Patent Application
20230003636
The present disclosure relates to the field of non-destructive contact-free characterization of a mechanical and/or physico-chemical property of a sample, by analysis of the acoustic wave produced by an interaction with pump and probe laser beams.
The technical field is more particularly that of photo-acoustics, in particular, the ultra-fast, non-destructive measurement of the mechanical, thermal, or optical properties of a sample (solid, liquid, or gaseous inert matter or living cell). “Ultra-fast measurement” means a measurement having a temporal resolution on the order of a picosecond.
The pump-probe method makes it possible to use very short laser pulses to measure ultra-fast phenomena in matter, such as the movement of atoms or the excitation of electrons.
For this, a very short and intense laser pulse, the “pump,” is sent to a sample to excite it. A second weaker pulse, the “probe,” is sent immediately after, which makes it possible to measure the effect of the initial excitation. By repeating and modifying the time between the first and the second pulse, it is thus possible to reconstruct the evolution of the excitation over time and to make a recording over a time window of a measurement of the light intensity in the photodetector or of the number of photons counted by the photodetector.
The absorption of a first so-called “pump” laser pulse causes sudden heating and therefore surface expansion, which, depending on the geometry of the object studied, either propagates in the form of an acoustic pulse lasting a few picoseconds or excites the resonant vibrations of the system. The acoustic pulses or vibrations are detected using a second time-delayed pulse (by optical method in the case of the homodyne only), “the probe,” by the interferometric measurement of the changes in the optical reflectivity coefficient induced by transient sound waves.
The field of picosecond acoustics has developed in recent years and has enabled the study of the elastic properties of thin films, multilayer systems, nanostructures and nanoparticles, the detection of stress and disorder phenomena at interfaces, adhesion measurement, etc.
Picosecond acoustics have also been used for the study of massive systems. At the same time, industrial applications in the field of non-destructive evaluation in the microelectronics industry have emerged and led to the development of commercial measuring devices.
In the prior art, the article “Measuring the mechanical properties of matter at the nanometer scale using picosecond acoustics” PHOTONIQUES No94, Nov. 1, 2018, page 30)33 XP055734848, the article by Allaoua Abbas “Development of a heterodyne pump-probe device” of May 9, 2013 XP055734712, as well as the thesis “Development of a heterodyne pump-probe device: application to picosecond acoustic imaging” defended by Allaoua Abbas on May 9, 2014 (https://tel.archives-ouvertes.fr/tel-00988758/document) are known.
This document describes the operating principle of a heterodyne bench. Two pump and probe pulse trains, issued from different lasers, are combined (RO) before being focused on the sample being studied by which they are reflected and then directed toward a photodetector using suitable optics. The rejection of the pump signal diffused by the sample is done either using polarizers, if the polarizations of the beams have been kept crossed, or using an interferometric filter, if the wavelengths of the pump and the probe are quite separated. Alternatively, a non-polarizing beam splitter may be used.
The difference Δf, called the beat frequency, between the repetition frequency of the probe fProbe and pump fProbe+Δf trains, increasingly delays the probe pulses with respect to the pump pulses. At each pump excitation, the pump-probe delay increases by the quantity δhet, the value of which is given by the relationship:
δhet=Tprobe−Tpump,
where τprobe and τpump are, respectively, the periods of the probe and pump lasers. δhet may also be expressed as a function of fprobe and Δf using the formula:
The delay increases until the pump and probe pulses again arrive at the same instant on the sample. This precise moment is called coincidence. The time between two coincidences is the time it takes for the probe to go through all the pump-probe delays. This time is called the beat period, and its value is given by the inverse of the beat frequency Δf. The transient signal obtained by studying the variation of the intensity of the probe collected by the photodetector during a beat period thus corresponds to the response of the sample dilated in time. This dilation of the time scale is similar to the stroboscopic effect, observable on a macroscopic scale, and which, under certain conditions, makes it possible to freeze or slow down movements of a periodic nature.
The relationship that binds the temporal scale of the dilated time τdiluted, which corresponds to the time scale of the acquisitions, and the physical time scale Tphysical, which corresponds to the true dynamics of the sample, is:
Two delays may be defined:
The interaction with the material results in the formation of acoustic waves with high spectral content, which allows the characterization of samples with nanometric dimensions by a photodetector. Differential detection makes it possible to partly overcome the fluctuation of the intensity of the probe laser. Moreover, this makes it possible to bring the levels of the average signals back to the neighborhood of 0 and to take advantage of the minimum step for quantifying the acquisition map.
This thesis also proposes a solution for imaging a surface of a sample by controlling the relative positions of the focal spots of the pump and the probe on the sample. This control makes it possible to mesh the space around an excitation in order to map the transient phenomena generated by the excitation.
Also known is the Neta TECH article “JAX-M1” January 2018 XP055735269 as well as the Japanese patent JP H05172737.
The French patent FR2892511 is also known, which describes a heterodyne optical sampling device equipped with two pulsed laser sources capable of emitting, respectively, a pump beam and a probe beam of respective repetition frequencies Fs and Fp with Fs≠Fp, an element combining the pump and probe beams intended to be sent to a sample and which comprises a signal channel including a system for the photodetection of the response signal of the sample and connected to this signal channel, and a system for acquiring the response signal. It is mainly characterized in that, as the acquisition system comprises an acquisition triggering element, it comprises, connected to this triggering element, a synchronization channel including a device for measuring the beat frequency IFs-Fpl.
Finally, the United States patent application publication US2004/196453 is known, which relates to a laser probe and a thermal pump to form a complete image used to detect defects under the surface.
The solution proposed in the prior art has multiple drawbacks resulting from the scanning used for the deflectometry and the generation of surface waves (measurement of the transverse properties of the material).
A first drawback, recognized in the document, is the difficulty of controlling the movements at such high frequencies, and of maintaining synchronization between the pulse trains and the photodetection.
A second drawback lies in the analysis times induced by these scans, which are indicated in the thesis as being able to reach several hours. Such durations are hardly compatible with industrial use, for example, for sample control in a production cycle.
Such durations are downright prohibitive for the analysis of samples having a very short lifespan, for example, an isolated living cell, the lifespan of which does not exceed a few minutes.
Finally, the point interaction leads to local information that does not take into account the mechanical, chemical, and physical links with the neighboring local areas and which, as a result, provide a type of characterization of a surface of the sample, which is both disturbed by the remanence of neighboring pulses and by the absence of consideration of the global characteristics of the analyzed surface.
In summary, the problem encountered by the person skilled in the art is to be able to construct one or more maps (1D in one dimension or 2D in two dimensions) representative of the physical properties (mechanical, thermal, or optical) of various samples, in time frames compatible with operational requirements such as: compliance with production line inspection rates, non-alteration of the sample under the transient conditions necessary for measurement, reduction of measurement times, increase in the quantities of data collected.
In order to remedy these drawbacks, the present disclosure relates, in its most general sense, to a method for the non-destructive contact-free physical characterization of a sample by repeated excitations of the surface of a sample with a sequence of laser pulses comprising at least one pump pulse by a first “pump” laser followed by a succession of L temporally offset pulses by a second “probe” laser, and the analysis of the beam emitted by the surface of the sample by an activated photodetector, for the acquisition of the signals delivered by the photodetectors during constant time windows,
Characterized in that
The duration of the laser pulses is between a picosecond and a femtosecond, and preferably on the order of a hundred femtoseconds.
Preferably, the digital processing comprises applying a transformation array MTR to the array of M×L signal values delivered by the photodetectors to determine the array Mcc.
According to an advantageous embodiment, the method includes steps for recalculating the array MTR through supervised learning.
According to a variant, the method includes the recording of a plurality of transformation arrays MTR, each corresponding to a particular physical characteristic.
According to another particular embodiment, the method further includes a step for the automatic optimization of the focus of the optics of the “pump” and “probe” beams comprised of controlling a sequence of variation of the focusing and measurement of a quality factor of the signal produced by the photodetector, and selecting the focusing corresponding to a maximization of the quality factor over all of the recorded values.
Another approach is measurement with a point pump centered or not on the probe line to allow the surface wave acquisition.
The present disclosure relates secondly to equipment for the non-destructive contact-free physical characterization of a sample comprising two pulsed laser sources for the emission, respectively, of a “pump” beam and a “probe” beam, as well as a detector characterized in that it further includes at least one device for shaping a beam to transform the distribution of the “pump” and “probe” beams into a uniform spatial distribution of the “top hat” type along N dimensions, N being equal to one or two, and in that the detector consists of an N-dimensional array of M photodetectors, each measuring the number of photons before and after the probe pulse, and before the following pump pulse, with M greater than 2.
According to a first variant, the pump and probe beams are coaxial in the zone of interaction with the sample.
According to a second variant, the pump beam is perpendicular to the plane of the zone of interaction with the sample, and the probe beam forms an angle other than 90° with the plane of the zone of interaction with the sample.
According to a particular embodiment, the equipment according to the present disclosure includes a computer for controlling the recording, for each sequence, of an array MPD of M×L signal values delivered by the photodetectors and for applying at least one digital processing to the array to establish a map of the zone of the sample analyzed by the detector in the form of an array MCC of the values of the physical characteristic observed for Q points of the analyzed zone, Q being between 1 and M, depending on at least one transformation array stored in a computer memory.
The present disclosure relates thirdly to a computer memory device for the customization of physical characterization equipment including a recording of a digital transformation array MTR to the array of M×L signal values delivered by the photodetectors to determine the array MCC.
Fourthly, the present disclosure relates to the application of the aforementioned method for N-dimensional mapping, with N equal to 1 or 2, of
Other features and advantages of the present disclosure will be evident from the following detailed description, given by way of non-limiting example and with reference to the accompanying drawings, in which:
The present disclosure combines the use of a pump beam for excitation and a probe beam for detection, which are generated by a synchronous or asynchronous system with a device for shaping the beams in a line or in a square array, then an acquisition means of the probe beam before and after disturbance by the sample. The acquisition is carried out by probe signal discretization of n points by a photodetector with an element array with subtraction from the signal after disturbance of the signal before disturbance, then by the digitization of the differential signal and the temporal reconstruction of the sample response.
The equipment for mapping a physical characteristic of a sample according to the present disclosure, described by way of example, comprises a system (100) of two laser sources producing two beams (110, 120), a pump beam (120) and a probe beam (110) offset temporally with possible adjustment of the delay.
The pump beam (120) has a form factor in a line or in a square array form, with a non-Gaussian form (“top hat beam”).
The probe beam (110) has a form factor in a line or in a square array form, identical to the pump beam.
A “top hat” type of beam exhibits an almost uniform fluence (energy density) in a circular disc. It is generally formed of diffractive optical elements from a Gaussian beam. It may be obtained by combining a beam profiler (“laser beam profiler”) with a picosecond laser or a femtosecond laser.
The system further includes a differential photodetector (200) with an element array (201 to 205) in line or in a square array form measuring at each of its points the difference of the signals before and after disturbance of the incident probe beam on the sample. It optionally allows, as a variant, the output of the respective average values of the outward and return signals in order to balance the respective averages of the outward and return signals on a single reference value. The electrical signals delivered by the elements (201 to 205) are transmitted to a pre-processing circuit (210) then to a digital analog digitizer (220) making it possible to record the response signals measured by the photodetector (200).
A preferred configuration illustrated in
In this case it is necessary to add a certain number of optical elements.
Similarly, when the angle between the probe (121 to 123) and pump (111 to 113) beams is 0°, it is necessary to add an optical element (130) allowing the beams to recombine with each other.
The system comprises a synchronous or asynchronous device (100) making it possible to generate two pulsed laser beams (110,120) offset temporally, a Probe beam (110) and a Pump beam (120). The wavelengths of the beams are independent; for the Probe beam (110) the wavelengths are comprised within the spectral band of the photodetector (200).
The Pump beam (120), coming from an asynchronous or synchronous device, passes through an optical device (125) for shaping the beam (120). At the output of the device, the beam has a shape adapted to the sensor array of the photodetector and a uniform distribution of energy over its entire surface (so-called “top hat” shape of the signal).
The Probe beam (110), coming from an asynchronous or synchronous device (100), passes through a sampling optic, which makes it possible to sample a fraction of power to send it to one of the element arrays of the differential photodetector (200) with an element array (201 to 205). Initially, the probe beam passes through an optical beam shaping device (115). At the output of the device, the beam has a shape adapted to the sensor array of the photodetector (200) and a uniform distribution of energy over its entire surface (so-called “top hat” shape of the signal).
Then, in the case of the configuration with a normal incidence of the probe beam (110) on the sample (150) illustrated in
Then, in the case of an angle of 0° between the pump (120) and probe (110) beams, the beams are recombined in a recombination optic making it possible to give them a single orientation on the sample (150).
Finally the two beams (110, 120) are focused on the sample via an optical focusing device. In the case of a configuration with non-normal incidence of the probe beam on the sample, the probe beam passes through two different optical focusing devices, the paths before and after the sample not being similar, the semi-reflecting optics are not used in this configuration.
The probe beam reflected or transmitted (151 to 153) by the sample (150) is then injected into the differential photodetector (200) with an element array (201 to 205). The element array (201 to 205) of the photodetector (200) discretizes the signal into n signals.
Afterwards, these signals are subtracted from the probe signal measured before reflection or transmission of the probe beam on the sample. The signals are conditioned and then digitized by the multi-channel A/D digitizer (220). In addition, the digitizer (220) is synchronized with the laser sources (100) to allow a temporal reconstruction of the n responses of the sample (150) by digital processing of the measurements.
In the case of a normal incidence of the probe channel on the sample (150), the probe beam reflected or transmitted (151 to 153) by the sample (150) is reflected by the semi-reflecting optics in the differential photodetector (200) with an element array. The element array (201 to 205) of the photodetector discretizes the signal into n signals. Then these signals are subtracted from the probe signal measured before reflection or transmission of the probe beam on the sample (150). The signals are conditioned and then digitized by the multi-channel A/D digitizer (220). In addition, the digitizer is synchronized with the laser sources to allow a temporal reconstruction of the n responses of the sample by digital processing of the measurements.
The differential photodetector (200) with an element array (201 to 205) comprises an array of photosensitive elements, which may range up to 64 elements distributed in a row or arranged in an 8×8 square. The spectral response of these photo elements (201 to 205) makes it possible to cover a spectral band ranging from 190 to 1700 nm. The functions covered by the photodetector are:
The probe (110) and pump (120) beams may be offset spatially on the sample (150) in order to measure transverse physical phenomena.
The scanner system may be made with two movable mirrors or by means of two lenses, the first of which is off-centered with respect to the second.
Photodetector
The array photodetector (200) may be equipped with channels for monitoring the optical powers making it possible to visualize the average optical power of the probe before and after sampling in order to balance the differential channels of the detector and to optimize the signals. The advantage of line/square sensors is that the elements have a common cathode.
The current passing through the cathode is the sum of all the currents generated by each photosensitive element (201 to 205). Current flowing through the resistor R (206) creates a voltage across its terminals. This voltage is amplified using an amplifier (207) to obtain a voltage proportional to the optical power on the in-line sensor. Only one monitoring will therefore be necessary for the line/square sensor.
The photodetector (200) is equipped with an array of n photosensitive elements on the return sample path and a single photosensitive element or an array of n elements on the sampling channel before the sample. The currents generated are then converted into voltage before obtaining the difference between the signals.
The signals may be multiplexed at the photodetector output in order to limit the number of digitizer channels.
Signal Processing
The digitized signals are recorded in a table made up of the values of each of the elements (201 to 205), for the different moments of the probe pulses, which may be represented in the form of a set of light intensity curves measured by the photodetector or the number of photons counted by the photodetector with respect to time. These have a maximum value corresponding to the zero offset between the pump pulse and the first probe pulse, then generally decreasing values.
These curves are processed to extract characteristic information such as singular points or the slope of certain segments.
This digital array is processed by a transformation array associating the digital values from the photodetector (200) with the values of the physical characteristic studied. This transformation array may be built empirically, or by supervised learning. It may be reassessed regularly as a function of the results of the measurements taken.
This transformation array may be recorded on a device to allow the customization of mapping equipment, for example, by access to an online memory or in the form of a physical memory that may be inserted into a connector provided for this purpose in the equipment.
Applications
The equipment according to the present disclosure is suitable for different applications:
| Number | Date | Country | Kind |
|---|---|---|---|
| FR1913490 | Nov 2019 | FR | national |
This application is a national phase entry under 35 U.S.C. § 371 of International Patent Application PCT/FR2020/052188, filed Nov. 26, 2020, designating the United States of America and published as International Patent Publication WO 2021/105622 A1 on Jun. 3, 2021, which claims the benefit under Article 8 of the Patent Cooperation Treaty to French Patent Application Serial No. FR1913490, filed Nov. 29, 2019.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/FR2020/052188 | 11/26/2020 | WO |
