WO2015156782A1

WO2015156782A1 - Spectral detection for imaging system including a micromirror array

Info

Publication number: WO2015156782A1
Application number: PCT/US2014/033336
Authority: WO
Original assignee: Pandata Research Llc
Priority date: 2014-04-08
Filing date: 2014-04-08
Publication date: 2015-10-15

Abstract

An optical system includes a micromirror array, conjugate to a sample and disposed in incident and/or return optical paths to/from the sample. Micromirrors in the array are imaged onto corresponding lateral locations on the sample. The micromirrors are modulated at unique assigned frequencies. Imaging optics produce a modulated image of the sample at a multi-pixel detector. The image is dispersed in one dimension by a spectrally dispersive element in the return optical path. The detector pixels produce time-varying electrical signals. Lock-in amplifiers or a Fast Fourier Transform module spectrally analyze the time-varying electrical signals. Signal amplitudes at the assigned frequencies determine how much light from a particular lateral location at the sample is present at each detector pixel. The signal amplitudes along the spectral direction, referenced to the lateral location, provide a wavelength spectrum of the sample at the particular lateral location.

Description

SPECTRAL DETECTION FOR IMAGING SYSTEM INCLUDING A MICROMIRROR ARRAY

The present disclosure relates generally to an optical detection system that converts a spectrally dispersed, modulated image of a sample into electrical signals, and from the electrical signals extracts a wavelength spectrum of the sample, as a function of lateral location on the sample.

BACKGROUND

Samples can be optically analyzed by directing light onto the sample, collecting light reflected from the sample, directing the collected light onto a two-dimensional multi-pixel detector, and analyzing the collected light. For measurements of the sample that vary as a function of two parameters, the multi- pixel detector can often produce electrical signals that allow the sample to be characterized all at once. For example, measuring a reflectivity of a sample as a function of lateral directions x and y can include forming an image of the sample on the multi-pixel detector, and recording the electrical signal outputs from the detector pixels. Such a reflectivity measurement can be taken essentially in real time, with reflectivity values being generated all at once from the multi-pixel detector.

An example of a system that performs measurements as a function of three parameters, rather than two parameters, is a pushbroom spectrometer. A typical pushbroom spectrometer measures intensity of a scene as a function of x- dimension, y-dimension, and wavelength. The pushbroom spectrometer images one thin slice of a scene onto a two-dimensional multi-pixel detector. The optics in the pushbroom spectrometer map values of x-dimension intensity and wavelength onto the two dimensions of the detector. The values of y-dimension intensity are obtained by scanning the scene past the spectrometer. Typical pushbroom spectrometers do not perform measurements of the scene all at once. SUMMARY OF THE DISCLOSURE

A system can optically characterize a sample by directing light onto the sample, and analyzing light reflected from the sample. The system can use modulation techniques to link a particular measurement to a particular lateral location on the sample. The system can use confocal techniques to restrict the analysis to one depth below the surface of the sample. The system can use spectrometer techniques in the return optical path to analyze the wavelength characteristics of the reflected light. The modulation, confocal, and

spectrometer techniques can be used for any suitable illumination wavelength, and any suitable physical principle upon which the measurements are based.

In some examples, an optical system includes a micromirror array, conjugate to a sample and disposed in incident and/or return optical paths to/from the sample. Micromirrors in the array are imaged onto corresponding lateral locations on the sample. The micromirrors are modulated at unique assigned frequencies. Imaging optics produce a modulated image of the sample at a multi-pixel detector. The image is dispersed, e.g., spectrally blurred, in one dimension by a spectrally dispersive element in the return optical path. The detector pixels produce time- varying electrical signals. Lock-in amplifiers or a Fast Fourier Transform module spectrally analyze the time- varying electrical signals. Signal amplitudes at the assigned frequencies determine how much light from a particular lateral location at the sample is present at each detector pixel. The signal amplitudes along the spectral direction, referenced to the lateral location, provide a wavelength spectrum of the sample at the particular lateral location.

Advantageously, the optical system can perform measurements of a sample, all at once, as a function of x-dimension, y-dimension, and wavelength spectrum.

The optical system can use any suitable illumination wavelength. In some cases, the measurements are first performed using a first illumination wavelength, then are performed a second time using a second illumination wavelength different from the first illumination wavelength. Particular measurements, taken at one or more wavelengths, can provide information that can be used to determine a characteristic of the sample. For instance, the measurements can be used to determine a concentration of a particular substance in the sample, such as glucose.

The optical system can use any suitable physical principle upon which the measurements are based. In one configuration, the measurements rely solely on reflectance at the illumination wavelength, where the reflected spectrum is substantially equal to an illuminating spectrum. In another configuration, the measurements rely on Raman scattering, which can produce shifts in wavelength upward and downward of about 10-100 nm. The downward shift in wavelength from Raman scattering is often very faint, and may not be observable under normal conditions. In another configuration, the measurements rely on fluorescence, which can produce shifts in wavelength upward of about 50-200 nm. In general, the amount of shift in wavelength depends on the excitation wavelength; the numerical ranges above correspond to excitation in the near-IR or visible wavelengths.

The actual wavelength(s) selected for the light source(s) can be based on a particular spectroscopic feature of a substance to be measured, low background absorption (e.g., in air or water, or other materials), and suitability for the physical principle upon which the measurements are based. For instance, the near infrared range of 700 nm to 800 nm can work well with Raman scattering, the near infrared range of 1.5 microns to 4 microns can have low background scattering and absorption, and the mid-infrared range of 8 microns to 12 microns can show good spectroscopic features. Other suitable wavelengths can also be used.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 A is a block diagram of an example of an optical system having a micromirror array in the incident optical path.

FIG. IB is a schematic drawing of the optical system of FIG. 1 A.

FIG. 2A is a block diagram of an example of an optical system having a micromirror array in the return optical path.

FIG. 2B is a schematic drawing of the optical system of FIG. 2A. FIG. 3A is a schematic drawing of a portion of the optical system of FIGS. 2A-2B, including a spectrally dispersive element disposed in the return optical path after the second imaging optics.

FIG. 3B is a schematic drawing of a portion of the optical system of FIGS. 2A-2B, including a spectrally dispersive element disposed in the return optical path before the second imaging optics.

FIG. 4A is a block diagram of an example of a confocal optical system having a micromirror array in both the incident and return optical paths.

FIG. 4B is a schematic drawing of the optical system of FIG. 4A. FIG. 5 is a schematic drawing of a portion of the confocal optical system of FIGS. 4A-4B, including an incident optical path, for light focused to a specified depth.

FIG. 6 is a schematic drawing of a portion of the confocal optical system of FIGS. 4A-4B, including a return optical path, for light focused to a specified depth.

FIG. 7 is a schematic drawing of a portion of the confocal optical system of FIGS. 4A-4B, including an incident optical path, for light focused to an erroneous depth.

FIG. 8 is a schematic drawing of a portion of the confocal optical system of FIGS. 4A-4B, including a return optical path, for light focused to an erroneous depth.

FIG. 8A shows an example of a configuration that can produce a zero- averaged electrical signal, which oscillates between negative and positive voltages.

FIG. 9 is a schematic drawing of an example of an electrical pathway compatible with the optical systems of FIGS. 1, 2, and 4.

FIG. 10 is a schematic drawing of a portion of the electrical pathway of FIG. 9, including a bandpass filter.

FIG. 11 is a schematic drawing of a portion of the electrical pathway of FIG. 9, including a plurality of lock-in amplifiers.

FIG. 12 is a schematic drawing of a portion of the electrical pathway of FIG. 9, including a Fast Fourier Transform module. FIG. 13 A is a schematic drawing of an example of a micro mirror array, in which each micromirror in the array is switched at a unique assigned frequency.

FIG. 13B is a schematic drawing of another example of a micromirror array, in which some adjacent micromirrors in the array are switched, together, at unique assigned frequencies.

FIG. 13 C is a schematic drawing of an example of a micromirror array, in which groups of adjacent micromirrors in the array are switched, together, at unique assigned frequencies within the single octave.

FIG. 14 is a schematic drawing of examples of a scene, a modulated scene, and a dispersed modulated scene for the optical systems of FIGS. 1, 2, and 4.

FIG. 15 A is a schematic drawing of a further example of a scene.

FIG. 15B is a schematic drawing of an example of a modulated scene, based on the scene of FIG. 15 A.

FIG. 15 C is a schematic drawing of an example of a dispersed modulated scene, based on the modulated scene of FIG. 15B.

FIG. 15D is an example of plots of extracted wavelength spectra for two locations within the scene of FIG. 15 A.

FIG. 16 is a flow chart of an example of a method of operation for an optical system.

DETAILED DESCRIPTION

FIG. 1 A is a block diagram of an example of an optical system 100 that can optically characterize a sample 190. The sample 190 is not part of the optical system 100. FIG. IB is a schematic drawing of the optical system 100 of FIG. 1A.

The optical system 100 includes a micromirror array 110. Micromirrors in the array 110 are switchable between first and second positions. The micromirror array 110 is configured to simultaneously reflect light from micromirrors of the micromirror array 110 that are in the first position. In the example of FIG. IB, the micromirror array 110 includes four micromirrors in a linear configuration; in practice, the micromirror array 110 can be two- dimensional, with micromirrors numbering in the hundreds or thousands on a side of the micro mirror array 110.

Driving electronics are configured to drive the micromirror array 110. In some example, the driving electronics are provided by a computer 180; in other examples, the driving electronics may be provided by one or more dedicated signal generators. The driving electronics direct trigger signals 182 to a set of one or more micromirrors in the array 110 to switch the set of micromirrors at unique assigned frequencies. The trigger signals 182 can have unique frequencies, which can optionally lie within a single octave. The trigger signals 182 are periodic, such as square waves (with a 50% duty cycle), rectangular waves (with a duty cycle other than 50%), sine waves, or other suitable waveforms. Switching of the micromirrors is typically initiated at zero- crossings of the trigger signals 182, so that signals having fast-rising and/or fast- falling edges are preferable. In the example of FIG. IB, the four unique assigned frequencies are denoted as fl, f2, f3, and f4.

First imaging optics 130 are disposed in an optical path between the micromirror array 110 and the sample 190. The first imaging optics 130 and the micromirror array 110 are arranged so that each micromirror in the array 110 is imaged onto a corresponding lateral location 192 at a specified depth at or below a surface of the sample 190. In other words, the micromirror array 110 and the specified depth are conjugate to each other, with respect to the first imaging optics 130. In the example of FIG. IB, the micromirrors modulated at assigned frequencies 184 denoted as fl, f2, f3, and f4 are imaged onto corresponding lateral locations 192 denoted as LI, L2, L3, and L4. The first imaging optics 130 can include one or more lenses, one or more mirrors, or a combination of at least one lens and at least one mirror. In some configurations, the computer 180 controls the spacing between the micromirror array 110 and first imaging optics 130, and/or the spacing between first imaging optics 130 and the sample 190, so that measurements can be performed at a selectable depth or at a series of depths. In some configurations, the sample 190 can be mounted at an angle with respect to the incident optical path. Such an angled mounting can ensure that the micromirrors in the array 110 are imaged to a plane that is parallel to a top surface of the sample under test 190. Such an angled mounting may be used when the micro mirrors in the array 110 are located at different distances from the first imaging optics 130, as in FIG. IB.

A light source 140 is configured to provide an incident light beam along an incident optical path. The incident optical path extends from the light source 140 to the first imaging optics 130 to the sample 190. The light source 140 can include one or more collimating or focusing elements, such as a lens or curved mirror. Examples of suitable light sources can include a single semiconductor laser, multiple semiconductor lasers having the same wavelength, multiple semiconductor lasers having different wavelengths, a single light emitting diode, multiple light emitting diodes having the same wavelength, multiple light emitting diodes having different wavelengths, one or more quantum cascade lasers, one or more superluminescent light sources, one or more amplified spontaneous emission sources, any combination of the above, or other suitable light sources. The computer 180 can also control the light source 140, including functions such as switching the light producing element on or off, adjusting an intensity of the light producing element, switching from one wavelength to another wavelength, and other suitable functions. In the configuration of FIGS. 1 A and IB, the micromirror array 110 is disposed in the incident optical path between the light source 140 and the first imaging optics 130, and arranged so that the micro mirrors in the array 110 are simultaneously illuminated from the light source 140.

A multi-pixel detector 150 is configured to receive a return light beam reflected from the sample 190 along a return optical path. The return optical path extends from the sample 190 to the first imaging optics 130 to the multi- pixel detector 150. The multi-pixel detector 150converts the return light beam to at least one electrical signal.

A spectrally dispersive element 160 is disposed in the return optical path between the first imaging optics 130 and the detector 150. The dispersive element 160 deflects light along a spectral direction, the deflection being wavelength-dependent. Examples of suitable dispersive elements can include one or more prisms, one or more diffraction gratings, or other suitable dispersive elements. In the configuration of FIGS. 1A and IB, a beamsplitter 142 is disposed in the incident and return optical paths adjacent to the first imaging optics 130. The beamsplitter 142 transmits light in the incident optical path, from the micromirror array 110 to the first imaging optics 130, and reflects light in the return optical path, from the first imaging optics 130 to the dispersive element 160. In other configurations, the transmitted and reflected portions may be switched, so that the beamsplitter 142 reflects incident light and transmits return light.

In the configuration of FIGS. 1A and IB, the micromirror array 110 and the sample 190 are conjugate to each other, with respect to the first imaging optics. In the configuration of FIGS. 1A and IB, the sample 190 and the multi- pixel detector 150 are conjugate to each other, with respect to the first imaging optics 130.

FIG. 2A is a block diagram of an example of another optical system 200 that can optically characterize a sample 290. FIG. 2B is a schematic drawing of the optical system 200 of FIG. 2A. Elements numbered 2xx in optical system 200 are similar in structure and function to corresponding elements lxx in optical system 100 of FIGS. 1A and IB.

One difference between the optical systems 100 and 200 is the location of the respective micromirror arrays 110, 210. In FIGS. 1A and IB, the micromirror array 110 is in the incident optical path; in FIGS. 2A and 2B, the micromirror array 210 is in the return optical path.

Another difference between the optical systems 100 and 200 is that the optical system 200 of FIGS. 2A and 2B includes second imaging optics 270 in the return optical path between the micromirror array 210 and the detector 250. The second imaging optics 270 are positioned so that the micromirrors in the micromirror array 210 are imaged onto the detector 250. In the configuration of FIGS. 2A and 2B, the micromirror array 210 and the sample 290 are conjugate to each other, with respect to the first imaging optics 230. In the configuration of FIGS. 2A and 2B, the micromirror array 210 and the multi-pixel detector 250 are conjugate to each other, with respect to the second imaging optics 270.

FIGS. 3A and 3B show two possible configurations for the return optical path in the optical system of FIGS. 2A and 2B. In FIG. 3A, the spectrally dispersive element 360A is disposed in the return optical path between the second imaging optics 370A and the multi-pixel detector 350A. In FIG. 3B, the second imaging optics 370B are disposed in the return optical path between the spectrally dispersive element 360B and the multi-pixel detector 350B. Both configurations produce dispersion, or spectral blurring, in one dimension at the multi-pixel detector 350A, 350B.

FIG. 4A is a block diagram of an example of another optical system 400 that can optically characterize a sample 490. FIG. 4B is a schematic drawing of the optical system 400 of FIG. 4A. Elements numbered 4xx in optical system 400 are similar in structure and function to corresponding elements 2xx in optical system 200 of FIGS. 2A and 2B.

Compared with the optical system 200 of FIGS. 2A and 2B, the beamsplitter 442 and light source 440 are repositioned to be in the optical path between the micromirror array 410 and the second imaging optics 470. As such, the micromirror array 410 is disposed in both the incident optical path, between the light source 440 and the first imaging optics 430, and the return optical path, between the first imaging optics 430 and the multi-pixel detector 450. Because light in the optical system 400 reflects twice from the micromirror array 410, once before and once after reflecting from the sample 490, the optical system 400 can provide the benefits of confocal detection. In contrast, the systems 100, 200 are non-confocal. For confocal configurations, the unique assigned frequencies can all be within a single octave.

Confocal detection provides signals arising from a specified depth at or below the surface of a sample, while rejecting signals arising from depths away from the specified depth. The optical system 400 uses the micromirror array 410, and its modulation at the unique assigned frequencies, to keep only signals arising from the lateral locations 492 at the specified depth in the sample 490. The confocal effects are explained further in FIGS. 5-8. FIGS. 5-6 show incident and return optical paths for light reflected from the specified depth, which contributes to a desired signal. FIGS. 7-8 show light reflected from an erroneous depth, which is ultimately excluded from the desired signal. In order to more clearly show the confocal effects, FIGS. 5-8 omit the beamsplitters and spectrally dispersive elements from the optical paths. In practice, a measurement system can measure wavelength spectrum, instead of or in addition to a wavelength-integrated reflectivity of a sample.

FIG. 5 shows a portion of the incident optical path, for light focused to a specified depth 594. The light rays in FIG. 5 strike a mirror in the micromirror array 510 having modulation frequency f3 ; similar rays exist for the other micromirrors on the array 510.

Initially, prior to striking the micromirror array 510, the incident light has an intensity 550 that is shown in FIG. 5 as being constant over time. In practice, the intensity may vary slowly over time, but is effectively time- invariant on the time scale of the modulation frequencies.

The micromirror modulates incident light with modulation 552, so that light reflected from the micromirror is the product of the intensity 550 with the modulation 552. In this example, the modulation frequency is f3, so that the period of the time- varying modulation is I / O. The intensity of the light, both before and after first imaging optics 530, is intensity 554. The first imaging optics 530 bring the incident light to a focus at the specified depth 594 within the sample 590. A portion of the incident light does reach the erroneous depth 596, but is not shown in FIG. 5.

FIG. 6 shows a portion of the return optical path, for light returning from the specified depth 694 within the sample 690. Light also does return from the erroneous depth 696, but is not shown in FIG. 6.

The sample 690 has a reflectivity R at the corresponding lateral location at the specified depth 694. The intensity 656 of reflected light has an amplitude that is proportional to reflectivity R. The intensity 656 is also modulated with assigned frequency f3, which corresponds to the lateral location.

Because the specified depth 694 is conjugate to the micromirror array 610, the light returning from the specified depth completely retraces its optical path through the optical system, through the first imaging optics 630, to the micromirror array 610. The returning light strikes micromirror f3 for a second time, which continues to have a modulation 658 at frequency f3. Light in the system can be well approximated as traveling infinitely fast in the system, so that when it strikes micromirror f3 a second time, the micromirror is still in the on position, and the light reflects to the detector. This second modulation of the light at frequency f3 , therefore, has no effect on the intensity, and the intensity 660 that reaches the detector has a modulation frequency of f3, and an amplitude that is proportional to the reflectivity R of the sample 690 at the lateral location corresponding to frequency f3 and at the specified depth 694. The spectrum 662 of the light reaching the detector shows a peak corresponding to frequency f3, and an amplitude of the peak being proportional to reflectivity R. The rays and spectrum shown in FIG. 6 are for just one micromirror; in practice, the spectrum 662 shows multiple peaks, each arising from a corresponding micromirror or set of micro mirrors in the array 610.

Whereas FIGS. 5-6 show the optical paths for light that contributes to the desired signal, FIGS. 7-8 show corresponding optical path for light that is ultimately excluded from the desired signal. FIG. 7 shows a portion of the incident optical path, for light focused to a specified depth 794 but arriving slightly out-of-focus to erroneous depth 796 within the sample 790.

The intensity 764 of the incident light, the modulation 766 of the micromirror in the array 710, the first imaging optics 730, and the intensity of the modulated incident light 768 are the same as respective elements in FIG. 5.

FIG. 8 shows a portion of the return optical path, for light returning from the erroneous depth 896 within the sample 890. The reflectivity of the sample 890 at the erroneous depth 896 is denoted as R' in FIG. 8. The intensity 870 of reflected light, from the sample 890, is modulated at frequency f3 with an amplitude proportional to R' .

The first imaging optics 830 are configured to image the specified depth 894 onto the micromirror array 810. As a result, the erroneous depth 896 is imaged to a location away from the micromirror array 810. FIG. 8 shows rays coming to a focus 809 between the micromirror array 810 and the first imaging optics 830. (FIG. 8 is not drawn to scale, with the separation between the focus 809 and the micromirror array 810 being exaggerated for clarity.)

Because the returning light focuses away from the micromirror array 810, the returning light forms a spot on the micromirror array 810 that extends laterally over more than one micromirror. The size of the spot is proportional to the distance away from the specified depth, so that the area of the spot increases quadratically with distance. The spot size can exceed the micromirror size by a factor of hundreds or thousands. As a result, the fraction of light that returns to the same mirror is reduced by the factor of hundreds or thousands, and a large fraction of the return light is directed onto micromirrors that differ from the micromirror in the incident path. This large fraction of return light, which reflects off one mirror in the incident path and reflects off a different mirror in the return path, is excluded from the desired signal.

In the example of FIG. 8, the light ray under consideration strikes the micromirror with assigned modulation frequency f4. The intensity 870 leaving the sample 890 is modulated with frequency f3, and the modulation 872 from the return micromirror is f4, so the intensity 874 of light leaving the micromirror has a complicated waveform in the time domain, as shown in FIG. 8. This light receives modulation at two different frequencies in the same octave, which heterodynes out of the octave. The spectrum 876 of this light has two spectral components, one at the sum of the modulation frequencies, f3+f4, and one at the difference of the modulation frequencies, If3-f4l. Both f3+f4 and If3-f4l lie outside the octave between F and 2F. For this light, from the erroneous depth 896, there is effectively no spectral component at f3 or f4. In this manner, the light from the erroneous depth makes it through the optical system to the detector, is converted into an electrical signal by the detector, has spectral components that lie outside of the single octave, and is effectively ignored by the lock-in amplifiers that extract information only within the single octave. Unlike known scanning pinhole confocal techniques, in which the light from erroneous depths is blocked by the opaque screen, the present technique passes the light to the detector, and filters out light from the erroneous depths in the electrical domain. Effectively, this forms an arbitrary number of virtual confocal apertures, arbitrarily close to each other.

In practice, light that is modulated twice at different frequencies may not completely heterodyne out of the octave, and may have residual components left at the modulated frequencies. Mathematically, the optical modulation does not average to zero, but averages to a particular DC value. The DC value of the optical modulation produces the residual components at the modulation frequencies. The residual components can be addressed by using an image sensor, and demodulating in the electrical domain or the software domain. By demodulating electrically or in software, the signal can go negative (unlike an intensity level for light, which is always positive or zero), and out-of-focus light can average to zero. By averaging to zero, the residual components at the modulation frequencies can be reduced or eliminated.

FIG. 8A shows an example of a configuration that can produce a zero- averaged electrical signal, which oscillates between negative and positive voltages. An incident beam 886 strikes a two-position micromirror 888. When the micromirror 888 is in a first of the two positions, the micromirror 888 directs a reflected beam 890 onto a first detector 892. The first detector 892 produces an electrical signal, A. When the micromirror 888 is in a second of the two positions, the micromirror 888 directs a reflected beam onto a second detector 894. The second detector produces an electrical signal, B. A subtracter 896 subtracts electrical signals A minus B to produce electrical signal 898. The electrical signal switches between a positive voltage and a negative voltage, depending on the position of the micromirror 888. If the micromirror 888 is incorporated into a micromirror array, and detectors 892, 894 are incorporated into respective detector arrays, the micromirror array and detector arrays can be used in a return optical path similar to that in FIGS. IB, 2B and 4B, but producing an electrical signal that has a zero DC component, rather than a non- zero DC component. Such a zero-DC electrical signal can reduce or eliminate residual components at the modulation frequencies.

FIG. 9 shows an example of an electrical pathway compatible with the optical systems of FIGS. 1, 2, and 4. Light propagating along a return optical path 902 strikes a multi-pixel detector 950. Each pixel in the detector 950 produces a respective electrical signal 952. An optional bandpass filter 954, for each electrical signal 952, can keep frequencies within the range of frequencies that extends over the unique assigned frequencies. In some examples, the range of frequencies extends over a single octave. A plurality of signal amplitude extractors 958, for each electrical signal 956, extract signal amplitudes 960 at each of the modulation frequencies for the micromirror array. The signal amplitudes 960, corresponding to the pixels in the multi-pixel detector 950, can be used individually or can be summed appropriately to produce wavelength spectrum values 962. FIG. 10 shows the effect of an optional bandpass filter 1054. Electrical signal 1052 includes desired signal amplitudes Al, A2, A3, A4 at respective frequencies fl, f2, f3, f4 in the octave between frequencies F and 2F, plus extraneous signals (for example, produced by nonlinearities in the system) at frequencies outside the octave. Bandpass filter 1054 transmits frequencies within the octave between frequencies F and 2F. Bandpass filter 1054 blocks frequencies outside the octave, such as frequencies less than F and/or frequencies greater than 2F. In some cases, the bandpass filter 1054 is formed as a single circuit; in other cases, the bandpass filter 1054 is formed as an edge filter at F, followed by another edge filter at 2F. Electrical signal 1056 includes the desired signal amplitudes Al, A2, A3, A4 within the octave between F and 2F, and has attenuated signals outside the octave. Bandpass filter 1054 is not necessary for extracting the signal amplitudes downstream, but can reduce noise in the extraction by filtering out frequencies that do not include the desired signal information.

FIG. 11 shows an example of using lock- in amplifiers to extract the signal amplitudes from the electrical signal 1156. Each of the unique assigned frequencies has its own lock- in amplifier. In the example of the FIG. 11, there are four assigned frequencies fl, f2, f3, f4, and four corresponding lock- in amplifiers 1158A, 1158B, 1158C, 1158D. In some examples, the number of lock-in amplifiers can equal the number of micromirrors in the micromirror array. In some other examples, the number of lock-in amplifiers can equal the number of sets of micromirrors in the micromirror array, where each set of micromirrors receives its own unique assigned frequency. A computer 1180, with a suitable signal generator, generates trigger signals 1182 at the assigned frequencies fl, f2, f3, f4. The trigger signals 1182 are sent to the micromirror array, and are also sent to respective lock-in amplifiers 1158A, 1158B, 1158C, 1158D. Each lock-in amplifier 1158A, 1158B, 1158C, 1158D returns a respective signal amplitude 1160A, 1160B, 1160C, 1160D, with values Al, A2, A3, A4, at the respective assigned frequencies fl, f2, f3, f4. In some examples, the lock-in amplifiers 1158A, 1158B, 1158C, 1158D are formed as identical circuits, each of which receives a different frequency input from a respective trigger signal 1182. FIG. 12 shows an example of using a Fast Fourier Transform module 1258 to extract the signal amplitudes 1260 from the electrical signal 1256. A typical Fast Fourier Transform module 1258 produces a spectrum of a specified time interval of the input electrical signal 1256 at evenly-spaced, discrete frequencies. It is straightforward to analyze the spectrum, find the peaks, and return the peak values. Compared with the lock-in amplifiers, using a Fast Fourier Transform has the advantages of not requiring the trigger signals as input and producing a relatively large number of signal frequency components, but has the disadvantage of including some noise from frequencies that are between the assigned frequencies, especially when the assigned frequencies are densely packed. Using lock- in amplifiers and Fast Fourier Transforms are but two example techniques for extracting the signal amplitudes from the electrical signals; other suitable extraction techniques can also be used.

FIG. 13A shows an example of a micromirror array 1308A having sixteen mirrors, where each mirror receives a different assigned frequency fl, f2, ..., fl6. The micromirrors in the array 1308A are switched at unique assigned frequencies, which can be within the single octave between frequencies F and 2F.

FIG. 13B shows an example of a micromirror array 1308B having sixteen mirrors, where some of the mirrors are modulated at the same assigned frequency. In this example, the central four mirrors in the array 1308B are modulated at frequency f6, while the twelve mirrors surrounding the central four are modulated at respective frequencies fl-f5 and f7-fl3. Modulating a group of mirrors at the same frequency can improve the signal amplitude, which can be desirable for light-starved optical systems, but can decrease the sharpness of the confocal effect. In some examples, the system can dynamically change the assigned frequencies to improve the signal amplitude, as needed.

FIG. 13C shows another example of a micromirror array 1308C having sixteen mirrors, where some of the mirrors are modulated at the same assigned frequency. Groups of adjacent micromirrors in the array 1308C are switched, together, at unique assigned frequencies within the single octave. The groups can be elongated, as with frequencies fl, f3, f5, and f6, or symmetric, as with frequencies f2 and f4. The elongation can be along a first direction (e.g., vertical in FIG. 13C), as with frequencies f5 and f6, or along a second direction (e.g., horizontal in FIG. 13C), as with frequencies fl and f3. Each group can include the same number of micromirrors, or different groups can include different numbers of micromirrors.

For the optical systems of FIGS. 1, 2, and 4, the specified depth in the sample 190, 290, 490, the micromirror array 110, 210, 410 and the multi-pixel detector 150, 250, 450 are all conjugate to one another, with respect to the first imaging optics 130, 230, 430 and the second imaging optics 270, 470. For example, if the specified depth is at an object plane of the first imaging optics 130, 230, 430, then the micromirror array 110, 210, 410 is at a corresponding image plane of the first imaging optics 130, 230, 430. It is instructive to consider a sequence of images that can be associated with at the specified depth in the sample 190, 290, 490, the micromirror array 110, 210, 410 and the multi- pixel detector 150, 250, 450. Such images can help show how wavelength spectra are extracted from the electrical signals produced by the pixels in the multi-pixel detector.

FIG. 14 is a schematic drawing of a sequence of images that can be associated with various locations in the optical systems of FIGS. 1, 2, and 4. In some examples, one or more of the images can be a real image that occurs at a location in the optical systems. In some examples, one or more of the images can be a mathematical construct, formed as data that relates to a particular location, without being a real image in the optical systems.

A scene 1402 can correspond to a light distribution at the specified depth at or below the surface of the sample. For instance, in FIG. 2B, light from the light source 240 is directed by the beamsplitter 242, through the first imaging optics 230, to the lateral locations 292 at the specified depth within the sample 290. If one were to photograph the light intensities reflected by the lateral locations 292, it would resemble scene 1402 in FIG. 14.

Element 1404 corresponds to modulation from a micromirror array, such as array 110, 210, 410 in FIGS. 1, 2, and 4. Each micromirror in the array, or group of micromirrors in the array, is modulated at a unique assigned frequency. In some examples, the unique assigned frequencies are all within a single octave. The scene 1402 and the modulation 1404 are combined to form a modulated scene 1406. The modulated scene can correspond to a light distribution leaving the micromirror array 210 in FIG. 2B. In some examples, the micromirror array modulates light that is incident on the scene 1402; in other examples, the micromirror array modulates light reflected from the scene 1402. In some examples that use confocal detection, the micromirror array modulates light incident on the scene 1402, and then further modulates light reflected from the scene 1402. If one were to take a photograph of the scene 1402, and switch each lateral location in the photograph on and off at a respective unique assigned frequency, it would resemble modulated scene 1406 in FIG. 14.

Element 1408 corresponds to a dispersion along one dimension from a spectrally dispersive element, such as dispersive element 160, 260, 460 in FIGS. 1, 2, and 4. The modulated scene 1406 and dispersion 1408 are combined to form a dispersed modulated scene 1410. The dispersed modulated scene 1410 can correspond to a light distribution at multi-pixel detector 150, 250, 450 in FIGS. 1, 2, and 4. The dispersed modulated scene 1410 has a wavelength- dependent blur along one dimension. If one were to pass the modulated scene 1406 through a prism, thereby smearing one direction with a wavelength- dependent blur from the prism, it would resemble dispersed modulated scene 1410 in FIG. 14. The dispersed modulated scene 1410 is directed onto a multi- pixel detector. Spectral values are extracted from the location and amplitude of the assigned frequencies on the multi-pixel detector.

FIGS. 15A-15D show a specific example of a scene 1502, a modulated scene 1506, a dispersed modulated scene 1510, and resulting wavelength spectra that are extracted from the dispersed modulated scene 1510. This example shows how light from a specific location in the scene 1502 gets distributed in the dispersed modulated scene 1510, and subsequently produces a wavelength spectrum for the specific location.

In the specific example of FIG. 15A, the scene 1502 includes just two small, bright spots 1504, 1505 on a dark background. In practice, the light distribution in a real scene includes light at most or all locations within the scene, not just at isolated locations within the scene. In FIG. 15B, applying modulation to the scene 1502 produces an example of a modulated scene 1506. In this example, the scene is partitioned into lateral locations, with each lateral location corresponding to a micromirror or set of micromirrors on a micromirror array. The small bright spots 1505, 1504 extend over respective single lateral locations 1509, 1508, which are modulated at respective unique assigned frequencies fl, £2. The light at lateral location 1509 has an intensity proportional to the intensity of bright spot 1505, and is modulated on and off at frequency f 1. The light at lateral location 1508 has an intensity proportional to the intensity of bright spot 1504, and is modulated on and off at frequency f2.

In FIG. 15C, dispersing the modulated scene 1506 along one dimension produces an example of a dispersed modulated scene 1510. In the example of FIG. 15C, the spectral dispersion is along the horizontal direction, although any suitable direction may be used. The light from lateral location 1509 is dispersed over a row 1512 of pixels. In some examples, the wavelength of light increases from left to right along the row 1512 of pixels in FIG. 15C. In other examples, the wavelength of light decreases from left to right along the row 1512 of pixels in FIG. 15C. For each pixel in the row 1512, the light is modulated at frequency fl. Similarly, the light from lateral location 1508 is spectrally dispersed over a row 1513 of pixels. For each pixel in the row 1513, the light is modulated at frequency f2.

In FIG. 15D, an example of a wavelength spectrum 1516 for lateral location 1509 is extracted from the row 1512 of pixels. Each value in the wavelength spectrum 1516 is proportional to the signal amplitude at frequency fl for the corresponding pixel in row 1512. Similarly, an example of a wavelength spectrum 1514 for lateral location 1508 is extracted from the row 1513 of pixels. Each value in the wavelength spectrum 1514 is proportional to the signal amplitude at frequency f2 for the corresponding pixel in row 1513. The wavelength spectra 1514, 1516 are referenced to their corresponding lateral locations 1508, 1509, as indicated by a dashed vertical line in FIG. 15D. In the specific example of FIG. 15D, the reference location for lateral location 1509 is the third pixel from the right, and the reference location for lateral location 1508 is the second pixel from the left. In general, the light from a particular lateral location may extend across more than one row of detector pixels. The detector pixels may number more or less than the lateral locations; in general, the resolution with respect to wavelength is determined by the number of detector pixels.

FIG. 16 is a flow chart of an example of a method of operation 1600 for an optical system, such as optical systems 100, 200, 400 in FIGS. 1, 2, and 4. The method can measure a wavelength spectrum of a sample as a function of lateral location on the sample. Step 1602 receives a modulated image that includes a plurality of lateral locations. Each lateral location can be modulated at a unique assigned frequency. Step 1604 disperses the modulated image in a spectral direction to form a dispersed modulated image. The dispersion is wavelength-dependent. Step 1606 detects the dispersed modulated image with a multi-pixel detector. Step 1608 converts the light received by each pixel in the detector to a respective time-varying electrical signal. Step 1610 analyzes each time- varying electrical signal to extract signal amplitudes at the assigned frequencies. Step 1612 compiles, for each assigned frequency and

corresponding lateral location, a wavelength spectrum from the extracted signal amplitudes.

The description of the invention and its applications as set forth herein is illustrative and is not intended to limit the scope of the invention. Variations and modifications of the embodiments disclosed herein are possible, and practical alternatives to and equivalents of the various elements of the embodiments would be understood to those of ordinary skill in the art upon study of this patent document. These and other variations and modifications of the embodiments disclosed herein may be made without departing from the scope and spirit of the invention.

Claims

What is claimed is:

1. An optical system for optically characterizing a sample, comprising: a micromirror array, the micromirrors in the array being switchable between first and second positions, the micromirror array being configured to simultaneously reflect light from micromirrors of the micromirror array that are in the first position;

driving electronics configured to drive the micromirror array, the driving electronics directing trigger signals to a set of one or more micromirrors in the array to switch the set of micromirrors at unique assigned frequencies;

first imaging optics disposed in an optical path between the micromirror array and the sample, the first imaging optics and the micromirror array arranged so that each micromirror in the array is imaged onto a corresponding lateral location at a specified depth at or below a surface of the sample;

a light source configured to provide an incident light beam along an incident optical path, the incident optical path extending from the light source to the first imaging optics to the sample;

a multi-pixel detector configured to receive a return light beam reflected from the sample along a return optical path, the return optical path extending from the sample to the first imaging optics to the multi-pixel detector, the multi- pixel detector converting the return light beam to at least one electrical signal, wherein the micromirror array is disposed on at least one of the incident and return optical paths so that the micromirrors in the array are simultaneously illuminated; and

a dispersive element disposed in the return optical path between the first imaging optics and the detector, the dispersive element deflecting light along a spectral direction, the deflection being wavelength-dependent.

2. The optical system of claim 1, wherein the micromirror array is disposed in the incident optical path between the light source and the first imaging optics.

3. The optical system of claim 1, wherein the micromirror array is disposed in the return optical path between the first imaging optics and the detector.

4. The optical system of claim 3, further comprising second imaging optics in the return optical path between the micromirror array and the detector, the second imaging optics being positioned so the micromirrors in the micromirror array are imaged onto the detector.

5. The optical system of claim 4, wherein the dispersive element is disposed in the return optical path between the second imaging optics and the multi-pixel detector.

6. The optical system of claim 4, wherein the dispersive element is disposed in the return optical path between the micromirror array and the second imaging optics.

7. The optical system of claim 1,

wherein the micromirror array is disposed in the incident optical path between the light source and the first imaging optics; and

wherein the micromirror array is disposed in the return optical path between the first imaging optics and the multi-pixel detector.

8. The optical system of claim 7, further comprising second imaging optics in the return optical path between the micromirror array and the detector, the second imaging optics being positioned so the micromirror in the array is imaged onto the detector.

9. The optical system of claim 8, wherein the dispersive element is disposed in the return optical path between the second imaging optics and the multi-pixel detector.

10. The optical system of claim 8, wherein the dispersive element is disposed in the return optical path between the micromirror array and the second imaging optics.

11. The optical system of claim 7, wherein the unique assigned frequencies are within a single octave.

12. The optical system of claim 1 , wherein each micromirror in the array is switched at a respective unique assigned frequency.

13. The optical system of claim 1, wherein groups of adjacent micromirrors in the array are switched, together, at respective unique assigned frequencies.

14. A method for measuring a wavelength spectrum of a sample as a function of lateral location on the sample, the method comprising:

receiving a modulated image that includes a plurality of lateral locations, each lateral location being modulated at a unique assigned frequency;

dispersing the modulated image in a spectral direction to form a dispersed modulated image, the dispersion being wavelength-dependent;

detecting the dispersed modulated image with a multi-pixel detector; converting the light received by each pixel in the detector to a respective time- varying electrical signal;

analyzing each time-varying electrical signal to extract signal amplitudes at the assigned frequencies; and

compiling, for each assigned frequency and corresponding lateral location, a wavelength spectrum from the extracted signal amplitudes.

15. The method of claim 14, wherein the wavelength-dependent dispersion directs different wavelengths to different pixels along the spectral direction.

16. The method of claim 14, wherein the wavelength-dependent dispersion is referenced with respect to a corresponding lateral location.

17. The method of claim 14, further comprising:

modulating light at a plane conjugate with the sample to form modulated light; and

imaging the modulated light to form the modulated image.