Enhanced photostability of photoluminescent dye

Optics Communications 271 (2007) 457–461 www.elsevier.com/locate/optcom

Enhanced photostability of photoluminescent dye-doped solutions and polymers with the addition of dielectric-oxide micro-particles Mohammad Ahmad

*

Laser Photonics Research Group, School of Physics and Astronomy, University of Manchester, Manchester, M13 9PL, UK Department of Materials, Prince Consort Road, South Kensington, Imperial College London, London, SW7 2AZ, UK IRC, Department of Materials, Queen Mary, University of London, Mile End, London, E1 4NS, UK Received 2 August 2006; received in revised form 11 October 2006; accepted 16 October 2006

Abstract Organic dye molecules in both solid polymethyl methacrylate and in solution have been found to display greatly enhanced photostability with the addition of micro-particles. Micro-particle doped poly(methyl methacrylate) samples and solutions were prepared doped with the laser dye pyrromethene 567 and rhodamine 6G. Study of the composite dye material as a gain medium in a laser provided a controlled and sensitive test of the photostability. The micro-particle doped samples used as the active laser medium demonstrated a doubling in service life to 0.4 million pulses compared with undoped samples before the output intensity was reduced to one-half. Further addition of a singlet oxygen quencher (DABCO) enhanced the photostability to 0.6 million pulses. Possible explanations discussed include cavity QED effects, surface photophysical and photochemical interactions and thermo-mechanical effects. These observations are relevant to photoluminescent devices and in general to the photostability of organic molecules. 2006 Elsevier B.V. All rights reserved. Keywords: Photostability; Photoluminescence; Solid-state dye laser; Pyrromethene; Rhodamine; Micro-particles; PMMA; Organic; Polymers

1. Introduction This paper reports observation of a substantial increase in the photostability of organic molecules used as the active elements in dye lasers and which is expected to be effective in many other photo-active devices. In liquid and solidstate dye lasers it has been found that the incorporation of dielectric-oxide micro-particles, including b-alumina (beta prime-prime alumina which is isomorphic form of aluminium oxide Al2O3), pyrex glass and silica gel, increases the photostability without compromising other laser performance properties. Photoluminescent organic materials are becoming an increasingly important class of materials. Their role as efficient light emitters has led to the success of the dye laser over the last three decades. Considerable research has con*

Tel.: +44 0207 594 6813; fax: +44 0207 5843194. E-mail address: [email protected]

0030-4018/$ - see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.optcom.2006.10.023

firmed the success in incorporating photoluminescent organics as active laser molecules into solid host materials [1]. A further notable success has been the development of electrically conducting hosts and conjugated polymers to produce the related electroluminescent class of materials [2,3]. This has led to the important commercial development of polymer LEDs and full colour displays. Organic materials are also important in many other areas of optoelectronics including non-linear optics [4] and photovoltaic devices [5,6]. Photodegradation is a fundamental problem affecting all aspects of organic optoelectronics. The requirement of high power densities in organic LEDs and lasers results in photodegradation that steadily destroys the device. Packing techniques are required to increase operational lifetimes by removing oxygen present. Solar energy, particularly in the ultra-violet region, degrades the performance of organic photovoltaic devices. The efficiency and cost of organic solar cells offer promise for domestic and industrial

458

M. Ahmad / Optics Communications 271 (2007) 457–461

use, but their lifetimes are currently too short. Few widespread techniques are available to enhance the photostability of solid photoluminescent materials. The most notable are the removal of oxygen, the avoidance of ultra-violet exposure and the use of anti-oxidants or singlet oxygen quenchers [7–11]. The novel approach described here may have far reaching applications beyond photoluminescent devices to photostability in dyestuffs. In this study a liquid or solid-state dye laser is used as a fast test of photostability due to the delicate balance of optical gain versus loss. The solid-state laser also has real-world application for medical lasers or it could be a convenient laboratory accessory to add wavelength tunability to the second harmonic Q-switched Nd:YAG laser or other similar lasers that are widespread. Details of the laser and photostability experiments undertaken are described. Possible explanations are discussed for the photostability enhancement that is observed, these include cavity QED mechanisms, surface photophysical and photochemical interaction and thermo-mechanical effects. 2. Dye and micro-particle doped samples The materials under study were two laser dyes, pyrromethene 567 and rhodamine 6G, in ethanol or doped into the solid polymer PMMA. The solid polymer samples were made from methyl methacrylate (MMA) monomer which was distilled to remove the polymerisation inhibitor hydroquinone monomethyl ether. Pyrromethene 567 was dissolved into the monomer at 3.4 · 10 4 M concentration and the mixture was placed in a water-filled ultrasonic bath until the dye was completely dissolved. In the case of R6G, 10% ethanol was added to aid solubility. Pieces of b-alumina (available from Ceramatec, Inc., USA; and Ionotec, Ltd., UK) crystals were ground into a powder and added to the dye solution along with 1 mg/ml 2,2-azobis 2-methylpropionitrile polymerisation initiator. Finally, the mixture was replaced in the ultrasonic bath for a few minutes. The dissolved mixture was then allowed to settle at room temperature so that the large particles of the glass could be separated by sedimentation. The remaining solution mixture was filtered with a 0.2 lm filter. The concentration of the micro-particles is approximately 5 mg/ml and further quantitative analysis is underway. The monomer solution, in sealed test tubes, was placed in a water bath at a temperature of 40 C for 2–3 days until a viscous liquid was formed. The tubes were then transferred to an oven where the temperature was increased step-wise at 5 C / day until it reached 90 C. Then the temperature was reduced over two days to room temperature. The glass tubes were broken to remove the polymerised samples, which were then cut into disks and polished. The solid samples were placed in an optical laser cavity and excited with a Q:switched frequency doubled Nd:YAG laser. The composites of PMMA containing particles of b-alumina are referred to be as b-PMMA. One set of b-PMMA composite samples were also doped with a singlet-oxygen quencher,

1,4-diazbicyclo[2,2,2]octane (DABCO), with the concentration of the quencher being 10 times that of the dye. 3. Tests of photostability The laser cavity was a compact plane–plane configuration, as used in Ref. [7]. The input mirror was dichroic with 90% transmission at 532 nm and 95% reflectivity between 560 nm and 600 nm. The output mirror was a 70% broadband reflector that was not necessarily optimum. A short cavity length of 15 mm was used to reduce the cavity losses due to a highly divergent output. The pump source was a Q-switched Nd:YAG laser operating at the second harmonic of 532 nm. This delivered up to 60 mJ/pulse in 6 ns at 1–10 Hz repetition rate or in a single pulse. A 20 mm focal length lens focused the pump beam onto the sample which was placed before the focus such that the diameter of the pump beam was 2 mm at the sample input face. The pump beam was aligned off-axis at a slight tilt angle of 16 to the resonator axis so that any transmitted pump light was not collinear with the output beam and did not fall onto the volume absorbing power meter. Photostability experiments on the solid samples were performed by varying the pump fluence from 0.16 to 3.0 J cm 2 and the repetition rate from 2 to 10 Hz. Through these experiments, different aspects of the dependence of laser performance on excitation of the solid samples have been studied, these ranged over high fluence, low average power or high average power, low fluence regimes to high average power and high fluence. The conversion efficiency of the laser was measured as a function of input energy and the number of input pulses. After testing, the samples were inspected for signs of laser damage to the bulk. The laser performance on liquid samples was evaluated using 1 ml of dye solution (1 · 10 4 M pyrromethene or 5 · 10 5 M rhodamine 6G) in a 1 cm optical path length cuvette. The pump laser pulse energy was 15.4 mJ at a 10 Hz repetition rate. Experiments on solid-state materials containing microparticles revealed substantially increased photostability. Data obtained from 3.4 · 10 4 M P567 doped PMMA with and without micro-particles is presented in Fig. 1 for a 2 Hz repetition rate and a pump fluence of 0.16 J cm 2. The number of pulses for the conversion efficiency to fall to one-half of its initial value is seen to increase from 0.2 million without particles to 0.4 million for samples containing micro-particles. Micro-particles had no effect on the laser efficiency of either solutions or dye-doped PMMA. It can be noted that the addition of micro-particles alone increases the photostability by an amount comparable to using DABCO alone [10], but the addition of both DABCO and micro-particles increased the photostability to exceptional levels for solid-state dye laser. In this case the operational life was increased to 0.6 million pulses for the extrapolated half-intensity value of the conversion efficiency, corresponding to a total absorbed pump energy of

M. Ahmad / Optics Communications 271 (2007) 457–461 80

7

Number of pulses to half-intensity (x105)

PMMA

Conversion efficiency (%)

459

Beta-alumina Beta-alumina + DABCO

70 60 50 40 30 20 10 0 0

1

2

3

4

PMMA Beta-alumina Beta-alumina + DABCO

6 5 4 3 2 1 0 0

2

4

5

6

8

10

12

Repetition rate (Hz)

Number of pulses (x10 ) Fig. 1. Conversion efficiency versus number of pulses for P567 doped into pure PMMA and PMMA with additives. The pump fluence was 0.16 J cm 2, the repetition rate was 2 Hz and the samples were 8 mm long and doped with a dye concentration of 3.4 · 10 4 M. The balumina + DABCO sample was 16 mm long.

Fig. 2. The number of pulses emitted by the SSDL before the emission peak intensity falls to one-half versus the repetition rate. The pump fluence was 0.16 J cm 2 in all cases and all the samples were 8 mm long and doped with a P567 dye concentration of 3.4 · 10 4 M.

365 GJ mol 1. That sample is one of the most stable SSDLs yet reported [11]. In all cases an increase in the repetition rate reduced the operation lifetime of the SSDLs. The same samples as those used for the 2 Hz repetition rate study were also tested at 5 Hz and 10 Hz. If the assumption is made that photochemical processes are complete in the 100 ms between pulses then this experiment tests some aspects of the thermal properties of the material. If a 2 Hz repetition rate is used then 10 mW of power is deposited in the active region. At 10 Hz, the figure is 50 mW. Fig. 2 shows the half-life of laser operation of the three materials at three different repetition rates. For each of the materials, the reduction in lifetime is comparable indicating that the thermal processes are similar. The dependence of the laser performance of the laser dyes in liquid solution on the addition of micro-particles was investigated. In ethanol the photostability increased by a factor of three upto to 18 GJ mol 1 for samples containing b-alumina micro-particles. There was no any noticeable effect on the laser efficiency with micro-particles

doping. The dye laser output wavelength was 565 nm. The normalised photostability of rhodamine 6G in ethanol increased from 20 GJ mol 1 to 60 GJ mol 1 and the output wavelength was 575 nm. To provide a genuine test of the materials capability as a gain medium for a solid-state dye laser the operational lifetime was measured; this was carried out at a 10 Hz repetition rate using pump fluences ranging from 0.16 to 3 J cm 2. The photostability is normalised in units of the total average pump energy absorbed by the sample per mole of the dye at which the laser intensity is reduced to one-half. Fig. 3 shows that increasing the pump fluence drastically reduces the operation lifetime as would be expected. Moreover, the reduction is non-linear so that even the normalised photostability falls with increasing pump fluence. The highest average pump power was 0.94 W which is approaching the average pump power that may typically be required in a high power application, for example in medical dermatology procedure and the microparticle doped SSDL lasted for 4000 pulses to half-intensity. In many respects this test is even more severe than

Number of pulses

1,000,000 PMMA beta-alumina

100,000 10,000 1,000 100 10 1 0

0.5

1

1.5

2

2.5

3

3.5

Pump fluence (Jcm-2) Fig. 3. Number of pulses and normalized photostability with variation of the pump fluence at 10 Hz repetition rate. The data providing number of pulses were all taken with 8 mm long samples and doped with a P567 dye concentration of 3.4 · 10 4 M.

460


in a real application as that would probably use a larger beam diameter and a method of translating the sample through the pump beam to use an area of unexposed dye. In all the tests described here a single site on the sample was used with no rotation or translation. If a simple extrapolation were used to take account of using the whole area of a 25 mm diameter optic then the half-life would be 625,000 pulses or 17 h of continuous operation at a 10 Hz repetition rate. All the data so far presented has been for P567. The second dye studied, R6G is generally an order of magnitude less stable than P567 [11]. However, the addition of micro-particles to a solid PMMA sample containing rhodamine 6G, provided the same proportion of enhancement to the photostability as for P567. No laser damage occurred in micro-particle doped solidstate PMMA at all the pump fluences used in this study. PMMA samples alone showed both surface and bulk damage when pumped with a fluence higher than 1.0 J cm 2. This addition of micro-particles also provide a higher laser damage threshold. Issues related to output laser beam quality, as observed with inclusion of nanoparticles [12–15] and Rayleigh [16,17] scattering have not been discussed here because of no direct link of these issues with the parameters measuring photostability of laser dyes used in this study. 4. The possible roles of micro-particles in dye-doped liquids and solids This paper describes the enhancement of the photostability of pyrromethene P567 in solution and in solid PMMA when co-doped with b-alumina micro-particles. Similar results have also been obtained with rhodamine 6G. The particles had been prepared simply by filtering with 0.2 lm filter, following grinding pieces of bulk material with a pestle and mortar. An explanation for this interesting and slightly unexpected phenomenon is speculative at the present time and several possibilities are currently being investigated. These include cavity quantum electrodynamic (cavity QED) effects [18], modification of the photophysical and photochemical interactions between the dye molecule and its environment and changes in the thermal or mechanical properties of the bulk host material [8–11,19]. The detailed microstructure of the composite media following the preparation procedure are not yet known, but the observation of photostability enhancement in both solutions and solid media indicates that the mechanism is not primarily related to the matter phase. The effect has also been observed with other micro-particles than b-alumina, these have included particles of pyrex glass and silica gel. The dependence on the material of the microparticle is also currently under investigation. The photodegradation mechanisms of P567 have been studied previously in detail [10] and a major degradation mechanism has been highlighted as self-sensitised photooxidation. During one cycle of excitation, the history of a dye molecule is as follows. In air-saturated solutions exci-

tons undergo intersystem crossing to the triplet state. For example, in air-saturated benzene 9 ± 1% of P567 molecules convert to the triplet. This intersystem-crossing yield is oxygen enhanced. The triplet states are then quenched by oxygen to give singlet oxygen with a 100% yield. The rate constant of chemical reaction between singlet oxygen and the ground state of the dye has been shown to be significant at 106 s 1. In a laser system the photodegradation yield is expected to be different due to effective competition between stimulated emission and intersystem crossing. Reducing intersystem crossing, additive-induced triplet quenching and singlet oxygen quenching have all been shown to be effective techniques of photostability enhancement. Another photodegradation route has been found to be via free radicals, which is strongly dependent on the solvent [11]. Oxygen also enhances the free radical mechanism, presumably due to complexion. If cavity QED effects are responsible for the increase in photostability then the emission properties of the excited dye molecules may be modified via a change in the density of optical modes available for emission by a nearby microparticle [12–15]. In this case, the interface between the dye molecule and the micro-particle may act as a microcavity, and a resonance may exist between the dye molecule exciton and a whispering gallery mode in the micro-particles. Previous studies of laser action in rhodamine 6G solutions containing TiO2 nano-crystals [20–23] showed that the threshold for laser oscillation was significantly reduced. The dye molecules may be adsorbed onto the micro-particle, so that there may be a weak coupling to the optical modes of the micro-particles. Account needs to be made that the micro-particles may not be spherical. Clearly in this study a range of micro-particle sizes up to 200 nm exist so only a fraction of the dye molecules will participate in a resonant interaction. Such an interaction may enhance the emission rate of the dye molecule excitons making light emission compete more favourably to non-radiative deactivation which is known to be the first step towards photodegradation. If the micro-particles are too small to produce an increase in optical scatter, however, it may not be possible for them to support optical modes at the emission wavelength. The presence of micro-particles may also block one or more of the steps outlined above that lead to photodegradation. An interaction between the micro-particles and the dye molecules, or possible adsorption of the dye onto the surface of a micro-particle is likely to affect the photophysics and photochemistry of the system. Any triplet state quenching effects of the dye molecules by micro-particles will reduce the degradation rate. Similarly stabilisation would occur if singlet oxygen were quenched by micro-particles. Other possibilities include adsorption of oxygen on the surface of the microspheres, thus reducing the oxygen concentration in the rest of the dye-doped medium. It is also known that charge transfer occurs between particles of TiO2 and dye molecules in photovoltaic cells [5,6]. Another category of possible explanations involves thermal or mechanical alterations to the host material. An


increase in the thermal conductivity of the host material will spread deposited heat across the entire sample more effectively. This will reduce the formation and the severity of a temperature increase in the gain region. An increase in temperature would induce thermal degradation to become important as it would increase the photodegradation rate constants. A dispersion of micro-particles may also affect the micro-hardness of the host material, especially if oxygen is adsorbed on the surface of the micro-particles. Oxygen is a polymerisation inhibitor, so its removal will result in a harder material. Curiously, however, in a previous report, an increase in the viscosity of the polymer has been shown to increase the photostability of the dye but reduce the laser damage threshold [24]. In the case described here an increase in both properties has been observed. In summary, the addition of micro-particles with diameter less than 0.2 lm to a dye-doped polymer has substantially increased the photostability of the dye, without compromise in the laser efficiency. Possible explanations under consideration are cavity QED effects, a modification of the chemical environment or changes to the physical properties of the host material. Interesting, some dye-sensitised photovoltaic devices already contain TiO2 micro-particles. Characterisation and optimisation of this effect may give photovoltaic in increase in their photostability that is in their intrinsic design. Along with beneficial effects to solid-state dye lasers and photovoltaic devices, this effect is of fundamental relevance to the behaviour all optoelectronic organic materials. More generally, the phenomenon may have significance for many other systems of organic molecules where photostability is involved.

References [1] M.D. Rahn, T.A. King, Appl. Opt. 34 (36) (1995) 8260.

461

[2] J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burns, A.B. Holmes, Nature 347 (6293) (1990) 539. [3] S.P. Lee, H. Choi, K.W. Lee, K.H. Mo, J.W. Jang, E. Lee, I.M. Kim, C.E. Lee, J. Korean Phys. Soc. 48 (1) (2006) 46. [4] N. Bloembergen, Non-linear optics, Fourth ed., World Scientific, Berlin, 1996. [5] B. Oregan, M. Gratzel, Nature 353 (6346) (1991) 737. [6] C.S. Karthikeyan, M. Thelakkat, M. Willert-Porada, Thin Solid Films 511 (2006) 187. [7] A.N. Fletcher, Appl. Phys. B 27 (2) (1982) 93. [8] M.D. Rahn, T.A. King, A.A. Gorman, I. Hamblett, Appl. Opt. 36 (24) (1997) 5862. [9] A.A. Gorman, I. Hamblett, T.A. King, M.D. Rahn, J. Photochem. and Photobio. A, Chem 130 (2–3) (2000) 127. [10] M. Ahmad, M.D. Rahn, T.A. King, Appl. Opt. 38 (30) (1999) 6337. [11] M.D. Rahn, T.A. King, in: R. Scheps (Ed.), Solid State Lasers VIII, Proc. SPIE, vol. 3613, 1999, p. 94. [12] F.J. Duarte, R.O. James, Opt. Letts. 28 (21) (2003) 2088. [13] H. Fujiwara, K. Sasaki, H. Masuhara, Chemphyschem 6 (11) (2005) 2410. [14] P. Sandeep, P.B. Bisht, Chem. Phys. Lett. 371 (3–4) (2003) 327. [15] H. Ohtaka, T. Koyama, Y. Taniguchi, Molecular crystals and Liquid Crystals 349 (2000) 395. [16] T. Itoh, Y. Ozaki, H. Yoshikawa, T. Ihana, H. Masuhara, Appl. Phys. Lett. 88 (8) (2006). Art. No. 0841012. [17] M.A.F. de Souza, A. Lencina, P. Vaveliuk, J. Appl. Phys. 100 (2) (2006). Art. No. 023113. [18] A.J. Campillo, J.D. Eversole, H.-B. Lin, Phys. Rev. Lett. 67 (4) (1991) 437. [19] M. Ahmad, T.A. King, D.K. Ko, B.H. Cha, J. Lee, Opt Comm. 203 (3–6) (2002) 327. [20] N.M. Lawandy, R.M. Balachandran, A.S.L. Gomes, E. Sauvain, Nature 368 (6470) (1994) 436. [21] H.Z. Wang, F.L. Zhao, Y.J. He, X.G. Zheng, X.G. Huang, M.M. Wu, Opt. Lett. 23 (10) (1999) 777. [22] M. Alvarez, F. Amat-Guerri, A. Costela, I. Garcia-Moreno, C. Gomez, M. Liras, R. Sastre, Appl. Phys. B 80 (8) (2005) 993. [23] A. Costela, I. Garcia-Moreno, C. Gomez, F. Amat-Guerri, M. Liras, R. Sastre, Appl. Phys. B 76 (4) (2003) 365. [24] A. Costela, F. Florido, I.G. Moreno, R. Duchowicz, A. Amat-Guerri, J.M. Figuera, R. Sastre, Appl. Phys. B 60 (4) (1995) 383.