An experimental study of the lasing characteristics of photonic crystal lasers based on the conjugated polymer 2-methoxy-5-(2′- ethylhexyloxy)-1,4-phenylenevinylene (MEH-PPV) is reported in this letter. One and two dimensional (1D, 2D) photonic crystal structures were patterned on a glass substrate through interferometric lithography on photoresist layers. A 1.5 µm layer of polymethylglutarimide (PMGI) was deposited to prevent photoxidation of the polymer. Lasing action was observed under optically pumped conditions. Instabilities associated with pumping geometries were demonstrated in the case of 2D photonic crystal laser. As a result, the laser spectrum and threshold gain were found to be strongly dependent on the excitation geometry. The broad spectrum of the amplified spontaneous emission (ASE) allows laser tunability by engineering the effective refractive index of the devices or by controlling the periodicity of the photonic crystal.
© 2005 Optical Society of America
Photonic crystals have many device applications because of their unique ability to confine light in multiple dimensions. The absence of optical modes in the photonic band gap can be used to suppress the spontaneous emission and create resonances that can be used in a variety of applications, including lasers. One dimensional (1D) periodic structures have been exhaustively studied, and are the principle behind of many devices such as distributed feedback lasers. A complete mathematical description of a 1D photonic crystal laser based on semiconductor materials was developed in the mid-70 by Striefer et al.  after the pioneering work of Kogelnik et al. .
Conjugated conducting polymers have been the subject of intense theoretical and experimental study due to their promising applications in optical and electronic devices. Among the conjugated polymers, poly[2-methoxy-5-(2′-ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) has become very attractive because it can be excited both electrically and optically, providing versatility and convenience. Applications such as light emitting diodes (LEDs) , photovoltaic cells , photodetectors , and field effect transistors  have been realized based on MEH-PPV. The commercial potential of these devices resides in their ease of fabrication since conjugated polymers can be deposited and patterned by inexpensive techniques such as spin coating or ink jet printing on a variety of substrates, including flexible ones . One of the first demonstrations of lasing action in MEH-PPV was reported by Lawandy et al.  from optically pumped MEH-PPV solutions containing coloidally suspended titanium dioxide (TiO2). In this device the titanium dioxide particles scatter the emitted photons in the MEH-PPV molecules in such a way that the feedback obtained from the scattering processes exceeds the losses above threshold. More recently Turnbull et al.  presented the first demonstration of lasing action from neat films of MEH-PPV. In this configuration the device was held under a dynamic vacuum to prevent photoxidation and the subsequent degradation of the device.
In this work we report laser instabilities associated with the pumping geometry for 2D photonic crystal structures. These instabilities can result in the degradation of the spectral quality of the laser emission and in an increase in the threshold gain.
2. Sample preparation and optical characterization
Gilch-type MEH-PPV was purchased from Aldrich with average molar number Mn=70,000-100,000 g/mol with a polydispersity value (Mw/Mn=1.52). MEH-PPV solutions were prepared from toluene at a concentration of 5 g/dm3. The dissolution process turned out to be difficult, due to the high molecular mass of MEH-PPV. To enhance the solubility, the solutions were placed in an ultrasonic bath for 6 hours and kept in a dark environment at room temperature until device fabrication.
For the optical characterization, thin films of MEH-PPV were prepared by spin casting the solution at 3000 rpm onto a glass substrate. The absorption spectrum was measured with an 8453 Hewlett-Packard spectrophotometer. For the photoluminescence (PL) spectrum a HR2000 Ocean Optics fiber-coupled CCD spectrometer was used with a Q-switched frequency-doubled Quanta Ray (Spectra Physics) Nd:YAG laser operating at 532 nm, 2.5 ns pulse length, with a repetition rate of 10 Hz as the excitation source. The measured absorption spectrum is shown in Fig. 1 from which we can infer two prominent absorption bands. The one in the visible region with a peak around 510 nm has the largest transition probability. The other peak is located in the UV region around 335 nm and corresponds to the phenyl group of the polymer. The PL profile also shows two emission peaks. The first one is located at 566 nm and corresponds to the polymer chains, and the second peak is located at 599 nm and is a consequence of both the polymer and the aggregates.
3. Results and discussion
3.1 1D photonic crystal structure
For the lasing application, the photonic crystal structures were patterned in a photoresist layer on a glass substrate through interferometric lithography. The PMMA (refractive index 1.52 at 632.8 nm) was diluted with Shipley Microposit EC-11 solvent at a ratio of 1:5, and spun on a quartz glass substrate at 3000 rpm. The resulting film thickness was 70 nm as measured by a Tencor P10 surface profilometer. Figure 2 shows the interferometric lithography system. The setup is similar to a Lloyd-mirror configuration. A CW argon laser operating at 244 nm (after second harmonic conversion) is used as the illumination source. The beam is expanded and then combined at the sample. The direct beam and the secondary beam are reflected off the mirror to produce an interference pattern with period given by:
After the photoresist was developed, the samples were soft baked at 100 C for 1 hour and then exposed to UV radiation (mercury lamp, 365 nm) for about 1 hour to increase the hardness of the photoresist. Figure 3 shows the atomic force microscope (AFM) image of the 1D photonic crystal structure. The period of the corrugation was about 400 nm with an average depth of 70 nm. The silica samples were not etched; instead photoresist-polymer interfaces were used as the photonic crystal corrugations. Compared to etching the quartz glass, this increases the confinement factor of the fundamental optical mode in the laser waveguide structure around the corrugation, which decreases the threshold gain of the device.
To incorporate the gain media into the laser structures, the MEH-PPV (refractive index 1.8 at 640 nm) was deposited from solution onto the samples to form a film thickness of 100 nm. To prevent photoxidation and to enable operation of the device in an oxygen rich environment, a 1.5 µm thick layer of polymethylglutarimide (PMGI) was spin coated on top of the MEH-PPV film. PMGI has a refractive index of 1.54 and is transparent to visible light .
Figure 4(a) shows the theoretical calculation of the fundamental mode in the polymer film waveguide. In this calculation the refractive index of the corrugation region was approximated as the average of the refractive indices of PMMA and MEH-PPV. The effective index of the structure was found to be around 1.60. A more rigorous analysis based on the Fourier expansion of the index of refraction around the grating structure and the Floquet’s theorem can be found in reference . The Bragg condition that determines the lasing wavelength of the photonic crystal structure is given by:
Where n is an integer, neff is the effective index of the structure, λB is the Bragg wavelength and Λ is the periodicity of the photonic crystal. For surface emission normal to the corrugation, a second order diffraction grating with n=2 was chosen. The periodicity Λ must be selected such that the Bragg wavelength overlaps with the amplified spontaneous emission (ASE) of MEH-PPV. In this particular case, a periodicity Λ of 400 nm was selected. Figure 4(b) shows the ASE spectra obtained by pumping an unpatterned film of MEH-PPV coated with PMGI. It is important to note that a film patterned with a photonic crystal structure will produce a PL spectrum that has an angular dependence due to the diffraction process in the photonic crystal. Therefore care must be taken to avoid misinterpreting the PL spectrum. Lasing action in the photonic crystal MEH-PPV was obtained by optically pumping the devices beyond the threshold gain of the laser. The resulting laser emission was normal to the surface at both faces of the polymer film. To test the directionality of the emission, the laser spectrum was recorded at 12 inches away from the sample. The recorded spectrum from the device is shown in Fig. 4(b). It shows considerable line-narrowing to ≤0.65 nm FWHM, the resolution limit of the spectrometer. Also notice that the emission of the device is around 640 nm, which is in agreement with the theoretical emission wavelength determined by equation (2) using an effective index of 1.6 which was calculated for the structure shown in Fig 4(a). Compared to most lasing materials, MEH-PPV has a very broad ASE spectrum. This allows the laser emission to be tuned over a large range by varying the periodicity of the photonic crystal structure or by tailoring the thickness of the MEH-PPV layer to change the effective index of the waveguide structure.
3.1 2D photonic crystal structure
One of the major drawbacks of 1D photonic crystal lasers is the existence of lateral modes that lead to multimode operation, thus limiting the operation of single mode high power lasers. A very elegant way to reduce the lateral modes and concentrate the laser emission in a narrow line is to create optical confinement of the propagating waves in all directions (in plane) through a 2D photonic crystal structure .
Whereas in 1D-PCL a corrugation periodicity of 400 nm is used, two perpendicular corrugations of 400 nm form the 2-D photonic crystal structure. In this configuration laser action is achieved by phase locking of the modes established in the two perpendicular resonators . However this laser configuration has a drawback of being dependent of the optical pumping geometry as will be shown next. To achieve a 2D structure, two successive exposures are performed after rotating the sample by 90 degrees. Thus the exposure time of each succesive exposure has to be half the exposure time required for the 1D structure. Figure 5 shows the atomic force microscope (AFM) image of the 2D photonic crystal structure on photoresist. After the 2D structure is obtained, the same process as described above is used to complete the laser structure by spin coating the MEH-PPV and the PMGI.
The observed laser emission was normal to the surface of the device. Figure 6 ilustrates the difference in the spectral characteristics obtained by pumping the laser with a circular and elliptical spot. Figure 6(a) shows the laser spectrum recorded by the spectrometer when the device is pumped with a collimated circular spot. However, when the structure is pumped with a non-circular beam, instabilities can take place as shown in Fig. 6(b). To understand the difference between the two spectra in Fig. 6, an illustration of the diffraction process taking place in the grating is presented in Fig. 7.
The circular excitation geometry allows both perpendicular resonators to have nearly equal threshold gains, and thus the locking process between the the two resonators is more readily achieved. In contrast, when the device is pumped with a non-circular spot, one of the resonator sees a longer pumping length and thus its threshold gain is lowered. Hence the longer stripe is more likely to lase compared to the other resonator with the shorter pump stripe This geometry makes the locking process between the two perpendicular resonators more difficult. The result is that the device behaves as a 1D PCL along the longer stripe, and as a simple diffraction grating along the shorter stripe. This can be infered from the diffraction slope around the lasing wavelength in Fig. 6 (b). The overall effect is an increase in loss due to the diffraction produced by the shorter resonator, that ultimately increases the threshold pumping power and deteriorates the spectral quality of the laser emission. Figure 7(a) shows the diffraction process for these geometries. Another disadvantage of this configuration is that Bragg reflection does not take place for TE  mode in identical grating period, so only TM polarization can be obtained A more proactive solution could be a fourth-order process in which both grating work in sinergy to couple the backward and forward traveling waves. Figure 7(b) shows this diffraction process in k-space. This configuration is inherently more stable since both grating are needed to complete the coupling process. It is also less suceptible to an arbitrary excitation geometry, and cross gratings can be designed with different periods for TE modes. This device is currently being fabricated and will be reported in a future publication.
Lasing action from both 1D and 2D photonic crystal MEH-PPV polymer was demonstrated and instabilities associated with pumping geometry for the 2D case was outlined. The devices were optically pumped and operated in an atmospheric environment using PMGI as an oxygen barrier layer. New barriers with very low permeation rate such as Barix are promising materials to increase the lifetime of these devices. To the best of our knowledge this is the first report of instabilities due to noncircular pumping geometries in a 2D PCL based on MEH-PPV. A fourth order diffraction process was proposed as a more stable resonator design.
This work was partially supported by the Dayton Area Graduate Studies Institute (DAGSI), AFOSR Summer Faculty Fellowship and AFOSR through in-house research grant LRIR-2305DW01. The authors acknowledge Dr. Graham A. Turnbull from The University of St. Andrews for valuable discussions related with the lasing characteristics of MEH-PPV and Sarah Blickenstaff for equipment support.
References and Links
1. W. Streifer, D. R. Scifres, and R. D. Burham, “Couple wave analysis of DFB and DBR Lasers,” J. Quantum Electron. 13134–141 (1977). [CrossRef]
2. H. Kogelnik and C. V. Shank, “Stimulated emission in a periodic structure,” Appl. Phys. Lett. 18152 (1971). [CrossRef]
3. Y. Shi, J. Liu, and Y. Yang, “Device performance and polymer morphology in polymer light emitting diodes : morphology dependent emission spectra,” Macromol. Symp. 154, 187–197 (2000). [CrossRef]
4. J. Gao, G. Yu, and A. J. Heeger, “Polymer p-i-n Junction Photovoltaic Cells,” Adv. Mater 10, 692–695 (1998). [CrossRef]
5. J. Gao, F. Hide, and H. Wang, “Efficient photodetectors and photovoltaic cells from composites of fullerenes and conjugated polymers: Photoinduced electron transfer,” Synth. Met. 84, 979–980 (1997). [CrossRef]
6. T. Sakanoue, E. Fujiwara, R. Yamada, and H. Tada, “Visible light emission from polymer-based field-effect transistors,” Appl. Phys. Lett. 84, 3037–3039 (2004). [CrossRef]
7. M. D. McGehee and A.J. Heeger, “Semiconducting (Conjugated) Polymers as Materials for Solid-State Lasers,” Adv. Mater , 22, 1655–1668 (2000). [CrossRef]
8. N.M. Lawandy, R.M. Balachandran, A.S.L. Gomez, and E. Sauvain, “Laser action in strongly scattering Media,” Nature , 368436–438 (1994). [CrossRef]
9. G. Turnbull, P. Andrews, W. Barnes, and I. Samuel, “Relationship between photonic band structure and emission characteristics of a polymer distributed feedback laser,” Phys. Rev. B 64125122-1 1-125122-6 (2001). [CrossRef]
10. M. D. McGehee, M. A. Diaz-Garcia, F. Hide, R. Gupta, and E. K. Miller, “Semiconducting polymer distributed feedback lasers,” Appl. Phys. Lett. 72, 1536–1538 (1998). [CrossRef]
11. M. Toda, “Proposed cross grating single-mode DFB laser,” J. Quantum Electron. 281653–1662 (1992). [CrossRef]
12. S. Riechel, C. Kallinger, U. Lemmer, J. Feldman, A. Gombert, V. Wittwer, and U. Sherf, “A nearly limited surface emitting conjugated polymer laser utilizing a two-dimensional photonic band structure,” Appl. Phys. Lett. 77, 2310–2312 (1998). [CrossRef]