Reversible switching of quantum cascade lasermodes using a pH-responsive polymeric cladding as transducer Bernhard Basnar,1,* Stephan Schartner,1 Maximilian Austerer,1 Aaron Maxwell Andrews,1 Thomas Roch,2 Werner Schrenk,1 and Gottfried Strasser1,3 1
Center for Micro- and Nanostructures, Vienna University of Technology, Floragasse 7, 1040 Vienna, Austria 2 Department of Experimental Physics, Comenius University, 84248 Bratislava, Slovakia 3 The State University of New York at Buffalo, 332 Bonner Hall, Buffalo, NY 14260-1920, USA * Corresponding author:
[email protected]
Abstract: We present a novel approach for the reversible switching of the emission wavelength of a quantum cascade laser (QCL) using a halochromic cladding. An air-waveguide laser ridge is coated with a thin layer of polyacrylic acid. This cladding introduces losses corresponding to the absorption spectrum of the polymer. By changing the state of the polymer, the absorption spectrum and losses change, inducing a shift of 7 cm-1 in the emission wavelength. This change is induced by exposure to acidic or alkaline vapors under ambient conditions and is fully reversible. Such lasers can be used as multi-color light source and as sensor for atmospheric pH. ©2008 Optical Society of America OCIS codes: (140.5965) Semiconductor lasers, quantum cascade; (160.5470) Polymers; (300.6340), Spectroscopy, infrared; (310,3915) Metallic, opaque, and absorbing layers; (310.4925) Other properties (stress, chemical, etc.).
References and links 1. 2. 3. 4. 5. 6. 7.
8. 9.
10.
11.
D. L. Woolard, E. R. Brown, M. Pepper, and M. Kemp, “Terahertz frequency sensing and imaging: a time of reckoning future applications?” Proc. IEEE 93, 1722-1743 (2005). J. Faist, F. Capasso, D. L. Sivco, C. Sitori, A. L. Hutchinson, and A. Y. Cho, “Quantum cascade laser,” Science 264, 553-556 (1994). C. Gmachl, D. L. Sivco, R. Colombelli, F. Capasso, and A. Y. Cho, “Ultra-broadband semiconductor laser,” Nature 415, 883-887(2002). D. Barate, R. Teissier, Y. Wang, and A. N. Baranov, “Short wavelength intersubband emission from InAs/AlSb quantum cascade structures,” Appl. Phys. Lett. 87, 051103 (2005). J. Ulrich, J. Kreuter, W. Schrenk, G. Strasser, and K. Unterrainer, “Long wavelength (15 and 23 µm) GaAs/AlGaAs quantum cascade lasers,” Appl. Phys. Lett. 80, 3691-3693 (2002). B. S. Williams, “Terahertz quantum-cascade lasers,” Nat. Photon. 1, 517-525 (2007). M. A. Belkin, F. Capasso, A. Belyanin, D. L. Sivco, A. Y. Cho, D. C. Oakley, C. J. Vineis, and G. W. Turner, “Terahertz quantum-cascade-laser source based on intracavity difference-frequency generation,” Nat. Photon. 1, 288-292 (2007). J. Faist, C. Gmachl, F. Capasso, C. Sirtori, C. L. Sivco, J. N. Baillargeon, and A. Y. Cho, “Distributed feedback quantum cascade lasers,” Appl. Phys. Lett. 70, 2670-2672 (1997) . C. Pflügl, M. Austerer, W. Schrenk, S. Golka, G. Strasser, R. P. Green, L. R. Wilson, J. W. Cockburn, A. B. Krysa, and J. S. Roberts, “Single-mode surface-emitting quantum-cascade lasers,” Appl. Phys. Lett. 86, 211102 (2005). R. Colombelli, K. Srinivasan, M. Troccoli, O. Painter, C. F. Gmachl, D. M. Tennant, A. M. Sergent, D. L. Sivco, and A. Y. Cho, F. Capasso, “Quantum Cascade Surface-Emitting Photonic Crystal Laser,” Science 302, 1374-1377 (2003). C. Pflügl, W. Schrenk, S. Anders, and G. Strasser, “Spectral dynamics of distributed feedback quantum cascade lasers,” Semicond. Sci. Technol. 19, S336-S338 (2004).
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8557
12.
13.
14.
15. 16.
17. 18. 19.
20. 21. 22. 23.
24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
S. Song, S. S. Howard, Z. Liu, A. O. Dirisu, C. F. Gmachl, and C. B. Arnold, “Mode tuning of quantum cascade lasers through optical processing of chalcogenide glass claddings,” Appl. Phys. Lett. 89, 041115 (2006). L. Diehl, B. Lee, P. Behroozi, M. Lončar, M. Belkin, F. Capasso, T. Aellen, D. Hofstetter, M. Beck, and J. Faist, “Microfluidic tuning of distributed feedback quantum cascade lasers,” Opt. Express 14, 1166011667 (2006). M. Lončar, B. G. Lee, L. Diehl, M. A. Belkin, F. Capasso, M. Giovannini, J. Faist, and E. Gini, “Design and fabrication of photonic crystal quantum cascade lasers for optofluidics,” Opt. Express 15, 4499-4514 (2007). M. Sarangan, W. P. Huang, G. P. Li, and T. Makino, “A ridge waveguide DFB laser model including transverse carrier and optical effects,” IEEE J. Quantum Electron. 32, 408-416 (1996). M. A. Belkin, M. Lončar, B. G. Lee, C. Pflügl, R. Audet, L. Diehl, F. Capasso, D. Bour, S. Corzine, and G. Höfler, “Intra-cavity absorption spectroscopy with narrow-ridge microfluidic quantum cascade lasers,” Opt. Express 15, 11262-11271 (2007). T. Bae, D. A. Scherson, and E. B. Yeager, “Infrared spectroscopic determination of pH changes in diffusionally decoupled thin-layer electrochemical cells,” Anal. Chem. 62, 45-49 (1990). Schenk, I. W. Marison, and U. von Stockar, “pH prediction and control in bioprocesses using mid-infrared spectroscopy,” Biotechnol. Bioeng. 100, 82-83 (2008). R. P. Green, L. R. Wilson, E. A. Zibik, D. G. Revin, J. W. Cockburn, C. Pflügl, W. Schrenk, G. Strasser, A. B. Krysa, J. S. Roberts, C. M. Tey, and A. G. Cullis, “High-performance distributed feedback quantum cascade lasers grown by metalorganic vapor phase epitaxy,” Appl. Phys. Lett. 85, 5529-5531 (2004). D. Hofstetter, J. Faist, M. Beck, and U. Oesterle, “Surface-emitting 10.1 µ m quantum-cascade distributed feedback lasers,” Appl. Phys. Lett. 75, 3769-3771 (1999). V. Moreau, M. Bahriz, R. Colombelli, P.-A. Lamoine, Y. DeWilde, and L. R. Wilson, “Direct imaging of a laser mode via midinfrared near-field microscopy,” Appl. Phys. Lett. 90, 201114 (2007). D. Sarkar and P. Somasundaran, “Conformatinal dynamic of poly(acrylic acid). A study using surface plasmon resonance spectroscopy,” Langmuir 20, 4657-4664 (2004). H. Hoffmann, M. L. T. Liveri, and F. P. Cavasino, “Electric birefringence and zero-shear vixcosity studies of the conformational changes of polyacrylic and polymethacrylic acids with pH and concentration,” J. Chem. Soc. Faraday Trans. 93, 3161-3165 (1997). http://webbook.nist.gov/chemistry/. F. Mohammad, M. Z. A. Rahman, and A. Kassim, “An experimental study on diffusion in electrically conducting polymers,” Chem. Environ. Res. 10, 41-46 (2001). P. Musto, L. Mascia, G. Mensitieri, and G. Ragosta, “Diffusion of water and ammonia through poyimidesilica bicontinuous nanocomposites: interactions and reactions” Polymer 46, 4492-4503 (2005). D. O. Hummel and F. Scholl, Atlas der Polymer- und Kunststoffanalyse Band 2, Kapitel 5.4. (Carl Hanser Verlag, München & VCH, Weinheim, 1988). F. L. Buchholz and N. A. Peppas, “Superabsorbent polymers: science and technology” ACS Symp. Ser. 573 (1994). J. Buchanan, B. Hird, and T. M. Letcher, “Primary hydration of poly(acrylic acid) sodium salts” Polym. Bull. 16, 493-498 (1985). S. Yang, Y. Zhang, Y. Guan, S. Tan, J. Xu, S. Cheng, and X. Zhang, “Water uptake behavior of hydrogenbonded PVPON-PAA LBL film,” Soft Matter 2, 699-704 (2006). Oprea and U. Weimar, “High sensitivity polyacrylic acid films for ammonia detection with field effect devices,” Sens. Actuat. B 111-112, 572-576 (2005). Sahm, A. Oprea, N. Bârsan, and U. Weimar, “Water and ammonia influence on the conduction mechanisms in polyacrylic acid films,” Sens. Actuat. B 127, 204-209 (2007). K. Dioumaev, “Infrared methods for monitoring the protonation state of carboxylic amino acids in the photocycle of bacteriorhodopsin,” Biochemistry (Moscow) 66, 1269-1276 (2001). Saito, T. Gojo, K. Yato, and M. Miyagi, “Optical constants of polymer coatings in the infrared,” Infrared Phys. Technol. 36, 1125-1129 (1995). Y. Wang, Y. Abe, Y. Matsuura, M. Miyagi, and H. Uyama, “Refractive indices and extinction coefficients of polymers for the mid-infrared region,” Appl. Opt. 37, 7091-7095 (1998)
1. Introduction Spectroscopy in the mid-infrared (mid-IR) and terahertz wavelength regime is of high importance in environmental analysis (trace gases, pollutants) and organic and pharmaceutical synthesis (compound characterization and quality control), and is a promising candidate for future high security application (airport security, hazardous chemicals identification) [1]. Quantum Cascade lasers (QCLs) are well established high power light sources in this wavelength regime [2-7]. Their gain - and hence the emission - spectrum can be designed to #95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8558
cover almost every desired wavelength in this region. Within the gain spectrum a distinct selection of the wavelength is possible via the laser cavity. Shallow DFB structures along the rib waveguide [8,9], deep etched PhC or DBR structures [10] as well as very small cavities are typical realizations that allow the selection of one single mode. A tunability of this single mode was shown via the effective refractive index neff. In case of a DFB laser the emitted free space wavelength λ is given by λ = 2Λneff (1) where Λ is the period of the grating. A temperature tuning of approximately −0.05 cm-1/K caused by the self heating of the laser in pulsed operation was used to resolve individual absorption lines of ammonia isotopes [11]. Another concept is the application of cladding materials of different refractive indices on top of a DFB grating. Only a small fraction of the light experiences the change in refractive index causing a small shift in the emitted laser line. Song et al deposited a chalcogenide cladding made from As2S3 onto a DBR grating to shift the emission by 3.8 cm-1 to longer wavelengths due to a change in the effective refractive index [12]. Subsequent illumination with UV light led to an additional shift of 1.31 cm-1 to longer wavelengths. These modifications are irreversible optically. However, heating of the cladding to a temperature above the glass transition temperature (185°C) or removal of the cladding material and recoating can be utilized to restore the refractive index, and thus the wavelength, to its original state. Utilizing the differences of the refractive indexes of various solvents, mode tuning of a DFB-QCL in a microfluidic set-up was reported by Diehl et al [13]. They achieved a variation of up to 1.15 cm-1 as compared to air by using solvents with refractive indices between 1.3 and 1.735. The same group also published a theoretical paper [14] estimating this effect to increase about 10-fold using photonic crystal QCLs. In both cases the changes are fully reversible by simply exchanging the solvent in the microfluidic channels. The introduction of losses can also be used to determine the wavelength of a laser. The first approach is the fabrication of an absorbing, periodic grating to yield a loss-coupled DFB structure [15]. This is similar to the above-mentioned methods as the wavelength is determined by the grating period. However, a second approach uses the intrinsic absorbance features of chemical compounds to introduce losses that vary with the wavelength. No grating is needed as the emission will take place at the wavelength exhibiting the least losses as compared to the gain of the laser. This behavior was found in an intra-cavity absorption spectroscopy (ICAS) experiment where different solvents caused changes in the threshold and in the wavelength of the emitted light [16]. Whilst such ICAS experiments utilize the absorbance of the analyte, we present a new paradigm harnessing the wavelength-specific losses of an organic cladding material which acts as a transducer. To this end, we use an organic compound that contains carboxylic acid functional groups. Such compounds have already successfully been used to determine the pH in different solutions by infrared absorption spectroscopy utilizing the spectral changes upon dissociation of the acid functionality [17,18]. We demonstrate the successful switching of the laser wavelength in ambient conditions upon exposure of a polyacrylic acid-cladding to the vapors of hydrochloric acid or ammonia, leading to a reversible shift of 7 cm-1 in the emission wavelength. We also compare the results with waveguide-calculations and confirm the good agreement of our results with theoretical considerations 2. Experimental 2.1. Chemicals Polyacrylic acid (medium viscosity, Fluka), hydrochloric acid (37-38%, J.T. Baker), potassium hydroxide (J.T. Baker), and ammonium hydroxide (28-38% ammonia in water, Acros organics) were used as purchased.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8559
2.2. Laser fabrication The active region of the QCL was grown by solid source molecular beam epitaxy and is made up by an In0.53Ga0.47As/In0.52Al0.48As 4-well design similar to that presented in [19]. Starting from the highly doped substrate (2×1018 cm-3) a 2.2 µm low doped (1×1017 cm-3) InP layer serves as a dielectric lower cladding, followed by a 100 nm InGaAs buffer layer (1×1017 cm-3). On top of that, 35 active stages are grown, forming a 1.84 µm thick active region, followed by another 600 nm InGaAs Buffer (1×1017 cm-3) and a 10 nm thick, highly doped (7.2×1018 cm-3) contact facilitating InGaAs layer. The upper part of the waveguide is realized as an air cladding in order to provide an interaction area with the polymer. Air windows are therefore left open on top of the laser ridges and the devices are contacted via extended contact pads with 2 µm broad lateral contact stripes on either sides along the ridges [20]. The contact layers are Ti/Au (10/500 nm) sputter deposited on a 300 nm SiNx insulation layer. 2.3. Spin-coating 0.15g of polyacrylic acid (PAA) were dissolved in 10 ml of 0.2 M KOH under heating. This stock solution was spin-coated at 2000 rpm for 35 s onto silicon wafer pieces (approx 1 x 1 cm2) for the transmission FTIR measurements, yielding on average a 2 µm thick film. For spin-coating of the laser bars, the solution was diluted 1:4 with distilled water. The spincoating was, then, carried out at 3000 rpm for 35 s, yielding an average layer thickness of approximately 500 nm on the laser bar, comparable to the extension of the evanescent field of the waveguide mode. Thickness of the polymeric layer was measured using reflectometry (F20, Filmetrics, USA) for large size samples and AFM (Dimension3100, Veeco-DI, USA) for the laser bar samples. As the facets of the laser can be covered with the polymer during the spin-coating procedure, a cleaning step was performed. For this purpose, a few drops of a diluted potassium hydroxide solution were put on a soft cleanroom-tissue and the laser facets were gently moved across the wet portion of the paper. The potassium hydroxide leads to a chemical dissolution of the PAA without chemically corroding the facets. 2.4. FTIR transmission measurements FTIR measurements were performed using a BRUKER Equinox-55 in free running mode with an internal deuterated triglycine sulfate (DTGS) detector. Standard transmission measurements were performed at 4 kHz with 0.2 cm-1 resolution and 3 scans were averaged for each spectrum. Rapid time resolved transmission measurements were undertaken at 10 kHz with 2 cm-1 resolution. Each spectrum was measured by a single scan, yielding a timeresolution of approx 1.5 s. QCL emission spectra were recorded at 4 kHz with 0.2 cm-1 resolution and 3 scans were averaged for each spectrum. The rapid time resolved emission spectra were recorded at 20 kHz with 1 cm-1 resolution, where each spectrum was measured by a single scan, only. The time resolution was 2 s. 2.5. Switching experiments The switching of the polymer between its acid and salt state was achieved by flushing the polymer with a constant stream of air, saturated with HCl or ammonia vapors, at 30 ccm/min. 2.6. LIV measurements The samples were indium soldered on copper submounts, wire bonded and fixed on a peltiercooler stabilized room temperature set up. The lasers are driven with 100 ns pulses at a repetition rate of 5 kHz. The output power was measured with the DTGS detector of the BRUKER Equinox-55 fourier-transform infrared spectrometer.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8560
3. Results and discussion We used a QCL with InAlAs/InGaAs active region grown lattice-matched on an InP substrate with an emission wavelength around 1300 cm-1. The processing of the laser was performed in such a way as to obtain an air-waveguide on top of the laser ridge (see Fig. 1), as was described by Hofstetter et al [20]. Such an air-waveguide enables an interaction of the small, evanescent part of the laser mode with the polymer deposited on top. This is due to the fact that the evanescent wave can exceed the solid-state boundary of the laser [21] and penetrate into the cladding material.
Fig. 1. Basic design of the PAA-coated QCLs
The air-waveguide was coated with a thin film of polyacrylic acid (PAA). This polymer exhibits a very pronounced IR absorption spectrum as is shown in Fig. 2. The top spectrum of Fig. 2a depicts PAA in its free acid state and the bottom one in its potassium salt state. Both spectra are dominated by as strong absorption, in one case at 1700 cm-1 in the second 1550 cm-1, which corresponds to the carbonyl and the carboxylate bond, respectively. In the area of the laser emission, i.e. around 1300 cm-1, both states exhibit a slope in absorption associated to absorption bands at 1250 and 1350 cm-1, respectively (Fig. 2b).
b)
a)
Fig. 2. (a) Absorbance FTIR spectra of PAA in its acid (red, dashed curve) and potassium salt state (blue, solid curve) measured in transmission on a 2 µ m thick film of PAA on silicon. (b) Close-up of the absorbance spectrum in the range between 1200 and 1400 cm-1. The gray rectangle marks the emission wavenumber range of the QCLs.
These spectral changes are accompanied by conformational changes [22,23]. In the free acid state the carboxylic acid moieties are capable of forming hydrogen bonds, causing a dense packing of the polymer chains. In the salt state, however, the carboxylate groups experience electrostatic repulsion due to the negative charges, causing a looser packing. This behavior is most pronounced in solutions with low ionic strength, whereas in the solid state this effect is less pronounced.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8561
3.1. Spectral changes In order to assess the influence of the polymeric cladding on the emission behavior of the QCL, we fabricated lasers with four different ridge widths (28, 38, 48, and 58 µm). After that, the waveguides were coated with a 500 nm thick film of PAA in its potassium salt state. Conversion to the free acid state was achieved by exposure to the vapors of hydrochloric acid. Figure 3 shows the emission spectra of the lasers before and after coating at 288 K. All lasers show slightly different Fabry-Perot spectra in their native state (black, dotted curves). Deposition of the organic cladding causes a narrowing of the laser bandwidth. All lasers show emission only between 1290 and 1303 cm-1, irregardless of the native spectrum of the laser. This is in good agreement with the absorption minimum of the PAA potassium salt. By subjecting the polymer cladding to the vapors of hydrochloric acid, the PAA potassium salt is transformed into its respective free acid state. As with the PAA potassium salt-coated lasers, a narrowing of the emission is observed. However, due to the high absorbance for wavenumbers below 1300 cm-1, the emission window is now between 1303 and 1320 cm-1.
Fig. 3. Emission spectra of QCLs with different ridge width, ranging from 28 to 58 µ m. Dotted black lines represent the native QCLs, solid blue lines represents the PAA/potassium salt coated QCLs, and dashed red lines represent the PAA in its acid-state. All lasers are 2 mm long and are operated at room temperature. The thickness of the PAA coating is 500 nm.
It is important to note that this reduction in bandwidth is not caused by simple absorption of the emitted laser light. Instead it is caused by selective light amplification of those wavelengths that show the least losses compared to the gain at that specific wavelength. This causes damping of less favored wavelengths and enhancement of preferred wavelengths. The laser with the 58 µm ridge width demonstrates this behavior most clearly. Before coating the laser emits light with wavenumbers between 1295 and 1306 cm-1. After coating with the potassium salt of PAA the emission is shifted to wavenumbers between 1290 and 1304 cm-1. However, conversion to the free acid state with HCl vapors leads to an emission #95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8562
between 1305 and 1316 cm-1, a wavelength domain where neither the native laser nor the PAA potassium salt coated laser showed any emission. Although this behavior is most pronounced for the laser with 58 µm ridge width, all of the lasers show a higher energy output in the acid state at their peak wavelength as compared to the output power at that wavelength before coating, confirming the assumption of selective light amplification. Another important fact is that the change in emission is not caused by the absorbance of the ammonia or hydrochloric acid vapors as these are present in different spectral regions [24]. Instead, this system is sensitive to the non-optical property of pH by converting the pH information into a spectral information using the polymer as transducer. 3.2. Switching experiments The spectral changes of the PAA during the conversion from one state to the other was followed by time-resolved FTIR spectroscopy. Figure 4 shows the transmissivity at the main absorption bands (1550 and 1725 cm-1) as a function of time. The switch from salt to acid state is depicted in Fig. 4a, where after 2 seconds exposure to HCl vapors was initiated and maintained for 18 seconds. As can be seen, the conversion is completed after approximately 3 seconds of exposure (5 seconds on the time scale).
Fig. 4. Time-resolved transmission FTIR-characterization of the pH-switching behavior of a 2 µm film of PAA at 1725 cm-1 and 1550 cm-1. (a) Change of the transmissivity of the salt-state PAA during a 20 s exposure to HCl vapours. (b) Change of the transmissivity of the acid-state PAA during a 20 s exposure to NH3 vapors. (c) Repeated switching between acid- and salt-state PAA every 15s. The related spectra were recorded every 1.5s with a spectral resolution of 2 cm-1 at room temperature.
For the back-conversion (Fig. 4(b), exposure to ammonia vapor initiated after 3 s), slower kinetics are observed, causing the switching process to extend over a period of 12 seconds. This can be explained by the relatively small diffusion coefficient of ammonia in polymers [25,26]. #95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8563
By subsequent exposure of the polymeric cladding to the vapors of hydrochloric acid and of ammonia, the cladding can be reversibly switched between its acid and its salt state. This is shown in Fig. 4(c), where every 15 seconds the flushing gas was switched. Repeatable switching with a switch rate of 4 conversions per minute was observed. Subjecting a laser with its cladding in the free acid state to the vapors of ammonia allows determination of the time constant for a single switching event as well as the spectral changes during the transition. Figure 5 shows the emission spectra of the 38 µm ridge laser as a function of time. Flushing with ammonia vapor was initiated after 5 s and continued for the length of the experiment. As can be seen, the laser switches between two different wavelengths, one around 1307 cm-1 corresponding to the acid state of the polymer and one around 1300 cm-1 corresponding to the salt state of the polymer, on the time scale of a few seconds. In the transition state, both wavelengths are emitted, with the one at 1307 cm-1 decreasing over time and the one at 1300 cm-1 simultaneously increasing. The requirement of a sufficiently high temporal resolution of 2 seconds per spectrum necessitated a reduction in the spectral resolution to 0.5 cm-1 and an increase in the scan speed of the FTIR. The resulting interference between the pulse frequency of the QCL and the scan frequency of the FTIR caused variations in the absolute values of the emitted power from spectrum to spectrum. In order to obtain high-resolution spectra of the switching event without the abovementioned instrumental artifacts, the conversion time needed to be increased. This was achieved utilizing the tendency of ammonia to effuse from the polymer matrix. This effusion causes the polymer to be converted back into its acid state on a time scale of half an hour. Due to this, high resolution spectra at low scan speeds could be recorded. Figure 5b shows the emission spectra of the same QCL for this switching from the salt to the acid state of the polymer. The emission follows the opposite trend as compared to the switching from acid to salt, i.e. the peak at around 1300 cm-1 starts to disappear whilst simultaneously a peak at around 1307 cm-1 appears. It is important to note that, although the peak associated with the salt state already disappeared after 20 min, the peak for the acid state still undergoes changes. This can be explained by the changes in the absorption spectrum of the polymer and the loss-induced lasing mechanism. The effusion of the ammonia causes a reduction in the absorption band at 1345 cm-1 (and, thus, in the losses around 1310 cm-1) and a concomitant increase in absorption at 1260 cm-1 which causes an increase in losses around 1300 cm-1. After about 20 minutes the losses around 1300 cm-1 are already so high so as to inhibit emission at this wavelength and any further increase in absorbance cannot be followed. However, the conversion is not completed and further effusion of the ammonia induces a further reduction in the absorption band at 1345 cm-1, influencing the emission peak at 1307 cm-1. Taking a close look at the emission spectra of the laser in its potassium salt and its ammonium salt state, distinct differences can be observed. Whilst the potassium salt coated laser exhibits its main emission peak at 1300 cm-1 and another emission at 1293 cm-1, the ammonium salt state is devoid of that second emission. Similarly, in its acid state the initial spectrum (obtained after exposure of the potassium salt to the vapors of hydrochloric acid) shows two emission peaks (1308 cm-1 and 1316 cm-1) whilst the acid state obtained from the ammonia salt of the polymer after hydrochloric acid gases lacks the smaller emission peak at 1316 cm-1. This indicates the sensitivity of this method to different molecules. The absorption of the PAA is attributed to the carbonyl (C=O) or carboxylate (COO−) bond for the acid or salt state, respectively. However, the nature of counter-ions influences the bond-strength of the carbon-oxygen bond and, thus, causes slight changes to the exact position and shape of the absorption band [27]. This, in return, can cause noticeable differences in the emission spectrum due to the high sensitivity of the loss-induced wavelength selection.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8564
Fig. 5. Time-resolved emission-characterization of the pH-switching behavior of the 38 µ m wide QCL. (a) Change in the emission spectrum of the acid-state laser upon exposure to ammonia vapors. Spectral resolution is 0.5 cm-1. Flushing of the laser with ammonia vapors was initiated after 4 s. (b) Change in the emission spectrum of the ammonium salt-state PAA due to the effusion of ammonia from the polymer. (c) Repeated switching between acid- and salt-state PAA by exposure to ammonia vapor (switch to even numbers) and effusion of ammonia from the polymer or exposure to HCl vapors (switch to odd numbers). Numbers represent the centre of a gaussian fit to the emission spectra peak. All measurements were performed at room temperature.
This switching can be repeated several times, leading to the reversible switching between the two characteristic wavelengths (1300 cm-1 and 1307 cm-1), as is depicted in Fig. 5(c). In this context it is also important to address the issue of long-term stability of the polymer. PAA is a well-known supersorbent that is capable of absorbing several hundredtimes its own weight in water when immersed in solution [28]. The sorption of water from the air is a function of the humidity and of the ratio between free and dissociated acid functional groups [29]. Both an increase in the number of dissociated acid groups and an increase in humidity cause higher sorption of water, reaching about a 1:1 weight-ratio of polymer:water. Long-term exposure of our PAA thin-films to saturated water vapor showed no degradation in the film structure. Infrared measurements confirmed the presence of absorbed water by an increase in absorbance at 3500 cm-1, a behavior that has already been reported elsewhere [30].
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8565
However, the water exhibited no absorption in the wavelength range of the QCL and, thus, had no direct influence on the switching. Nonetheless, literature indicates enhanced diffusion of ammonia into the polymer under increased humidity [31,32], indicating that our experimental conditions probably influence the speed of the switching behavior. In order to evaluate this, a more detailed study is necessary. 3.3. Output power characteristics As the output power of a laser is dependent on the losses, the deposition of an absorbing coating is expected to reduce the optical output of a laser. Figure 6 shows the peak power for the different states of the laser. It is immediately evident, that the I-V characteristics do not change with changes in the cladding. This is to be expected as the voltage is applied between the top metal contacts and the backside-contact, thus not passing through the polymer.
Fig. 6. IV-characteristics and power output of the 38 µ m wide QCL for different modifications. The dotted black curves represent the IV and the peak power for the native QCL, the solid blue lines represent the same laser after PAA/potassium salt coating, and the dashed red line represents the same laser after a 15 s exposure to HCl vapors. All measurements were performed at room temperature.
However, the emitted power varies strongly for the different modifications. The native laser has a lasing threshold of about 2.59 A and emits about 120 mW peak power at 5 A. After spin-coating with the PAA potassium salt, the lasing threshold increases to about 3.34 A, with the peak power dropping to less than 20 mW. This is caused by the increased losses due to absorption. Converting the cladding to its acid-state by exposure to HCl vapors leads to a reduction in the threshold to 3.15 A and an increase in the peak power to about 60 mW at 5 A. This is equivalent to 50% of the optical power of the native laser. The reason for this increase in the emitted power can be attributed to the reduced losses due to reduced absorption of the organic film. Restoration of the salt state by exposure to ammonia yields a surprising increase in the output power which reaches values equivalent to the acid state, i.e. 50% of the original output power. The reason for this unexpected behavior is found in the original potassium salt coating. The coating is prepared from an approximately equimolar mixture of acrylic acid units within the PAA and potassium hydroxide by spin-coating. Taking a close look at Fig 2 one can see that, in spite of the equimolar ratio in solution, a peak at 1700 cm-1 is visible which can be attributed to the presence of monomer units in their acid state. Thus, the PAA potassium salt cladding introduces losses in the whole gain region of the laser, causing a reduction in the output power. Because of the predominance of the salt absorption, the emission wavelength is still shifted to lower wavenumbers. When subjecting the laser to ammonia vapors, the remaining carboxylic acid groups are being converted to carboxylates. This eliminates the C=O band at 1260 cm-1, reduces the absorption losses and increases the output power. The reduction of the output power was also observed during the switching events. The total emitted power dropped with the start of the switching event and increased again towards #95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8566
the end of the switching event, reaching a maximum when the switching was finished. This is in agreement with the above-mentioned explanation, as in the transition state, the laser incurs losses for both the carboxylic acid and the carboxylate. 3.4. Modeling For waveguide simulations the value of the refractive index of the cladding material is necessary. The imaginary part, related to the absorbance, can be estimated from IR transmission experiments. We measured a 3 µm thick layer of polyacrylic acid on a silicon wafer in transmission and recorded the absorbance spectrum. Taking into account the molecular mass of the acrylic acid monomer (72 g/mol) and the density of the polymer (1.1 g/cm3) we obtain a concentration of carboxylic acid groups in the polymer of 15 mol/l. This value can be entered into the formula of the molar extinction coefficient ε A ε= (2) d ⋅c with A being the absorbance, d the thickness of the layer, and c the concentration. Thus we obtain a value for ε of 262 mol-1cm-1. This is in good agreement with literature, where values between 190 and 950 mol-1cm-1 are reported [33]. Having confirmed the validity of our approach, we can now use the absorbance at 1307 cm-1 (corresponding to the acid state of the polymer) to determine the losses α which are obtained by the equation A (3) α= d where A and d are the same as in Eq. (2). Thus we obtain α = 170 cm-1. As α is directly proportional to the imaginary part of the refractive index, we can calculate the imaginary part of n as Imag(n) = 0.01. Because of the small value, only a minimal dispersion at the emission wavelength of the laser is expected and this should not influence the real part of the refractive index in a significant fashion. The lack of literature values for the real part of the refractive index of polyacrylic acid in the mid-infrared wavelength regime requires an approximation. As several polymers, including the closely related poly(methyl methacrylate), show basically no change in their refractive indices at the mid-infrared wavelengths as compared to at wavelengths in the visible light regime [34,35], it can be assumed that the same holds true for PAA. Thus we can use the literature value for the real part of n in the visible range and we obtain a complex refractive index of nPAA = 1.527 + 0.01i. For the laser measurements we used a coating of 500 nm thickness as determined by AFM measurements. This value, together with nPAA can be used to perform waveguide calculations. Calculated waveguide modes with and without the polymer are presented in Fig. 7. The modal overlap with the active region, given by the confinement factor Γ as well as the effective refractive index neff change by 4×10-3 and 1×10-3, respectively, and are, therefore, not degrading laser performance significantly. Waveguide losses αW in turn increase from 8.8 to 10.6 cm-1, giving rise to an increase in lasing threshold. With mirror losses of αM = 6.5 cm-1 for a 2.02 mm long Fabry-Peròt cavity the change in total losses leads to an expected increase in threshold current by 12%. The afore mentioned threshold currents of 2.59 and 3.15 A (Fig. 6) for the native laser and the laser with the polymer in its acid state are typical values and correspond to an increase of 22 %, which is in reasonable agreement with the lasing thresholds obtained through the waveguide calculations.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8567
Fig. 7. (a) and (b) show the refractive index line ups together with calculated waveguide modes for the native laser and for a laser with a 500 nm layer of polymer on top, respectively. The effective refractive index as well as the confinement to the active region stay almost constant whereas losses increase by 20%. This can also be seen in (c) presenting the influence of the polymer thickness on the confinement factor Γ (dashed line) and the waveguide losses αw (solid line).
Finally, in Fig. 7(c) we present the calculated influence of the PAA layer thickness on the main waveguide parameters αW and Γ. It can be seen that Γ increases slightly with the thickness until the thickness exceeds 500 nm. After that, the confinement is not influenced anymore. A slightly different behavior is observed for the αW. The waveguide losses increase strongly from 8.8 to 10.8 cm-1 for a thickness increase from 0 to 700 nm. Above that only minor changes occur and the losses saturate at a constant level of 11 cm-1. Due to this behavior, the output power can be significantly increased by slightly decreasing the thickness of the cladding. It should be noted that this system not only represents a novel way for switching the emission wavelength of a laser, but it is also a potentially sensitive sensor for various signals, which are normally not detectable by a laser. In the present system, the laser shows a full switching cycle upon conversion of 600 pmol of carboxylic acid groups in the polymer. Using thinner films it is expected that this detection limit can be reduced to the 10 pmol range. 4. Conclusion In conclusion, we have succeeded in generating a grating-less QCL with a polymeric transducer as cladding material that can be switched by a pH trigger. A two-color laser with emission at 1300 cm-1 and 1307 cm-1 was obtained that shows fully reversible switching on a 15 second time scale. Such a chromic switching in air under ambient conditions is a new paradigm that can be applied to photochromic, thermochromic, electrochromic, tribochromic or other transducers, making this approach highly versatile. Furthermore, this approach can be used as sensor. Most of the triggering mechanism (light irradiation, pH, pressure, current, voltage, friction, etc.) are not directly detectable by optical means but the transducer layers make laser spectroscopy on such parameters possible. Of special interest is the possibility to perform highly sensitive measurements as the sensing volume (waveguide surface area times the penetration depth of the evanescent field) is very small.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8568
Acknowledgments We are grateful for R. Liska (Institute of Applied Synthesis, Vienna University of Technology) for providing the polyacrylic acid. Special thanks to P.L. Souza (Laboratorio de semiconductors, CETUC, Pontificia Universidade Catolica do Rio de Janeiro, Brazil) and M.P. Pires (Instituto de Fisica, Universidade Rederal do Rio de Janeiro, Brazil) for the MOCVD growth of the lower InP claddings. Financial support through the Austrian research projects PLATON and ADLIS is gratefully acknowledged.
#95675 - $15.00 USD Received 1 May 2008; revised 21 May 2008; accepted 21 May 2008; published 27 May 2008
(C) 2008 OSA
9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 8569
