QUANTUM CASCADE LASERS FOR BIOSENSING

APPLICATIONS

Pietro Regoliosi

Lehrstuhl für Nanoelektronik, Technische Universität München, Arcistrasse 21, 80333 Munich, Germany

INFN – Sezione di Trieste, Via A.Valerio 2, 34127 Trieste, Italy

Andrea Vacchi

INFN – Sezione di Trieste, Via A.Valerio 2, 34127 Trieste, Italy

Giuseppe Scarpa, Paolo Lugli

Lehrstuhl für Nanoelektronik, Technische Universität München, Arcistrasse 21, 80333 Munich, Germany

Keywords: Nanotechnology, optical biosensors, infrared spectroscopy, quantum cascade lasers.

Abstract: Quantum cascade lasers represent nowadays a mature technology to obtain laser sources in the medium and

far infrared region (up to THz). Several advantages with respect to other coherent sources make this kind of

lasers particularly attractive: the emission frequency can be selected by properly designing the growth

structure, the emission wavelength is tunable with a very good precision and a high optical power in the IR

range can be emitted in a spot of small size. These properties make them suitable for several applications,

including gas spectroscopy in the IR range. In this work we introduce different types of quantum cascade

lasers and we provide a description of their performances and properties, showing that they are suitable

candidates for biosensing applications.

1 INTRODUCTION

In recent years, the first realization and the further

development of quantum cascade lasers (QCL)

(Faist, 1994; Hofstetter, 2001; Pfügl, 2003;

Faugeras, 2005)) have provided reliable, powerful

and tunable sources for the IR region, whose

wavelength emission can be selected along a wide

range by properly designing the constituting layered

structure. Therefore, possible applications of such a

source have been already explored in different fields

of IR spectroscopy, like environmental monitoring

(McManus, 2005), medical diagnostics (Roller,

2002), atomic spectroscopy (Vacchi, 2006), plasma

diagnostics (Röpcke, 2006). QCL look quite

profitable, since they could provide a compact and

unique source also in the IR region, avoiding the

requirement to use sophisticate and complicate laser

systems.

In the present work, we present the development and

testing of QCL sources designed to be used in

atomic spectroscopy, with the aim to describe the

characteristics which make them usable also for

other applications. In Section 2 we describe the

peculiar structure of quantum cascade lasers and we

introduce the lasers tested in order to define the

achievable performances.

In Section 3 it is presented the experimental

setup which allowed us to perform the

measurements aimed to define the lasers

characteristics: the results are shown in Section 4.

Finally, in Section 5 a discussion of the possible

applications of QCL in the field of biosensing is

provided taking into account the obtained results and

presenting some of the worldwide already running

activities.

2 DEVICES UNDER TEST

In conventional semiconductor lasers light is

generated by stimulated emission across band-gap

Regoliosi P., Vacchi A., Scarpa G. and Lugli P. (2008).

QUANTUM CASCADE LASERS FOR BIOSENSING APPLICATIONS.

In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 87-92

DOI: 10.5220/0001048600870092

 SciTePress

(inter-band emission); in the case of QCL radiation

is generated by intersubband transitions across

designed energy gap inside the same band (intra-

band emission). Thus the emission transition can be

by choosing the thickness of the different layers of

the materials composing the active region of the

device. A scheme of the basic principle is shown in

Figure 1.

Figure 1: Scheme of the principle of an injectorless QCL.

The QCL structures we designed have been

thought for high power and reduced instability in the

laser pulse. The scheme is called bound-to-

continuum (Pfügl, 2003): the upper laser state is a

bound one as in the basic case of QCL structure,

while the lower state is spread on a continuum of

state, which helps the laser emission, the fast

injection of the following active layer and thus

inversion of the population. The optimal structure as

derived by simulations is presented in Figure 2.

Three different groups of lasers have been

designed and fabricated with a lattice matched

technique in order to avoid strain problems

(morphological defects). The different designs

aimed to check the possibility to span the emission

over a wider range around 7μm.

Figure 2: Bound-to-continuum scheme for a QCL with

injection region.

Such kind of lasers are basically Fabry-Perot

structures, thus expected to have multimode

emission (i.e., to emit with several different near

peaks). For spectroscopy purposes, single-mode,

narrow linewidth lasers with well-defined, precise

tunability can be required. In order to achieve these

goals, QCL are fabricated with a periodic structure

built in the cavity (Figure 3). This periodic variation

of the refractive index or of the gain leads to a

certain amount of coupling between the back- and

forth-travelling waves. The coupling becomes

strongest if the periodicity is an integer multiple of

half the laser wavelength in the cavity. Because

feedback occurs along the whole cavity and not only

on the mirrors, these devices are called distributed

feedback lasers (DFB). They show usually an

excellent single-mode behaviour, can be precisely

tuned with temperature or current, and deliver a

reasonable amount of output power. In order to test

the DFB monomode behaviour, we commissioned

such a structure from Alpes Lasers SA, Switzerland,

with specification able to match our structures.

Figure 3: Scheme of the distributed feedback structure,

with the grating superimposed to the active region (from

University of Neuchatel).

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

3 EXPERIMENTAL SETUP

The lasers are fixed on the cold finger of a cryostat,

whose chamber can be evacuated and cooled down

to a temperature as low as liquid nitrogen

temperature (77K). The emitted light is coupled out

of the cryostat through a CaF window (more than

90% transmission in the 2-9 μm range), then

collimated and focused on a IR detector by two

parabolic reflectors (equivalent f/# ~ 3.9). The

measurements are performed by applying a pulsed

voltage on the QCL, and reading out the optical

power on the detector by mean of an oscilloscope.

In case of spectral measurements, the focused

beam has been coupled to the input of a

monochromator, and the detector has been

positioned at the output of the monochromator itself.

We performed both power and spectral

measurements at different temperature, in order to

define the achievable power and the possible tuning

of the laser emission. A scheme of the whole setup is

depicted in Figure 4.

In order to perform both the kind of

measurements in the same time, it has been explored

another possible setup, by mean of inserting a ZnSe

beam splitter in the optical path. The beam splitter

divides the emitted light in two parts: one is focused

directly on a photo detector to measure the optical

power, the other is coupled with the monochromator

for a contemporaneous spectral measurement. The

configuration is also able to provide an online

feedback on the drifting of the emission wavelength,

which shows to be useful in case of spectroscopy

measurements.

Figure 4: Setup for the optical power measurements: (a for

power estimation, b for spectral measurements).

4 RESULTS

In Figure 5 an example of the I-V characteristics of

our set of lasers is shown as a function of the

temperature together with the corresponding

emission peak powers. The measurements gave the

idea of the achievable power: more than 1W at

cryogenic temperatures; about half of such power is

still present at room temperature. The laser could be

operated up to 400K. Considering that the laser facet

has an area of about 60-100 μm

, it corresponds to a

power density of 10

W/cm

if the whole power is

perfectly focused. The amount of power allows the

lasers to be exploited in spectroscopy of gas, for

example by mean of absorption measurements.

Figure 5: I-V curves and peak power of QCL of our set of

laser in the range between 77K and 360K.

Figure 6: QCL Spectra between 78K and 275K.

The corresponding spectra for one of the three

groups of lasers are shown in Figure 6 as a function

of the temperature. Due to the high emitted power, it

is possible to observe fine structures: none of the

spectra could be optimally fit with a simple

Gaussian curve, due to the expected multimode

emission. The convoluted spectra have widths

spanning from 70 nm to about 150 nm. Taking into

account all the three groups, the overall range that is

possible to span with such lasers is very wide, from

6900 nm up to 7800 nm, demonstrating the

possibility to tailor the structure in order to span a

desired wavelength range. This feature is another

important one in terms of spectroscopy application,

QUANTUM CASCADE LASERS FOR BIOSENSING APPLICATIONS

since it could allow to scan wide spectral ranges

with a limited number of different devices.

In order to define the achievable tunability, we

performed studies on the DFB structure: since it

emits in monomode behaviour, its tunability is more

well defined. The power achievable with the DFB

structure is lower than the one emitted by our

structures (even if still about 1W at cryogenic

temperatures). However the emission is clearly

monomode in a certain range of temperatures as it is

possible to estimate from the measurements shown

in Figure 7 and it is tuneable with very nice

precision by mean of changing the voltage applied

by the pulser. Figure 8 reports the study of the

tunability, performed at 90K: the peak wavelength

increases with increasing voltage (due to the

increase of the temperature of the active region), and

the behaviour is linear. The achievable tunability is

as low as (30±1) pm/V around the emission, and the

estimation is limited by the resolution of the

spectrometer (the actual width of the laser line

should be as narrow as 5 pm).

Figure 7: Monomode emission of DFB structure and its

shift with the temperature.

Figure 8: Linear fit of the variation of the central emission

wavelength as a function of the pulser voltage.

The very nice achievable precision is a feature

strictly required for atomic spectroscopy, where the

fine definition of the transition energies is quite

important, but it can be also exploited in general

spectroscopy.

5 QCL FOR BIOSENSING

The presented results describe the achievable

performances with a QCL source: high IR power

localized in the space, spanning on a very wide

range of wavelengths or with monomode emission

depending on the chosen structure, providing the

availability of fine tuning by mean of very precise

temperature changes. All these properties look

profitable for spectroscopy in the IR region, and they

can be applied as well to biosensing applications.

Indeed, most of the chemical species have

distinctive absorption lines in the range between 3

and 20 μm (500 – 4000 cm

-1

), due to the vibration

modes of the most important organic bonds (see the

panel in Figure 9). For example, the QCL structures

presented so far emit around 7 μm which means

about 1425 cm

-1

(they have been designed in order

to provide sources for atomic applications), thus

they could be used to detect the C-H bending in the

alkanes. Moreover as mentioned above it is possible

to choose the emission wavelength only by changing

or using different materials without modifying the

achievable performances.

Therefore, molecular detection in terms of

vibrational spectroscopy is possible, and it is

currently under development. Most of the activities

is oriented on the utilisation of DFB structures, since

spectroscopic applications usually require single

mode operation. The laser emission is collimated

and focused in a gas cell (usually a tube terminated

with antireflection coated windows), which provides

both the volume where the absorption takes place

and a way to increase the optical path of the

radiation and thus to increase the sensitivity of the

system. The method has been explored with

different kinds of laser to detect different gases, like

and H

0 (640 cm

-1

, Kosterev, 2002), C

and

(1000 cm

-1

, Weidmann, 2004), NO (1920 cm

-1

Weber, 2002), providing detection limit down to <

1ppbv, which could be already exploited in medical

diagnostics. Recently, more sophisticated

approaches as the off-axis integrated cavity output

spectroscopy (OA-ICOS) have been applied to

deeply investigate the achievable sensitivity

(Bakhirkin, 2006). Another explored way to realize

QCL-based gas sensors has made use of photonic

bandgap fibers, which are able to transmit radiation

in the IR range. The QCL emission is coupled into

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

the fiber using a coupling cell: a sensor able to

detect C-H stretch band of ethyl chloride gas has

been realized in this way (975 cm

-1

, Charlton, 2005).

THz lasers, which have higher wavelengths, have

been used for gas phase spectroscopy (Hübers,

2006).

For laser spectroscopy of larger molecules with

broad absorption features narrow linewidth is not

required, but it is more important to be able to tune

the emission over a wide wavelength range. In this

case, the bound-to-continuum structures we

presented provide the required broad spectrum (the

lower state of the transition is a relatively broad

continuum). The fine tuning is achieved by mean of

the utilization of an external cavity configuration:

the emission of the laser is collimated and reflected

off a diffraction grating so to create a resonant

cavity. The first order diffraction from the grating

provides the laser feedback, while the output of the

Figure 9: Absorption regions in the infrared.

external cavity is obtained by the zeroth order. The

tunability is provided by moving and rotating the

grating with piezo actuated positioners. The method

has been proved to be fruitful (Wysocki, 2005) and

it is currently exploited for measuring absorption

spectra of large molecules, like Freon (around 1150

-1

, Phillips, 2007).

Another idea which can be exploited comes from

the realization of surface-emitting quantum cascade

micro cavity lasers: the active region of the laser is

covered with a patterned surface, like a photonic

crystal. The presence of the crystal provides both

feedback for the laser action and a selection of the

polarization of the emitted light (Colombelli, 2003).

This structure could allow also the possibility to

allocate defects on the photonic crystal which can

absorb chemical species: the change in emission due

to the absorption of molecules can be monitored as a

change of the laser emission and the whole would

perform as a compact and sensitive biosensing

device.

6 CONCLUSIONS

QCL present themselves as reliable sources for IR

range: they provide high optical IR power, their

emission can be tailored to span a selected range and

by mean of a superimposed grating it could be made

monomode. The emission is also tuneable with very

fine precision, either intrinsically by changing the

laser temperature or externally by using a coupled

diffraction grating Their performances make them

suitable for a series of application, since most of the

molecular species present distinctive absorption

lines in the medium infrared.

Different methods to realize QCL-based

biosensor have been already tested or are currently

in development, principally in the field of the gas

sensing for medical diagnostics applications or

environmental control: several substances have been

already proved to be detectable through the

utilization of QCL based sensing devices, which

start to provide significant and interesting results.

Such a development is a very interesting field

which looks to be double faced. From one side a

new class of biosensors could profit of the unique

properties of QCL in the IR range: the versatility of

such structures could allow a very wide range of

design and applications. On the other hand the study

of such sensors could provide a further push to the

design and development of even more efficient QCL

structures properly aimed to better match the

requirements of biosensor field.

ACKNOWLEDGEMENTS

The present work has been performed in cooperation

with the SISTER project of the AREA Science Park

of Trieste, Italy.

REFERENCES

Faist J. (1994) ‘Mid-infrared field-tunable intersubband

electroluminescence at room temperature by phonon-

assisted tunneling in coupled-quantum wells’, Applied

Physics Letters, vol. 64 no. 9, February, pp. 1144-

1146.

Hofstetter D. (2001) ‘High temperature operation of

distributed feedback quantum-cascade lasers at 5.3

µm’, Applied Physics Letters, vol. 78 no. 4, January,

pp. 396-398.

QUANTUM CASCADE LASERS FOR BIOSENSING APPLICATIONS

Pflügl C. (2003) ‘High-temperature performance of GaAs-

based bound-to-continuum quantum-cascade lasers’,

Applied Physics Letters, vol. 83 no. 23, December, pp.

4698-4700.

Faugeras C. (2005) ‘High-power room temperature

emission quantum cascade lasers at =9 m’, IEEE

Journal of Quantum Electronics, vol. 41, no. 12,

December, pp. 1430-1438.

McManus J.B. (2005) ‘A high precision pulsed quantum

cascade laser spectrometer for measurements of stable

isotopes of carbon dioxide’, Journal of Modern

Optics, vol. 52 no 16, November, pp. 2309-2321.

Roller C. (2002) ‘Simultaneous NO and CO

measurements in human breath with a single IV-VI

mid-infrared laser’, Optics Letters, vol. 27 no.2,

January, pp. 107-109.

Vacchi A. (2006) ‘Steps towards the development of a

QCL laser for muH experiments’, Book of Abstracts,

International Conference on Precision Physics of

Simple Atomic Systems, Venice, Italy, p. 33.

Röpcke J. (2006) ‘Application of mid infrared tuneable

diode laser absorption spectroscopy to plasma

diagnostics: a review’, Plasma Sources Science and

Technology, vol.15, October, pp. S148-S168.

Kosterev A. (2002) ‘ Chemical sensing with pulsed QC-

DFB lasers operating at 15.6 m’, Applied Physics B,

vol. 75, August, pp. 351-357.

Weidmann D. (2004) ‘Monitoring of ethylene by a pulsed

quantum cascade laser’, Applied Optics, vol. 43, June,

pp. 3329-3334.

Weber W. (2002) ‘Using a wavelength-modulated

quantum cascade laser to measure NO concentrations

in the parts-per-billion range for vehicle emissions

certification’, Applied Spectroscopy, Vol. 56, June, pp.

706-714.

Bakhirkin Y. (2006) ‘Sub-ppbv nitric oxide concentration

measurements using cw thermoelectrically cooled

quantum cascade laser-based integrated cavity output

spectroscopy’, Applied Physics B, vol. 82, November,

pp. 149-154.

Charlton C. (2005) ‘Midinfrared sensors meet

nanotechnology: Trace gas sensing with quantum

cascade lasers inside photonic band-gap hollow

waveguide’, Applied Physics Letters, vol. 86, May,

194102.

Hübers H.-W. (2006) ‘High-resolution gas phase

spectroscopy with a distributed feedback terahertz

quantum cascade laser’, Applied Physics Letters, vol.

89, August, pp. 061115.

Wysocki G. (2005) ‘Widely tunable mode-hop free

external cavity quantum cascade laser for high

resolution spectroscopic applications’, Applied Physics

B, vol. 81, September, pp. 769-777.

Phillips M. (2007) ‘External cavity quantum cascade laser

for quartz tuning fork photoacoustic spectroscopy of

broad absorption features’, Optics Letters, vol. 32,

May, pp. 1177-1179.

Colombelli R. (2003) ‘Quantum cascade surface-emitting

photonic crystal laser’, Science, November, pp. 1374-

1377.

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices