that the stimulus by the pump light is observed by appropriately setting the feedback response so that the signal induced by the stimulus is not canceled by the feedback is demonstrated.

  20 (a) and 20 (b) are cross-sectional views taken along X = 4.25 mm in FIGS. 19 (b) and 19 (e), respectively. This figure shows the effect that the collapse of the ratio between the intensity A of the probe light and the intensity B of the reference light being measured is automatically corrected by feedback control. FIG. 20A shows a case where no feedback is applied, and the signal intensity is increased due to stimulated Raman scattering in the region where the polystyrene (PS) sphere 602 exists. However, although the PS sphere 602 has a spherical shape, the signal intensity greatly fluctuates depending on the location. This is because the optical noise of the probe light is not sufficiently canceled because the condition of A = B set by optical adjustment at the start of measurement is disturbed by passing through the PS sphere 602.

  On the other hand, FIG. 20B shows a case where feedback is applied, and the signal intensity is large due to stimulated Raman scattering in the region where the PS sphere 602 exists. Moreover, a large fluctuation in the signal intensity in the region where the PS sphere 602 exists as seen in FIG. 20A is not observed. This is because in FIG. 20B, the disturbance of the condition of A = B due to transmission through the PS sphere 602 is automatically corrected by feedback, and the optical noise of the probe light is canceled to the maximum. . The principle that the collapse of A = B (or intensity ratio) during measurement is automatically corrected by the phase adjustment method using feedback control of the sixth embodiment has been demonstrated.

  This application claims its priority based on Japanese Patent Application No. 2013-135212 filed on June 27, 2013, and is incorporated herein by reference in its entirety. Incorporate.

10A, 10B, 100, 200, 300, 400, 500 Optical measuring device 11, 201 Light source 18, 18-1 to 18-n, 211 Detector 12, 15, 52, 55, 119, 122, 504, 507 Polarized beam Splitter (optical element)
20, 210 Sample 30, 230, 430 Balancer 34 Comparator (square wave generator)
38,231 Multiplier 51, 110 White light source 58, 311 Spectrometer 59 Multi-channel lock-in amplifier 127 Stage 150 Measurement system 202, 252 Intensity modulator 232 Phase shifter 435 Integrator 450 Lock-in detector 451 Reference signal reduction 501 Pulse light source L _P probe pulse light _{L R} reference pulse light _{L S} stimulation or pumping pulse light

Claims (12)

A light source;
A first optical element that splits light from the light source into probe light and reference light;
A first path for guiding the probe light to the sample;
A second path of the reference light in which an optical path length is adjusted so as to give a relative delay between the reference light and the probe light on a time axis;
A detector that outputs the detection signal by detecting the probe light that irradiates the sample, and the reference light, the optical path length of which has been adjusted, with a common detection element;
Among the detection signals, a signal derived from the probe light and a signal derived from the reference light are balanced by applying opposite signs to the signal derived from the probe light and the signal derived from the reference light, respectively. A balancer that outputs the difference between the measurement results of the sample,
An optical measuring device comprising: The balancer is
A multiplier for multiplying the detection signal by a square wave having an amplitude of ± 1 synchronized with the light from the light source;
The optical measurement apparatus according to claim 1, wherein the difference is obtained from the multiplication result. The balancer is
A bias source for applying a bias to the square wave;
Further comprising
The optical measurement apparatus according to claim 2, wherein a detection intensity ratio between the probe light and the reference light is adjusted by the bias. The balancer is
A phase shifter for adjusting the phase of a sine wave signal synchronized with the angular frequency of light from the light source;
A multiplier for multiplying the detection signal by the sine wave signal having the adjusted phase;
The optical measurement apparatus according to claim 1, wherein the difference is obtained from the multiplication result.   The balancer includes a feedback system that feeds back a DC component of the output of the multiplier to the phase shifter, and uses the feedback to balance the detection intensity ratio of the probe light and the reference light. The optical measuring device according to claim 4. The resonator according to claim 4, further comprising a resonator inserted between the detector and the balancer, wherein a frequency component of a specific band of the detection signal is input to the balancer. Optical measuring device. A stimulus source for generating a stimulus in synchronization with the probe light;
Further comprising
The sample receives the stimulation in synchronization with the probe light,
The optical measuring apparatus according to claim 1, wherein the balancer outputs a change in the state of the sample induced by the stimulus as the measurement result. A reference signal source for generating a reference signal;
A lock-in detector that locks in the output of the balancer at the frequency of the reference signal;
Further comprising
The optical measurement apparatus according to claim 7, wherein the intensity of the stimulus is modulated by a frequency of the reference signal. A spectrometer disposed between the sample and the detector;
Further comprising
The light source is a white light source;
The first optical element divides light from the light source into white probe light and white reference light,
The second optical element guides the sample by superimposing the white probe light and the white reference light,
The white probe light and the white reference light transmitted through the sample are split into a plurality of wavelength components by the spectrometer.
2. The optical measurement according to claim 1, wherein the detector is arranged corresponding to each of the dispersed wavelengths, and detects the probe light and the reference light having a corresponding wavelength component in common. apparatus. Split the light from the light source into probe light and reference light,
Irradiate the sample with the probe light,
Giving a relative delay between the reference light and the probe light on the time axis;
Detecting the probe light irradiated on the sample and the reference light given the delay with the same detection element,
Among the detected signals, the signal derived from the probe light and the signal derived from the reference light are balanced by applying opposite signs to the signal derived from the probe light, and the signal derived from the probe light Obtaining a difference between signals derived from the reference light as a measurement result of the sample,
An optical measurement method.   11. The balancing includes a process of multiplying a signal detected by the detection element by a square wave having an amplitude of ± 1 synchronized with light from the light source to which a DC bias is applied. An optical measurement method as described in 1.   The optical measurement method according to claim 10, wherein the balancing includes a process of multiplying a signal detected by the detection element by a phase-adjusted sine wave synchronized with light from the light source.