17 April 2017 Breath analysis using external cavity diode lasers: a review
Author Affiliations +
J. of Biomedical Optics, 22(4), 040901 (2017). doi:10.1117/1.JBO.22.4.040901
Most techniques that are used for diagnosis and therapy of diseases are invasive. Reliable noninvasive methods are always needed for the comfort of patients. Owing to its noninvasiveness, ease of use, and easy repeatability, exhaled breath analysis is a very good candidate for this purpose. Breath analysis can be performed using different techniques, such as gas chromatography mass spectrometry (MS), proton transfer reaction-MS, and selected ion flow tube-MS. However, these devices are bulky and require complicated procedures for sample collection and preconcentration. Therefore, these are not practical for routine applications in hospitals. Laser-based techniques with small size, robustness, low cost, low response time, accuracy, precision, high sensitivity, selectivity, low detection limit, real-time, and point-of-care detection have a great potential for routine use in hospitals. In this review paper, the recent advances in the fields of external cavity lasers and breath analysis for detection of diseases are presented.
Bayrakli: Breath analysis using external cavity diode lasers: a review



Technologies for clinical diagnoses and monitoring are always needed for discriminating patients from healthy subjects. To identify diseases and to monitor therapy and/or the progression of diseases, most used methods—such as biopsy, computer tomography, magnetic resonance imaging, endoscopy, and blood analysis—are time-consuming, expensive, invasive, painful, stressful, and difficult to perform. Therefore, noninvasive tools are needed to monitor the diseases in their early stages and to holistically measure the physiological status of patients. There are different potential noninvasive techniques, such as breath, nail, hair, mucus, saliva, sweat, and/or tear analysis. Among them, the detection of volatile organic compounds (VOCs) present in breath at trace concentrations with an acceptable accuracy and precision is most promising. This method may provide noninvasive, point-of-care (PoC), easily repeatable, painless, and real-time diagnosis of a number of disease states and therefore is preferable. Breath biomarkers can be analyzed either after administration of a drug or substrate or without any prior administration of a chemical.

A number of breath molecules have been detected and quantified so far using the different methods, such as proton transfer reaction mass spectrometry (PTR-MS), selected ion flow tube mass spectrometry (SIFT-MS), gas chromatography-mass spectrometry (GC-MS), e-noses, and laser-based sensors. Laser spectroscopic detection techniques—such as tunable diode laser absorption spectroscopy (TDLAS), cavity-enhanced absorption spectroscopy (CEAS), integrated cavity output spectroscopy (ICOS), cavity ringdown spectroscopy (CRDS), photoacoustic spectroscopy (PAS), quartz-enhanced photoacoustic spectroscopy (QEPAS), cavity leak-out spectroscopy, and optical frequency comb cavity-enhanced absorption spectroscopy—provide the features of high-sensitivity, high-selectivity, low costs, real-time measurements, and PoC function. Therefore, these methods are candidates for performing accurate breath diagnostics in hospitals.

To identify and quantify the molecules in breath, narrow linewidth (<1  MHz), mode-hop-free (1 to 2  cm1) wavelength-tunable, single-mode light sources are needed. These features can be obtained using external cavity lasers (ECLs). ECL-based spectroscopic techniques may become very suitable for monitoring diseases by analyzing exhaled breath air samples. Many review papers about breath analysis using laser-based techniques have been published. Ciaffoni et al.1 focused on the current status of volatile sulfur compounds sensing via laser absorption spectroscopy. The review article of Wang and Sahay2 is related to breath analysis of almost all of the biomarkers that have been analyzed in actual human breath using high-sensitivity laser spectroscopic techniques. Wojtas et al.3 presented recent advances in human breath analysis using optical methods. Risby and Tittel4 reviewed the current status of mid-infrared quantum and interband cascade lasers for clinical breath analysis. The review article of Bielecki et al.5 presented an overview on high-sensitivity laser absorption spectroscopic techniques used for trace gas detection. Mid-infrared spectroscopic sensing techniques were reviewed for potential application in breath diagnostics.6 In the review paper,7 techniques such as mass spectrometry-based (GC-MS, PTR-MS, SIFT-MS), laser absorption spectroscopy-based (CRDS and TDLAS), and other spectroscopic techniques for breath analysis applications were compared in terms of advantages/disadvantages, versatilities and plausibility to be transformed in clinical applications. McCurdy et al.8 summarized some of the recent advances in laser absorption spectroscopy based on semiconductor lasers and optical detection techniques for clinically relevant exhaled gas analysis in breath, specifically molecular biomarkers such as nitric oxide, ammonia, carbon monoxide, ethane, carbonyl sulfide, formaldehyde, and acetone.

Regarding the issues above mentioned, the present review article focuses on the latest status of breath analysis using the ECL-based methods. The article is structured as follows: first, we will explain the respiratory system and breath sampling. The absorption of electromagnetic waves by breath molecules is discussed next. The latest developments and functional principle of ECLs are described in detail as well. Finally, the laser-based techniques used for breath analysis and limitations of breath analysis are presented in detail. Regarding this content, the present review article differentiates itself from previous reviews related to the laser-based breath analysis.


Respiratory System

The lung is primarily responsible for gas exchange between the atmosphere and the blood. The molecules present in blood can diffuse into the alveoli and subsequently appear in the exhaled breath. Transport of metabolites from the blood to the exhaled breath air can make it possible to utilize breath analysis as a noninvasive diagnostic tool. Low and high blood-soluble gases present in the breath air:9 Gases with low blood-solubility like oxygen (λb:a0.7) and carbon dioxide (λb:a3) exchange in the alveoli. For these gases, there is a simple relationship between the breath and blood, which depends predominately on two factors: the blood solubility of the gas (λb:a) and the distribution of ventilation-to-perfusion in all gas exchange units. Gas exchange of high blood-soluble gases like ethanol (λb:a1756) takes place in the airways, and there is a more complex relationship between end-exhaled breath measurement and blood. Airway gas exchange is related to multiple factors, such as airway temperature, bronchial blood flow, and blood–air partition coefficient. Soluble gases with a blood:air partition coefficient less than 10 (λb:a<10) exchange almost solely with the alveoli, whereas gases with a blood:air partition coefficient greater than 100 (λb:a>100) exchange almost exclusively with the airways. Gases in between these two extremes (10<λb:a<100) exchange partially with the airways and partially with the alveoli.9,10

Figure 1 shows expirograms in which gases that exchange in the alveoli and those that exchange in the airways can be seen: phase 1 shows that the anatomic dead space empties. When fresh air is inhaled, low blood-soluble gases in the blood are transfered into the inspired air entering the alveoli. This action continues throughout the entire breathing cycle (inspiration and expiration). When the air is exhaled, molecules present in the conducting airways exit the lung first (phase 1). The airways take part in the gas exchange. Therefore, phase 1 does not exist in the expirogram for gases that exchange completely in the airways (e.g., ethanol). Therefore, an anatomical dead space cannot be given for such gases. For these gases, the end-exhaled breath concentration is always less than that in the alveoli. Phase 2 represents the transition from air residing in the dead space to alveolar air containing high partial pressures of gas. Phase 3 corresponds to air from the alveoli for low soluble gases and airways for high soluble gases. Alveolar air is expired via the conducting airways (phase 3). Molecules with high blood solubility are absorbed from the tissue and mucus lining the airway wall. In contrast to low blood-soluble gases, therefore, these gases present in large concentrations in the airway tissue and mucus for a given partial pressure. The partial pressure of soluble gas in the end-exhaled breath is always less than that in the alveoli due to the absorption–desorption process, which is the major mechanism of pulmonary gas exchange for any gas with a blood–air partition coefficient greater than 100.10

Fig. 1

Phases of an expirogram from a low blood-soluble gas that exchanges in the alveoli (red line) and from a high blood-soluble gas that exchanges in the airways (blue line).9



Breath Sampling

Breath sampling procedures are threefold: alveolar, dead-space, and mixed sampling. Dead space air contains the first 150 mL of expiration air, and here no gas exchange takes place. Alveolar breath is the part of exhaled air coming after the dead-space air, whereas end-tidal air is the last fraction of expired air. Alveolar gas exchange is dependent on ventilation, pulmonary perfusion, and the blood:air partition.10 The term “end-exhaled breath” should be utilized for highly blood soluble VOCs, whereas the term “alveolar breath” may be used for low blood-soluble compounds.

It is often tacitly assumed that end-tidal air will reflect the alveolar concentration CA, which is related to the concentration of the VOC in mixed venous blood Cv, the substance-specific blood:gas partition coefficient λb:air, alveolar ventilation V˙A, and cardiac output Q˙c:11,12


where λb:air states the diffusion equilibrium between alveoli and capillaries, V˙A describes the volume of VOC per unit time that arrives to the alveoli, and Q˙c describes the blood volume, which is pumped by the heart.

In mixed sampling, total breath (alveolar + dead space) is collected. Concentrations of endogenous VOCs in alveolar air are two to three times higher than those found in mixed expiratory samples because there is no dilution by dead space gas.13

There may be discrepancies between the true alveolar breath and the measured levels of water-soluble VOCs because their concentrations may be diluted on their way (a dilution effect). A single breath or multiple breath cycles can be applied for breath sampling.14 Breaths may considerably change from each other. The alveolar gas sample may not be represented by the composition of a single breath. Therefore, multiple breaths may be preferable.


Absorption of Electromagnetic Waves by Molecules

Analysis of the breath molecules using laser-based methods is based on the absorption of electromagnetic waves by molecules under investigation. Therefore, we will explain the absorption phenomena in this section. The overall excited energy of a molecule is a sum of the electronic (Eel), vibrational (Evib), and rotational transitions (Erot) energy:



Rotational spectra originate from the transitions between rotation levels of a given vibration level in an electronic state [Fig. 2(a)]. In this process, only the rotational quantum number (j) changes. These spectra lie in the spectral range from the microwave to the far infrared. The rotational–vibrational spectra consist of the transitions from rotation levels of a vibration level to rotation levels of another vibration level in the same electronic state [Fig. 2(b)]. The rotation and vibration (ν) quantum number change in these transitions. These spectra lie in the infrared spectral region. Electronic spectra consist of the transitions between the rotation levels of different vibration levels of an electronic state and rotation or vibration levels of another electronic level [Fig. 2(c)]. There are changes in all three quantum numbers (J, ν, and electronic quantum number). These spectra lie in the near-infrared or visible range or in the UV range.

Fig. 2

Electronic, vibrational, and rotational quantum states and corresponding transitions: (a) rotational transition, (b) rotational–vibrational transition, and (c) electronic transition.


The number f of normal vibrations of a molecule is given as


f=3N5(for linear molecules),


f=3N6(for nonlinear molecules),
where N is the number of internal degrees of freedom. For water molecules, N is equal to 3: there are three modes of normal vibrations (Fig. 3). Transitions from the ground state to the first excited state are known as fundamental transitions (e.g., 0,0,0 → 1,0,0). There are also other transition types: harmonic transitions (overtones) from the ground state (e.g., 0,0,0 → 2,0,0), combination transitions from the ground state (e.g., 0,0,0 → 1,2,1), harmonic transitions between excited states only (e.g., 1,0,0 → 3,0,0), and combination transitions between excited states only (e.g., 1,2,0 → 2,3,1).

Fig. 3

Normal vibration modes of a water molecule and its characteristic ground and first excited vibrational energy levels.


The ratio of the transmitted intensity I and initial intensity I0 of laser beam transmitting through an absorbing medium at a particular frequency is given using the Beer–Lambert equation:




where α is absorption coefficient, ϕ is lineshape function, P is total pressure of the medium, x is mole fraction of the absorbing species, L is pathlength, n is the number-density of the absorbing species, and Si (S) is linestrength, which is defined as the total absorption per molecule. The linestrength of an absorption transition depends on the population in the lower quantum state, which is a function of the Boltzmann fraction, and the probability of the transition.15 The linestrength Slu for a transition lu normalized to the total number density N is expressed (in cgs units) by16


where Nl and Nu are the number density of molecules in the states l and u, respectively, with their corresponding statistical weights gl and gu, h denotes the Planck constant, |μlu| is the transition dipole moment, and νlu (in cm1) is the frequency of the transition. In the case of thermal equilibrium of the system at temperature T, the equation above can be rearranged using the Maxwell–Boltzmann distribution law (SI units):


with the energy of the lower state El and the total internal partition function Q(T). The equation for nu/gu is analogue. This leads to a form of S (in cm/molecule), which accords with the notation of the HITRAN database:


where Ia is the abundance. Since the linestrength in HITRAN is defined for the reference temperature Tref=296  K, S(T) must be corrected using the tabulated values S(Tref), El, νlu, and Q(T):16


where S is in units of [cm1/(mol·cm2)]. For units of [cm2atm1], the following temperature scaling can be used:15



The conversion between S(T) and Si(T) is given as follows:15



A typical lineshape of an isolated absorption line is centered at ν0 and has a linewidth Δν. The absorption line can be broadened due to natural broadening, Doppler broadening, and collision broadening. The homogeneous natural and collision broadening result in Lorentzian line shapes, whereas the Doppler broadening gives rise to a Gaussian line profile. The natural linewidth is determined by the lifetimes of the energy states (τ1 and τ2):



The linewidth can be broadened due to the Doppler effect. The distribution of random velocities of molecules in gas is described by the Maxwellian velocity distribution function. Each velocity group gives rise to its own Doppler shift. The Doppler broadening is given by15


with Doppler halfwidth [full-width at half-maximum (FWHM)]:


ΔνD=ν0(8  kTln2mc2)1/2.

The perturbations that shorten energy level lifetimes cause collisional broadening. The linewidth broadening occurs due to the Heisenberg uncertainty principle; shortening the energy level lifetime results in greater uncertainty of the energy of molecule, and thus the absorption lineshape becomes broader. The lineshape for collisional broadening takes the form of a Lorentzian function:15


with collision halfwidth (FWHM):


where m is the mass of molecule, d is the diameter of molecule, and p is the pressure of gas.

At low pressure, the near-Gaussian line shape is obtained because thermal motion is the dominant broadening mechanism. The line shape becomes near-Lorentzian with the increasing pressure. When the Doppler contribution can be neglected (ΔνcpΔνD), the line shape can be considered pure Lorentzian. In the case of ΔνcpΔνD, a pure Gaussian line shape is received. Molecular ro-vibrational transitions exhibit pressure broadened FWHM of 3  MHz (0.0001  cm1) at low pressures <5  mbar and 1.5  GHz (0.05  cm1) at atmospheric pressure, where Doppler widths of 30 … 90 MHz (0.0010.003  cm1) at room temperature are typically observed for molecular transitions in the infrared.16 When the condition ΔνcpΔνD is fulfilled (between 5 and 100 Torr for most lightweight gases), the line shape will be a combination of the Gaussian and Lorentzian, which is known as Voigt profile.


External Cavity Diode Lasers

Different methods can be utilized to realize a wavelength-tunable single-mode operation: tunable CO2 lasers, tunable dye lasers, tunable solid-state lasers, tunable excimer lasers, nonlinear optical frequency conversion lasers, tunable free-electron lasers, and tunable external cavity semiconductor lasers. Within the framework of this work, the focus will only be on the tunable external cavity semiconductor lasers (ECLs). Further information on the other tunable lasers can be found in, e.g., Ref. 17.

Grating-coupled ECLs can be realized in Littrow or Littman configurations.18 In the Litrow configuration [Fig. 4(a)], there are a gain chip, a lens (or a parabolic mirror), and a diffraction grating. The gain chip serves as the gain medium for ECL setup. The lens (or parabolic mirror) is used to collimate the beam emitted from the gain chip. The collimated beam is diffracted by a grating. The first order of the diffracted beam is fed back into the gain medium. The wavelength is tuned by rotating the grating. Because two optical components (mirror or lens and grating) are used, the alignment is easy and the feedback is strong.

Fig. 4

(a) Littrow and (b) Littman configuration.


In the Littman configuration [Fig. 4(b)], a mirror is added to the setup. The diffracted beam is sent back to the grating by a mirror and diffracted again. Finally, the beam reaches to the gain chip and it is amplified. The tuning is usually achieved by rotating the mirror. Because of the double pass of the beam on the grating, the optical feedback strength is reduced. However, the Littman setup has an advantage of the increased wavelength selectivity.

Generally, the Littrow setup is preferable for maximizing the tuning range. To optimize the optical feedback, the incident beam and first order of the diffracted beam have to be colinear. The directions of the various orders of the diffracted beam with respect to the normal are determined by


where φi is the angle of the incident beam, φr,m is the angle of the diffracted beam of the m, order, d is the period of the grating, and λ is the wavelength. The wave whose wavelength satisfies the condition:


is selected by the grating and is amplified. For the other wavelengths, it can be written as follows:


where Δφ is the angle between incident and diffracted beam [see Fig. 4(a)] and is defined as



Finally, the wave, which makes an angle Δφ, is focalized on the facet at a distance:


where f is the focal length of the lens.

The reflectivity of the facet inside the external cavity setup is very important for a properly operating ECL. This reflectivity should be as low as possible to effectively suppress the Fabry–Perot modes of the gain chip, so a pure single mode can be selected and amplified. The reflectivity of the outcoupling facet of the gain chip can be reduced by various ways, such as the tilted gain stripes, the buried facets, and the antireflection (AR) coating. The tilting of the waveguide decreases in the internal Fresnel reflection from the facet and therefore increases the losses.19,20 In the case of buried facets, a semi-insulating material between the end of the guide and the facet, which reduces the coupling of the beam back into the active region, is utilized.21 When a facet is AR-coated, a phase shift of π and, therefore, an optical path difference of λ/2 occur after the reflection in the optical thick medium. The reduction of the reflectivity of the AR-coated facet is achieved by the destructive interference between the waves reflected from the front and rear sides of the layer. The reflectivity reduction causes a decrease in the competition between the modes of the small cavity (gain chip) and of the large cavity (ECL) for oscillation. This situation leads to enhancement of the coupling efficiency to the EC, to increased output power of the ECL, and to enlarged coarse/fine tuning range and the range of the stable operating region.


Wavelength Selection

Tuning mechanism of an ECL can be explained by considering a sketch for the gain and losses in the ECL setups (Fig. 5).17 As one can see in Fig. 5, there are three loss mechanisms: (1) Filter loss due to the wavelength selectivity property of the wavelength selective element (e.g., grating), (2) coupling and output loss between the guided waves in the gain chip and vacuum in ECL, and (3) facet etalon loss, which is modulated at the period of the internal Fabry Perot mode spacing Δνint. The facet etalon loss is caused by the interference between the reflections from facets.

Fig. 5

Gain and losses for ECLs.


The injection current is controlled to tune the gain profile of the gain chip and the internal cavity mode frequency. To vary the external mode frequency, the grating can be rotated and/or the EC length can be changed by applying the voltage to a piezoelectric transducer controlling the grating position. For a perfect operation of the ECLs, the facet etalon loss ripple should be weak and the bandwidth of the wavelength selective element loss should be narrow compared with the period of the ripple. The bandwidth of the grating can be estimated by the following equation:


where λ is the wavelength, a is the diameter of the collimated laser beam, and θ is the incident angle of the beam. Only the EC mode, which is the closest to the loss minimum of the wavelength selective element, can oscillate. The other modes see higher losses and cannot be amplified. Tuning to the next EC mode can be achieved by varying the loss minimum of the wavelength selective element. In the case of the grating-coupled ECLs in Littrow configuration, this variation occurs by rotating the grating [Fig. 6(a)] and/or changing the cavity length [Fig. 6(b)]. A strong facet etalon loss and a wider bandwidth of the wavelength-selective element loss lead to tuning with EC mode jumps approximately equal to Δνint. This discontinuity is called “tuning gap” [Figs. 6(a) and 6(b)]. The shift in the external mode frequency exceeding half the free spectral range of the external cavity gives rise to mode hops. At this point, the adjacent external mode has moved closer to the peak of the internal mode, and the laser oscillation hops to this new mode.22 To achieve the tuning without any gap (tuning on the EC mode, mode-hop-free tuning) [Fig. 6(c)], there are different methods: (1) the injection current, the grating angle, and/or the external cavity length should be simultaneously controlled;23 (2) the grating has to be mounted on a pivot arm, which is rotated about an axis lying near the intersection of the grating and laser output facet plane;24 and (3) an electro-optically tunable single-mode ECL has to be built.25

Fig. 6

Possibility of changing the grating position and corresponding wavelength tuning. (a) The grating is rotated around its the center, (b) the grating is moved back and forth, and (c) the grating is rotated around an axis lying near the intersection of the grating and laser output facet plane.



Spectral Tuning Characteristics

To consider the gain chip and the ECL as a combined laser system with a compound cavity, the intracavity facet should have a low reflectivity, and the coherence length should be longer than the distance from the gain chip to the grating. The tuning range is limited by the gain bandwidth of the used gain chip that is an inherent characteristic of the gain medium. The separation between the FP and EC modes must be larger than the bandwidth of the grating for a single-mode operation. This can be achieved, for instance, using an appropriate AR coating.

The theoretically reachable tuning range of an ECL for wavelength is given by26


and, equivalently, the frequency tuning range:


where λ is the emission wavelength, 2γgain is the FWHM of the modal gain spectrum, E0 is the photon energy at the gain maximum, g0 is the apex of the modal gain spectrum, h is the Planck constant, L is the chip length, RG is the first-order efficiency of the grating, and TL is the lens transmission. The relative tuning is given by Δλ/λc, where λc=(λmin+λmax)/2 is the center wavelength and λmin and λmax are the shortest and the largest achievable wavelengths, respectively, in the ECLs. From these equations, one can conclude that the following should be considered to increase the tuning range: a gain chip with a broad gain bandwidth should be used in an ECL setup, and reflectivity from the facet inside the ECL should be decreased by one of the methods mentioned above. In Table 1, some ECLs reported in the spectral range between 1.0 and 11.5  μm are listed.

Table 1

Some ECLs in the spectral range between 1 and 11  μm.

Wavelength (μm)Tuning range (cm−1)Temperature (K)LinewidthReference
1567RT400 Hz (FS)27
1.3780RT2.9 kHz (FS)28
1.5720RT140 kHz (FR)29
1.5870RT5 kHz (FR)30
8.4 and 9.6265nRTNS34
7.3, 8.5, 9.4, 10.4, and 11.5432RTBetween 2 and 35
8.2 and 9.3292RTNS36

Note: RT, room temperature; nRT, near room temperature; FR, free running; FS, frequency stabilized; NS, not stated.



The laser linewidth (FWHM of the lineshape) describes the spectral purity of a laser (Fig. 7). The linewidth of a semiconductor laser is given by the Schawlow–Townes formula modified by Henry (Ref. 37 and references therein):


where ν is the lasing frequency, α is the total losses of the cavity, αm is the mirror losses of the output facet, P is the output power, Δνc=αc/(2πn) is the FWHM of the cavity resonance, n1t and n2t are the populations of the upper and lower states of the laser transition at threshold, respectively, and αe=(dnr/dN)/(dni/dN) is the linewidth enhancement factor introduced by Henry. The imaginary (nr) and real parts (ni) of the refractive index change during the relaxation oscillations of the field intensity. Varying of the imaginary part of the index causes the gain to change, whereas the change in the real part of the refractive index gives rise to an additional phase shift. Therefore, the laser linewidth is broadened by a factor (1+αe2) compared with the Schawlow–Townes linewidth. The lower energy level of semiconductor lasers is not empty, which is different from most other lasers. As laser photons in the valence band may reradiate by spontaneous emission, phase fluctuations of the laser field increase. This is another phenomenon that leads to the linewidth broadening of semiconductor lasers.

Fig. 7

Spectral line.


Loh et al.38 investigated experimentally the influence of the grating reflectivity, grating resolution, and AR coating on the intrinsic linewidth of a grating-coupled ECL. They found that the linewidths are little affected by the presence of an AR coating on the diode facet and that the grating resolution has a larger effect on the linewidth than the grating reflectivity.38

The laser linewidth can be determined using several methods, such as optical cavity measurements,29 heterodyne measurements,37 homodyne/heterodyne interferometry using a long interferometer delay,39 measurements from the frequency noise power spectral density,40 measurements from root-mean-square frequency deviation,41 and saturation spectroscopy.42

Tkach and Chraplyvy43 reported that there are five distinct regimes of feedback effects, with well-defined transitions between them as a function of feedback power ratio. At the lowest levels of feedback down to 90  dB (regime I), linewidth narrowing or broadening is observed depending on the phase of the feedback. Regime II is seen at a feedback level (up to around 45  dB), which depends on the distance to the external reflector, and frequency splitting of the mode and mode hopping are observed. Regime III is found at a level (about 39 to 45  dB), which does not depend on the distance to the reflection. In this regime, the mode hopping is suppressed, and the laser is observed to operate on a single narrow line. Regime IV starts at 39  dB and is independent from the distance to the reflection. In this regime, line broadening is observed due to the coherence collapse effect. Relaxation oscillation sidebands that present in each mode in the optical spectrum increase in strength giving rise to the line broadening. Consequently, coherence time decreases. Regime V is independent of distance too and observed at the highest levels of feedback (usually greater than 10  dB). AR coating of output facet is needed to achieve this regime. In this regime, the operation of the laser occurs on a single longitudinal mode with narrow linewidth for all phases of the feedback and the laser is relatively insensitive to additional external optical perturbations.43


Frequency Stabilization

There are different noise sources, which can affect on the laser’s linewidth: (1) fundamental noises (white, shot, and thermal noises): the intrinsic Lorentzian linewidth can be reduced using optical feedback44 and (2) flicker (1/f) noises and technical noises (noises from the laser driver, piezo drift, mechanical vibrations, temperature fluctuations, acoustic noises, air pressure fluctuations). These noise sources broaden the laser linewidth. This Gaussian linewidth can be measured using the methods given above. For reduction of these noises, with other word, stabilization of a laser’s frequency over time and consequently narrowing the linewidth of laser, there are different stabilization techniques, such as side-of-fringe stabilization,45 on-peak-locking, Paul–Drever–Hall (PDH) method.46 Using these techniques, the sub-kHz-linewidth47 and sub-Hz-linewidth48 ECLs have been reported.


Rate Equations

A set of rate equations can describe the time evolution of the photon flux S and the electronic populations of the upper and lower states of the radiative transition (n3,n2). Yang et al.26 conducted rate equations analysis of EC quantum cascade lasers. They considered a four-level quantum cascade laser structure. The EC system was simplified to an effective system with both the internal cavity and the EC having the same cavity length Leff. To consider the longer round-trip time of the EC, they used the effective speed of light introduced by Maulini.49 The EC can be simplified to have an effective cavity length of Leff but with a reduced speed of light c/ρcav, where ρcav=(τint+τext)/τint, τint and τext are the round-trip times of the internal cavity and the EC, respectively. The rate equations then are given by26,49,50








where J is the injection current density, e is the unit charge, n3 and n2 are the sheet carrier density of upper laser level 3 and lower laser level 2, τij represents the nonradiative scattering lifetime from level i to level j (i, j=1, 2, 3), 1/τ3=1/τ32+1/τ31, τ2=τ21, τsp is the spontaneous emission time (calculated to be 1.4×107  s), β0.001 is the fraction of the spontaneous light emitted in the lasing mode, SFP is the photon flux density (per unit length per second) generated by the laser chip facets, SEC is the photon flux density for a certain mode selected by the external cavity, αFP and αEC represent the total losses corresponding to the frequencies of FP mode and the EC mode, gFP and gEC denote the values of the gain cross-section corresponding to the frequencies of FP mode and the EC mode and are determined by26




where z322 is the matrix dipole element of the optical transition, Γ is the optical confinement factor of the laser, ε0 is the free-space permittivity, Lp is the period length of the QCL, λ0 is the emission wavelength, 2γ32 is the FWHM of the gain spectrum, ν is the lasing frequency, and h is the Planck constant.


External Cavity Laser-Based Techniques for Breath Analysis

The ECL-based techniques can be effectively used for the detection and quantification of breath VOCs at concentrations ranging from parts per million by volume (ppmv) to the parts per trillion (pptv) level. The laser sources should have high sensitivity and selectivity for breath molecule sensing. For selectivity, an absorption line that is free of interference from other species in gas mixture should be used. For the best sensitivity, a strong molecular absorption line should be chosen. To achieve a higher signal-to-noise ratio (SNR), either the noise levels should be decreased and/or the signal should be enhanced. For decreasing the noise levels, there are different methods, such as wavelength/frequency modulation spectroscopy. The signal can be further increased by obtaining an effective long pathlength using multipass cells or cavity enhancement methods. ECL-based spectroscopic detection techniques, such as multipass cell absorption method (MCAB), CRDS, CEAS, ICOS, PAS, and QEPAS, provide the features of high-sensitivity, high-selectivity, low costs, real-time measurements, and PoC function. Furthermore, laser-based techniques can be implemented without the need for radioactive labels. Therefore, these methods are candidates for performing accurate breath diagnostics in hospitals. In Table 2, some important breath molecules, detection wavelength, detection limit, and methods are summarized.

Table 2

Some important breath molecules, detection wavelengths, detection limits, and methods.

Breath moleculesDetection wavelength (μm)Detection limitMethods
Acetaldehyde (5.7980 ppbMPAS
Acetone (0.26657 ppbCRDS
8.20.17 ppmCEAS
Ammonia (1.5118 ppbCEAS
1.530.65 ppmQEPAS
10.346 ppbQEPAS
Carbon dioxide (1.593 ppmCRDS
Carbonyl disulfide (4.5928 ppbQEPAS
Carbon monoxide (CO)1.564900 ppbCEAS
2.343.3 ppmQEPAS
Carbonyl sulfide (OCS)5.262250 ppbCEAS
Ethane (3.340.3 ppbCEAS
3.341 ppbMPAS-WMS
Ethanol (9.430 ppbPAS
3.8157 ppbCRDS
Ethylene (10.5330 ppbPAS
9.2230 ppbPAS
10.58 ppbQEPAS
Formaldehyde (HCHO)3.66 ppbMPAS-WMS
Hydrogen sulfide (1.576 ppmPAS
2.6500 ppbQEPAS
Isoprene (0.2664 ppmCRDS
Methane (3.40.1 ppmMPAS-WMS
3.31.4 ppbMPAS
Nitric oxide (NO)5.20.7 ppbCRDS
5.40.2 ppbCRDS

Note: CRDS, cavity ringdown spectroscopy; CEAS, cavity-enhanced absorption spectroscopy; (QE)PAS, (quartz-enhanced) photoacoustic absorption spectroscopy; MPAS, multipass absorption spectroscopy; WMS, wavelength modulation spectroscopy.


Multipass Absorption Spectroscopy

In the MCAB,75 the absorption signal can be enhanced via longer optical pathlengths using optical multipass cells. There are four types of multipass cells most commonly applied: White cells,76,77 Herriott cells,78 Chernin cells,79 and astigmatic mirror multipass cells.80 Beer–Lambert law of linear absorption is used to determine the absorption coefficient α(ν):


where I(ν) is transmitted intensity of the radiation after a distance Leff through the medium, I0 is initial intensity of the radiation, Leff is effective path-length, σ(ν) is absorption cross-section of the absorbing species, N is molecular concentration (the number density of the absorbing species), S is linestrength (i.e., the total absorption per molecule) of the absorbing species at temperature T, ϕ(νν0) is line shape function for the particular absorption line.

Using Eq. (33), the absolute number density N can be calculated by integrating the absorption coefficient:



The Herriott cell consists of two mirrors (Fig. 8). The input mirror has a small hole. A laser beam is injected into a Herriott cell through the hole and it is reflected multiple times between the two mirrors. The light path might be lengthened 10 or 100 times inside the cell. The laser beam leaving the cell through the same or the other hole reaches the detector. The sensors employing a multipass gas cell have been reported for many trace gases, such as acetaldehyde,51 ethane,61 formaldehyde,67 and methane.71,72

Fig. 8

Herriott multipass cell.



Cavity Ringdown Spectroscopy

CRDS81 is based on the observation of the decay rate of a laser beam stored in an optical cavity formed from ultra-high reflective mirrors (Fig. 9) and is first demonstrated by O’Keefe and Deacon.82 In this case, the absorption coefficient is determined using the following equation:


where τ0 and τ are the ringdown time without and with the absorbing gas present, c is the light velocity.

Fig. 9

Cavity ringdown absorption cell.


A laser beam, which is coupled into the optical cavity, is reflected back and forth inside the cavity. When beam is reflected every time, a small fraction of this light leaks out of the cavity. Many trace gases, such as acetone,52 carbon dioxide,57 ethanol,63 isoprene,70 and nitric oxide,73,74 have been analyzed using the CRDS technique.


Integrated Cavity Absorption Spectroscopy and Cavity-Enhanced Absorption Spectroscopy

ICOS83 and CEAS84 are the same method. ICOS is used in the context of pulsed excitation, and CEAS is used in the context of continuous-wave cavity excitation. This method is based on analyzing the intensity of light transmitted through a high finesse optical cavity as a function of wavelength (Fig. 10). To reduce the noise levels and therefore to improve the detection limit, the laser beam can be coupled into the optical cavity at a small angle with respect to the cavity axis or away from the central axis of the optical cavity. This is known as “off-axis alignment,” which has advantage over on-axis alignment: it is almost insensitive to vibrations. In the case of ICOS and CEAS, Eq. (36) is applied to extract the absorption coefficient:


where R is mirror reflectivity and d is the distance between mirrors.

Fig. 10

High finesse optical cavity for ICOS or CEAS.


The gases—such as acetone,53 ammonia,54 carbon monoxide,54 carbonyl sulfide,60 and ethane60—have been analyzed using the ICOS or CEAS techniques.


Photoacoustic Spectroscopy and Quartz-Enhanced Photoacoustic Spectroscopy

In the PAS,85 the acoustic waves resulting from the absorption of radiation (photoacoustic effect) are detected and analyzed using a sensitive microphone (Fig. 11). The PAS response S to optical absorption is given by Eq. (37):


where α is the absorption coefficient per unit concentration of the target species, C is the concentration of the target species, P is the optical power, f is the PA sound frequency, A is the resonator cross-section area, and k is a constant describing the microphone transfer function and other system parameters. Usually, PA resonators are designed for f values in the 500 to 4000 Hz range and have Q factors of 20 to 200.86

Fig. 11

Photoacoustic cell.


Another approach to photoacoustic detection of trace gases is QEPAS,86,87 which utilizes a quartz tuning fork (QTF) as a sharply resonant acoustic transducer. The QTF is a piezoelectric element. To further increase the QEPAS signal, a microresonator can be added to the QEPAS sensor system.88 Also, to achieve the best QEPAS detection sensitivity and selectivity, it is important to choose the optimum operating pressure for a QEPAS sensor system. The PAS- or QEPAS-based sensors have been used for the detection of many gases, such as ammonia,55,56 carbonyl disulfide,58 carbon monoxide,59 ethanol,62 ethylene,6465.66 and hydrogen sulfide.68,69


Breath Biomarkers

Owing to its noninvasiveness, ease of use, easy repeatability, and painlessness, breath analysis has attracted considerable attention for scientific and clinical studies. Until now, many studies have demonstrated that the human breath air contains more than 1000 different molecules.2,89 The matrix of exhaled breath air is much less complicated compared to serum or urine. A few atmospheric molecules in % concentration (N2, O2, H2O, CO2), inorganic gases (e.g., NO, NH3, and CO), VOCs at the levels of parts per million (ppm), parts per billion (ppb), or parts per trillion (ppt) (e.g., CH4, HCO2, N2O), and nonvolatile substances measured in breath condensate (e.g., isoprostanes, cytokines, leukotrienes) have been found to be present in exhaled human breath (Table 3). Spectral fingerprints of these molecules span from the UV to the IR spectral regions. The mixture is exhaled at temperatures between 34°C and 37°C while relative humidity may range from 91% to 96% in oral exhalations and from 82% to 85% in nasal exhalations.90 The mixture of exhaled breath air varies from person to person both quantitatively and qualitatively.91 Analysis of some VOCs, which present in everyone, can provide a safe method for studying human metabolism, detecting diseases, clinical diagnosis, therapeutic monitoring, and investigation of potential adverse effects of new drug therapy. In Table 4, some important breath molecules, concentration, physiological basis, and related diseases are summarized. The 872 breath VOCs from apparently healthy humans have been reported in the review article of de Lacy Costello et al.89

Table 3

Main breath molecules.3

MoleculesInhaled air (%)Expired air (%)
Carbon dioxide0.045.00
Water vapor1.00–2.005.00
Other (1000 different molecules)0.01Different for each person

Table 4

Some important breath molecules, concentration, physiological basis, and related diseases.

Breath moleculesConcentrationPhysiological basisRelated diseases
Acetaldehyde (ppbEthanol metabolismAcute respiratory distress syndrome
Acetone (ppmDecarboxylation of acetoacetateDiabetes
Ammonia (ppmProtein metabolismHelicobacter pylori
Kidney diseases
Butanol (pptButyryl-CoALung cancer
Carbon dioxide (%Product of respirationHelicobacter pylori
Carbonyl disulfide (ppbGut bacteriaCoronary and artery disease
Cystic fibrosis
Carbon monoxide (CO)ppmHeme oxygenaseAsthma
Respiratory tract infections
Carbonyl sulfide (OCS)ppbGut bacteriaLiver-related diseases
Ethane (ppbLipid peroxidationScleroderma
Ethanol (ppbGut bacteriaColorectal cancer
Ethylene (ppbLipid peroxidationSchizophrenia
Renal failure
Formaldehyde (HCHO)ppmLipid peroxidationLung cancer
Hydrogen cyanide (HCN)ppbPseudomonas aeruginosaChronic airway infection with Pseudomonas aeruginosa in adults with cystic fibrosis
Hydrogen sulfide (ppmAnerobic bacterial metabolism of thiol proteinsAlzheimer
Isoprene (ppbCholesterol biosynthesisBlood cholesterol
Methane (ppmGut bacteriaColonic fermentation
Methanethiol (ppbMethionine metabolismStreptococcus pneumoniae
Methanol (ppbMetabolism of fruitCirrhotic liver
Lung cancer
Methyl nitrate (pptType 1 diabetes
Nitric oxide (NO)ppbNitric oxide synthaseObstructive sleep apnea
Airway inflammation
Lung inflammation
Pentane (ppbLipid peroxidationObstructive sleep apnea


Limitations of Breath Analysis

It has been known for a long time that the smell of exhaled breath is an indicator for diseases.122 The work of Pauling et al., in which hundreds of VOCs were determined in normal human breath at the levels of parts per billion (ppb) or lower, has been accepted as the beginning of the modern breath analysis. In spite of the long investigation time, this method has not yet been introduced as a standard tool into clinical diagnosis. Very few diseases are actually diagnosed using exhaled breath air analysis; for example, breath urea analysis. There are several potential confounders and issues blocking the introduction of breath analysis into clinical applications. To widely use the breath test for more specific diseases in hospitals, the following issues should be addressed:123124.125

  • The ease of use, low response time, low detection limit, high sensitivity, high selectivity, portability, compactness, small weight, reliability for long time periods, low energy consumption, and low construction and maintenance cost devices with good precision and accuracy being able to perform the online measurements, point-of-care detection, operating at room temperature are very important for widespread clinical use of the breath tests. Such devices are currently being developed. The availability of such devices can represent a breakthrough for clinical use of breath analysis.

  • The correlation between diseases and breath biomarkers should be understood in depth. The issues of biointerferences should be addressed: a molecule present in breath can be established as a biomarker for more than one disease; one specific disease can be detected by more than one breath molecule.

  • Exhalation physiology of VOCs should be understood exactly. The sources of the breath biomarkers inside the human body should be known very well. Also, we should be able to explain whether the breath molecules are exogenous (not biochemically produced in the body and therefore not a part of metabolic processes; e.g., taken up via skin absorption, inhalation from the ambient air, ingested foods, and beverages) or endogenous (generated by metabolic processes at various organs within the human body, excreted into blood, exchanged with the inhaled air in the lung via the alveolar interface, and ultimately exhaled) in origin. The other alternative sources of VOCs are bacteria in the gut, viruses, fungi, saliva, and mucus created in the respiratory tract. VOCs may also be generated in the airways and the oral cavity by bacterial infections. Cells or tissues in the mouth, nose, sinuses, airway, and the gastrointestinal tract can also generate molecules in exhaled breath that are not found in circulating blood (such as nitric oxide).123 Therefore, the concentrations of breath biomarkers can measure higher when there are also exogenous sources, which contribute molecules to exhaled breath. The levels of some molecules are dynamic, e.g., the levels of ammonia, which is the by-product of protein metabolism. Because protein metabolism can change based on many factors, such as pH, food/beverage consumption, medications, and exercise, the concentrations of ammonia are changeable depending on the protein metabolism.

  • The acceptable standardization and normalization of procedures for sampling, analysis, and background correction should also be determined. The correction for background concentrations of VOCs is necessary for distinguishing endogenous substances from exogenous contaminants, which have no diagnostic value. There are different approaches for background correction. One approach is to calculate “alveolar gradients” (expiratory concentrations—ambient or inspiratory concentrations).126 Calculating alveolar gradients is easy to realize, but this method does not consider the complexity of the inhalation/exhalation function, as given above and below. Another approach is to try washing out the lung from ambient concentrations by subjects breathing pure air for a certain time before measurement.127 However, the washout of the entire body may take days or weeks, depending on the identity of the molecule.4 There is no consensus for a standard method to enable the background levels to be subtracted.4 Currently, accepted standardized techniques for breath air sampling and analysis do not exist. Therefore, some results obtained by different groups are not always consistent. The American Thoracic Society published the first guidelines in 1999128 and updated guidelines in 2005129 for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide in adults and children.

  • An acceptable unit should be used. It is common to express the concentration of breath VOCs as ppm by volume (ppmv), ppbv, or pptv. An alternative is to express exhaled concentrations in mol or weight per volume.

  • Increased or decreased levels of biomarkers in breath can be an indicator for a large number of diseases. Therefore, the range between disease or wellness states should be defined exactly. For this purpose, a larger number (several hundreds) of human subjects should be investigated.

  • How the concentrations of breath molecules under investigation depend on the parameters, such as ethnicity, gender, pulmonary function, age, body mass index, smoking and alcohol consumption, etc, should also be defined. Dependence of diffusion of VOCs from blood to alveolar air across the alveolar–capillary membrane on their physicochemical properties, such as polarity, solubility in fat, Henry partition constant, and volatility, should also be considered.130 The low blood-soluble gases exchange in the alveoli, while the gas exchange of highly blood-soluble VOCs takes place in the airways rather than alveoli.9 A lipid-soluble molecule could be stored in tissues, not well perfused by blood, and be released more slowly than a similar molecule with hydrophilic properties that is not stored.123 The physiological parameters, such as blood pressure, heartbeat rate, blood:gas partition coefficient, cardiac output, ventilation/perfusion ratio of the lung, and alveolar ventilation, may also have an effect on the exhalation of breath biomarkers.131 These effects may be negligible when the subject is healthy. However, in patients with severe hemodynamic or pulmonary dysfunction, the normalization or correction of data can be needed by means of end tidal and/or arterial pCO2 measurement.13

  • Data interpretation is a crucial step in breath analysis.125 Confounding variables, which have a real statistical correlation with the disease and a breath marker, are environmental compounds, medical history, gender, age, weight, physiological parameters like cardiac output and respiratory flow, effects of breath sampling, sample storage, and the analytical technique. These effects have to be considered and addressed during data interpretation.125

To address all these issues and to move breath analysis forward, collaboration among optical engineers, experts in breath analysis, clinicians, and spectroscopists is required.


Conclusion and Outlook

Several issues should be addressed for advances in the field of breath monitoring and for widespread usage of this method as a routine clinical tool. Breath analysis has not yet been used in hospitals as a clinical tool due to lack of generally accepted evaluation criteria for data. Multiple potential confounders, such as interference from other molecules, smoking and alcohol consumption, dermal absorption, inhalation from the ambient air, and so on, should be carefully considered when performing measurements of breath biomarkers.

Breath research may be performed as (1) exploratory research (the discovery of new breath molecules, e.g., breath markers for chronic kidney disease), (2) focused research (the in-depth evaluation of a molecule whose biochemistry is already well-understood, e.g., breath acetone), and (3) breath condensate (the analysis of nonvolatile molecules).123

Breath analyses conducted by the techniques, such as GC-MS, SIFT-MS, and PTR-MS, are still in laboratory research. The development of inexpensive, real-time, portable breath monitors is a key factor in the advancement of breath tests. Such devices should be moved from laboratory research to commercial reality. Laser-based spectroscopic instruments are a good candidate for this field. The current laser-based methods are not satisfactory for medical applications as routine diagnostics tool. Technologies in laser physics and optical engineering should be further improved. The development of absorption spectroscopy instruments with lasers may provide certain advantages for sensitive examination of not only small molecules but also heavy molecules with relatively broad spectral structures. The issue of spectral interferences can be eliminated using single-mode tunable laser. Another type of interference, which should be considered, is biointerferences; for instance, ammonia is a biomarker for both asthma and chronic kidney diseases. This challenging field requires the extensive interdisciplinary collaborations.

The availability of a cheap breath analyzer can even enable patients to measure their own biomarker levels at home. This situation can result in performing more large cross-sectional and longitudinal in-depth studies with many data points, instead of making just small pilot studies limiting the gain of widespread usage of breath analysis. One may expect that the issues mentioned in the present manuscript may be soon addressed and breath analysis may be introduced into clinical practice.


The author has no relevant financial interests and no other potential conflicts of interest to disclose.


The author thanks Yasar Kemal Erdogan for helping to compile Table 2.



L. Ciaffoni, R. Peverall and G. A. D. Ritchie, “Laser spectroscopy on volatile sulfur compounds: possibilities for breath analysis,” J. Breath Res. 5, 024002 (2011).http://dx.doi.org/10.1088/1752-7155/5/2/024002Google Scholar


C. Wang and P. Sahay, “Breath analysis using laser spectroscopic techniques: breath biomarkers, spectral fingerprints, and detection limits,” Sensors 9, 8230–8262 (2009).SNSRES0746-9462http://dx.doi.org/10.3390/s91008230Google Scholar


J. Wojtas et al., “Ultrasensitive laser spectroscopy for breath analysis,” Opto-Electron. Rev. 20, 26–39 (2012).OELREM1230-3402http://dx.doi.org/10.2478/s11772-012-0011-4Google Scholar


T. H. Risby and F. K. Tittel, “Current status of mid-infrared quantum and interband cascade lasers for clinical breath analysis,” Opt. Eng. 49, 111123 (2010).http://dx.doi.org/10.1117/1.3498768Google Scholar


Z. Bielecki et al., “Application of quantum cascade lasers to trace gas detection,” Bull. Pol. Acad. Sci. Tech. Sci. 63, 515–525 (2015).http://dx.doi.org/10.1515/bpasts-2015-0059Google Scholar


S. S. Kim et al., “Potential and challenges for mid-infrared sensors in breath diagnostics,” IEEE Sens. J. 10, 145–158 (2010).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2009.2033940Google Scholar


K. K. Chow, M. Short and H. Zeng, “A comparison of spectroscopic techniques for human breath analysis,” Biomed. Spectrosc. Imaging 1, 339–353 (2012).http://dx.doi.org/10.3233/BSI-120029Google Scholar


M. R. McCurdy et al., “Recent advances of laser-spectroscopy-based techniques for applications in breath analysis,” J. Breath Res. 1, 014001 (2007).http://dx.doi.org/10.1088/1752-7155/1/1/014001Google Scholar


J. C. Anderson and M. P. Hlastala, “Breath tests and airway gas exchange,” Pulm. Pharmacol. Ther. 20, 112–117 (2007).http://dx.doi.org/10.1016/j.pupt.2005.12.002Google Scholar


J. C. Anderson, A. L. Babb and M. P. Hlastala, “Modeling soluble gas exchange in the airways and alveoli,” Ann. Biomed. Eng. 31, 1402–1422 (2003).ABMECF0090-6964http://dx.doi.org/10.1114/1.1630600Google Scholar


L. E. Farhi, “Elimination of inert gas by the lung,” Respir. Physiol. 3, 1–11 (1967).http://dx.doi.org/10.1016/0034-5687(67)90018-7Google Scholar


J. King et al., “A mathematical model for breath gas analysis of volatile organic compounds with special emphasis on acetone,” J. Math. Biol. 63, 959–999 (2011).JMBLAJ0303-6812http://dx.doi.org/10.1007/s00285-010-0398-9Google Scholar


W. Miekisch, J. K. Schubert and G. F. E. Noeldge-Schomburg, “Diagnostic potential of breath analysis—focus on volatile organic compounds,” Clin. Chim. Acta 347, 25–39 (2004).CCATAR0009-8981http://dx.doi.org/10.1016/j.cccn.2004.04.023Google Scholar


M. E. O’Hara et al., “Development of a protocol to measure volatile organic compounds in human breath: a comparison of rebreathing and on-line single exhalations using proton transfer reaction mass spectrometry,” Physiol. Meas. 29, 309–330 (2008).PMEAE30967-3334http://dx.doi.org/10.1088/0967-3334/29/3/003Google Scholar


M. E. Webber, “Diode laser measurements of NH3 and CO2 for combustion and bioreactor applications,” Dissertation, Stanford University (2001).Google Scholar


S. Welzel, “New enhanced sensitivity infrared laser spectroscopy techniques applied to reactive plasmas and trace gas detection,” Dissertation, Ernst-Moritz-Arndt Universität Greifswald (2009).Google Scholar


F. J. Duarte, Eds., Tunable Lasers Handbook (Optics and Photonics), Eastman Kodak Company, Rochester, New York (1995).Google Scholar


M. G. Littman and H. J. Metcalf, “Spectrally narrow pulsed dye laser without beam expander,” Appl. Opt. 17, 2224–2227 (1978).APOPAI0003-6935http://dx.doi.org/10.1364/AO.17.002224Google Scholar


D. R. Scifres, W. Streifer and R. D. Burnham, “GaAs/GaAlAs diode lasers with angled pumping stripes,” IEEE J. Quantum Electron. 14, 223–227 (1978).http://dx.doi.org/10.1109/JQE.1978.1069780Google Scholar


M. C. Robbins, J. Buus and D. J. Robbins, “Analysis of antireflection coatings on angled facet semiconductor laser amplifiers,” Electron. Lett. 26, 381 (1990).ELLEAK0013-5194http://dx.doi.org/10.1049/el:19900248Google Scholar


N. K. Dutta et al., “Fabrication and performance characteristics of buried-facet optical amplifiers,” J. Appl. Phys. 67, 3943–3947 (1990).JAPIAU0021-8979http://dx.doi.org/10.1063/1.345003Google Scholar


S. Dutta, D. S. Elliott and Y. P. Chen, “Mode-hop-free tuning over 135 GHz of external cavity diode lasers without antireflection coating,” Appl. Phys. B 106, 629–633 (2012).http://dx.doi.org/10.1007/s00340-011-4841-4Google Scholar


A. T. Schremer and C. L. Tang, “External-cavity semiconductor laser with 1000 GHz continuous piezoelectric tuning range,” IEEE Photonics Technol. Lett. 2, 3–5 (1990).IPTLEL1041-1135http://dx.doi.org/10.1109/68.47023Google Scholar


F. Favre and D. Le Guen, “82 nm of continuous tunability for an external cavity semiconductor laser,” Electron. Lett. 27, 183–184 (1991).http://dx.doi.org/10.1049/el:19910117Google Scholar


L. Levin, “Mode-hop-free electro-optically tuned diode laser,” Opt. Lett. 27, 237 (2002).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.27.000237Google Scholar


Q. K. Yang et al., “Rate equations analysis of external-cavity quantum cascade lasers,” J. Appl. Phys. 107, 043109 (2010).http://dx.doi.org/10.1063/1.3309764Google Scholar


I. Bayrakli, “Frequency stabilization at the sub-kilohertz level of an external cavity diode laser,” Appl. Opt. 55, 2463–2466 (2016).Google Scholar


I. Bayrakli, “Actively frequency-stabilized external cavity diode laser with a linewidth of 2.9 kHz,” Opt. Quantum Electron. 48, 1–8 (2016).OQELDI0306-8919http://dx.doi.org/10.1007/s11082-015-0305-0Google Scholar


I. Bayrakli and H. Akman, “Ultrasensitive, real-time analysis of biomarkers in breath using tunable external cavity laser and off-axis cavity enhanced absorption spectroscopy,” J. Biomed. Opt. 20(3), 037001 (2015).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.20.3.037001Google Scholar


S. Bennetts et al., “External cavity diode lasers with 5 kHz linewidth and 200 nm tuning range at 1.55  μm and methods for linewidth measurement,” Opt. Express 22, 10642–10654 (2014).OPEXFF1094-4087http://dx.doi.org/10.1364/OE.22.010642Google Scholar


S. Tan et al., “Low-threshold, high SMSR tunable external cavity quantum cascade laser around 4.71  μm,” Opt. Quantum Electron. 45, 1147–1155 (2013).OQELDI0306-8919http://dx.doi.org/10.1007/s11082-013-9731-zGoogle Scholar


G. Wysocki et al., “Widely tunable mode-hop free external cavity quantum cascade laser for high resolution spectroscopic applications,” Appl. Phys. B 81, 769–777 (2005).http://dx.doi.org/10.1007/s00340-005-1965-4Google Scholar


T. Dougakiuchi et al., “Broadband tuning of external cavity dual-upper-state quantum-cascade lasers in continuous wave operation,” Appl. Phys Express 4, 102101 (2011).APEPC41882-0778http://dx.doi.org/10.1143/APEX.4.102101Google Scholar


R. Maulini et al., “External cavity quantum-cascade laser tunable from 8.2 to 10.4  μm using a gain element with a heterogeneous cascade,” Appl. Phys. Lett. 88, 201113 (2006).APPLAB0003-6951http://dx.doi.org/10.1063/1.2205183Google Scholar


A. Hugi et al., “External cavity quantum cascade laser tunable from 7.6 to 11.4  μm,” Appl. Phys. Lett. 95, 061103 (2009).APPLAB0003-6951http://dx.doi.org/10.1063/1.3193539Google Scholar


A. Wittmann et al., “Heterogeneous high-performance quantum-cascade laser sources for broad-band tuning,” IEEE J. Quantum Electron. 44, 11 (2008).IEJQA70018-9197http://dx.doi.org/10.1109/JQE.2008.2001928Google Scholar


R. Maulini et al., “Continuous-wave operation of a broadly tunable thermoelectrically cooled external cavity quantum-cascade laser,” Opt. Lett. 30, 2584 (2005).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.30.002584Google Scholar


H. Loh et al., “Influence of grating parameters on the linewidths of external-cavity diode lasers,” Appl. Opt. 45, 9191–9197 (2006).APOPAI0003-6935http://dx.doi.org/10.1364/AO.45.009191Google Scholar


T. Okoshi, K. Kikuchi and A. Nakayama, “Novel method for high resolution measurement of laser output spectrum,” Electron. Lett. 16, 630–631 (1980).ELLEAK0013-5194http://dx.doi.org/10.1049/el:19800437Google Scholar


G. Di Domenico, S. Schilt and P. Thomann, “Simple approach to the relation between laser frequency noise and laser line shape,” Appl. Opt. 49, 4801–4807 (2010).APOPAI0003-6935http://dx.doi.org/10.1364/AO.49.004801Google Scholar


D. S. Elliott, R. Rajarshi and S. J. Smith, “Extracavity laser band-shape and bandwidth modification,” Phys. Rev. A 26, 12–18 (1982).http://dx.doi.org/10.1103/PhysRevA.26.12Google Scholar


N. Mukherjee et al., “Linewidth measurement of external grating cavity quantum cascade laser using saturation spectroscopy,” Appl. Phys. Lett. 92, 111116 (2008).APPLAB0003-6951http://dx.doi.org/10.1063/1.2901038Google Scholar


R. W. Tkach and A. R. Chraplyvy, “Regimes of feedback effects in 1.5-μm distributed feedback laser,” J. Lightwave Technol. 4, 1655–1661 (1986).JLTEDG0733-8724http://dx.doi.org/10.1109/JLT.1986.1074666Google Scholar


P. Samutpraphoot et al., “Passive intrinsic-linewidth narrowing of ultraviolet extended-cavity diode laser by weak optical feedback,” Opt. Express 22, 11592–11599 (2014).OPEXFF1094-4087http://dx.doi.org/10.1364/OE.22.011592Google Scholar


M. Kourogi and M. Ohtsu, “Novel optical frequency discriminator for FM noise reduction of semiconductor lasers,” Opt. Commun. 81, 204–208 (1991).OPCOB80030-4018http://dx.doi.org/10.1016/0030-4018(91)90639-UGoogle Scholar


R. W. P. Drever et al., “Laser phase and frequency stabilization using an optical resonator,” Appl. Phys. B 31, 97–105 (1983).http://dx.doi.org/10.1007/BF00702605Google Scholar


Y. Zhao et al., “100-Hz linewidth diode laser with external optical feedback,” IEEE Photonics Technol. Lett. 24, 1795–1798 (2012).IPTLEL1041-1135http://dx.doi.org/10.1109/LPT.2012.2214029Google Scholar


S. Hirata et al., “Sub-hertz-linewidth diode laser stabilized to an ultralow-drift high-finesse optical cavity,” Appl. Phys. Express 7, 022705 (2014).APEPC41882-0778http://dx.doi.org/10.7567/APEX.7.022705Google Scholar


R. Maulini, “Broadly tunable mid-infrared quantum cascade lasers for spectroscopic applications,” PhD Thesis, Université de Neuchâtel (2006).Google Scholar


J. Faist et al., Semiconductors and Semimetals Quantum Cascade Laser, , H. C. Liu and F. Capasso, Eds., Vol. 66, Academic Press, San Diego (2000).Google Scholar


P. C. Kamat et al., “Measurement of acetaldehyde in exhaled breath using a laser absorption spectrometer,” Appl. Opt. 46, 3969–3975 (2007).APOPAI0003-6935http://dx.doi.org/10.1364/AO.46.003969Google Scholar


M. Sun et al., “A fully integrated standalone portable cavity ringdown breath acetone analyzer,” Rev. Sci. Instrum. 86, 095003 (2015).RSINAK0034-6748http://dx.doi.org/10.1063/1.4930121Google Scholar


L. Ciaffoni et al., “Demonstration of a mid-infrared cavity enhanced absorption spectrometer for breath acetone detection,” Anal. Chem. 85, 846–850 (2013).ANCHAM0003-2700http://dx.doi.org/10.1021/ac3031465Google Scholar


M. J. Thorpe et al., “Cavity-enhanced optical frequency comb spectroscopy: application to human breath analysis,” Opt. Express 16, 2387–2397 (2008).OPEXFF1094-4087http://dx.doi.org/10.1364/OE.16.002387Google Scholar


A. A. Kosterev and F. K. Tittel, “Ammonia detection by use of quartz-enhanced photoacoustic spectroscopy with a near-IR telecommunication diode laser,” Appl. Opt. 43, 6213–6217 (2004).APOPAI0003-6935http://dx.doi.org/10.1364/AO.43.006213Google Scholar


R. Lewicki et al., “Real time ammonia detection in exhaled human breath using a distributed feedback quantum cascade laser based sensor,” Proc. SPIE 7945, 79450K (2011).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.874887Google Scholar


E. R. Crosson et al., “Stable isotope ratios using cavity ring-down spectroscopy: determination of 13C/12C for carbon dioxide in human breath,” Anal. Chem. 74, 2003–2007 (2002).ANCHAM0003-2700http://dx.doi.org/10.1021/ac025511dGoogle Scholar


J. P. Waclawek, H. Moser and B. Lendl, “Compact quantum cascade laser based quartz-enhanced photoacoustic spectroscopy sensor system for detection of carbon disulfide,” Opt. Express 24, 6559–6571 (2016).OPEXFF1094-4087http://dx.doi.org/10.1364/OE.24.006559Google Scholar


M. Yufei et al., “Sensitive detection of carbon monoxide based on a QEPAS sensor with a 2.3  μm fiber-coupled antimonide diode laser,” J. Opt. 17, 055401 (2015).JOOPDB0150-536Xhttp://dx.doi.org/10.1088/2040-8978/17/5/055401Google Scholar


J. Wojtas, “Application of cavity enhanced absorption spectroscopy to the detection of nitric oxide, carbonyl sulphide, and ethane-breath biomarkers of serious diseases,” Sensors 15, 14356–14369 (2015).SNSRES0746-9462http://dx.doi.org/10.3390/s150614356Google Scholar


C. Li et al., “Compact TDLAS based optical sensor for ppb-level ethane detection by use of a 3.34  μm room-temperature CW interband cascade laser,” Sens. Actuators B 232, 188–194 (2016).SABCEB0925-4005http://dx.doi.org/10.1016/j.snb.2016.03.141Google Scholar


I. R. Ivascu et al., “CO2 laser photoacoustic measurements of ethanol absorption coefficients within infrared region of 9.2-10.8  μm,” Spectrochim. Acta Part A 163, 115–119 (2016).http://dx.doi.org/10.1016/j.saa.2016.03.038Google Scholar


S. Zhou, Y. Han and B. Li, “Simultaneous detection of ethanol, ether and acetone by mid-infrared cavity ring-down spectroscopy at 3.8  μm,” Appl. Phys. B 122, 187 (2016).http://dx.doi.org/10.1007/s00340-016-6465-1Google Scholar


M. Navas, A. Jiménez and A. Asuero, “Human biomarkers in breath by photoacoustic spectroscopy,” Clin. Chim. Acta 413, 1171–1178 (2012).CCATAR0009-8981http://dx.doi.org/10.1016/j.cca.2012.04.008Google Scholar


R. Cernat et al., “Laser photoacoustic spectroscopy method for measurements of trace gas concentration from human breath,” Rom. Rep. Phys. 62, 610–616 (2010).RORPED1221-1451Google Scholar


Z. Wang, J. Geng and W. Ren, “Quartz-enhanced photoacoustic spectroscopy (QEPAS) detection of the ν 7 band of ethylene at low pressure with CO2 interference analysis,” Appl. Spectrosc. 1–8 (2017).APSPA40003-7028http://dx.doi.org/10.1177/0003702817690406Google Scholar


W. Ren, L. Luo and F. K. Tittel, “Sensitive detection of formaldehyde using an interband cascade laser near 3.6  μm,” Sens. Actuators, B 221, 1062–1068 (2015).http://dx.doi.org/10.1016/j.snb.2015.07.078Google Scholar


A. Szabo et al., “In situ and wide range quantification of hydrogen sulfide in industrial gases by means of photoacoustic spectroscopy,” Meas. Sci. Technol. 24, 065501 (2013).http://dx.doi.org/10.1088/0957-0233/24/6/065501Google Scholar


S. Viciani et al., “A quartz-enhanced photoacoustic sensor for H2S trace-gas detection at 2.6  μm,” Appl. Phys. B 119, 21–27 (2015).http://dx.doi.org/10.1007/s00340-014-5991-yGoogle Scholar


P. Sahay, S. Scherrer and C. Wang, “Measurements of the weak UV absorptions of isoprene and acetone at 261–275 nm using cavity ringdown spectroscopy for evaluation of a potential portable ringdown breath analyzer,” Sensors 13, 8170–8187 (2013).SNSRES0746-9462http://dx.doi.org/10.3390/s130708170Google Scholar


J. Mikolajczyk et al., “System of optoelectronic sensors for breath analysis,” Metrol. Meas. Syst. 23, 481–489 (2016).Google Scholar


L. Dong et al., “Compact CH4 sensor system based on a continuous-wave, low power consumption, room temperature interband cascade laser,” Appl. Phys. Lett. 108, 011106 (2016).APPLAB0003-6951http://dx.doi.org/10.1063/1.4939452Google Scholar


A. A. Kosterev et al., “Cavity ringdown spectroscopic detection of nitric oxide with a continuous-wave quantum-cascade laser,” Appl. Opt. 40, 5522–5529 (2001).APOPAI0003-6935http://dx.doi.org/10.1364/AO.40.005522Google Scholar


B. W. M. Moeskops, S. M. Cristescu and F. J. M. Harren, “Sub-part-per-billion monitoring of nitric oxide by use of wavelength modulation spectroscopy in combination with a thermoelectrically cooled, continuous-wave quantum cascade laser,” Opt. Lett. 31, 823–825 (2006).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.31.000823Google Scholar


D. S. Baer et al., “Multiplexed diode-laser sensor system for simultaneous H2O, O2, and temperature measurements,” Opt. Lett. 19, 1900–1902 (1994).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.19.001900Google Scholar


R. G. Pilston and J. U. White, “A long path gas absorption cell,” J. Opt. Soc. Am. 44, 572–573 (1954).JOSAAH0030-3941http://dx.doi.org/10.1364/JOSA.44.000572Google Scholar


J. U. White, “Long optical paths of large aperture,” J. Opt. Soc. Am. 32, 285–288 (1942).JOSAAH0030-3941http://dx.doi.org/10.1364/JOSA.32.000285Google Scholar


D. Herriott, H. Kogelnik and R. Kompfner, “Off-axis paths in spherical mirror interferometers,” Appl. Opt. 3, 523–526 (1964).APOPAI0003-6935http://dx.doi.org/10.1364/AO.3.000523Google Scholar


S. M. Chernin and E. G. Barskaya, “Optical multi-pass matrix systems,” Appl. Opt. 30, 51–58 (1991).APOPAI0003-6935http://dx.doi.org/10.1364/AO.30.000051Google Scholar


J. B. McManus, P. L. Kebabian and M. S. Zahniser, “Astigmatic mirror multi-pass absorption cells for long-path length spectroscopy,” Appl. Opt. 34, 3336 (1995).APOPAI0003-6935http://dx.doi.org/10.1364/AO.34.003336Google Scholar


M. Mazurenka et al., “Cavity ring-down and cavity enhanced spectroscopy using diode lasers,” Ann. Rep. Prog. Chem. Sect. C: Phys. Chem. 101, 100–142 (2005).http://dx.doi.org/10.1039/b408909jGoogle Scholar


A. O’Keefe and D. A. G. Deacon, “Cavity ring-down optical spectrometer for absorption measurements using pulsed laser sources,” Rev. Sci. Instrum. 59, 2544–2551 (1988).RSINAK0034-6748http://dx.doi.org/10.1063/1.1139895Google Scholar


A. O’Keefe, “Integrated cavity output analysis of ultra-weak absorption,” Chem. Phys. Lett. 293, 331–336 (1998).CHPLBC0009-2614http://dx.doi.org/10.1016/S0009-2614(98)00785-4Google Scholar


R. Engeln et al., “Cavity enhanced absorption and cavity enhanced magnetic rotation spectroscopy,” Rev. Sci. Instrum. 69, 3763–3769 (1998).RSINAK0034-6748http://dx.doi.org/10.1063/1.1149176Google Scholar


D. Hofstetter et al., “Photoacoustic spectroscopy with quantum cascade distributed-feedback lasers,” Opt. Lett. 26, 887–889 (2001).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.26.000887Google Scholar


A. A. Kosterev and F. K. Tittel, “Applications of quartz tuning forks in spectroscopic gas sensing,” Rev. Sci. Instrum. 76, 043105 (2005).RSINAK0034-6748http://dx.doi.org/10.1063/1.1884196Google Scholar


A. A. Kosterev et al., “Quartz-enhanced photoacoustic spectroscopy,” Opt. Lett. 27, 1902–1904 (2002).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.27.001902Google Scholar


P. Patimisco et al., “Quartz-enhanced photoacoustic spectroscopy: a review,” Sensors 14, 6165–6206 (2014).SNSRES0746-9462http://dx.doi.org/10.3390/s140406165Google Scholar


B. de Lacy Costello et al., “A review of the volatiles from the healthy human body,” J. Breath Res. 8, 014001 (2014).http://dx.doi.org/10.1088/1752-7155/8/1/014001Google Scholar


T. Hibbard and A. J. Killard, “Breath ammonia analysis: clinical application and measurement,” Crit. Rev. Anal. Chem. 41, 21–35 (2011).CCACBB1040-8347http://dx.doi.org/10.1080/10408347.2011.521729Google Scholar


P. Gouma and M. Stanacevic, “Selective nanosensor array microsystem for exhaled breath analysis,” Procedia Eng. 25, 1557–1560 (2011).http://dx.doi.org/10.1016/j.proeng.2011.12.385Google Scholar


L. D. Bos et al., “Exhaled breath metabolomics as a noninvasive diagnostic tool for acute respiratory distress syndrome,” Eur. Respir. J. 44, 188–197 (2014).http://dx.doi.org/10.1183/09031936.00005614Google Scholar


C. Wang, A. Mbi and M. Shepherd, “A study on breath acetone in diabetic patients using a cavity ringdown breath analyzer: exploring correlations of breath acetone with blood glucose and glycohemoglobin A1C,” IEEE Sens. J. 10, 54–63 (2010).http://dx.doi.org/10.1109/JSEN.2009.2035730Google Scholar


I. Bayrakli, A. Turkmen and M. C. Kockar, “Examination of feasibility of using breath ammonia analysis based on off-axis cavity-enhanced absorption spectroscopy with external cavity diode laser for non-invasive real-time diagnosis of Helicobacter pylori,” Appl. Spectrosc. 70, 1269–1277 (2016).APSPA40003-7028http://dx.doi.org/10.1177/0003702816654052Google Scholar


I. Bayrakli et al., “Applications of external cavity diode laser-based technique to non-invasive clinical diagnosis using expired breath ammonia analysis: chronic kidney disease, epilepsy,” J. Biomed. Opt. 21, 087004 (2016).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.21.8.087004Google Scholar


A. Amano et al., “Monitoring ammonia to assess halitosis,” Oral Surg. Oral Med. Oral Pathol. 94, 692–696 (2002).http://dx.doi.org/10.1067/moe.2002.126911Google Scholar


G. Song et al., “Quantitative breath analysis of volatile organic compounds of lung cancer patients,” Lung Cancer 67, 227–231 (2010).http://dx.doi.org/10.1016/j.lungcan.2009.03.029Google Scholar


R. H. Eggers et al., “A methodological analysis of the C13-urea breath test for detection of Helicobacter pylori infections: high sensitivity and specificity within 30 min using 75 mg of C13-urea,” Eur. J. Gastroenterol. Hepatol. 2, 437–444 (1990).EJGHES0954-691XGoogle Scholar


M. Phillips, “Detection of carbon disulfide in breath and air: a possible new risk factor for coronary artery disease,” Int. Arch. Occup. Environ. Health 64, 119–123 (1992).IAEHDWhttp://dx.doi.org/10.1007/BF00381479Google Scholar


M. A. Kamboures et al., “Breath sulfides and pulmonary function in cystic fibrosis,” Proc. Natl Acad. Sci. U. S. A. 102, 15762–15767 (2005).http://dx.doi.org/10.1073/pnas.0507263102Google Scholar


K. Zayasu et al., “Increased carbon monoxide in exhaled air of asthmatic patients,” Am. J. Respir. Crit. Care Med. 156, 1140–1143 (1997).AJCMED1073-449Xhttp://dx.doi.org/10.1164/ajrccm.156.4.96-08056Google Scholar


M. Yamaya et al., “Increased carbon monoxide in exhaled air of subjects with upper respiratory tract infections,” Am. J. Respir. Crit. Care Med. 158, 311–314 (1998).AJCMED1073-449Xhttp://dx.doi.org/10.1164/ajrccm.158.1.9711066Google Scholar


S. S. Sehnert et al., “Breath biomarkers for detection of human liver diseases: preliminary study,” Biomarkers 7, 174–187 (2002).http://dx.doi.org/10.1080/13547500110118184Google Scholar


K. A. Cope et al., “Abnormal exhaled ethane concentrations in scleroderma,” Biomarkers 11, 70–84 (2006).http://dx.doi.org/10.1080/13547500500515046Google Scholar


P. Paredi, S. A. Kharitonov and P. J. Barnes, “Elevation of exhaled ethane concentration in asthma,” Am. J. Respir. Crit. Care Med. 162, 1450–1454 (2000).AJCMED1073-449Xhttp://dx.doi.org/10.1164/ajrccm.162.4.2003064Google Scholar


H. Amal et al., “Breath testing as potential colorectal cancer screening tool,” Int. J. Cancer 138, 229–236 (2016).http://dx.doi.org/10.1002/ijc.29701Google Scholar


C. Popa, M. Petrus and A. M. Bratu, “Ammonia and ethylene biomarkers in the respiration of the people with schizophrenia using photoacoustic spectroscopy,” J. Biomed. Opt. 20, 057006 (2015).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.20.5.057006Google Scholar


C. Popa et al., “The level of ethylene biomarker in the renal failure of elderly patients analyzed by photoacoustic spectroscopy,” Laser Phys. 23, 125701 (2013).LAPHEJ1054-660Xhttp://dx.doi.org/10.1088/1054-660X/23/12/125701Google Scholar


A. Wehinger et al., “Lung cancer detection by proton transfer reaction mass-spectrometric analysis of human breath gas,” Int. J. Mass Spectrom. 265, 49–59 (2007).IMSPF81387-3806http://dx.doi.org/10.1016/j.ijms.2007.05.012Google Scholar


F. J. Gilchrist et al., “Hydrogen cyanide concentrations in the breath of adult cystic fibrosis patients with and without Pseudomonas aeruginosa infection,” J. Breath Res. 7, 026010 (2013).http://dx.doi.org/10.1088/1752-7155/7/2/026010Google Scholar


M. Whiteman et al., “The novel neuromodulator hydrogen sulfide: an endogenous peroxynitrite ‘scavenger’?” J. Neurochem. 90, 765–768 (2004).JONRA90022-3042http://dx.doi.org/10.1111/j.1471-4159.2004.02617.xGoogle Scholar


T. Karl et al., “Human breath isoprene and its relation to blood cholesterol levels: new measurements and modeling,” J. Appl. Physiol. 91, 762–770 (2001).Google Scholar


L. Le Marchand et al., “Use of breath hydrogen and methane as markers of colonic fermentation in epidemiologic studies: circadian patterns of excretion,” Environ. Health Perspect. 98, 199–202 (1992).EVHPAZ0091-6765http://dx.doi.org/10.1289/ehp.9298199Google Scholar


W. Filipiak et al., “Characterization of volatile metabolites taken up by or released from Streptococcus pneumoniae and Haemophilus influenzae by using GC-MS,” Microbiology 158, 3044–3053 (2012).MIBLAO0026-2617http://dx.doi.org/10.1099/mic.0.062687-0Google Scholar


R. F. del Rio et al., “Volatile biomarkers in breath associated with liver cirrhosis—comparisons of pre- and post-liver transplant breath samples,” EBioMedicine 2, 1243–1250 (2015).http://dx.doi.org/10.1016/j.ebiom.2015.07.027Google Scholar


M. Hongying et al., “Analysis of human breath samples of lung cancer patients and healthy controls with solid-phase microextraction (SPME) and flow-modulated comprehensive two-dimensional gas chromatography (GC × GC),” Anal. Methods 6, 6841–6849 (2014).AMNEGX1759-9679http://dx.doi.org/10.1039/C4AY01220HGoogle Scholar


B. J. Novak et al., “Exhaled methyl nitrate as a noninvasive marker of hyperglycemia in type 1 diabetes,” Proc. Natl. Acad. Sci. U. S. A. 104, 15613–15618 (2007).http://dx.doi.org/10.1073/pnas.0706533104Google Scholar


C. O. Olopade et al., “Exhaled pentane and nitric oxide levels in patients with obstructive sleep apnea,” Chest 111, 1500–1504 (1997).CHETBF0012-3692http://dx.doi.org/10.1378/chest.111.6.1500Google Scholar


S. A. Kharitonov and P. J. Barnes, “Nitric oxide, nitrotyrosine, and nitric oxide modulators in asthma and chronic obstructive pulmonary disease,” Curr. Allergy Asthma Rep. 3, 121–129 (2003).http://dx.doi.org/10.1007/s11882-003-0024-7Google Scholar


S. A. Kharitonov and P. J. Barnes, “Nitric oxide in exhaled air is a new marker of airway inflammation,” Monaldi. Arch. Chest Dis. 51, 533–537 (1996).Google Scholar


M. A. Birrell et al., “Utility of exhaled nitric oxide as a noninvasive biomarker of lung inflammation in a disease model,” Eur. Respir. J. 28, 1236–1244 (2006).http://dx.doi.org/10.1183/09031936.00048506Google Scholar


J. Huh, D. Yi and B. Gam, Korea Traditional Medicine Book, Nae Yi Won, Seoul, Korea (1613).Google Scholar


T. H. Risby and S. F. Solga, “Current status of clinical breath analysis,” Appl. Phys. B 85, 421–426 (2006).http://dx.doi.org/10.1007/s00340-006-2280-4Google Scholar


W. Cao and Y. Duan, “Current status of methods and techniques for breath analysis,” Crit. Rev. Anal. Chem. 37, 3–13 (2007).CCACBB1040-8347http://dx.doi.org/10.1080/10408340600976499Google Scholar


W. Miekisch, J. Herbig and J. K. Schubert, “Data interpretation in breath biomarker research: pitfalls and directions,” J. Breath Res. 6, 036007 (2012).http://dx.doi.org/10.1088/1752-7155/6/3/036007Google Scholar


M. Phillips, “Method for he collection and assay of volatile organic compounds in breath,” Anal. Biochem. 247, 272–278 (1997).http://dx.doi.org/10.1006/abio.1997.2069Google Scholar


T. H. Risby and S. S. Sehnert, “Clinical application of breath biomarkers of oxidative stress status,” Free Radical Biol. Med. 27, 1182–1192 (1999).http://dx.doi.org/10.1016/S0891-5849(99)00212-9Google Scholar


A. T. Society, “Recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide in adults and children-1999,” Am. J. Respir. Crit. Care Med. 160, 2104–2117 (1999).AJCMED1073-449Xhttp://dx.doi.org/10.1164/ajrccm.160.6.ats8-99Google Scholar


A. T. Society, “ATS/ERS Recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, 2005,” Am. J. Respir. Crit. Care Med. 171, 912–930 (2005).AJCMED1073-449Xhttp://dx.doi.org/10.1164/rccm.200406-710STGoogle Scholar


B. Buszewski et al., “Human exhaled air analytics: biomarkers of diseases,” Biomed. Chromatogr. 21, 553–566 (2007).BICHE20269-3879http://dx.doi.org/10.1002/(ISSN)1099-0801Google Scholar


R. Schubert et al., “Metabolic monitoring and assessment of anaerobic threshold by means of breath biomarkers,” Metabolomics 8, 1069–1080 (2012).http://dx.doi.org/10.1007/s11306-012-0408-6Google Scholar


Ismail Bayrakli is an associate professor at Suleyman Demirel University. He received his MSc degree in 2005 and his PhD in physics in 2008 from Technical University of Berlin and Humboldt University, Berlin, Germany. His research interests include tunable single- and double-mode diode lasers, ultra-sensitive and ultra-resolution laser-based biomedical sensors, and breath analysis.

Ismail Bayrakli, "Breath analysis using external cavity diode lasers: a review," Journal of Biomedical Optics 22(4), 040901 (17 April 2017). http://dx.doi.org/10.1117/1.JBO.22.4.040901
Submission: Received 17 December 2016; Accepted 28 March 2017


Semiconductor lasers


Biological research




Back to Top