Spectroscopic Terahertz Imaging at Room Temperature ... - MDPI

sensors Article

Spectroscopic Terahertz Imaging at Room Temperature Employing Microbolometer Terahertz Sensors and Its Application to the Study of Carcinoma Tissues Irmantas Kašalynas 1, *, Rimvydas Venckeviˇcius 1 , Linas Minkeviˇcius 1 , Aleksander Sešek 2 , 1 , Bogdan Voisiat 1 , Dalius Seliuta 1 , Gintaras Valušis 1 , ¯ Faustino Wahaia 3 , Vincas Tamošiunas 2 2 Andrej Švigelj and Janez Trontelj 1

2 3

*

Department of Optoelectronics, Center for Physical Sciences and Technology, Savanoriu Ave. 231, Vilnius 02300, Lithuania; [email protected] (R.V.); [email protected] (L.M.); [email protected] (V.T.); [email protected] (B.V.); [email protected] (D.S.); [email protected] (G.V.) Faculty of Electrical Engineering, University of Ljubljana, Trzaska 25, Ljubljana 1000, Slovenia; [email protected] (A.S.); [email protected] (A.Š.); [email protected] (J.T.) Instituto de Investigacao e Inovacao em Saudeand, Instituto de Engenharia Biomedica, University of Porto, Rua do Campo Alegre, 823, Porto 4150-180, Portugal; [email protected] Correspondence: [email protected] or [email protected]; Tel.: +370-5-231-2418

Academic Editors: Dragan Indjin and Vincenzo Spagnolo Received: 6 February 2016; Accepted: 18 March 2016; Published: 25 March 2016

Abstract: A terahertz (THz) imaging system based on narrow band microbolometer sensors (NBMS) and a novel diffractive lens was developed for spectroscopic microscopy applications. The frequency response characteristics of the THz antenna-coupled NBMS were determined employing Fourier transform spectroscopy. The NBMS was found to be a very sensitive frequency selective sensor which was used to develop a compact all-electronic system for multispectral THz measurements. This system was successfully applied for principal components analysis of optically opaque packed samples. A thin diffractive lens with a numerical aperture of 0.62 was proposed for the reduction of system dimensions. The THz imaging system enhanced with novel optics was used to image for the first time non-neoplastic and neoplastic human colon tissues with close to wavelength-limited spatial resolution at 584 GHz frequency. The results demonstrated the new potential of compact RT THz imaging systems in the fields of spectroscopic analysis of materials and medical diagnostics. Keywords: compact THz sensors and components; THz imaging systems; multispectral THz imaging; medical THz imaging

1. Introduction Radiation of terahertz (THz) frequency offers non-destructive and non-ionizing ways of imaging and spectroscopy stimulating the development of THz technologies for security, medicine, biochemical, and materials science [1,2]. Recent achievements in the THz field have triggered new applications in biology and biomedicine with the particular aim of exploring the specificity of fingerprint spectra of materials [3,4]. Development of stand-alone, real-time, and frequency sensitive imaging schemes is of prime interest due to the measurement time, portability, and price issues of currently available THz systems [5]. Recently a compact room temperature (RT) imaging system developed for security needs demonstrated the ability to screen objects inside packages without opening them and to draw a materials map via principal components analysis [6,7]. Most practical applications need real time

Sensors 2016, 16, 432; doi:10.3390/s16040432

www.mdpi.com/journal/sensors

Sensors 2016, 16, 432

2 of 15

measurements and a high signal to noise ratio (SNR), therefore, frequency-selective THz sensors with fast response time, high dynamic range, and low noise-equivalent-power (NEP) are required. Compact, sensitive, and large-format THz cameras delivering images in real time have been developed for many practical applications [8–10]. In Reference [8] a THz bolometer camera was able to achieve fast scanning of a large field of view of opaque scenes in a complete body scanner prototype. As an option, the uncooled microbolometer THz focal plane array (FPA) was reported by the company NEC [9]. Values of minimum detectable powers per pixel were comparable with those of other compact THz detectors, such as uncooled field effect transistor (FET) THz sensors and cooled bolometer arrays demonstrating large potential in the frequency range of 0.3–4.3 THz. On the other hand, the French Institut National d’Optique (INO) has developed a THz imaging system capable of detecting concealed weapons or hidden objects behind drywall, and for non-destructive testing military applications [11]. The imaging system was based on the THz cameras built at INO, for example, the THz-optimized IRXCAM-THz-160 and IRXCAM-THz-384 cameras that support the 160 ˆ 120 and 384 ˆ 288 uncooled microbolometer pixel array with a pixel pitch of 52 µm and 35 µm, respectively. ‘ The sensitivity of INO’s THz detectors was determined by NEP values reaching up to 25 pW/ Hz ‘ and 76 pW/ Hz at frequencies of 4.25 THz and 2.54 THz, respectively [11]. The NEP dependence on the wavelength was explained by differences in pixel size, detector bandwidth, and pixel responsivity. As an alternative, a simple mass producible THz detection array has been developed within standard complementary metal-oxide-semiconductor (CMOS) technology [12]. The sensors based on a thin-film absorber on a membrane and process-integrated thermopiles have provided a 5 ms thermal time ‘ constant, together with a wavelength independent NEP of 1 nW/ Hz. Such a THz detection array enabled real-time imaging at 50 frames/s with a signal-to-noise ratio of 10 for an optical intensity of 30 µW/cm2 . Recently, a significant advance toward compact, low-cost real-time THz imaging systems have been proposed integrating metamaterial absorbers with bolometric vanadium-oxide sensors [13]. The absorber was realised directly in the layers of a standard 0.18 µm CMOS process but the micro-bolometer sensors were defined by post-processing procedures. An absorption magnitude ‘ of 57% at 2.5 THz, a minimum NEP of 37 pW/ Hz and a thermal time constant of 68 ms for the sensor were experimentally assessed. Very recently, direct comparison of commercial thermal detector arrays for off-axis THz holography and real time THz imaging have been performed employing a far infrared gas laser system as a powerful THz radiation source [14]. The results revealed that during the same experiments the SNR of the pyroelectric camera was significantly lower in comparison to the bolometeric one at around 3 THz. Moreover, THz cameras have not yet reached the high lateral resolution of the thermal micro-bolometers both in number of pixels and in pixel pitch. Thus, a search for a new physical mechanism for efficient THz radiation detection and the performance optimisation of single THz sensor is necessary. FET-based THz sensors have been proposed for efficient rectification of THz waves [15,16]. Currently, the most promising device for THz detection at RT seems to be the THz antenna-coupled FETs (TeraFETs) consistently developed for multispectral THz imaging up to 4.25 THz [7,17]. The theory of plasma rectification suggests that the response of the TeraFETs extends into THz region far beyond the cut-off frequency of Si transistor [16,18]. However, in practice optimal TeraFET performance was observed at frequencies of about 600–700 GHz due to electrical losses in parasitic RC components and interconnects [7,18] as well as due to optical losses in the substrate due to antenna effects [19]. The main advantage of Si technology is that the THz devices can be fabricated within a standard CMOS process and supplementary modules such as amplifiers and multiplexers can be integrated on the same chip [10,20]. As an alternative, InGaAs-based bow-tie diodes—thermoelectric-based RT THz sensors—were developed for the frequency range up to 1 THz [21]. The NEP value of such THz sensors was found ‘ below 4 nW/ Hz [22] and an attractive possibility to fabricate a monolithic THz detectors array was demonstrated [23]. Furthermore, suitability of the bow-tie diodes for spectroscopic needs was confirmed and compared versus a commercial pyro-electric THz sensors by measuring packed samples

Sensors 2016, 16, 432

3 of 15

at discrete fingerprint frequencies in the range of 0.58–2.52 THz [22]. The measured THz absorbance of the samples was found to be in a good agreement with Fourier spectroscopy data allowing the authors to perform principal components analysis of the admixtures. At that time THz imaging system size reduction possibilities were limited by the usage of a bulky THz source—an optically pumped gas laser. Separate group presented a stand-alone, portable system for high resolution real-time THz imaging based on the quantum cascade laser (QCL) emitting at 3.4 THz in continuous-wave mode at a cryogenic temperature of 50 K with an output power of 1 mW [24]. Real time THz imaging capability with a spatial resolution of 2.5 times the wavelength was demonstrated in the system with a commercial uncooled microbolometer camera. A confocal microscopy THz system based on a cryogenically cooled 2.9 THz QCL providing a large contrast enhancement via a lateral and axial resolution better than 70 µm and 400 µm, respectively, was demonstrated [25]. On the other hand, antenna-coupled titanium (Ti) microbolometric sensors have been proposed for fast and sensitive THz detection at RT [26,27]. Our group developed dipole-type THz antennas on a thin silicon-nitride-oxide (SiNO) membrane and used them for efficient THz radiation coupling to the air-bridged microbolometer [28]. A typical narrow band microbolometer THz sensor (NBMS) equipped with a 300 GHz frequency dipole antenna exhibited the response time of 1 µs, sensitivity ‘ of 300 V/W, and NEP as low as 14 pW/ Hz. The performance of the NBMS in a vacuum was up to three times better as compared to the operation at room environment [26,28]. In this work, particular attention has been given to the investigation of spectral selectivity of the NBMSs. For this purpose an air-bridged Ti microbolometer was coupled to a double-dipoleor cross-dipole-type antenna. The frequency response was measured in a wide frequency range of 0.1–1.5 THz via the recently proposed quasi-optical THz detectors characterization technique [29]. The NBMS was found to be a very sensitive and frequency selective device. Therefore, a compact multi-frequency THz imaging system based on the NBMSs was developed and applied for inspection of plastic packages and for principal components analysis of explosive simulators. Over the past few years improvement in the compact RT THz sensors brought very sophisticated THz imaging systems into being [1,5]. However, the total size of the system is still limited by the dimensions of commercially available THz components like mirrors, beam splitters, waveguides, and lenses. The Fresnel zone plates being thinner, lighter and in some cases more effective in comparison with identical diameter and focal length refractive lenses can be used to reduce the system size. Recently, a compact focusing component—the THz zone plate with integrated resonant filter apertures (TZP)–has been developed for 0.76 THz frequency [30,31]. Quite complex setup based on an optically pumped THz laser has been used at that time to proof the concept and to illustrate the operation principles of the TZP. And it was not possible to demonstrate wavelength limited operation without a stable THz source emitting Gaussian mode beam. Nevertheless, such a novel diffractive lens was found to be more efficient in terms of frequency selection and high aspect-ratio focusing in a single device. Moreover, the laser-ablated zone plates can be integrated directly into the bottom surface of the semiconductor substrate on which the THz sensors were fabricated [32]. Integration of THz components into a single device has the advantages of size, price, permanent stable construction and alignment with the THz sensor. Terahertz science and technology provided new ways for supplementary diagnosis and therapy of the skin, colon, and gastric cancer [33–35]. In general, cancer environment causes increased blood supply to affected tissues and an increase of water content [34,36]. This fact acts as a natural contrast mechanism for THz imaging [37]. Moreover, a structural change occurring in the affected tissues was also demonstrated as a contributing factor to the THz contrast [33,35,38]. In this work thin diffractive optics was developed for efficient THz beam focusing at a frequency of 584 GHz. The TZP lens with a 16.5 mm diameter and a 10 mm focal length was fabricated on a 30 µm thick metal foil. The focusing performance was obtained measuring two-dimensional profiles of THz beam along an optical axis by the recently proposed technique [39]. The numerical aperture (NA) for the TZP was of about 0.62 which allowed us to increase the spatial resolution of THz images

Sensors 2016, 16, 432

4 of 15

in comparison to that measured with commercial parabolic mirrors (PMs) roughly by 25%. Finally, 4 of 15 imaging system enhanced with novel TZP was proposed for biomedical microscopy applications. For this purpose dehydrated human colon tissues were imaged at a frequency of 584 GHz. Higher contrast and close to wavelength limited spatial resolution were observed in the measured Higher contrast and close to wavelength limited spatial resolution were observed in the measured THz images comparing non-neoplastic control and neoplastic tumor tissues. THz images comparing non-neoplastic control and neoplastic tumor tissues. Sensors 2016, 16,THz 432 the compact

2. 2. Antenna AntennaCoupled CoupledTitanium TitaniumMicrobolometer MicrobolometerSensors Sensors Narrow band THz THz antennas antennas were were developed developed for for multispectral multispectral THz THz imaging imaging applications applications in Narrow band in the the frequency range from 0.2 THz to 2 THz. A schematic view of the antenna-coupled microbolometer frequency range from 0.2 THz to 2 THz. A schematic view of the antenna-coupled microbolometer THz THz is shown in1.Figure 1. Selectivity was by enhanced by the adjustment the dipole antenna sensorsensor is shown in Figure Selectivity was enhanced adjustment dipole antenna geometry and geometry and resonant-cavity design, i.e., the back side reflection mirror was positioned at the resonant-cavity design, i.e., the back side reflection mirror was positioned at the quarter wavelength quarter distance. An air-bridged and Ti-microbolometer andprocessed THz antenna on distance.wavelength An air-bridged Ti-microbolometer THz antenna were on awere few processed microns thin aSiNO few microns thin SiNO membrane in order to increase the sensitivity via reduction of thermal losses membrane in order to increase the sensitivity via reduction of thermal losses of the device. of metalized the device.bottom A metalized bottom plate under the SiNO membrane acted as a perfect reflector and A plate under the SiNO membrane acted as a perfect reflector and enhanced the enhanced the spectral selectivity of the THz sensor [26,28]. spectral selectivity of the THz sensor [26,28]. The types of of dipole antennas of The THz THz sensors sensors were were fabricated fabricated on onaa4′′ 411size sizeSiSisubstrate. substrate.Several Several types dipole antennas various dimensions and complexity were was of various dimensions and complexity weredesigned designedThe Theresponse responsespectrum spectrumof of the the NBMS NBMS was measured with a custom-designed Fourier transform infrared (FTIR) spectrometer in vacuum at measured with a custom-designed Fourier transform infrared (FTIR) spectrometer in vacuum atRT. RT. The mercury-arc lamp lamp of of the the FTIR FTIR spectrometer spectrometer served served as as the the THz THz radiation radiation source source [29]. [29]. Measured Measured The mercury-arc response shown in in Figure Figure 2. 2. As it was was expected expected the the NBMS NBMS with with double double dipole dipole antenna antenna response spectra spectra are are shown As it (DA) design demonstrated the maximum sensitivity at a specified resonant frequency, namely 300 (DA) design demonstrated the maximum sensitivity at a specified resonant frequency, namely 300 GHz GHz or 600 GHz. GHz. Experimental obtained using using or 600 Experimental results results were were compared compared with with gain gain calculations calculations of of the the DA DA obtained the ANSYS HFSS HFSS computer computer program. program. Data Figure 2a. 2a. A reasonably good good the ANSYS Data comparison comparison is is shown shown in in Figure A reasonably agreement between modeling and experimental data in the vicinity of the fundamental frequency agreement between modeling and experimental data in the vicinity of the fundamental frequency of of the antenna was achieved. Moreover, higher order resonances of the antenna coupled THz sensor the antenna was achieved. Moreover, higher order resonances of the antenna coupled THz sensor were were observed the spectrum as indicated by vertical in2a. Figure Note the simulated observed in the in spectrum as indicated by vertical arrows arrows in Figure Note2a. that the that simulated antenna antenna gain spectrum nicely fitted the experiment data in the whole range up to 1.3 THz frequency. gain spectrum nicely fitted the experiment data in the whole range up to 1.3 THz frequency.

Figure Figure 1. 1. A A schematic schematic view view of ofthe theantenna-coupled antenna-coupled Ti-microbolometer Ti-microbolometer sensor sensor (top); (top); microscope microscope image image of the fabricated fabricated Ti-microbolometer Ti-microbolometer and and central central part part of of THz THz antenna: antenna: flat-top of the flat-top view view (bottom (bottom left) left) and and side-3D view (bottom right). Note a tiny air-bridged Ti wire positioned in the center of the antenna. side-3D view (bottom right). Note a tiny air-bridged Ti wire positioned in the center of the antenna.

Sensors 2016, 16, 432 Sensors 2016, 16, 432

Sensors 2016, 16, 432

5 of 15 5 of 15

5 of 15

Figure 2. A normalized response spectrum of the THz sensor with a double dipole antenna optimized

Figure 2. A A normalized response spectrum ofTHz thesensor THz with sensor with adipole double dipoleoptimized antenna Figure normalized response the a double antenna for2.operation at a frequency ofspectrum 300 GHz:of the results of experiment and calculation (a); Frequency optimized for operation at a frequency of 300 GHz: the results of experiment and calculation (a); for operation at the a frequency 300the GHz: theantenna results (DA) of experiment and calculation (a); design Frequency response of THz sensorofwith dipole and cross-dipole antenna (CDA) Frequency theGHz THz sensor with the dipole and cross-dipole antenna (CDA) optimized theof600 frequency (b); Note that theantenna amplitude scale is linear (a) and logarithmic response ofresponse theforTHz sensor with the dipole antenna (DA) and(DA) cross-dipole antenna (CDA) design design optimized for the 600 GHz frequency (b); Note that the amplitude scale is linear (a) and (b), respectively. optimized for the 600 GHz frequency (b); Note that the amplitude scale is linear (a) and logarithmic logarithmic (b), respectively. (b), respectively. Measured response spectra of the THz sensor with different THz antenna designs optimized for 600 GHz frequency are shown inthe Figure 2b. Note that the results are shown in the semi log scale. Measured Measured response response spectra spectra of of the THz sensor with different THz antenna designs optimized for Although the NBMS with DA design has a quite large side peak at around 1.3 THz, the CDA design 600 GHz frequency are shown in Figure 2b. Note Note that that the the results results are shown shown in the semi log scale. provided a single-peak response characteristic without any side peaks. Although the NBMS with DA design has aa quite large side peak around 1.3 the CDA AlthoughThe thefirst NBMS design has quite large peak at atfor around 1.3 THz, THz, theradiation CDA design design dualwith frequencies linear THz camera wasside developed low intensity THz provided a single-peak response characteristic without any side peaks. provided a single-peak response characteristic without any by side detection at RT. The NBMS camera was successfully applied thepeaks. company (Luvitera Ltd., Vilnius, The first THz was for low THz radiation first dual dual frequencies linear THz camera camera was developed developed forin low intensity THz radiation Lithuania) for frequencies beam profilelinear monitoring of the pulse-emitters used theintensity THz time domain detection atRT. RT.The The NBMS camera successfully applied the (Luvitera Ltd., at NBMS camera waswas successfully by camera theby company Ltd., Vilnius, spectrometers (TDS) [40]. Figure 3 shows a photo of theapplied linear THz of 2 ×company 16 (Luvitera pixels with each line Vilnius, Lithuania) for beam profile monitoring of the pulse-emitters used in the THz time domain optimized for 300 GHz and 600 GHz frequencies, respectively. The main advantages of the camera were Lithuania) for beam profile monitoring of the pulse-emitters used in the THz time domain high sensitivity of[40]. 300 V/W and33the NEP aas low asof 14 pW/√Hz. The pixel pitchofand size were 2 each spectrometers Figure shows aphoto photo ofthe thelinear linear THz camera of ˆ pixel 16 pixels with (TDS) Figure shows THz camera 2 ×2the 16 pixels with each line mm and mm × 0.6 mm,600 respectively. The relative detectivity D of main the which equals the square line optimized for 300 GHz and 600 GHz frequencies, respectively. Thepixel, main advantages ofcamera the camera optimized for0.6 300 GHz and GHz frequencies, respectively. The advantages of the were ‘ 4.3  109 cm /√Hz/W. These values root the absorber areaV/W divided bythe theNEP NEP, was estimated to be were highofsensitivity 300 and 14 pW/ Thepitch pixeland pitch thesize pixel size2 high sensitivity of 300ofV/W and the NEP as lowasaslow 14 as pW/√Hz. TheHz. pixel theand pixel were compare well with the detectivity of other RT THz detectors [12,14] and the ˚NBMS camera was good were 2 mm ˆrespectively. 0.6 mm, respectively. Thedetectivity relative detectivity D ofwhich the pixel, which equals mm and 0.6 and mm 0.6 × 0.6mm mm, The relative D of the pixel, equals the square enough to monitor the beam profile of the photoconductive THz antenna emitting power of 10 ‘ μW 9 cm 9 the square root of the absorber area divided by the NEP, was estimated to be 4.3 ˆ 10 / Hz/W. root of the absorber area divided by the NEP, was estimated be 4.3  10 cm /√Hz/W. These values in real time without any additional optical components [28,40].toMoreover, such a THz camera can be These values compare well with the detectivity of other RT THz detectors [12,14] and the NBMS camera compare well with the detectivity of other RT THz detectors [12,14] and the NBMS camera was 2 applied for real-time imaging using minimum intensity of the THz optical field of 2 μW/cm at a 30good was good monitor ofratio. the photoconductive THz antenna power of enough toenough monitor the profile ofprofile the photoconductive antenna emitting power more of 10 μW Hz refresh rateto with abeam ten tothe onebeam signal to noise Although THz the THz camera acquiresemitting images 10 µW in real time without any additional optical components [28,40]. Moreover, such a THz camera in real time anyinadditional optical components [28,40]. Moreover, such athe THz camera can be than tenwithout times faster comparison to the single pixel raster-scan technique, further sample pixel 2 at a2 30 can bebyapplied formethod real-time minimum intensity the THz optical 2 [22]. µW/cm at pixel scan wasimaging implemented in order to avoid unsearchable discretization applied for real-time imaging using using minimum intensity of theofTHz optical fieldfield ofeffects 2 of μW/cm

aHz 30refresh Hz refresh rate with tosignal one signal to noise Although thecamera THz camera acquires images rate with a tenatoten one to noise ratio.ratio. Although the THz acquires images more more than ten times in comparison the single raster-scan technique, further the sample than ten times fasterfaster in comparison to theto single pixel pixel raster-scan technique, further the sample pixel pixel by pixel scan method was implemented in order to avoid unsearchable discretization effects by pixel scan method was implemented in order to avoid unsearchable discretization effects [22].[22].

Figure 3. A photo of dual frequency 2 × 16-pixels THz camera, where the top line of pixels is designed for 600 GHz and the bottom line—for 300 GHz. The pixel pitch is 2 mm.

3. Multispectral THz Imaging High antenna coupled NBMS were for the multispectral THz imaging Figure 3. Asensitivity photo of dual frequency 2 × 16-pixels THz employed camera, where the top line of pixels is designed Figure 3. A photo of dual frequency 2 ˆ 16-pixels THz camera, where the top line of pixels is designed experiment The test samples were prepared as white pellets composed of polytetrafluoroethylene for 600 GHz and the bottom line—for 300 GHz. The pixel pitch is 2 mm. for 600 GHz and the line—for 300 GHz. Thedetails pixel pitch is 2 mm. procedure can be found (PTFE) powder andbottom different admixtures. The on fabrication elsewhere [6,7]. TheImaging samples were packed in a plastic container as shown in Figure 4. Transmittance 3. Multispectral THz 3. Multispectral Imaging spectra of the THz samples obtained by FTIR spectroscopy in vacuum are shown in Figure 5. The pellets High sensitivity antenna coupled NBMS were employed for the multispectral THz imaging High sensitivity antenna coupled NBMS were employed for the multispectral THz imaging experiment The test samples were prepared as white pellets composed of polytetrafluoroethylene experiment The test samples were prepared as white pellets composed of polytetrafluoroethylene (PTFE) powder and different admixtures. The details on fabrication procedure can be found (PTFE) powder and different admixtures. The details on fabrication procedure can be found elsewhere [6,7]. The samples were packed in a plastic container as shown in Figure 4. Transmittance spectra of the samples obtained by FTIR spectroscopy in vacuum are shown in Figure 5. The pellets

Sensors 2016, 16, 432

6 of 15

elsewhere [6,7]. The samples were packed in a plastic container as shown in Figure 4. Transmittance spectra of the samples obtained by FTIR spectroscopy in vacuum are shown in Figure 5. The pellets with the content of lactose and tartaric acid demonstrated different absorption signatures, namely a sharp absorption peak at frequencies of 0.55 THz and 1.1 THz, respectively. Therefore, the NBMS can be Sensors 2016, 16, 432 6 of 15 applied for spectroscopic THz imaging in a similar manner as it was proposed for THz beam profiling Sensors 2016, 16, 432 6 of 15 of thewith photoconductive pulsed-emitters Our developed 300absorption GHz frequency resonant the content of lactose and tartaric [28,40]. acid demonstrated different signatures, namelysensors a can serve reference signal measurement and a 600 frequency sensors covering sharp absorption peak at frequencies of acid 0.55 THz andGHz 1.1 THz, respectively. Therefore, thespectrum NBMS canarange with for the content of lactose and tartaric demonstrated different absorption signatures, namely be applied spectroscopic THz imaging in a similar as it wascomponents, proposed THz beam sharp absorption peakthe at frequencies of 0.55between THz and different 1.1 manner THz, respectively. Therefore,for thesuch NBMS 0.5–0.7 THz canfor provide discrimination chemical ascan lactose, profiling of the photoconductive pulsed-emitters [28,40]. Our developed 300 GHz frequency resonant be acid, applied for spectroscopic tartaric sucrose, etc. [6,7]. THz imaging in a similar manner as it was proposed for THz beam sensors can serve forthe reference signal measurement and 600 GHz frequency sensors profiling of the photoconductive [28,40]. Oura developed 300 frequencies GHz frequency resonant The samples inside plasticpulsed-emitters container were measured at resonant ofcovering the NBMS’s. spectrum range 0.5–0.7 THz can provide the discrimination different chemical components, sensors can serve for reference signal measurement and between a 600 GHz frequency sensors covering The results are shown in Figure 6. The pellet containing lactose had higher absorbance both at 300 GHz such as lactose, sucrose, etc. the [6,7]. spectrum rangetartaric 0.5–0.7 acid, THz can provide discrimination between different chemical components, and 600 GHz frequencies. While other pellets with tartaric acid demonstrated higher absorption only at samples insideacid, the plastic container suchThe as lactose, tartaric sucrose, etc. [6,7].were measured at resonant frequencies of the NBMS’s. the frequency ofare 600 GHz in the accordance with the resultslactose obtained by FTIRfrequencies spectroscopy. This abnormal The results shown in Figure 6. The pellet containing had higher absorbance both at NBMS’s. 300 GHz The samples inside plastic container were measured at resonant of the absorbance of lactose pellets seen 6. only the 300tartaric GHz sensors was attributed toabsorption sideatoptical effects and GHz While other pellets with acidhad demonstrated higher The 600 results arefrequencies. shown in Figure Thewith pellet containing lactose higher absorbance both 300 only GHz such at as defocusing and scattering being more pronounced in smaller diameter samples and thicker the frequency of 600 GHz in other accordance the results by FTIR spectroscopy. and 600 GHz frequencies. While pellets with tartaric acidobtained demonstrated higher absorption This only samples (see Figure 4). abnormal absorbance of GHz lactose seen only the 300obtained GHz sensors wasspectroscopy. attributed to This side at the frequency of 600 in pellets accordance with with the results by FTIR optical effects such as defocusing and scattering being more pronounced in smaller diameter abnormal absorbance ofanalysis lactose pellets seen only with the 300 measured GHz sensors wastransmission attributedsamples to side Principal component was performed by using THz images. and thicker samples Figure 4).and of optical effects such as(see defocusing scattering pronounced in smaller samples Obtained spatial content distribution lactosebeing and more tartaric acid is shown indiameter Figure 7. The blue Principal component analysis was performed by using transmission thicker samples (see Figure 4). sample color and represents the amount of the across pellet as measured describedTHz elsewhere [6,7].images. Note that Obtained spatial content distribution of lactose and tartaric acidmeasured is shown THz in Figure 7. The blue color Principal component analysis was performed by using transmission admixture maps were obtained without opening the plastic container. In this way theimages. suitability represents the amount of distribution the sample across pellet astartaric described Note7.that Obtained spatial content of lactose and acidelsewhere is shown [6,7]. in Figure Theadmixture blue color of the antenna-coupled microbolometer sensors was demonstrated for multispectral THz imaging maps were the obtained without the plastic container. In this way the [6,7]. suitability theadmixture antennarepresents amount of the opening sample across pellet as described elsewhere Note of that applications. Further research will was be oriented towards THz antennaTHz technology development to coupled microbolometer sensors demonstrated for multispectral imagingofapplications. maps were obtained without opening the plastic container. In this way the suitability the antennaincrease resonant frequencies to 2.5 THz [8,9] and integration of compact THz components into a single Further will be oriented THz antenna for technology development to increase resonant coupledresearch microbolometer sensorstowards was demonstrated multispectral THz imaging applications. device [32]. research frequencies to 2.5will THzbe[8,9] and integration of compact THz components into a single deviceresonant [32]. Further oriented towards THz antenna technology development to increase frequencies to 2.5 THz [8,9] and integration of compact THz components into a single device [32].

Figure 4. A photo of test pellets packed inside a plastic container. Samples were prepared by mixing

Figure 4. A photo with of test pellets packed inside a plastic container. Samples were prepared by mixing PTFE tartaric acid, container. 3—Mixture of 5% of tartaric acid and 5% of Figurepowder 4. A photo of1—actose, test pellets2%–10% packed of inside a plastic Samples were prepared by mixing PTFEsucrose. powder with 1—actose, 2%–10% of tartaric acid, 3—Mixture of 5% of tartaric acid and 5% of The reference PTFE pellet labeledofastartaric 4 and a acid, small3—Mixture hex nut were calibration purposes. PTFE powder with 1—actose, 2%–10% ofadded 5% offor tartaric acid and 5% of sucrose. The reference PTFE pellet labeled as 4 and a small hex nut were added for calibration purposes. sucrose. The reference PTFE pellet labeled as 4 and a small hex nut were added for calibration purposes. Frequency (THz)

Transmittance Transmittance

0.3 1 0.3 1

0.5

0.6 Frequency 0.8 0.9(THz) 1.1

1.2

1.4

0.5

0.6

1.2

1.4

40

45

40

45

0.8

0.9

1.1

0.1 0.1 0.01

Lactose Tartaric Lactose acid Sucrose Tartaric acid PTFE Sucrose

0.01 1E-3 10 1E-3 10

15 15

PTFE30 25 35 -1 Wavenumber 20 25 30 (cm35) -1 Wavenumber (cm )

20

Figure 5. Transmittance spectrum of the samples with lactose, tartaric acid, sucrose, and reference PTFE. Figure 5. Transmittance spectrum of the samples with lactose, tartaric acid, sucrose, and reference PTFE.

Figure 5. Transmittance spectrum of the samples with lactose, tartaric acid, sucrose, and reference PTFE.

Sensors 2016, 16, 432 Sensors 2016, 16, 432

Sensors 2016, 16, 432

7 of 15

Sensors 2016, 16, 432

7 of 15 7 of 15

7 of 15

(a)

(b)

Figure 6. The (a) THz transmission image of plastic container measured with the (b) system based on the

Figure 6.NBMS The designed THz transmission imageofof measured with the system based on the for the frequencies 300plastic GHz (a)container and 600 GHz (b), respectively Figure The THzfortransmission image of plastic container with the system based on the NBMS6.designed the frequencies of 300 GHz (a) and 600 measured GHz (b), respectively. NBMS designed for the frequencies of 300 GHz (a) and 600 GHz (b), respectively (a) (b) Figure 6. The THz transmission image of plastic container measured with the system based on the NBMS designed for the frequencies of 300 GHz (a) and 600 GHz (b), respectively Figure 7. Content distribution map of the lactose (Left) and tartaric acid (Right) in the pellets packed in a plastic container as shown in Figure 4. The intensity scale is linear and represents the amount of the component across the pellet.

Figure 7. Content distribution map of the lactose (Left) and tartaric acid (Right) in the pellets packed Figure 7. ContentTHz distribution mapfor of High the lactose (Left) and tartaric acid (Right) in the pellets packed 4. Diffractive Components Spatial Resolution Imaging in a plastic container as shown in Figure 4. The intensity scale is linear and represents the amount of in a plastic container as shown in Figure 4. The intensity scale is linear and represents the amount of The experimental setup the component across the pellet.of the compact all-electronic RT THz imaging system is shown in Figure Figure 7. Content the component across distribution the pellet. map of the lactose (Left) and tartaric acid (Right) in the pellets packed

8. The source of THz radiation was an electronic multiplier chain (Virginia Diodes, Inc., in a plastic container as shown in Figure 4. The intensity scale is linear and represents the amount of Charlottesville, VA, USA) delivering of about 0.8 mW power at aImaging frequency of 584 GHz. Emitted THz 4. Diffractive THz Components for High Spatial Resolution the component across the pellet. 4. Diffractive Components High Resolution(PE) Imaging radiationTHz was collimated with for 12 cm focalSpatial length polyethylene lens L1. The THz beam reflected by 2 inch diameter flatof mirror M was focused with theRT THz lensimaging TZP andsystem directedistoshown sampleinS.Figure The experimental setup the compact all-electronic THz 4. Diffractive THz setup Components for Highall-electronic Spatial Resolution Imaging The experimental of the compact RT THz imaging system is shown in Figure 8. was collimated 6 cm focal length PE lens L2 and focused onto THz Inc., 8. The Transmitted source of THz THzradiation radiation was an with electronic multiplier chain (Virginia Diodes, The source of experimental THz radiation was an electronic multiplier chain (Virginia Charlottesville, The setup the compact RT arrangement THz imagingofDiodes, system isInc., shown in Figure detector D with PM (P1) of 5of cm focal length.all-electronic The photo of the the optical components Charlottesville, VA, USA) delivering of about 0.8 mW power at a frequency of 584 GHz. Emitted THz VA, USA) delivering of aboutinradiation 0.8 mW8 power a frequency of 584 GHz. Emitted radiation samples shown Figure (on the right). The samples were raster-scanned byDiodes, position8. and The source isof THz was anat electronic multiplier chain (VirginiaTHz Inc., was radiationsynchronized was collimated with 12 cm focal length polyethylene (PE) lens L1. The THz technique beam reflected measurements in Cartesian coordinates [22]. sensitive detection collimated with 12 VA, cm focal delivering length polyethylene (PE) lensAat L1. The lock-in THzofbeam reflected 2 inch Charlottesville, USA) of about 0.8 mW power a frequency 584 GHz. Emittedby THz by 2 inch diameter flat mirror M was focused with the THz lens TZP and directed to sample S. was used with modulation frequency and time constant being set to 1.46 kHz and 10 ms, respectively. radiation was collimated 12 cm focalthe length (PE)directed lens L1. The THz beam diameter flat mirror M was with focused with THzpolyethylene lens TZP and to sample S. reflected Transmitted

Transmitted THz radiation collimated with 6 with cm focal length PE lens L2directed and focused ontoS.THz by 2 inch diameter flatwas mirror M6was the TZP and sample THz radiation was collimated with cm focused focal length PE THz lens lens L2 and focused onto to THz detector D detector D with PM (P1) of 5 cm focal length. The photo of the arrangement of the optical components Transmitted THz radiation was collimated with 6 cm focal length PE lens L2 and focused onto THz with PM (P1) of 5 cm focal length. The photo of the arrangement of the optical components and and samples shown in Figure (onlength. the right). Theofsamples were raster-scanned by positiondetector is D with PM (P1) of 5 cm 8focal The photo the arrangement of the optical components samples is shown in Figure 8 (on the right). The samples were raster-scanned by position-synchronized and samples is shown inin Figure 8 (on coordinates the right). The samples were raster-scanned by positionsynchronized measurements Cartesian [22]. A sensitive lock-in detection technique measurements in Cartesian coordinates [22]. A sensitive lock-in detection technique was used with synchronized measurements in Cartesian coordinates [22]. Aset sensitive detection technique was used with modulation frequency and time constant being to 1.46 lock-in kHz and 10 ms, respectively. modulation frequency and time constant being to 1.46 kHzsetand 10 ms, was used with modulation frequency and time set constant being to 1.46 kHzrespectively. and 10 ms, respectively.

Figure 8. THz imaging setup (Left), where M is a flat mirror; L1 and L2—PE lenses, TZP—a diffractive lens [30,31], S—a sample on three-axes translation stage, P1—a parabolic mirror, D—compact THz detectors. The photograph (Right) shows the arrangement of the components in the experiment.

We have designed the TZP component to manipulate the Gaussian beam of 584 GHz frequency. The diffractive lens design was similar to the conventional Fresnel lens with the main difference being the integration of the resonant cross shape apertures inside open regions [30,31]. Resonant apertures 8. THz imaging setup (Left), where M is a flat mirror; L1 and L2—PE lenses, TZP—a diffractive FigureFigure THz imagingsetup setup(Left), (Left),where whereMMisisa aflat flatmirror; mirror;L1 L1and andL2—PE L2—PElenses, lenses,TZP—a TZP—adiffractive diffractive Figure 8.8.THz imaging lens [30,31], S—a sample on three-axes translation stage, P1—a parabolic mirror, D—compact THz lens [30,31], S—a sample on three-axes translation stage, P1—a parabolic mirror, D—compact THz lens [30,31], S—a sample on three-axes translation stage, P1—a mirror, D—compact THz detectors. The photograph (Right) shows the arrangement of the parabolic components in the experiment. detectors. The photograph (Right) shows the arrangement of the components in the experiment. detectors. The photograph (Right) shows the arrangement of the components in the experiment. We have designed the TZP component to manipulate the Gaussian beam of 584 GHz frequency. The diffractive lens the design was similar to the lens with the main difference being We have designed theTZP TZP component manipulateFresnel theGaussian Gaussian beam 584 GHzfrequency. frequency. We have designed component totoconventional manipulate the beam ofof584 GHz the integration of the resonant cross shape apertures inside open regions [30,31]. Resonant apertures Thediffractive diffractivelens lens design was similar to conventional the conventional Fresnel lens the with the difference main difference The design was similar to the Fresnel lens with main being

the integration of the resonant cross shape apertures inside open regions [30,31]. Resonant apertures

transmittance at the desired frequency [41]. The focal distance and diameter of the TZP were selected to be 10 mm and 16.5 mm, respectively. The diffractive lens was fabricated from 30 μm thick molybdenum foil by the direct laser writing. Typical performances of the developed diffractive component are shown in Figure 9. The NA of the TZP was measured by three-dimensional Gaussian beam profiling [39]. The NA value was estimated to be about 0.62. Sensors 88of Sensors 2016, 2016, 16, 16, 432 x of15 15 The performance of the THz system based on the TZP lens versus a commercial off-axis PM was compared by imaging a spatial resolution target. The diameter and focal length of the PM were of 2 We have designed the TZP component to manipulate the Gaussian beam of 584 GHz frequency. being integration of the resonant cross apertures openon regions [30,31]. Resonant inchesthe leading to the NA = 0.45. This was theshape highest NA thatinside was found the components market. The diffractive lens design was similar to the conventional Fresnel lens with the main difference being apertures of the length K = 260 µm, width M = 30 µm, and period L = 300 µm were used to obtain Imaging results of the resolution target are shown in Figures 10 and 11. It is seen that the system the integration of the resonant cross shape apertures inside open regions [30,31]. Resonant apertures the peak ofwith transmittance at the desired frequency Thebetter focal distance and diameter of the TZP equipped diffractive component TZP provided[41]. a much spatial resolution. Periodic stripes of the length K = 260 µ m, width M = 30 µ m, and period L = 300 µ m were used to obtain the peak of were to be 10 ifmm 16.5was mm,not respectively. The was fabricated fromthat 30 µm were selected distinguishable theand period smaller than 0.6diffractive mm in thelens case of TZP lens; note the transmittance at the desired frequency [41]. The focal distance and diameter of the TZP were selected thick molybdenum foil by the direct laser writing. Typical performances of the developed diffractive resolution was limited wavelength of used THz radiation. And in the case of off-axis PM, the to be 10 mm and 16.5 mm, respectively. The diffractive lens was fabricated from 30 µ m thick component are shown in Figure 9. The NAofofabout the TZP three-dimensional smallest period of stripes was measured 0.8 was mm.measured Thus, theby imaging system withGaussian the TZP molybdenum foil by the direct laser writing. Typical performances of the developed diffractive beam profiling [39]. The NAinvalue was estimatedoftoup betoabout lens exibited improvement spatial resolution 25%. 0.62. component are shown in Figure 9. The NA of the TZP was measured by three-dimensional Gaussian beam profiling [39]. The NA value was estimated to be about 0.62. The performance of the THz system based on the TZP lens versus a commercial off-axis PM was compared by imaging a spatial resolution target. The diameter and focal length of the PM were of 2 inches leading to the NA = 0.45. This was the highest NA that was found on the components market. Imaging results of the resolution target are shown in Figures 10 and 11. It is seen that the system equipped with diffractive component TZP provided a much better spatial resolution. Periodic stripes were distinguishable if the period was not smaller than 0.6 mm in the case of TZP lens; note that the resolution was limited by the wavelength of used THz radiation. And in the case of off-axis PM, the smallest period of stripes was measured of about 0.8 mm. Thus, the imaging system with the TZP lens exibited improvement in spatial resolution of up to 25%.

Figure 9. of of collimated (a) and focused with the TZP (b)lens THz(b) beam of 584 GHz Figure 9. Shape Shape collimated (a) and focused with thelens TZP THz beam of frequency. 584 GHz Note the difference scales; (c) along the optical axisoptical after the diffractive lens; (d) Crossfrequency. Note the in difference inBeam scales;shape (c) Beam shape along the axis after the diffractive lens; section of collimated and focused THz radiation at peak intensity area. In all measurements the pixel (d) Cross-section of collimated and focused THz radiation at peak intensity area. In all measurements 2 2 , except×(a)—200 500 μm2ˆ . 500 µm2 . sizepixel was size 50 × was 50 μm the 50 ,ˆexcept 50 µm(a)—200

y (mm)

Photo PM TZP The performance of the THz system based on the TZP lens versus a commercial off-axis PM was (a) (b) (c) compared by imaging a spatial resolution target. The diameter and focal length of the PM30were of 2 inches leading to the NA = 0.45. This was the highest NA that was found on the components market. Imaging results of the resolution target are shown in Figures 10 and 11. It is seen that the20system equipped with diffractive component TZP provided a much better spatial resolution. Periodic stripes wereFigure distinguishable the period not smaller than 0.6lens mm(b)inTHz the beam case of lens;frequency. note10 that the 9. Shape of if collimated (a) was and focused with the TZP of TZP 584 GHz resolution was limited by the wavelength of used THz radiation. And in the case of off-axis PM, Note the difference in scales; (c) Beam shape along the optical axis after the diffractive lens; (d) Cross- the 0 smallest period of stripesand was measured about 0.8 mm.intensity Thus, the imaging system with the thepixel TZP lens section of collimated focused THz of radiation at peak area. In all measurements 0 10 20 30 40 50 0 10 20 30 40 50 2. to 25%. exibited spatial resolution up size improvement was 50 × 50 µ m2in , except (a)—200 × 500of µm x (mm) x (mm) PM Photo TZP Figure 10. A photo of the resolution target consisting of a set of periodic metal stripes with the period (a) (c)at 584 GHz frequency obtained by indicated by number in (b) mm (a). THz image of the resolution target 30

10

y (mm)

using commercial PM (b) and novel diffractive lenses (c). Black color in the THz images corresponds to the maximum of transmittance. 20

0 0

10

20

30

x (mm)

40

50

0

10

20

30

40

50

x (mm)

Figure 10. 10. A consisting of of aa set set of of periodic metal stripes stripes with with the the period period Figure A photo photo of of the the resolution resolution target target consisting periodic metal indicated by number in mm (a). THz image of the resolution target at 584 GHz frequency obtained indicated by number in mm (a). THz image of the resolution target at 584 GHz frequency obtained by by using commercial commercialPM PM(b) (b)and andnovel noveldiffractive diffractivelenses lenses(c). (c).Black Blackcolor color the THz images corresponds using inin the THz images corresponds to to the maximum of transmittance. the maximum of transmittance.

Sensors 2016, 16, 432 Sensors 2016, 16, 432

9 of 15 9 of 15

Signal (a.u.)

TZP

PM

3

(a)

2

1

y=32 mm

y=32 mm

y=20 mm

y=20 mm

(b)

y=8 mm

y=8 mm

0 0

10

20

30

40

50

0

10

x (mm)

20

30

40

50

x (mm)

Figure 11. 11. A A cross cross section section profile profilealong alongthe thex-axis x-axisof ofthe theTHz THzimage imageat atpositions positionsyy== 8, 8, 20, 20, and and 33 33 mm mm Figure measuredusing usingdifferent differentlenses: lenses:PM PM(a) (a)and andTZP TZP(b). (b). measured

5. A 5. A Compact Compact THz THz Imaging Imaging System System for for Medical Medical Applications Applications Firstdark-field dark-fieldTHz THz imaging with the intention of enhancing image contrast the First imaging with the intention of enhancing image contrast through through the analysis analysis of scattering and diffraction signatures was demonstrated back in 2001 [33]. That time the of scattering and diffraction signatures was demonstrated back in 2001 [33]. That time the samples samples contained obvious area ofwith fat, skin with hairs, connective and the tumor. Dark-field contained obvious area of fat, skin hairs, connective tissue andtissue the tumor. Dark-field imaging imaging discriminated between scattering/diffraction and absorption and demonstrated discriminated between scattering/diffraction losses andlosses absorption losses andlosses demonstrated enhanced enhanced imageatcontrast at boundaries edges. The development of microscopy confocal microscopy THz image contrast boundaries and edges.and The development of confocal THz systems systems deserved attentionwith because with the confocal spatial filtering it was possiblethe to deserved particularparticular attention because the confocal spatial filtering it was possible to improve improve the resolution and contrast of the image. Confocal THz systems based on powerful THz resolution and contrast of the image. Confocal THz systems based on powerful THz sources such as gas sources such gas THz lasers andQCLs cryogenically QCLs have As been proposed As THz THz lasers andascryogenically cooled have beencooled proposed [25,42]. THz radiation[25,42]. may penetrate radiation maytopenetrate into to a depth 10 mm or deeper, theconfocal development of THz systems confocal into the skin a depth of 10the mmskin or deeper, theofdevelopment of THz microscopy microscopy systems for medicine is of prime interest. The THz-TDS spectroscopy results of a human for medicine is of prime interest. The THz-TDS spectroscopy results of a human basal cell carcinoma basal cellthat carcinoma increased refractive index and absorption coefficient were the revealed increasedrevealed refractivethat index and absorption coefficient were the sources of contrast at THz sources of contrast THzchanges frequencies and that these were consistent with an increase of frequencies and that at these were consistent with changes an increase of water content in the tissue [37]. water content in the tissue [37]. In addition, the THz TDS spectra of dehydrated and paraffinIn addition, the THz TDS spectra of dehydrated and paraffin-embedded human colon tissues revealed embedded human colon tissues revealed that a higher ofonly watercontributing in cancerousfactor tissues that a higher percentage of water in cancerous tissuespercentage was not the to was the not the only contributing factor to the contrast [35]. Higher refractive index and absorption coefficient contrast [35]. Higher refractive index and absorption coefficient were observed in neoplastic tissues by were observed in neoplastic byline measuring spectra at several points reinforced in line along sample. measuring spectra at several tissues points in along the sample. Such research thethe feasibility Such research reinforced the stage feasibility THz techniques stage colon cancersystems detectionwere and of THz techniques for early colonofcancer detection for andearly various THz imaging various THz imaging systems were developed for this purpose [38,43]. Contrast up to 23% between developed for this purpose [38,43]. Contrast up to 23% between the neoplastic and non-neoplastic the neoplastic and non-neoplastic tissues observed in THz absorption and reflection images tissues was observed in THz absorption andwas reflection images averaged over the specified regions. averaged over the specified regions. 5.1. Demonstration of Compact THz Imaging System Performance 5.1. Demonstration of Compact THz Imaging System Performance In order to show suitability of a compact THz imaging system for medical applications, we used In ordersections to showwith suitability of a compact imaging system medical applications, we used histological colon cancer for theTHz experiments [43,44]. for Digital photos of the embedded histological sectionsand withfree-laying colon cancer forare theshown experiments [43,44]. photos of the in in paraffin samples ones in the top lineDigital of Figures 12 and 13embedded respectively. paraffin samples free-laying are (#H13.23034) shown in the by topthe lineright of Figures 12 and 13, respectively. The samples wereand obtained of theones patient hemicolectomy procedure via The samples obtained the patient (#H13.23034) by the procedure via selection of thewere specimens andoffurther fixation in formalin and afterright that hemicolectomy in paraffin blocks as described selection of the specimens andanalysis furtherclassified fixation sample in formalin and after that in paraffin blockspT3 as elsewhere [43,44]. The histologic #H13.23034T as colon adenocarcinoma described elsewhere [43,44]. histologic analysis colon N0 Mx in the TNM system withThe the original tumor size of classified about 4 cmsample ˆ 3.5 cm#H13.23034T ˆ 1.5 cm. Theassecond adenocarcinoma pT3 was N0 Mx in the of TNM system with thetumor original sizesurgery of about 4 cm  3.5and cm sample #H13.23034N obtained the non-neoplastic via tumor the same procedures  1.5studied cm. The sample #H13.23034N was obtained of the non-neoplastic tumor via the same was forsecond comparison purposes. surgery and measured was studied comparisongeometry, purposes.i.e. the radiation of THz frequency Theprocedures samples were in for transmission The samples were measured in transmission geometry, i.e.the theL2radiation THz frequency scattered and transmitted through the sample was collected with lens and of detected with the scattered and transmitted through sample was collected with the L2 lens and detected withNA the NBMS detector D (see also Figure 8).the In this experiment a commercial off-axis PM with the highest NBMS detector 8).note In this a commercial off-axis PM applied. with the highest was used insteadDof(see the also TZP Figure lens. We thatexperiment the confocal spatial filtering was not NA was used instead of the TZP lens. We note that the confocal spatial filtering was not applied.

Sensors 2016, 16, 432

10 of 15

Sensors 2016, 16, 432

10 of 15

H13.23034N

H13.23034T

Sensors 2016, 16, 432

10 of 15

H13.23034N

(a) (a)

(c)

H13.23034T

(b) (b)

(d)

Figure 12. A digital photograph (a,b) and THz image (c,d) of the histopathologic sections of non-

(c) (a,b) and THz image (c,d) of(d) Figure 12. A digital photograph the histopathologic sections of neoplastic and neoplastic colon tissue placed in a paraffin block. The control sample is presented on non-neoplastic and neoplastic colon tissue placed in a paraffin block. The control sample is presented 12. side A digital photograph (a,b) and THz the image (c,d) of the sections of color nontheFigure left-hand (a,c) and adenocarcinoma—on right-hand sidehistopathologic (b,d) column. The white in neoplastic and neoplastic colon tissue placed in a paraffin block. The control sample is presented on on the left-hand side (a,c) and adenocarcinoma—on the right-hand side (b,d) column. The white color the THz image corresponds to higher THz absorbance. The blue line indicates the contour of the tissue side (a,c) andto adenocarcinoma—on the right-hand side (b,d)line column. The white in in the THz image corresponds higher absorbance. The blue indicates thecolor contour of the andthe theleft-hand number—an averaged value of THz THz absorption inside the contour. the THz image corresponds to higher THz absorbance. The blue line indicates the contour of the tissue tissue and the number—an averaged value of THz absorption inside the contour. and the number—an averaged value of THz absorption inside the contour.

H13.23034N

H13.23034N

(a) (a)

(c) (c)

H13.23034T

H13.23034T

(b) (b)

(d) (d)

Figure 13.13. A digital photograph (c,d) of of the thesame samehistopathologic histopathologic sections Figure A digital photograph(a,b) (a,b)and andTHz THz image image (c,d) sections as as in in

Figure 13. A digital photograph (a,b)taken andout THz image and (c,d)placed of the histopathologic sections as in Figure 12 12 butbut the samples placed onsame mm thickPEPE plate and Figure the sampleswere weretaken outof of paraffin paraffin and on aa22mm thick plate and thethe Figure 12 but the samples were taken out ofthe paraffin and on a 2 ismm thick PE plate and THz imaging was performed employing TZP The controlsample sample ison onthe the left-hand side (a,c)the THz THz imaging was performed employing the TZP lens. lens. Theplaced control left-hand side (a,c) adenocarcinoma—on the side column. sample The white white color theTHz THz image and adenocarcinoma—on theright-hand right-hand side (b,d) (b,d) column. The inin the image imagingand was performed employing the TZP lens. The control is color on the left-hand side (a,c) and corresponds to to higher THz Suspicious regions were indicated ablue blue line. number corresponds higher THzabsorbance. absorbance. Suspicious regions indicated line. AA number adenocarcinoma—on the right-hand side (b,d) column. Thewere white color by inbyathe THz image corresponds inside each region presentsthe theaveraged averagedvalue value of of THz THz absorption. inside each region presents absorption. to higher THz absorbance. Suspicious regions were indicated by a blue line. A number inside each region presents the averaged value ofembedded THz absorption. THz image theparaffin paraffin embedded samples at in in Figure 12.12. TheThe THz image ofof the samples at 584 584 GHz GHzfrequency frequencyisisshown shown Figure mean value the THzabsorption absorptionwas was obtained obtained via thethe range TheThe mean value of of the THz via averaging averagingmeasurements measurementsinside inside range of interest (ROI) region indicatedby byaablue blueline line contour. contour. An averaged and standard deviation values of interest (ROI) region indicated An averaged and standard deviation values The THz image of the paraffin embedded samples at 584 GHz frequency is shown in Figure 12.

The mean value of the THz absorption was obtained via averaging measurements inside the range of interest (ROI) region indicated by a blue line contour. An averaged and standard deviation values were obtained for each ROI region and were shown in Figure 12. The adenocarcinoma regions can be observed as more THz absorbing regions even in dehydrated tissues due to increased radiation scattering caused by structural changes [43]. In this case, an averaged values of the THz absorption for samples H13.23034T and H13.23034N were 3.8 ˘ 1.6 and 3.0 ˘ 1.5, respectively. Systematic

Sensors 2016, 16, 432

11 of 15

were obtained for each ROI region and were shown in Figure 12. The adenocarcinoma regions can be observed more THz absorbing regions even in dehydrated tissues due to increased radiation Sensors 2016, as 16, 432 11 of 15 scattering caused by structural changes [43]. In this case, an averaged values of the THz absorption for samples H13.23034T and H13.23034N were 3.8  1.6 and 3.0  1.5, respectively. Systematic investigation investigation allows allows drawing drawing statistically statistically significant significant assessments assessments distinguishing distinguishing non-neoplastic non-neoplastic and and neoplastic tissues [43]. Therefore one sample measurement even with high NA optics neoplastic tissues [43]. Therefore one sample measurement even with high NA optics is is not not always always sufficient. 12,12, thethe tissue contour and and different absorption regions in the in THz sufficient.As Asshown shownininFigure Figure tissue contour different absorption regions theimage THz were blurred out and did not obviously correspond to the visible image most probably due to image were blurred out and did not obviously correspond to the visible image most probably duethe to interference fromfrom the paraffin blocks and and surroundings of the holder. the interference the paraffin blocks surroundings ofsample the sample holder. To To overcome overcome these these limits, limits, the the paraffin paraffin embedded embedded sample sample was was placed placed on on aa 22 mm mm thick thick PE PE plate plate ˝ C just enough to melt the paraffin away. A digital photo of and heated up to temperature of 50 and heated up to temperature of 50 °C just enough to melt the paraffin away. A digital photo of the the sample sticked to PE the plate PE plate via residual paraffin is shown in Figure 13. measurements The measurements sample sticked to the via residual paraffin is shown in Figure 13. The were were performed with the THz imaging system taking preference for the TZP lens with NA = performed with the THz imaging system taking preference for the TZP lens with NA = 0.62.0.62. The The samples oriented normal to the beam to the lens. Measured images samples werewere oriented normal to the THzTHz beam andand faceface to the TZPTZP lens. Measured THzTHz images are are shown in Figure 13.this In case this the caseinterference the interference was removed PEbehind plate behind the shown in Figure 13. In effecteffect was removed as the as PEthe plate the tissues tissues had constant thickness that facilitated identification of physical boundaries between different had constant thickness that facilitated identification of physical boundaries between different tissue tissue regions. Suspicious regions of increased THz absorption were selected and marked by a blue regions. Suspicious regions of increased THz absorption were selected and marked by a blue line. A line. meanofvalue of the THz absorption was via found via measurements averaging inside ROI meanAvalue the THz absorption was found measurements averaging inside the ROIthe region. region. Thevalues mean values are shown in Figure 13.topographical The topographical thickness the sample The mean found found are shown in Figure 13. The thickness of theofsample was was measured a three-dimensional digital microscope(Hirox (HiroxEurope EuropeLtd., Ltd., Limonest, Limonest, France). measured withwith a three-dimensional digital microscope France). Results Results are are shown shown in in Figure Figure 14. 14. Obviously, Obviously, the the change change in in the the THz THz absorption absorption of of the the neoplastic neoplastic tissue tissue (H13.23034T) does not correlate with variation in the sample thickness and morphology (H13.23034T) does not correlate with variation in the sample thickness and morphology differently differently from Moreover, the the THz from the the case caseof ofthe thenon-neoplastic non-neoplasticsample sample(H13.23034N). (H13.23034N). Moreover, THz image image of of non-neoplastic non-neoplastic sample demonstrated separate regions where averaged THz absorption value changed sample demonstrated separate regions where averaged THz absorption value changedfrom from2.6 2.6˘ 0.6 0.6 up up to to 4.9 4.9˘  1.1. 1.1. Note Note that that the the main main indication indication of of the the adenocarcinoma adenocarcinoma tissues tissues is is increased increased absorption absorption of of THz THz radiation radiation [43,44]. [43,44]. For For proper proper medical medical diagnostics diagnostics one one needs needs to to perform perform more more systematic systematic research research with samplesalso alsoaccounting accounting results of topographical measurements. However, that is with more more samples thethe results of topographical measurements. However, that is beyond beyond the of this paper. the scope ofscope this paper.

Figure 14. 14. Thickness Thickness and and surface surface profiles profiles of of sample sample H13.23034N H13.23034N (left-hand (left-hand side) side) and and H13.23034T H13.23034T Figure (right-hand side) obtained with a three-dimensional digital microscope. Note the differences between (right-hand side) obtained with a three-dimensional digital microscope. Note the differences the sample structure and THz absorption regions at a virtual cross section layer shown in Figure 13 between the sample structure and THz absorption regions at a virtual cross section layer shown in and supplementary materials for the H13.23034N (Animation1.gif) H13.23034T Figure 13 and Supplementary Materials forsample the sample H13.23034N (Animation1.gif)and and H13.23034T (Animation2.gif), respectively. (Animation2.gif), respectively.

5.2. Towards Confocal THz Microscopy 5.2. Towards Confocal THz Microscopy Exploring the performance of the THz imaging system such as high signal to noise ratio Exploring the performance of the THz imaging system such as high signal to noise ratio provided provided by the NBMS’s and high spatial resolution provided by the TZP, there was an interest to by the NBMS’s and high spatial resolution provided by the TZP, there was an interest to emulate emulate functionality of the confocal THz microscope as complementary tool for pathology. The axial functionality of the confocal THz microscope as complementary tool for pathology. The axial resolution resolution of the THz system based on the TZP lens was estimated by the Rayleigh length criterion [39]. of the THz system based on the TZP lens was estimated by the Rayleigh length criterion [39]. ‘ found to be of about The distance, along which the focused beam waist increased by factor of 2, was The distance, along which the focused beam waist increased by factor of 2, was found to be of 1.4 mm (see Figure 9c). The samples on PE plate were imaged in transmission mode with only slight about 1.4 mm (see Figure 9c). The samples on PE plate were imaged in transmission mode with only focusing adjustment by moving the sample in axial-steps of 1 mm. In such a way a set of THz images slight focusing adjustment by moving the sample in axial-steps of 1 mm. In such a way a set of THz images at different focal planes were obtained at 584 GHz. The results for the sample H13.23034N and H13.23034T were animated and presented in a section of Supplementary Materials Animation1.gif

Sensors 2016, 16, 432

12 of 15

and Animation2.gif, respectively. We demonstrate a virtual cross sections, roughly 1 mm deep, within the sample due to varying the ratio between transmitted and scattered THz radiation by the sample. Fine tissue structure and especially regions with different THz contrast were clearly observed. In the next stage, a confocal spatial filtering will be implemented to provide high quality THz images with fine details and better contrast within the objects. 6. Conclusions The spectral performance of the THz sensors based on air-bridged Ti-microbolometers coupled with a dipole antenna has been measured experimentally by FTIR spectroscopy at room temperature. These sensors were optimized for operation at a selected frequency of 300 GHz or 600 GHz and were used to develop an all-electronic THz imaging system suitable for plastic package inspection and spectroscopic-spatial analysis of materials. Next to it, a thin and lightweight focusing component called the TZP lens with the numerical aperture of 0.62 has been developed and used to enhance the performance of the THz imaging system. This was demonstrated by imaging carcinoma-affected biological tissues with close to wavelength limited spatial resolution. The proposed compact THz imaging systems prove the maturity of THz technology for medicine and security applications. Supplementary Materials: The following are available online at http://www.mdpi.com/1424-8220/16/4/432/s1, the confocal THz imaging data for the sample H13.23034N (Animation1.gif) and H13.23034T (Animation2.gif) measured at a frequency of 584 GHz. Acknowledgments: The authors acknowledge financial support for this work from the BM1205 COST Action for Short Term Scientific Mission of research Grants. Author Contributions: I.K. conceived the experiments, was involved in the development of compact THz components and methods; R.V. performed THz spectroscopy, was involved in the development THz systems; L.M. designed diffractive lenses, performed the TZP experiments; R.V. and L.M. performed the THz imaging experiments; F.W. provided biomedical samples, made their histopathological research; B.V. fabricated diffractive optic components; V.T. conceived diffractive optics designs, was involved in TZP development; A.S., A.Š. and J.T. contributed to concept of compact THz sensors and fabricated micorobolometer THz sensors; G.V., I.K., D.S., and V.T. were involved in the concept development; I.K. wrote the manuscript with contributions from all the authors. Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations The following abbreviations are used in this manuscript: THz RT NEP FET TeraFET QCL PTFE PE FTIR NBMS THz TDS TZP CMOS DA CDA PM D˚ SNR ROI

Terahertz Room temperature Noise-equivalent-power Fielf effect transistor Terahertz antenna-coupled fielf effect transistor Quantum cascade laser Polytetrafluoroethylene Polyethylene Fourier transform infrared Narrow band microbolometer terahertz sensor Terahertz time domain spectrometer Terahertz zone plate with integrated resonant filter apertures Complementary metal-oxide semiconductor Double dipole antenna Cross dipole antenna Parabolic mirror Relative detectivity Signal to noise ratio Range of interest

Sensors 2016, 16, 432

13 of 15

References 1. 2. 3. 4. 5.

6. 7.

8.

9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

Chan, W.L.; Deibel, J.; Mittleman, D.M. Imaging with Terahertz Radiation. Reports Prog. Phys. 2007, 70, 1325–1379. [CrossRef] Siegel, P.H. Terahertz Technology in Biology and Medicine. IEEE Trans. Microw. Theory Techn. 2004, 52, 2438–2447. [CrossRef] Yu, C.; Fan, S.; Sun, Y.; Pickwell-Macpherson, E. The Potential of Terahertz Imaging for Cancer Diagnosis: A Review of Investigations to Date. Quant. Imaging Med. Surg. 2012, 2, 33–45. [PubMed] Mickan, S.P.; Menikh, A.; Liu, H.; Mannella, C.A.; MacColl, R.; Abbott, D.; Munch, J.; Zhang, X.C. Label-Free Bioaffinity Detection Using Terahertz Technology. Phys. Med. Biol. 2002, 47, 3789–3795. [CrossRef] [PubMed] Dean, P.; Valavanis, A.; Keeley, J.; Bertling, K.; Lim, Y.L.; Alhathlool, R.; Burnett, A.D.; Li, L.H.; Khanna, S.P.; Indjin, D.; et al. Terahertz Imaging Using Quantum Cascade Lasers—a Review of Systems and Applications. J. Phys. D. Appl. Phys. 2014, 47, 374008. [CrossRef] Kasalynas, I.; Venckevicius, R.; Valusis, G. Continuous Wave Spectroscopic Terahertz Imaging With InGaAs Bow-Tie Diodes at Room Temperature. IEEE Sens. J. 2013, 13, 50–54. [CrossRef] Bauer, M.; Venckeviˇcius, R.; Kašalynas, I.; Boppel, S.; Mundt, M.; Minkeviˇcius, L.; Lisauskas, A.; Valušis, G.; Krozer, V.; Roskos, H.G. Antenna-Coupled Field-Effect Transistors for Multi-Spectral Terahertz Imaging up to 4.25 THz. Opt. Express 2014, 22, 19235–19241. [CrossRef] [PubMed] Simoens, F.; Meilhan, J.; Nicolas, J.-A. Terahertz Real-Time Imaging Uncooled Arrays Based on Antenna-Coupled Bolometers or FET Developed at CEA-Leti. J. Infrared, Millimeter, Terahertz Waves 2015, 36, 961–985. [CrossRef] Oda, N.; Kurashina, S.; Miyoshi, M.; Doi, K.; Ishi, T.; Sudou, T.; Morimoto, T.; Goto, H.; Sasaki, T. Microbolometer Terahertz Focal Plane Array and Camera with Improved Sensitivity in the Sub-Terahertz Region. J. Infrared, Millimeter, Terahertz Waves 2015, 36, 947–960. [CrossRef] Al Hadi, R.; Sherry, H.; Grzyb, J.; Zhao, Y.; Förster, W.; Keller, H.M.; Cathelin, A.; Kaiser, A.; Pfeiffer, U.R. A 1 K-Pixel Video Camera for 0.7-1.1 Terahertz Imaging Applications in 65-Nm CMOS. IEEE J. Solid-State Circuits 2012, 47, 2999–3012. [CrossRef] Dufour, D.; Marchese, L.; Terroux, M.; Oulachgar, H.; Généreux, F.; Doucet, M.; Mercier, L.; Tremblay, B.; Alain, C.; Beaupré, P.; et al. Review of Terahertz Technology Development at INO. J. Infrared, Millimeter, Terahertz Waves 2015, 36, 922–946. [CrossRef] Kasalynas, I.; Adam, A.J.L.; Klaassen, T.O.; Hovenier, J.N.; Pandraud, G.; Iordanov, V.P.; Sarro, P.M. Design and Performance of a Room-Temperature Terahertz Detection Array for Real-Time Imaging. IEEE J. Sel. Top. Quantum Electron. 2008, 14, 363–369. [CrossRef] Grant, J.; Escorcia-Carranza, I.; Li, C.; Mccrindle, I.J.H.; Gough, J.; Cumming, D.R.S. A Monolithic Resonant Terahertz Sensor Element Comprising a Metamaterial Absorber and Micro-Bolometer. Laser Photonics Rev. 2013, 7, 1043–1048. [CrossRef] Hack, E.; Valzania, L.; Gäumann, G.; Shalaby, M.; Hauri, C.; Zolliker, P. Comparison of Thermal Detector Arrays for Off-Axis THz Holography and Real-Time THz Imaging. Sensors 2016, 16, 221. [CrossRef] [PubMed] Dyakonov, M.I.; Shur, M.S. Plasma Wave Electronics: Novel Terahertz Devices Using Two Dimensional Electron Fluid. IEEE Trans. Electron Devices 1996, 43, 1640–1645. [CrossRef] Lisauskas, A.; Pfeiffer, U.; Öjefors, E.; Bolívar, P.H.; Glaab, D.; Roskos, H.G. Rational Design of High-Responsivity Detectors of Terahertz Radiation Based on Distributed Self-Mixing in Silicon Field-Effect Transistors. J. Appl. Phys. 2009, 105, 114511. [CrossRef] Schuster, F.; Coquillat, D.; Videlier, H.; Sakowicz, M.; Teppe, F.; Dussopt, L.; Giffard, B.; Skotnicki, T.; Knap, W. Broadband Terahertz Imaging with Highly Sensitive Silicon CMOS Detectors. Opt. Express 2011, 19, 7827. [CrossRef] [PubMed] Boppel, S.; Lisauskas, A.; Mundt, M.; Seliuta, D.; Minkeviˇcius, L.; Kašalynas, I.; Valušis, G.; Mittendorff, M.; Winnerl, S.; Krozer, V.; et al. CMOS Integrated Antenna-Coupled Field-Effect Transistors for the Detection of Radiation from 0.2 to 4.3 THz. IEEE Trans. Microw. Theory Tech. 2012, 60, 3834–3843. [CrossRef] Sakhno, M.; Gumenjuk-Sichevska, J.; Sizov, F. Modeling of the Substrate Influence on Multielement THz Detector Operation. J. Infrared, Millimeter, Terahertz Waves 2014, 35, 703–719. [CrossRef]

Sensors 2016, 16, 432

20. 21.

22. 23.

24. 25.

26. 27. 28. 29.

30.

31.

32.

33. 34. 35.

36.

37. 38. 39.

14 of 15

Zdaneviˇcius, J.; Bauer, M.; Boppel, S.; Palenskis, V.; Lisauskas, A.; Krozer, V.; Roskos, H.G. Camera for High-Speed THz Imaging. J. Infrared, Millimeter, Terahertz Waves 2015, 36, 986–997. [CrossRef] ¯ ¯ Seliuta, D.; Kašalynas, I.; Tamošiunas, V.; Balakauskas, S.; Martunas, Z.; Ašmontas, S.; Valušis, G.; Lisauskas, A.; Roskos, H.G.; Köhler, K. Silicon Lens-Coupled Bow-Tie InGaAs-Based Broadband Terahertz Sensor Operating at Room Temperature. Electron. Lett. 2006, 42, 825. [CrossRef] Kašalynas, I.; Venckeviˇcius, R.; Seliuta, D.; Grigelionis, I.; Valušis, G. InGaAs-Based Bow-Tie Diode for Spectroscopic Terahertz Imaging. J. Appl. Phys. 2011, 110, 114505. [CrossRef] ¯ Minkeviˇcius, L.; Madeikis, K.; Kašalynas, I.; Venckeviˇcius, R.; Seliuta, D.; Tamošiunas, V.; Valušis, G. Discrete spectrum terahertz imaging using bow-tie diodes: Optimized antenna designs and arrays. Proc. SPIE 2013. [CrossRef] Amanti, M.I.; Scalari, G.; Beck, M.; Faist, J. Stand-Alone System for High-Resolution, Real-Time Terahertz Imaging. Opt. Express 2012, 20, 2772. [CrossRef] [PubMed] De Cumis, U.S.; Xu, J.-H.; Masini, L.; Degl’Innocenti, R.; Pingue, P.; Beltram, F.; Tredicucci, A.; Vitiello, M.S.; Benedetti, P.A.; Beere, H.E.; et al. Terahertz Confocal Microscopy with a Quantum Cascade Laser Source. Opt. Express 2012, 20, 21924. [CrossRef] [PubMed] Trontelj, J.; Sešek, A. Micro-machined millimeter wave sensor array for FM radar application. Proc. SPIE 2012. [CrossRef] Simoens, F.; Meilhan, J. Terahertz Real-Time Imaging Uncooled Array Based on Antenna- and Cavity-Coupled Bolometers. Philos. Trans. A. Math. Phys. Eng. Sci. 2014, 372, 1–12. [CrossRef] [PubMed] Trontelj, J.; Valušis, G.; Venckeviˇcius, R.; Kašalynas, I.; Sešek, A.; Švigelj, A. A high performance room temperature THz sensor. Proc. SPIE 2014. [CrossRef] Kasalynas, I.; Svigelj, A.; Adamonis, J.; Kornijcuk, V. Performance of the antenna coupled microbolometers characterized by the quasi-optical measurements at frequencies 0.1–1.0 THz. In Proceedings of the 2013 38th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), Mainz, Germany, 1–6 September 2013. Minkeviˇcius, L.; Voisiat, B.; Mekys, A.; Venckeviˇcius, R.; Kašalynas, I.; Seliuta, D.; Valušis, G.; Raˇciukaitis, G.; ¯ Tamošiunas, V. Terahertz Zone Plates with Integrated Laser-Ablated Bandpass Filters. Electron. Lett. 2013, 49, 49–50. [CrossRef] ¯ Minkeviˇcius, L.; Madeikis, K.; Voisiat, B.; Kašalynas, I.; Venckeviˇcius, R.; Raˇciukaitis, G.; Tamošiunas, V.; Valušis, G. Focusing Performance of Terahertz Zone Plates with Integrated Cross-Shape Apertures. J. Infrared, Millimeter, Terahertz Waves 2014, 35, 699–702. [CrossRef] ¯ Minkeviˇcius, L.; Tamošiunas, V.; Madeikis, K.; Voisiat, B.; Kašalynas, I.; Valušis, G. On-Chip Integration of Laser-Ablated Zone Plates for Detection Enhancement of InGaAs Bow-Tie Terahertz Detectors. Electron. Lett. 2014, 50, 1367–1369. [CrossRef] Löffler, T.; Bauer, T.; Siebert, K.; Roskos, H.; Fitzgerald, A.; Czasch, S. Terahertz Dark-Field Imaging of Biomedical Tissue. Opt. Express 2001, 9, 616–621. [CrossRef] [PubMed] Woodward, R.M.; Wallace, V.P.; Arnone, D.D.; Linfield, E.H.; Pepper, M. Terahertz Pulsed Imaging of Skin Cancer in the Time and Frequency Domain. J. Biol. Phys. 2003, 29, 257–261. [CrossRef] [PubMed] Wahaia, F.; Valusis, G.; Bernardo, L.M.; Almeida, A.; Moreira, J.A.; Lopes, P.C.; Macutkevic, J.; Kasalynas, I.; Seliuta, D.; Adomavicius, R.; et al. Detection of Colon Cancer by Terahertz Techniques. J. Mol. Struct. 2011, 1006, 77–82. [CrossRef] Png, G.M.; Choi, J.W.; Ng, B.W.-H.; Mickan, S.P.; Abbott, D.; Zhang, X.-C. The Impact of Hydration Changes in Fresh Bio-Tissue on THz Spectroscopic Measurements. Phys. Med. Biol. 2008, 53, 3501–3517. [CrossRef] [PubMed] Wallace, V.P.; Fitzgerald, A.J.; Pickwell, E.; Pye, R.J.; Taday, P.F.; Flanagan, N.; Ha, T. Terahertz Pulsed Spectroscopy of Human Basal Cell Carcinoma. Appl. Spectrosc. 2006, 60, 1127–1133. [CrossRef] [PubMed] Doradla, P.; Alavi, K.; Joseph, C.; Giles, R. Detection of Colon Cancer by Continuous-Wave Terahertz Polarization Imaging Technique. J. Biomed. Opt. 2013, 18, 090504. [CrossRef] Minkevicius, L.; Madeikis, K.; Voisiat, B.; Tamosnmas, V. Compact diffractive optical components for terahertz beam manipulation. In Proceedings of the 2015 40th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), Hong Kong, China, 23–28 August 2015.

Sensors 2016, 16, 432

40.

41.

42. 43.

44.

15 of 15

Kasalynas, I.; Venckeviˇcius, R.; Adomaviˇcius, R.; Valušis, G. Antenna-coupled microbolometer-based THz detectors for room temperature beam profile imaging of the photoconductive THz pulse emitters. In Proceedings of the 2014 39th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), Tucson, AZ, USA, 14–19 September 2014. Voisiat, B.; Raˇciukaitis, G.; Kašalynas, I. THz components and plasmonic structures fabricated by direct laser patterning of metals. In Proceedings of the 2014 39th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), Tucson, AZ, USA, 14–19 September 2014. Salhi, M.A.; Pupeza, I.; Koch, M. Confocal THz laser microscope. J. Infrared Millim. Terahertz Waves 2010, 31, 358–366. [CrossRef] Wahaia, F.; Kasalynas, I.; Venckevicius, R.; Seliuta, D.; Valusis, G.; Urbanowicz, A.; Molis, G.; Carneiro, F.; Carvalho Silva, C.D.; Granja, P.L. Terahertz absorption and reflection imaging of carcinoma-affected colon tissues embedded in paraffin. J. Mol. Struct. 2016, 1107, 214–219. [CrossRef] Wahaia, F.; Kasalynas, I.; Seliuta, D.; Molis, G.; Urbanowicz, A.; Carvalho Silva, C.D.; Carneiro, F.; Valusis, G.; Granja, P.L. Study of paraffin-embedded colon cancer tissue using terahertz spectroscopy. J. Mol. Struct. 2015, 1079, 448–453. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).