See discussions, stats, and author profiles for this publication at: [Link]

net/publication/234052558

Photothermal converters for quantitative 2D and 3D real-time TeraHertz

imaging

Article in Quantitative InfraRed Thermography Journal · December 2010

DOI: 10.3166/qirt.7.217-235

CITATIONS READS

15 192

7 authors, including:

Christophe Pradere Daniel Balageas

Institut de Mécanique et d'Ingénierie de Bordeaux Ecole Nationale Supérieure d'Arts et Métiers
99 PUBLICATIONS 545 CITATIONS 312 PUBLICATIONS 2,790 CITATIONS

SEE PROFILE SEE PROFILE

Bruno Chassagne Jean-Christophe Batsale

ALPhANOV Centre technologique optique et lasers MINES ParisTech
32 PUBLICATIONS 197 CITATIONS 179 PUBLICATIONS 1,565 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

water content measurement at local scale View project

thermal properties of carbon fibers View project

All content following this page was uploaded by Daniel Balageas on 01 March 2014.

The user has requested enhancement of the downloaded file.

 page 1

This is an Author's Accepted Manuscript of an article published in Quantitative InfraRed Thermography

Journal, Volume 7, N°2, December 2010 [copyright Taylor & Francis], available online at:
[Link]

Photothermal converters for quantitative 2D and 3D real-time TeraHertz imaging

by Christophe Pradere* — Jean-Pascal Caumes** — Daniel Balageas*** — Simon Salort** — Emmanuel
Abraham**** — Bruno Chassagne ** — Jean-Christophe Batsale*

* Lab. TRansfert Ecoulements FLuides et Energétique (TREFLE), Univ. Bordeaux 1, UMR CNRS-ENSAM-UB1 8508,
33405 Talence, France - {[Link], [Link]}@[Link]
** ALPhANOV, Centre Technologique Optique et Lasers, 33405 Talence, France - {[Link], [Link],
[Link]}@[Link]
*** ONERA (The French Aerospace Lab), Dept. Composites Materials and Structures, 92320 Châtillon, France -
balageas@[Link]
**** CPMOH, Univ. Bordeaux / CNRS, 33405 Talence, France - [Link]@[Link]

Abstract
Recent advances for the measurement of TeraHertz (THz) radiation by using original IR temperature flux
sensors are presented. The bolometer principle is used for designing simple thermal converters for THz radiations
(measurement of the temperature increase of a sensitive absorber). Most of these sensors are efficient, sensitive and
fast enough for quantitative measure¬ment of THz source power as well as for 2D and 3D THz imaging. By combining
optical and thermal technologies, we extend and adapt the use of thermal sensors to large THz wave¬length till 3 mm
(0.1 THz). A large variety of mono- or arrayed- thermal sensors is used and opti¬mized for real-time room temperature
THz imaging using adapted IR focal-plane micro-bolometers array (FPMA) camera. Optimisation and adaptation of such
FPMA is discussed and a new arrayed prototype device of THz-Thermal Converter, “TTC”, for full-field real-time THz
imaging is presented. This small size, low cost and efficient prototype design is discussed from the thermal point of view
and is characterized using a compact power¬ful THz source. Their sensitivity is evaluated and the obtained 2D and 3D
images clearly illus¬trates the high potential of this new kind of THz camera. Finally, it is shown that non-arrayed
ex¬tended plane TTCs (EMIR sensitive screens) coupled to FPA cameras produce THz images free of diffraction
phenomena.

1. Introduction

THz wavelength is recognized as a very far IR wavelength covering a spectral range from 30 µm to 3 mm. One
of the most interesting optical properties of such long-wavelength radiation is the fact common objects opaque in the
visible and infrared spectral domains are transparent for them. Another important point is the non-ionising properties of
such electromagnetic wave: indeed, the low photon energy of this radiation (~meV) is considered as an important aspect
for applications on the contrary of X-ray for example. For all this reasons, THz technology is clearly interesting by its
various potentialities in large domains such as physics, biology, or medicine (Chan et al., 2007; Pickwell et al., 2006).
Apart from their well-know spectroscopic potential (Chan et al., 2007; Pickwell et al., 2006), imaging techniques are
growing rapidly, particularly in laboratory and industrial applications (i.e. online non-destructive inspection and monitoring
of processes). To access this high potential, crucial technological gaps need to be overcome. The first one concerns the
difficulty to enhance the power and robustness of the THz sources (CW or pulse). The free electron laser (power up to
several hundreds of watts (Vinokurov et al., 2006)) excepted, the main drawback comes from the very low power of THz
sources. Usually such sources are cooled to provide more than some tens of µW. Nowadays, the typical delivered power
goes from 1 pW to 100 mW (Coutaz et al., 2007). Consequently, the second crucial technological point is the
development of sensitive sensors, particularly for the design of real-time arrayed detectors (THz camera).
Our study mainly deals with the second point, where we focused on the design of a large variety of 1D and 2D
sensitive room-temperature sensors dedicated to THz metrology and active imaging. For that, the most important
contribution of our work is to detect wide-band THz radiation by using a large kind of classical IR sensors (Figure 1). In
the literature, the most sensitive imaging system is achieved by optical or electrical methods (Chan et al., 2007; Lee et
al., 2005) based on single integrated detectors. In the “thermal detector” category, microbolometers and pyrolectric
mono-detectors have been largely studied. The first advantage is that the technology is already mature and integrated
 page 2

for the different thermal bands. Usually, for frequencies lower than 30 THz, the detectors are cooled (T < 77 K) to
increase the sensor sensitivity and to decrease the thermal noise. More promising and developed systems (Lee et al.,
2005; Bründermann et al., 2006; Behnken et al., 2008; Caumes et al., 2009) devoted to 2D imaging are uncooled
infrared focal-plane micro-bolometer arrays (FPMAs). Even mass-production commercial systems specifically dedicated
to the thermal band 7–14 µm wavelength (20–40 THz) already exhibit some low sensitivity in the THz domain (0.1-5
THz). One technological development is held for monolithic 2D arrayed antenna-coupled or absorber deposition
bolometer based on standard Si technologie [Bolduc et al., 2010; Simoens et al., 2010).
Combining optical and thermal technology to large band infrared sensors, we extend and adapt the use of
thermal sensor to larger wavelength till 3 mm (0.1 THz) (Figure 1). The first sensor presented is a thermal fluxmeter
based on the Seebeck effect effect. This fluxmeter is used to measure the quantitative absolute power of a large variety
of CW THz sources. The second one is a device used for 2D and 3D THz imaging. This device is performed with fast
and sensitive IR pyrometric monosensors or quantum thermal detector (HgCdTe). The last one demonstrates the
concept of TeraHertz thermal imagers (TTC) (Pradere et al., 2009): it consists in the adaptation of FPMAs in the THz
range using arrayed and non-arrayed TeraHertz Thermal converters, whose function is to absorb the THz radiation and
convert it in a black-body radiation thanks to the photothermal effect. So, due to the high IR-FPMA sensitivities, it
becomes possible to precisely measure the induced temperature increase map and then to quantify the intensity
distribution of the THz incident radiation.

Figure 1. Technological development described in this work: adaptation and characterization of a wide panel of thermal
detectors for THz frequencies.

2. Experimental set-up

The general THz setups (Figure 2) developed to test the different thermal sensors in the millimetric domain are
presented. This setup has been used for beam metrology as well as for 2D and 3D THz imaging of opaque object
presented in this work.
The easy-to-align and portable experimental setup presented in Figure 2 offers all the essential characteristics
for in-situ THz metrology and characterization. A mW scale compact millimeter wave Gunn diode (…) with a power of 40
mW at 0.11 THz and 0.3 mW at 0.24 THz is used as a THz source. The THz beam is then collimated by a 150 mm gold
coated off axis parabolic (OAP) mirrors and then focused on the sample thanks to a 60 mm focal length plano-convex
lens made of commercial Teflon (a material transparent for THz radiations). All the optics have a 2” diameter offering a
numerical aperture of 0.5. The beam-waist has been measured to be diffraction-limited around 3 mm. This focal volume
is then imaged by the symmetric optical system (OAP and TEFLON lens) – onto several infrared temperature flux
sensor. Four types of thermal sensors are characterized:
- an absolute auto-calibrated TeraHertz fluxmeter based on the Seebeck effect (The TeraConverter),
- a commercial low-cost infrared temperature monosensor based on the pyroelectric technology,
- a quantum thermal detector HgCdTe cooled at 77 K,
- TTC converters coupled to an IR-FMPA camera.
 page 3

Figure 2. Experimental setup used for the detector characterization and for 2D and 3D THz imaging. A CW THz
source is collimated by a parabolic mirror, and focused by a HDPE lens. This focal point is then imaged by the
symmetrical optical system to test the different IR temperature flux sensors.

With this setup, it is also possible to realize 2D and 3D imaging of materials. For that, images were recorded by
characterizing the beam intensity attenuation in transmission of an object placed at the focal point of the system. A raster
scanned motorized system was used, at a velocity rate of 40 pixels/s, with a 1 mm step size. As a consequence, the
2
imaging acquisition time for a (100 x 100) mm object takes less than 5 minutes. However, the transverse imaging
resolution of our complete setup is diffraction limited between 1 and 3 mm due to the wavelength of the millimeter (or
THz) wave radiation; This is not the case for the last TTC constituted by an extended non-arrayed sensitive film.

3. Development of thermal fluxmeter for the power measurement of THz sources

The aim of this part is to present a new-measuring device - called TeraConverter - dedicated to optical terahertz
sources, in quasi isothermal conditions. The main idea is to implement a new sensor using a TeraHertz Thermal
Converter (TTC).This patented system (Pradere et al., 2009) consists in absorbing an incident optical terahertz source
and the induced radiated heat flux is measured by a thermal thermopile. In this system a differential measurement and
an in-situ calibration of the heat flux is realised. This avoids the influence of the surroundings and allows obtaining a
quantitative measurement of the incident optical Terahertz source power. The main advantages of this device are: (i), its
small size, (ii), the low cost of the TTC and the thermopile adapted for industrial tests, and (iii), the possibility to use this
device for online calibration metrology of Terahertz beams.

3.1. TeraConverter powermeter device

We have developed a differential TeraHertz sensor called TeraConverter presented on Figure 3. By using two
thermal sensors, we offer the possibility to simultaneously characterize the terahertz optical sources but also to provide a
thermal reference and calibration using an electrical heater. The thermal heat flux is measured from the differential
voltage due to the Seebeck effect of two thermopiles (elements 4 and 4’ in Figure 3b). Such thermopiles are a multilayer
media composed of two ceramic plates (in the top and bottom) and 127 thermocouples (Te2Bi3) connected in series in
the middle.
For the reference and the THz incident radiation measurements, a thin TTC absorber is placed directly on the
thermopile (4). As presented also in section 5, the TTC material was a 250 µm - thick carbon sheet with a diameter of 20
mm. The absolute value of the converted heat flux is then measured by this thermopile (a 3.2 mm-thick Melcor Peltier
module CP1.4-127-06L). The differential voltage is measured with a HP microvoltmeter, model 3140 A. For this study, the
optical Terahertz source is a continuous GUNN diode with a power of 40 mW and a wavelength of 2.8 mm (0.11 GHz). A
copper block is used as temperature reference between the two thermopiles. Due to the high mass density and the specific
heat of the copper, the temperature of the block is assumed to be isothermal. The whole measuring system is separated
from the surroundings, and particularly from the external thermal radiation, by an insulating foam. Furthermore, an electrical
heater placed in the reference side of the sensor could be used for an auto-calibration of the sensor (see calibration
section). A PC using a Labview program controls all these steps via RS232 port.
3.2. Calibration of the device

The voltage acquired by the fluxmeter through the Seebeck effect is proportional to the incident thermal heat
flux induced by the TeraHertz radiation absorbed by the TTC. The relation between the heat flux and the voltage is given
by Equation [1]:
 page 4

Figure 3. TeraConverter powermeter prototype: (a) Photography (b) Cross-section in the symmetry vertical plan: on the
left side, the electrical heater (1) is designed in order to obtain an auto-referenced and calibrated measurement; on the
right, the TTC (2) converts the THz radiation into a thermal flux; the copper block (3) gives a temperature reference; the
elements 4 and 4’ are thermopiles; the assembly constituted of the elements 1,2,3,4,4’ are thermally insulated by a
thermal insulator (5) and a foam (6). The whole is put in the cylindrical container (7) made of aluminium.

kG kG
φ= U = αU with α = [1]
β β

-1 -1 -1
where φ is the flux (W), k is the thermal conductivity of the thermopile (W.m .K ), β is the Seebeck coefficient (V.K ), G
is the €shape factor (m) and U is the voltage (V). Therefore, we have the possibility to calibrate the TeraConverter by
Joule effect using the electrical heat source. For that purpose, a calibration side is manufactured with an electrical
resistance (left side on figure 3b). This configuration was chosen in order to have a differential measurement of the
power with only one reference (copper bulk), and also for the compactness of the system. Typical calibration result is
presented on figure 4. The dissipated heat flux is measured by the thermopile as a function of time (Figure 4a). To calibrate the
system, various electrical powers from 100 µW to 1 W are applied, at room temperature (inset of Figure 4.a), thanks to a power
supply (Agilent E3631A). It has been check that the detector is room-temperature independant over ± 10°C. The
electrical power was set on, until the thermal steady state was reached.

a b

Figure 4. a) Measured voltages [Link], acquired by the TeraConverter excited by different electrical powers applied to
the heater (dashed curve is the second thermopile voltage signal, inset: typical sequence of applied voltage applied to
the heater). b) Measured voltage in the steady state in function of the electrical power applied to the heater. Linear
regression is done to obtain the calibration coefficient α.
 page 5

This procedure allows the measurement of two values of the heat flux, one during the rising part of the response,
the other one during the decreasing part. In this work, we choose to take the measured voltage in the steady state regime
and then to compare it with the imposed flux (Figure 4b). From Equation [1], the calibration coefficient is determined and we
-1
obtained: α25°C = 6.803 W V . Due to the mechanical and thermal symmetry of our system (Figure 3b), when a THz source
is set on the TeraConverter, the voltage measure will be convert into a heat flux using this coefficient (α) and allows to
deduce the power (W) of the incident THz radiation.
3.3. Power measurements of monochromatic THz sources using the TeraConverter

Here, the TeraConverter device is used to make the metrology of different monochromatic sources largely used
in the THz community. These measurements are detailed on the Figure 5. First, the measured temperature elevation
converted into voltage versus time by the TeraConverter illustrates the power metrology of a Gunn diode emitting at
0.11 THz (Figure 5a). The source is switched on at t = 0 and the voltage increases till a steady state around 5.47 mV.
This means a power of 38 mW using the calibrated proportionality factor α in good agreement with the Gunn diode
specification given by the constructor (40 mW). Several TeraConverter metrologies have then been tested using different
powers (inset of figure 5a) and different CW monochromatic THz sources.
Figure 5b sums up an experimental campaign made on a wide panel of monochromatic THz sources available
in the THz community. This campaign aims to demonstrate the large spectral band -from 0.1 to 10 THz- and the high
dynamic power sensitivity - from 0.1 to 100 mW - of the TeraConverter. The errors on these measurements are less than
5%. One must also note that the calibration coefficient α constant on the entire THz band tested. The minimum
measurable signal equivalent to 10 µW demonstrated that this device is suited to characterize THz and sub-THz sources
such as electronic diodes (Gunn, IMPATT, TUNNETT), BWO (Backward-Wave Oscillators), QCLs (Quantum Cascade
Lasers), Molecular lasers or Free Electron Lasers. This versatile and easy to use device clearly brings an add-value for
THz and sub-THz assistance and metrology for several THz setup and particularly for raster scanned 2D and 3D imaging.

a b

Figure 5. Typical voltage measurement made from the TeraConverter : a) Voltage versus acquisition time for a 40 mW
Gunn diode emitting at 0.11 THz (inset is voltage versus acquisition time for a 0.2 mW Gunn diode emitting at 0.24 THz).
b) TeraConverter metrology campaign on a wide panel of monochromatic sources from 0.1 to 10 THz (0.1-60 mW).

4. Illustration of thermal mono-sensors for 2D and 3D TeraHertz active imaging

As discussed before, our experimental setup (section 2) was suited for metrology of THz sources or new
detectors in this particular spectral domain. In this section, we demonstrate the ability of these detectors as versatile tools
for 2D and 3D imaging. For that, varied applications for non-destructive evaluation or remote inspection and detection of
hidden object are presented. We also show the interest of THz imaging for different industrial applications as art
conservation, anthropology, aeronautics or global security.
As discussed before, our experimental setup (section 2) was suited for metrology of THz sources or new
detectors in this particular spectral domain. In this section, we demonstrate the ability of these detectors as versatile tools
for 2D and 3D imaging. For that, varied applications for non-destructive evaluation or remote inspection and detection of
hidden object are presented. We also show the interest of THz imaging for different industrial applications as art
conservation, anthropology, aeronautics or global security.
The use of a thermal monosensor for 2D and 3D THz imaging requires making a raster scanning of the
object of interest. The setup is represented in Figure 6a. The sample is placed at the focal point of the THz beam and
 page 6

then imaged on a 250 µm-thic carbon sheet TTC. A thermal mono-sensor then measures the induced temperature
flux. The THz beam transmission through the object is hence measured for each scanned point where several optical
properties (absorption, reflexion…) could be quantified. For that, two-axis motorized translation stages ensure the fast
and precise scan of the object. Moreover, an additional motorized rotation stage allows making 3D THz tomography of
large object (100 x 100 m²). Three kinds of thermal sensors have been tested for 2D and 3D THz imaging using the
TTC. Each sensor provides different interesting characteristics for active imaging:
- a classical low-cost, compact and easy integrated IR pyrometer from Texense ([Link]/
documents/081029031637_infrared_temperature_sensors.pdf) brings a high sensitivity for THz detection (~ 0.1
mV/mW). In this case, neither mechanical modulation of the THz beam, nor lock-in amplifier are required, enabling easy
to use and continuous measurements. A drawback of this sensor comes from the low acquisition rate due to the time
response (~ 10 ms), Nevertheless, with such device, more than 30 dB SNR 2D and 3D THz imaging have been
performed for remote detection of hidden objects.
- a quantum type photoconductive HgCdTe monosensor (Hamamatsu P7752-10) with a wavelength range of 6-14
µm, a NEP of 3.10-12 [Link]-0.5 and a frequency range of 5 to 140 kHz has demonstrated fast and sensitive
measurements. In this case, and due to the frequency range of the sensor, the measurement is different because the
THz sources need to be modulated by a mechanical chopper enabling to use a lock-in amplifier. Due to the short
response time of these detectors, and the use of liquid N2 (77 K) to cool them, it is possible to reach a SNR as high as
~100 dB (McMillan, 2005).

b c

d e

Figure 6. Illustration of thermal mono-sensors application to 2D and 3D THz active imaging for NDE and remote
detection of hidden object by comparison of visible photographies and 2D/3D THz images: a) Experimental setup
schematized for different thermal detectors (see text), b) 2D spatial detection of plastic toy, c) 3D tomographic
reconstruction of two Russian dolls made in wood, d) Metallic corkscrew drilled into a cork stopper for wine bottle, e)
Fluorescent light bulb.
 page 7

These two detectors have been implemented and characterized for 2D THz imaging and tomography. A sample
of our active imaging is summed up on figure 6. This illustrates the high potential of the THz radiation for non destructive
evaluation and inspection of hidden objects. We have chosen to hide objects of interest behind various containers such
as the ones shown in Figure 6.
Figure 6b presents a THz image of a plastic toy and Figure 6e a fluorescent light bulb, revealing the internal
structures of these objects opaque for the bare eye. Here, 2D THz imagings are performed by using, for Figure 6b, an
infrared pyrometer (experimental duration ~ 5 min with a scan step of 1 mm). The use of compact and easy to integrate
pyrometer sensor hence opens the access to sensitive, fast and integrated imaging system for cultural heritage inspection.
Figure 6e demonstrates the interest of such system for non-destructive evaluation of opaque objects for visible
wavelength. The 77 K cooled quantum HgCdTe detector coupled with a lock-in amplifier used for this measurement
enables a 100 dB SNR (modulation rate of 110 Hz). These detectors offers the possibility to get faster response (> kHz)
promising faster imaging system applied to NDE.
Finally, the Figures 6c and 6d demonstrate the high sensitivity and acquisition velocity of these innovative
thermal detections for 3D THz tomography (Ferguson et al., 2002). Due to the fast acquisition rate > 10 Hz, it is timely
reasonable to made 2D raster scans of the objects with the supplementary rotation stage leading to the 3D
information on hidden objects. For the images presented, 18 object rotations have been done and the reconstruction
of the 3D object from this set of 36 projections has been done using a back-projection algorithm. One example (Figure
6d) is a metallic corkscrew drilled into a cork stopper for wine bottle. The images represent the 3D shape of the
sample surface obtained with a surface rendering software and the internal structures of the cork stopper with the
penetrating corkscrew. Another example (Figure 6c) is a wooden Russian doll ‘Matriochka’ with a second smaller doll
positioned inside the main doll. The 3D THz image (left image) clearly reveals the presence of the smaller wooden
doll inside the main doll, as also visible in the 2D image (upper right image). We also inserted a third smaller doll
inside the second one and its presence is revealed by the 2D image (lower right image) even if the wood absorption in
this part of the sample becomes larger.
These results clearly illustrate the efficient coupling between thermal technology and the THz potential. The THz
domain appears as an extension of the far infrared range. Most of the sensors developed in the thermal band exhibit
sensitivity and applicability to non-destructive evaluation. As a consequence, it becomes obvious to perform similar tests
and characterizations for 2D thermal detectors such as microbolometers or IR-FPMA. Demonstration of 2D THz imaging
with uncooled infrared focal-plane micro-bolometer arrays (FPMAs) have been published in the last few years (Lee and
Hu, 2005; Bründermann et al., 2006; Behnken et al., 2008; Caumes et al., 2009). Even mass-production commercial
systems specifically dedicated to the thermal band 7–14 µm wavelength (20–40 THz) already exhibit some low sensitivity
in the THz domain (0.1-30 THz). In section 5, from the Terahertz Thermal Converter (TTC) concept presented before, we
defined an arrayed device for 2D full-field THz imaging. Coupled with an IR-FPMA, we demonstrate the possibility to
design a THz camera able to characterize beams generated by various THz monochromatic sources. In section 6, we
demonstrate the possibility to use a photothermal screen coupled with an infrared camera to perform THz imaging.
®
These screens were initially developed for imaging microwave fields (centimetric wavelengths) by the EMIR
(ElectroMagnetic InfraRed) technique.

5. New auto-calibrated 2D arrayed THz camera for full-field 2D beam metrology (0.1 to 30 THz)

The main innovation was to adapt the FPMA using an arrayed TTC to directly obtain full-field 2D images in the
THz domain. Again, this TTC absorbs the THz radiation, converting the electromagnetic flux in a black-body radiation.
Hence, a sensitive IR-FPMA quantitatively measures the induced temperature map of the TTC (Figure 7a). To optimize
different TTC materials and thermal configuration, we used our compact and versatile THz setup presented in section 2
(Pradere et al., 2009). The TTC material, tested in isothermal conditions, was a 2.54 cm - dia., 250 µm - thick carbon
sheet, chosen because it absorbs more than 50% of the THz flux. It has been arrayed by laser drilling to form a 2 x 2
mm² pixels matrix (pixel separation of 1 mm hold by a ceramic wafer transparent in the THz domain). The inter pixel
space is only made by air separation between the pixels. Even with such precautions taken to decrease the effect of the
thermal diffusion between each pixels of the sensor, the image of Figure 7c gives the erroneous impression that the fill
factor is higher than that corresponding to these dimensions. This is due the heat diffusion through the ceramic
substrate.
The TTC was integrated to the IR camera (a commercial FPMAs camera with NETD ~ 0.1°C, using a
germanium objective) in an extension ring configuration (see the photo of the prototype in Figure 7b). One must note the
important point that this removable optical module implemented on this standard thermographic camera is versatile
enough to allow the TTC fixation on a large variety of available commercial camera. The surface temperature increase of
the TTC was then imaged by this modified camera which becomes a THz camera.
The characterization aimed to evaluate the sensitivity and the high potential of the present prototype. The first
acquisitions are represented on figure 7. The protocol consists to take the temperature map of the TTC without any
illumination. It is the reference pattern for ambient thermal radiation of our setup. This reference is then subtracted to the
one measured in presence of the THz radiation. The figure 7c presents the measured performed at the focal point in the
 page 8

setup described before. This acquisition has been done with an illumination time of 10 s. The maximum temperature
increase recorded on the TTC was 1.6 °C for an incident flux of around 40 mW. Some thermal diffusion appeared
between the pixels and the ceramic support. This point can be clearly optimized by the design of isoperibolic TTC
configuration. From the measured arrayed cartography (figure 7c), a basic imaging software enables to integrate the
temperature on each TTC pixel and then to reconstruct the 2D spatial beam profile (figure 7d). This reconstruction
resolution corresponds to the TTC pixel size (here: 2 mm). The low influence of the measured diffusion is noticed and the
beam reconstruction provides a fully coherent measurement of the beam size (~ 6 mm) compared to the diffraction limit
(compared to an Airy spot of 6.7mm) and the numerical aperture of the used optics (NA = 0.5).
No diffraction with the bolometric camera elements is also noticed, contrary to what is observed in the
experiment presented further (Figure 10) with a direct imaging of a THz beam by a bolometric camera. The difference
between the two experiments is the following: in the present case, obviously a part of the THz beam is transmitted by the
TTC (the absorption is not complete in the absorbing elements and there is almost a total transmission in between these
elements). Nevertheless, these THz photons are received by the bolometers with a sensitivity which is low compared to
the sensitivity for the IR photons coming from the heated TTC.
In conclusion, an absolute determination of the Gunn Diode power is possible with a 2 % error from the
temperature measurements calibrated from the known TTC absorption and emissivity. A thermal model will be integrated
to the device in order to obtain a competitive sensitivity (~ 10 µW) for an absolute THz power estimation.

Figure 7. Typical beam metrology using the 2D arrayed TTC: (a) Basic principle of a 2D arrayed TTC THz camera. (b)
Design prototype of the versatile optical module implemented on a standard thermographic camera. (c) Typical temperature
cartography of the TTC during the THz illumination. (d) 2D Reconstruction of the THz beam profile using the measured
TTC mapping.
®
6. New prospects for 2D THz imaging using non-arrayed TTC made of sensitive screens used in the EMIR
technique

®
6.1. Recall on the EMIR technique

Thermal converters have been already developed for imaging microwave fields (X and Ku bands in particular).
®
The thermal converter in this approach called EMIR (for ElectroMagnetic InfraRed) is a screen made of a thin polymeric
 page 9

film, some tens of microns thick, with a controlled electric conductivity (Balageas et al., 1993; Balageas and Levesque,
1998; Levesque et al., 2005). The absorption of the incident electromagnetic field can be volumic or confined at the film
surface depending on the technology adopted for making the film conductive. Once placed in a microwave field, the
screen converts a part of the electric field energy into heat (see Figure 8). This induces a temperature increase which
is easily monitored by an infrared camera, giving a thermal image representative of the intensity of the field in the plane
of the film.
Generally to avoid disturbances in the image due to natural convection, especially when films of large
2
dimensions (up to several m ) are used, an amplitude modulation of the analyzed field is produced and lock-in
®
thermography used (Balageas et al., 1993). The EMIR technique is also applicable to the characterization of pulsed
microwave fields (Levesque et al., 2005; Balageas and Levesque, 2010). In this case the measurement concerns the
energy density deposited by the incident field on the film instead of the power density (or intensity).
Figures 8a-8d present the technique as it is commonly used for monitoring microwave electric fields (mainly in the
domain ranging from the S to Ku bands). Lock-in thermography is often used (Figure 8a) and the electric properties of the
film are often adjusted at the maximum of absorption which correspond to a square resistivity (or sheet resistance) of 188
Ω/ (Figure 8b). Figures 8c-8d present a typical measurement configuration with a large screen: the tomography of the
electric field generated by a pyramidal horn at 8 GHz in the vertical symmetry plane of the horn. The camera is located
aside. The edge of the film is in contact with the end of the horn. The amplitude image obtained is given (Figure 8d),
showing the complexity of the near field and its rapid evolution with the distance from the entrance plane of the horn. This
image demonstrates one of the main advantages of this type of imaging: the resolution of the image is not limited by the
incident radiation wavelength, here 3.57 cm, but by the space resolution of the lock-in thermographic system which is far
better.

a b

®
Figure 8. The EMIR method for microwave fields imaging. a) Set-up for characterization of a microwave field, b)
Influence of the electric square resistivity of a thin film on its interaction with an electric field, c) Tomography of the
vertical symmetry plane of the electric field at the entrance of a horn (X-band): experimental configuration, d) near field
image of the field obtained by the EMIR technique.
 page 10

The use of this type of photothermal converter film for THz imaging has been tried by Russian researchers
using a carbon paper sheet as a converter (Cherkassky et al., 2005) for visualizing pulse repetitive high-power THz
radiations generated by a free electron laser (λ = 0.12 – 0.18 mm). The average power of the beam was 200 W. They
obtained near infrared images (λ = 2.6 – 3.1 µm) and power density distribution of the THz beam. In the present case,
the conditions for imaging are different since the beam power is only 40 mW and the THz wavelenght is further from
infrared (λ = 2.7 mm).
The ratio of the incident beam wavelength to the wavelength of the radiations re-emitted by the converter and
used for imaging, λTHz/λIR can be considered as a gain in spatial resolution, making attractive the photothermal method.
Table 2 compares the spatial resolution gains of the various EMIR technique applications.

Table 1. Spatial resolution gain in the various applications of the EMIR technique.

®
6.2. THz beam profiling by the EMIR technique

®
A few exploratory experiments with this type of screens have been performed with existing EMIR screens
coupled with a cooled IR FPA camera FLIR SC 7000 working at 3 - 5 µm. Figure 9 presents so-obtained images of the
THz beam generated by the Gunn diode of 0.11 THz, with a power of 40 mW. These screens allow to monitor with a
cooled InSb 256 x 256 pixels camera the continuous space distribution of the energy in the beam, in the plane of the
screen, with sensitivities depending on both THz absorption and IR emission properties. It is clearly shown that EMIR
thermal converters for microwave length are also very suitable for the THz wavelength of this diode. The converter
containing a Carbon loading (Figure 9a) is the most sensitive, with a maximum level at the beam center corresponding to
300 digital levels of the IR camera. Such results are very interesting in order to obtain a multi wavelength or multi
spectral 2D sensor using an IR camera as detector.
6.3. Comparison to direct measurements with bolometric cameras

®
Figure 10 highlights the interest of the EMIR imaging which is not limited by the diffraction of the THz
radiations as it is the case when performing the measurement directly with a microbolometer camera, without any TTC
®
and IR lens (Caumes et al., 2009). In effect, the Gaussian shape of the beam is well measured by the EMIR approach,
whereas the direct attack on the microbolometer elements produces diffraction patterns difficult to interpret. The problem
of the diffraction due to the size of the wavelength of the THz radiations compared to the size of the camera components
is a general problem which increases the difficulty to perform 2D THz imaging.

7. Conclusion

These first measurements offer a particularly high potential for THz imaging. Their results clearly illustrate the
efficient coupling between thermal technology and the THz potential. In this sense, the THz domain appears as an
extension of far infrared range. Most of the sensors developed in the thermal band exhibit sensitivity and offer
applications for non-destructive control. The large panel of Thermal sensors already affordable make possible to develop
innovative metrology of a wide panel of THz and sub-THz sources as different as electronic diodes (Gunn, IMPATT,
TUNNETT), BWO (Backward-wave oscillators), QCLs (Quantum cascade lasers), Molecular lasers or Free Electron
Laser. They appear to be very versatile, cheap and easy to use devices, clearly adapted for assistance and metrology for
several THz setup and 2D and 3D imaging.
 page 11

a b

c d

Figure 9. THz beam metrology using 2D EMIR TTCs. Electro-conductive polymeric films: (a) 106 Ω/, (b) 1150 Ω/, (c)
260 Ω/, (d) 1600 Ω/ Teledeltos paper.

Figure 10. Typical beam metrology using a bolometric IR camera without the TTC (Caumes et al., 2009).
 page 12

Coupled with an IR-FPMA, it is demonstrated that an absolute THz camera enables to characterize a 2D THz
spatial beam profile related to various THz monochromatic sources. The next step will be to optimize the IR-FMPA
camera and the TTC thermal efficiency in order to present 2D fullfield imaging. Some improvements are already in progress
to perform a complete quantitative thermal analysis of the TTC presented in section 5 in isothermal and isoperibolic
configuration. This will give access to the beam profile more rapidly and with more sensitivity. For example, using a
cooled commercial camera will improve the sensitivity and the spectral resolution.
®
Finally, the extension of the application domain of the EMIR technique from the microwave domain to the THz
domain is easily feasible as shown by preliminary tests. This paves the way to a wide variety of applications for the THz
imaging, with the important advantage of having a space resolution not limited by the diffraction of the THz radiations as
its occurs with the measurements performed directly with microbolometer cameras. This space resolution will be also
enhanced thanks to the use of lock-in detection.

8. References

Balageas D., Levesque P., “EMIR: a photothermal tool for electromagnetic phenomena characterization”, Rev. Générale
de Thermique, Vol. 37, 1998, pp. 725-739.
Balageas D., Levesque P., Déom A., “Characterization of electromagnetic fields using lock-in IR thermography”,
Thermosense XV, Proc. SPIE 1933, 1993, pp. 274-285.
Behnken B.N., Karunasiri G., Chamberlin D.R., Robrish P.R., Faist J., “Real-time imaging using a 2.8 THz quantum
cascade laser and uncooled infrared microbolometer camera”, Opt. Lett., Vol. 33, No 5, 2008, pp. 440-442.
Bolduc M., Marchese L., Tremblay B., Doucet M., Oulachgar H., Le Noc L., Williamson F., Alain C., Jerominek H.,
Bergeron A., “Video-rate THz Imaging Using a Microbolometer-based Camera“, International Conference on Infrared,
Millimeter, and Terahertz Waves, IRMMW-2010, Roma, Italy, Sept. 8, 2010, Paper We-C1.4.
Bründermann E., Ränsch J., Krauß M., Kunsch J., “Erste THz-Videos mit einer Silizium-basierten IR-Kamera”, Photonik,
No 6, 2006, pp. 60-63.
Caumes J.-P., Chassagne B., Coquillat D., Teppe F., Knap W., “Focal-plane micro-bolometer arrays for 0.5 THz spatial
room-temperature imaging”, Electron. Lett.,Vol. 45, No 1, 2009, pp. 34-35.
Chan W.L., Deibel J., Mittleman D.M., “Imaging with TeraHertz radiation”, Rep. Prog. Phys., Vol. 70, 2007, pp. 1325-
1379.
Cherkassky V.S., Knyazev B.A., Kubarev V.V., Kulipanov G.N., Kuryshev G.L., Matveenko A.N., Petroc A.K., Popik V.M.,
Scheglov M.A., Shevchenko O.A., Vinokurov N.A., “Imaging techniques for a high-power THz free electron laser”,
Nuclear Instruments and Methods in Physics Research A 543, 2005, pp. 102-109.
Coutaz J.-L., Garet, Gaborit F.G., Duvillaret L., “Potentialité des ondes THz en imagerie”, REE, 2007, No 9.
Ferguson B., Wang S., Gray D., Abbot D., Zhang X.C., “T-ray computed tomography”, Opt. Lett., Vol. 27, No 15, 2002,
pp. 1312-1314.
Lee A.W., Hu Q., “Real-time, continuous-wave terahertz imaging by use of a microbolometer focal-plane array”, Opt.
Lett. , Vol. 30, 2005, pp. 2563-2565.
Levesque P., Brémond P., Lasserre J.L., Paupert A., Balageas D.L., “Performance of FPA IR cameras and of their
improvement by time, space and frequency data processing – Part II: Application to thermographic measurement of
microwave fields”, QIRT Journal, Vol. 2, No 2, 2005, pp. 237-250.
McMillan R.W., “Terahertz imaging millimeter-wave radar”, Advances in sensing with Security Application, Proceeding
NATO, 2005.
Pickwell E., Wallace V.P., “Biomedical applications of terahertz technology”, J. Phys. D: Appl. Phys, Vol. 39, No 7, 2006,
R301.
Pradere C., Caumes J.-P., Chassagne B., Batsale J.-C., Imagerie Térahertz à convertisseur infrarouge perfectionné,
Patent N° FR 09 52097, Avril 2009, CNRS.
Pradere C., Caumes J.-P., Chassagne B., Batsale J.-C., Imagerie Térahertz à convertisseur infrarouge perfectionné,
Patent N° FR 09 57586, Avril 2009, CNRS.
Simoens F., Pocas S., Meilhan J., Ouvrier-Buffet J.L., Maillou T., Gellié P., Barbieri S., “THz uncooled microbolometer
array development for active imaging and spectroscopy applications”, International Conference on Infrared,
Millimeter, and Terahertz Waves, IRMMW-2010, Roma, Italy, Sept. 8, 2010, Paper Tu-B1.4.
Vinokurov N.A., Knyazev B.A., Kulipanov G.N., Matveenko A.N., Popik V.M., Cherkassky V.S., Scheglov M.A.,
‘Visualization of radiation from a high-power Terahertz free electron laser with a thermosensitive interferometer”,
Technical Physics, Vol. 52, No 7, 2007, pp 911-919.

View publication stats