This invention relates to imaging technology, and more particularly an imaging process using terahertz electro-magnetic waves. The invention is an apparatus designed for scanning and collecting for the imaging information.
Electro-magnetic waves, in the frequency range located between infrared and microwave, are one of the latest developed areas of spectrum. This band is often referred to as the terahertz (THz) band. These waves are transparent for most non-polar dielectrics, such as plastic, paper, stone, wood, oil, smoke, et. al. This makes THz wave imaging an ideal tool as a complement to x-ray and ultrasound imaging in security inspection and quality control applications. Additionally, THz wave imaging provides spectroscopic information on the target, and it can be used to identify the target. THz wave imaging is safe to both the sample and the operator as THz photons have very low energy (meV), which will not ionize molecules.
THz wave imaging technology such as described in U.S. Pat. No. 5,623,145 and U.S. Pat. No. 5,710,430 to M. Nuss, demonstrated the capability of seeing through plastic and mapping metal electrodes underneath. Most THz wave imaging apparatuses developed since then use a raster scanning mode to image the target. THz radiation emitted from the source is focused on the target imaged, and is recorded using a point detector after interacting with the target through transmission or reflection. The target is imaged via scanning the target crossing the THz wave focal spot in an X-Y plane or alternatively scanning the imager in opposite directions. Raster scanning fully utilizes THz waves generated from the source and results in a high measurement dynamic range, which is especially important in the THz regime where lack of intense sources and sensitive detectors pose a problem.
Raster scanning an image requires linearly scanning either the target or the imager within the entire image area, which is not only time consuming but also inconvenient. 2D focal plane imaging, which uses a 2D extended detector (such as an electro-optical (EO) crystal) or a detector array (such as a microbolometer array) instead of a point detector to record the distribution of the THz field at the image plane, was developed to improve imaging speed. An example is reported in “Two-dimensional electro-optic imaging of THz beams,” Appl, Phys. Lett., 69, 1026-1028 (1996) by Q. Wu et. al., where a single crystal EO sensor was employed as the extended THz wave sensor. An intense THz beam, which was extended illuminating the entire target, was generated using a femtosecond (fs) laser amplifier through an optical rectification or an optical switching process. The THz wave image of the target was created using an imaging lens and the EO crystal was placed at the image plane. An extended probing beam was used to read out the THz field distribution on the EO crystal, which was the THz wave image of the target.
Detector arrays, and 2D focal plane imaging methods, have also been developed for cw THz radiation under certain circumstances. One example was reported in “Real-time, continuous-wave terahertz imaging by use of a microbolometer focal-plane array,” Opt. Lett. 30, 2563-2565 (2005), by A. W. Lee et. al., where a microbolometer array, which was designed for middle infrared radiation, was demonstrated to have the capability to record 2D images at a high THz frequency (2.52 THz). Other examples include U.S. Pat. No. 6,242,740 to A. Luukanen et. al., and U.S. Pat. No. 6,943,742 to S. Holly.
Prior art 2D focal plane THz wave imaging systems, however are only available under limited conditions. Especially, these detector arrays are only sensitive at limited frequency ranges. They also require a relatively more intense THz source than point detectors due to the low sensitivity of detector arrays and dilution of the THz radiation on multi pixels at one time. Alternative methods were therefore developed to improve imaging speed without using an extended detector or detector array. Prior art U.S. Pat. No. 6,909,094 and U.S. Pat. No. 6,909,095, to P. Tran et. al., use a distributed waveguide technique to receive THz wave image at the image plane and using a point detector to receive signals from all pixels sequentially. “Terahertz wave reciprocal imaging,” Appl. Phys. Lett. 88, 151107 (2006), by J. Xu et. al, reported using a source array rather than a detector array to present a focal plane image in a reciprocal way. These arts avoid using 2D detector arrays, which are not available or not ideal for THz waves, however there are still other challenges when using a waveguide array or a source array.
U.S. Pat. No. 6,815,683 to J. Federici adapted an interferometric imaging technique from the microwave regime to the THz regime. Using only a few detectors and without using an imaging lens, this technique promises to present large scale THz wave imaging without moving imager or target. On the other had, this art utilizes THz radiation in an inefficient way, as no collection optics is allowed in this technology.
Compared to arts using distributed detection or emission components, the raster scanning method is still the predominantly used imaging method for THz radiation because it provides higher measurement dynamic range, gives better image quality and is available for all kinds of THz wave sources and detectors.
Efforts have also been taken to improve the speed of THz wave raster scanning image. An example was described in WO 02/057750 by B. Cole et. al., where a THz wave was guided into an imaging head, which contains only a limited number of optics (and THz optics). It was the imaging head rather than the entire imager being scanned in then imaging process. The speed of imaging was improved because less mass was scanned.
The detectivity of THz waves currently available is not sufficient to utilize these waves for an imaging device in a comparable fashion to other well developed areas of wavelength ranges. Furthermore, the lack of a strong source to overcome the detectivity problem is also challenging. The immature source technology also leads to difficulties in manipulating the wave to be accommodated into other techniques. As a result, detection of scattered light out of a target to be imaged is not a practical method, due to the lack of strong source. Equally, realization of a modulation technology to enhance the detectivity of a weak source is not feasible due to immature technology surrounding the source. Speed of imaging is another important factor in considering an effective imaging system.
Currently, no prior art is known for an imaging apparatus in the range of extended THz frequencies, e.g., 1 GHz˜100 THz, disclosing a concept for realization of practically usable sensitivity and speed. This disclosure addresses these two major points with an exemplary embodiment focusing on two dimensional raster scanning of an image by scanning the THz beam across each spot of the target using a wave in the frequency range of 10 GHz˜3 THz.
The present invention is directed to an scanning image system, in particular a THz wave imaging system. Rastering a target to be imaged with a pulsed THz wave centered at 1 THz is disclosed to be an exemplary embodiment, although the frequency range may be between about GHz and 100 THz. Some embodiments may utilize a frequency range of 1 GHz to about 10 THz. Other embodiments may utilize a frequency range of about 1 GHz to about 20 THz. Corresponding wavelengths may be up to at least a few tens of microns, for example about 20 microns. The THz source of the present THz wave imager could be generated coherently or incoherently.
Sources can be a THz pulse utilizing ultrafast optical rectification or an optical switch in combination with an ultrashort pulse laser. Furthermore, a cw THz electronic transmitter, or a THz wave laser, or even an incoherent thermal emitter can be incorporated within this disclosure.
On the other hand, in the detection scheme, a photo-conductive switch, electro-optic sample, bolometric detection, or other devices can be incorporated within the invention. For phase-sensitive detection, often called coherent detection, both pulsed and continuous wave can be combined within this disclosure to utilize the suppression of thermal noise in the detection technique. A time-domain point detector with relatively high response speed, for high speed imaging, is disclosed to be an exemplary embodiment.
Instead of linearly scanning the target or the imager, a mirror/lens set is used for steering the THz beam across each spot of the target. The steering mirror set could be driven using a galvanometer (or a pair of galvanometers for 2D scanning), which supports a line scanning speed up to 300 Hz (up to 600 lines per second with a zigzag scanning mode). This speed is sufficient for a real time imaging application. Especially, the incident beam could be collimated and the steering mirror could rotate about two axes, whose crossing point locates at a focal point of the focusing lens. The target is located at the focal plane on the opposite side of the imaging lens and is normal to the optical axis of the lens. Under this condition, the THz beam will be retro-reflected by the target and full THz radiation will be utilized in the imaging process.
The imaging wave can be obtained primarily by reflection or transmission. However, scattering or refraction arranged to a specular direction can be incorporated into the invention. Alternatively, the present invention could be used in a passive imaging configuration, where no incident THz radiation is required and the mirror/lens set is used to steer THz radiation emitted or reflected from an external source not included in the imaging system from each spot of the target to the point detector.
The spectral information carried by reflection, transmission, emission, refraction and scattering is analyzed by a time-domain measurement.
a is a diagram of an embodiment of the invention in a reflection configuration.
b is a diagram of an embodiment of the invention in a transmission configuration.
a is a diagram of a steering mirror setup in one embodiment of the invention.
b is a diagram of another steering mirror setup in one embodiment of the invention.
a indicates parallel beam walking for two different mirrors at different sides of the focal point in
b shows beam traces when the incident beam has a displacement to the focal point.
c indicates returning beam spots from the incident one on the steering mirror.
a shows beam traces in the 3-mirror configuration.
b shows the temporal variation when the steering mirror has offset to the focal point of the focusing lens.
a shows THz field distribution in the image area at a fixed delay time.
b shows THz amplitude distribution in image area.
c is the distribution of THz pulses' peak timing in image area.
d shows calibrated distribution of THz amplitude in the image area.
a gives a THz wave image of a metallic razor blade recorded according to the contrast to the background.
b is a THz wave image of the same blade using THz pulses reflected from its surface.
c gives a topographic image of the blade.
a compares THz waveforms with (dashed curve) and without (solid curve) target (a polyethylene pellet).
b is a THz wave image of this pellet recorded according to the contrast to the background.
c is a topographic image of the pellet using the time delay of transmitted THz pulses.
Furthermore, a curved mirror such as paraboloidal mirror can also be incorporated in replacement of the lens, with a proper adjustment of the target location.
a shows an important embodiment of the present invention, where a collimated THz beam 10 is incident into the mirror-lens set and it intersects with the rotation axis of the steering mirror 20 at one focal point of the lens 30. As a collimated THz beam 10 comes from its focal point, the lens 30 focuses the beam normally onto its focal plane on the opposite side. The target 40 lies on the focal plane. Therefore, each spot on the target is a function of the incident angle. If the THz beam is specularly reflected by the target, which is mostly true as THz waves have long wavelength, for example up to at least a few tens of microns, the returning beam counter propagates with the input beam. As a result, the entire radiation transmitted from the source can be utilized to image the target.
The embodiment shown in
Another interesting concept is the measurement of the angular radiation distribution of radiation out of a point source if the incident light is replaced with the radiation out of the target itself.
Furthermore, the focusing lens, which is identical to the collecting optics, can move along the optical axis, so that information to be imaged can be obtained at the surface or inside the target, if the target is sufficiently transparent.
The wave to be imaged can be extended in its property beyond the reflection shown in this exemplary embodiment. As long as the geometry of the focusing and collecting optics together with the steering optics is fulfilled as described in earlier sections, the origin of the wave out of the target can be scattered in the specular direction, or diffracted or even refracted to the specular direction by the structure in or on the sample. Randomly distributed nano-structures on and inside the target can be an example of an arrangement where this is possible.
It should be noted that, all embodiments discussed above and hereafter can be easily modified for use in a passive imaging configuration. The simplest modification is just replacing the transmitter with a point receiver. Rotating the steering mirror 20, THz radiation emitted (reflected) from each spot of the target 40 will be guided into the receiver sequentially.
The diameter of the imaging area in embodiments shown in
DI=2F tan(α), (1)
where F is the focal length of the focusing lens 30 and α denotes the maximum beam spanning angle driven by the steering mirror 20. In most cases, the THz beam spanning angle is two times the steering mirror's rotating angle.
On the other hand, the maximum image size is also limited by the size of the focusing lens and the size of the THz beam. In
DI=DL−DB, (2)
where DL and DB are diameters of the focusing lens 30 and the THz beam 302, respectively.
The spatial resolution of such an image is limited by the diffractive limitation of the THz beam:
Λ=Fλ/DB, (3)
where λ is wavelength of the imaging wave. To keep the same spatial resolution and scanning angle while obtaining a larger imaging area, one needs to scale up the size of the lens, its focal length, and the diameter of the THz beam on the lens.
a and
The configuration shown in
b shows beam traces of THz beams with a certain incident angle. If the incident THz beam (10a) passes through the focal spot of the focusing lens 30, the beam is bent by the lens 30 normally incident onto the target 40. The target then retro-reflects the incident beam. If the incident beam (10b) is away from the focal point with a displacement of H, the returning beam will be on the opposite side of the focal point with a counter propagation direction and will have the same displacement to the focal point. Consequently, the returning beam will be collected by the same focusing optics (such as 50 in
Although parallel beam walking does not mislead the returning beam, it does reduce overlap between the returning beam and the input beam.
A galvanometer setup with three galvanometers can be used to create 2D beam scanning and to compensate the parallel beam walking in the apparatus. An exemplary embodiment is presented in
a gives the relationship between rotation angles of the first steering mirror and the second steering mirror. The incident beam is guided into the steering mirror set at the point A, which is the rotation center of the 1st mirror. The rotation center of the 2nd mirror is at O. When the 1st mirror is at its original position, the input beam is steered following AO, and it will be reflected by the 2nd mirror to the 3rd mirror following OB, where B is located on the rotation axis of the 3rd mirror. For any angle of φ, which the incident beam changes from its original direction steered by the 1st mirror, there is an angle of β. If the 2nd mirror rotates β, the incident beam follows the path of AO'B, and ends at the same point B on the 3rd mirror. Rotation of the 1st mirror and the 2nd mirror together results in a scanning angle φ of the output beam. β and φ as functions of φ are in Eq. 4.
Another important feature of this configuration is the beam walking distance on the 2nd mirror, which determines the size of the second mirror.
The parallel beam walking can be compensated using a three-galvanometer configuration. However there is another issue created by the beam path changing which must be taken into account, especially when the imaging process has fine temporal resolution, such as in pulsed THz wave imaging. This issue is the optical path variation, which results in a temporal variation in temporal resolved imaging, with beam scanning.
ΔOP=2h[1−cos(φ)]. (6)
It is positive when the rotation axis is closer to the lens, and is negative otherwise.
Optical path as a function of rotation angle in three-galvanometer configuration can be resolved according to
If OA=OB=L is satisfied, Eq. 7a can to be simplified as:
OP=2L cos(β), (7b)
Unlike the temporal variation shown in Eq. 6, the temporal variation in this case is determined by the structure of the three-galvanometer setup, and it cannot be minimized by optimizing alignment. When the distance between the 1st mirror and the 2nd mirror (same as between the 2nd mirror and the 3rd mirror) is 40 mm, the maximum temporal variation for a 20 degree beam span is 5 mm (33 ps temporal variation).
Temporal variation causes no effect in an imaging process with no temporal resolution, or with a temporal resolution much longer than the temporal variation. However for some imaging processes, which have a fine temporal resolution, such as pulsed THz wave imaging, one needs to retrieve the imaging after recording images at all temporal positions, in order to compensate the temporal variation in scanning.
Overall, if an oversized target needs to be imaged, the exemplary embodiment described so far can be extended so that the all required optics can be translated laterally together. In this way, imaging of area-by-area is possible. The time for the translation can be utilized for data processing of the area imaged prior to the translation.
M1 and M2 are two scanning mirrors (X-mirror 201a and Y-mirror 201b) driven by a pair of galvanometers. M1′ is the image of M1 in M2 in the original direction. CL1 and CL2 are two cylindrical lenses. The focal length of each lens (301 and 302) equals half of the distance between the corresponding mirror (201a for 301, and 201b for 302) to the target 40. The distance between the target 40 to each of the cylindrical lens (301 or 302) is the focal length of the lens. CL1 and CL2 focus a collimated beam onto the target, and the lateral position of the focal spot is determined by the direction of the incident beam.
As presented in
δ=2D sin φ(1−cos θ), (8a)
ΔOP=2Δ(1−cos φ)=2D(1−cos φ)(1−cos θ). (8b).
If the distance between M1 and M2 is 25 mm and the maximum scanning angle is +/−20°, the maximum beam walking is 1 mm and maximum optical path variation is 185 μm.
a shows the distribution of THz field in the image area at a fixed delay time between pumping and probing pulses. The result was measured by using a metallic flat mirror as the target. Due to the temporal variation in scanning, the distribution of the THz field at a fixed timing shows ring structures like an interference pattern.
THz wave imager was also used to image dielectric targets.
Although the above discussion uses THz waves as imaging waves, embodiments of the invention may be utilize a frequency range of 1 GHz to 100 THz. Further, this technology can be applied to electromagnetic waves at other frequencies, sound, electron beams, et. al., which can be steered and focused, and can be used to image a target using this method.
This application claims priority from U.S. Provisional Application No. 60/907,639 filed Apr. 12, 2007, and incorporates the earlier application by reference herein.
Number | Name | Date | Kind |
---|---|---|---|
4733609 | Goodwin et al. | Mar 1988 | A |
5048904 | Montagu | Sep 1991 | A |
5168386 | Galbraith | Dec 1992 | A |
5237444 | Schermer | Aug 1993 | A |
5623145 | Nuss | Apr 1997 | A |
5710430 | Nuss | Jan 1998 | A |
6242740 | Luukanen et al. | Jun 2001 | B1 |
6815683 | Federici et al. | Nov 2004 | B2 |
6909094 | Tran et al. | Jun 2005 | B2 |
6909095 | Tran et al. | Jun 2005 | B2 |
6943742 | Holly | Sep 2005 | B2 |
7274451 | Bustamante et al. | Sep 2007 | B2 |
7449695 | Zimdars et al. | Nov 2008 | B2 |
7633036 | Gross | Dec 2009 | B2 |
7633663 | Hirakawa | Dec 2009 | B2 |
7643869 | Dabiri et al. | Jan 2010 | B1 |
20030066968 | Ziolo | Apr 2003 | A1 |
20030149346 | Arnone et al. | Aug 2003 | A1 |
20050161444 | Kitai et al. | Jul 2005 | A1 |
20050230625 | Zhang et al. | Oct 2005 | A1 |
20060049356 | Shen et al. | Mar 2006 | A1 |
20060176912 | Anikitchev | Aug 2006 | A1 |
20070235658 | Zimdars et al. | Oct 2007 | A1 |
20070242274 | Cluff | Oct 2007 | A1 |
Number | Date | Country |
---|---|---|
WO 02057750 | Jul 2002 | WO |
Number | Date | Country | |
---|---|---|---|
20080251720 A1 | Oct 2008 | US |
Number | Date | Country | |
---|---|---|---|
60907639 | Apr 2007 | US |