- Email: [email protected]
Excitation energy transfer in partly ordered polymer films differing in donor and acceptor transition moments orientation
Optical Materials xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Excitation energy transfer in partly ordered polymer films differing in donor and acceptor transition moments orientation A. Synak a,⇑, P. Bojarski a, M. Sadownik a, L. Kułak b, I. Gryczynski a, B. Grobelna c, S. Rangełowa-Jankowska a, D. Jankowski a, A. Kubicki a a b c
University of Gdansk, Faculty of Mathematics, Physics and Informatics, Institute of Experimental Physics, Wita Stwosza 57, 80-952 Gdansk, Poland Gdansk University of Technology, Faculty of Technical Physics and Applied Mathematics, Narutowicza 11/12, 80-952 Gdansk, Poland University of Gdansk, Faculty of Chemistry, Wita Stwosza 63, 80-308 Gdansk, Poland
a r t i c l e
i n f o
Article history: Received 27 October 2015 Received in revised form 19 January 2016 Accepted 1 March 2016 Available online xxxx Keywords: Energy transfer Emission anisotropy PVA films Partly ordered systems Monte Carlo simulations
a b s t r a c t Based on spectroscopic measurements selected properties of nonradiative Förster energy transport are studied in uniaxially stretched polyvinyl alcohol thin films for three systems differing in donor and acceptor transition moments orientation relative to the axis of stretching. In particular, donor – acceptor emission anisotropy spectra yield completely different regularities for these systems in uniaxially stretched films, whereas they are similar in unstretched films. In particular it is shown that acceptor fluorescence can be either strongly polarized after nonradiative energy transfer in stretched films or depolarized depending on the angular distribution of acceptor transition moments in the matrix. Donor and acceptor emission anisotropy decays exhibit similar regularities to those of steady-state measurements. The obtained results are analyzed with the help of Monte Carlo simulations. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction Polymer films have numerous applications in various industrial and biomedical sectors such as functional and protective coatings, biocompatible medical implants, advanced membranes, sensors [1,2]. Ordered polymer films doped with fluorophores are well known materials used in polarization spectroscopy, to determine transition moments directions of molecules, their conformations or to study properties of nonradiative energy transport process [3–5]. Excitation energy transport has been widely studied for donor and acceptor transition moments randomly distributed in space [6,7]. It is well known that for such disordered systems sensibilized fluorescence emitted by acceptors excited exclusively through energy transfer from donors is almost totally depolarized [8]. This is because only primarily excited molecules (by linearly polarized light) contribute to emission anisotropy. However, as shown recently, quite different observations have been reported for uniaxially oriented polymer films, where emission anisotropy of molecules excited via energy migration or transfer remain high [9–11]. As a result of mechanical uniaxial stretching of the polymer film along fixed axis, the directions of transition dipole moments of ⇑ Corresponding author.
elongated fluorophores loose their random orientation and exhibit certain nonrandom distribution relative to the axis of stretching. This means, in turn, that energy transfer takes place from an initially excited donor to an acceptor, transition moment direction of which is correlated with that of donor. On a macroscale, we can consider the value of the averaged orientation factor as a certain measure of these correlations. The averaging over all molecular configurations can lead to a much more pronounced change of the orientation factor in such a macroscopically nonrandom system compared to the random systems or systems which are locally nonrandom. The angular correlations between the directions of transition dipole moments are mostly responsible for the effect of emission anisotropy preservation in uniaxially stretched one component polymer films, where the excitation energy visits many chemically identical sites an arbitrary number of times [9–11]. Unlike one – component systems, one can imagine different physical situations in uniaxially oriented polymer matrices doped with donors and acceptors of excitation energy. Chemically nonidentical donors and acceptors can be selected in such a way that their transition moments are located differently towards respective long molecular axes. For example, if transition moments of both donor and acceptor are parallel to their long molecular axes, then they orientate quite efficiently upon matrix stretching towards this axis of stretching. On the other hand, if transition moment of the donor is perpendicular to long molecular axis, but
E-mail address: [email protected] (A. Synak). http://dx.doi.org/10.1016/j.optmat.2016.03.005 0925-3467/Ó 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
2
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
acceptor transition moment is parallel to its long molecular axis, then upon matrix stretching both orientations of donor and acceptor transition moments are totally different. As a result, quite different emission anisotropy courses and the values of orientation factors should be obtained in both discussed cases. Such manipulations on donor – acceptor orientations may be interesting as they can deliver valuable lacking information on the mechanism of energy transfer processes in partly ordered systems. The aim of this work is to partly fill this gap and investigate energy transfer mechanism in two-component donor – acceptor systems in uniaxially oriented polymer films. To perform this both steady state and time resolved emission anisotropy measurements were carried out for disordered and uniaxially stretched polymer matrices. The analysis of the experimental results is supported by Monte Carlo simulations of emission anisotropy upon energy transfer and the orientation factor between the interacting species. The Monte Carlo method applied for the analysis of energy transfer mechanism has been in detail described previously [12,13]. For investigations three differently orientating donor – acceptor systems were selected: system I: Diethyloxycyanine Iodide (donor, DOCI) – Diethylthiacyanine Iodide (acceptor, DTCI) with parallel orientation of both donor and acceptor transition moments towards respective long molecular axis; system II: Diethyloxycyanine Iodide (donor, DOCI) – Rhodamine 101 (acceptor, R101) with parallel orientation of donor transition moments and no specific orientation of acceptor transition moments (plane, not elongated R101 molecule); and system III: Acridone (donor) – DOCI (acceptor) with the donor transition moment perpendicular towards the long molecular axis and acceptor transition moment parallel to long molecular axis. Molecular objects used in this study as donors and acceptors of excitation energy are highly fluorescent, stable and important probes in many applications. They are for example known as membrane potential probes, play a role of sensitizers in photodynamic therapy and are used in optical materials and laser technology to name just a few [14,15].
2. Experimental Analytically pure acridone, DOCI, DTCI, R101 and poly (vinyl alcohol) (PVA) were obtained from Fluka. The dyes were dissolved in 5% water – ethanol solution of PVA at temperature T = 323 K to obtain homogeneous solution. Then samples were left in a clean place to allow water evaporation and seasoned for about a week before stretching and measurements. Uniaxial stretching of the films was performed at a temperature T = 313 K using a simple device designed in our laboratory. The fluorescence signal was always recorded from the central small area of the sample. The optical density of the film was low enough to neglect the inner filter effects [16]. Absorption spectra were measured using Shimadzu 1650 spectrophotometer. Fluorescence spectra were measured upon so called front face excitation and observation of sample fluorescence using spectrofluorometer constructed in our laboratory and described previously [17]. Steady-state emission anisotropy was measured with two-channel single photon counting apparatus [18]. Time resolved emission anisotropy measurements were performed in the usual way with our pulsed spectrofluorometer especially dedicated to measurements from thin layers. Similarly to other measurements, the emission anisotropy decays were performed in the front face mode upon the excitation kexc = 375 nm and at kexc = 445 nm with the use of laser head LDH-D-C-375 with controller PDL 800-D, PCI-board for TCSPC TimeHarp 200 (PicoQuant, Germany). Fluorescence light was recorded by the H10721P-01 photomultiplier (Hamamatsu Photonics K.K., Japan)
combined with the slit of Czerny-Tuner spectrograph Shamrock 303i-B (Andor Technology, UK). Analysis was obtained with FluoFit Pro version (PicoQuant, Germany). The measurements were also independently performed with Fluotime 200 spectrofluorometer (PicoQuant).
3. Results and discussion Fig. 1abc shows absorption, fluorescence spectra recorded for (1a) DOCI and DTCI, (1b) DOCI and R101, (1c) acridone and DOCI. The fluorescence spectra were obtained for the following excitation wavelengths: DOCI (kexc = 460 nm), DTCI (kexc = 520 nm), R101 (kexc = 520 nm), acridone (kexc = 380 nm). The spectra presented in Fig. 1abc were taken in disordered polymer matrix. The process of stretching did not induce significant change in the steady-state absorption and fluorescence spectra at low concentration. Fig. 1abc presents also emission anisotropy spectra recorded for disordered (RS = 1–wheels) and uniaxially stretched PVA films (RS = 3 – triangles and RS = 5 – squares,) for (1a) DOCI – DTCI (1b) DOCI – R101 and (1c) acridone – DOCI systems. RS is the stretching factor of the film along the selected direction of stretching. The emission anisotropy spectra were obtained for the following excitation wavelengths: DOCI–DTCI (kexc = 450 nm), DOCI–R101 (kexc = 450 nm), acridone – DOCI (kexc = 380 nm). Mention should be made that the overlaps between the fluorescence spectrum of acceptor and absorption spectrum of the donor for DOCI–R101 and acridone –DOCI systems are negligible in each case considered, therefore the effect of reverse energy transfer from acceptor to donor does not occur [7,18]. Moreover, in the case of DOCI (donor) –DTCI (acceptor) system the overlap between the emission spectrum of DTCI and absorption spectrum of DOCI is also insignificant. The critical concentration (i.e. the concentration at which the probability of transfer is 1/2) for reverse energy transfer resulting from this overlap exceeds 0.1 M which means that at donor dye concentration in this work (CD = 0.001 M) the reverse energy transfer is unlikely to occur effectively. It can be seen from Fig. 1abc that for all disordered films emission anisotropy decreases rapidly when passing from the donor to the acceptor fluorescence band and at long wavelengths, where mostly acceptors emit fluorescence, it attains very low values (r ffi 0.05). This fact results from random distribution of fluorophores to which energy transfer from initially excited donors takes place in unstretched polymer matrix. This observation is in satisfactory agreement with former theoretical predictions [8,19]. It has been calculated that acceptors excited via single step energy transfer emit practically depolarized light r1 0.01. Our results obtained by high precision Monte Carlo technique are similar. For acceptors excited after single step energy transfer we obtained r1 = 0.0104 and after two steps further depolarization takes place r2 = 0.0011. Somewhat higher (r 0.05) experimental emission anisotropies recorded in the acceptor fluorescence band result from residual contribution of highly polarized donor fluorescence. However, emission anisotropy spectra differ completely for each uniaxially stretched system. Instead of fluorescence depolarization one can see in Fig. 1a that the emission anisotropy is preserved in the acceptor fluorescence band both for RS = 3 (moderately ordered system) and RS = 5 (strongly ordered system). The effect of emission anisotropy preservation after energy transfer results from the preferential orientation of transition dipole moments of elongated donors and acceptors taking part in energy transfer. It is well known that the transition moments of most of the elongated molecules are parallel to the long molecular axis and tend to orientate towards the direction of matrix stretching [20,21]. Slight repolarization effect observed in the acceptor
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
Fig. 1. Absorption, fluorescence spectra recorded for (1a) DOCI and DTCI (1b) DOCI and R101, (1c) acridone and DOCI. Figure 1a-c presents also emission anisotropy spectra recorded for disordered and uniaxially stretched PVA films for (1a) DOCI – DTCI (1b) DOCI – R101 and (1c) acridone – DOCI systems. The arrows illustrate respectively increasing or decreasing of the value of the emission anisotropy with increasing Rs. Measurements have been done for the following concentrations: (1a) CDOCI = 0.001 mol/l, CDTCI = 0.00025 mol/l, (1b) CDOCI = 0.0016 mol/l, CR101 = 0.0005 mol/l, (1c) Cacridone = 0.00055 mol/l, CDOCI = 0.001 mol/l.
fluorescence band results in this case from somewhat better ability of acceptor transition moments to orientate in the matrix as revealed by linear dichroism data (linear dichroism ratio RD = 1.8 for DOCI and RD = 3.2 for DTCI at Rs = 5) [22]. Simulated donor emission anisotropy for Rs = 5 at C = 0.001 M yield rD = 0.54,
3
whereas after single step energy transfer to acceptor rA = 0.65, which qualitatively reflects the tendency observed for experimental data (Fig. 1a). Fig. 1b shows in turn the results obtained for DOCI – R101 system. Transition moment of DOCI (the same donor as previously) exhibits quite effective orientation towards the direction of stretching. However, R101 is a nonlinear molecule which does not orientate effectively in the polymer matrix upon its stretching (RD = 1.2 at RS = 5). It means that despite matrix stretching the angular distribution of transition moments of all acceptors is only slightly nonrandom. Therefore, energy transfer from highly ordered set of transition moments of DOCI molecules takes place to almost randomly distributed transition moments of R101 resulting in pronounced fluorescence depolarization in the acceptor fluorescence band. In the case of rhodamine 101 the absorption and fluorescence transition moments are also practically parallel. This is clear as the measurement of limiting emission anisotropy of R101 in anhydrous glycerol gives r = 0.39 (C = 0.00001 M, T = 283 K), which is close to the fundamental value r = 0.4. The arrows in Fig. 1b describe the changes of emission anisotropy in the donor and acceptor emission bands. Due to much better ability of DOCI transition moments to orientate with the increasing matrix stretching, donor emission anisotropy increases much more significantly (longer left hand side arrow). However, it is worth of notice that the fluorescence observed in the acceptor emission band is partly polarized despite weak acceptor ability to orientate in the matrix. This fact is mostly due to the contribution of highly polarized residual donor fluorescence. It results also from high sensitivity of emission anisotropy to even relatively small order effects of transition moments due to the matrix stretching. Our results of Monte Carlo simulations yield that even for RD = 1.2 emission anisotropy increases in the case studied about 30% compared to disordered system (RD = 1). Fig. 1c shows in turn distinct fluorescence repolarization in the band of acceptor emission both for RS = 3 and RS = 5 in the system acridone (donor) – DOCI (acceptor). In this case energy transfer takes place from donor molecules with transition moments localized perpendicular to long molecular axis (parallel to short molecular axis) to acceptors, transition moments of which are localized parallel to long molecular axis. Again, long molecular axes of acridone and DOCI tend to orientate towards the axis of matrix stretching, however, transition moments of both species orientate quite differently upon matrix stretching. It can be seen from Fig. 1c that emission anisotropy in the donor emission band decreases with the increase of matrix stretching in view of more and more effective orientation of acridone transition moment in the direction perpendicular to the direction of stretching. This regularity is symbolically indicated by down pointing arrow in Fig. 1c and it is quite opposite compared to those presented in Fig 1a and b. The long up pointing arrow (on the right hand side of the figure) shows how distinct is the effect of fluorescence repolarization in the acceptor emission band when passing from disordered to uniaxially stretched polymer matrix. As a result, information on the orientation of the electric vector of the exciting light is partly regained when the observation is made in the acceptor fluorescence band. Again, simulated emission anisotropy of acceptor molecules receiving excitation energy after energy transfer from acridone are 0.37 and 0.39 for RS = 3 and RS = 5, respectively, which exceed strongly respective values of the donor 0.22 and 0.17. The comparison performed for the three systems shows the major role of donor and acceptor transition moments orientation in the emission anisotropy spectra course of two – component partly ordered systems in which energy transfer takes place. From the discussion made it is also clear that contrary to most of disordered systems the measurements of acceptor emission anisotropy in macroscopically ordered systems can deliver valuable
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
4
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
information on the mechanism of energy transport and fluorophores distribution. Table 1 presents the values of energy transfer parameters for the three systems of interest. In order to conduct systematically energy transfer studies it is necessary to determine so called critical distance R0 . The physical meaning of R0 is that this is the distance between molecules at which the transfer probability is equal to 1/2 [23]. The values of critical distances R0 can be easily obtained from the measurements of donor and acceptor emission and absorption spectra and determined from the following formula:
R60 ¼
9hj2 iðln 10Þg0D Iv ðD; AÞ 128p5 n4 N0
where
Z
Iv ðD; AÞ ¼
1
0
f D ðv ÞeA ðv Þ
ð1Þ
dv
ð2Þ
v4
is the overlap integral and goD is the absolute donor fluorescence quantum yield in the absence of acceptors, N0 is the Avogadro number per one mM (6.02 1020); hj2 i is the orientation factor averaged over all molecular configurations (averaged orientation factor), where j2 is a real orientation factor defined for a given donor– acceptor pair as:
j2 ¼ ðcos uDA 3 cos uD cos uA Þ2
ð3Þ
where uDA denotes the angle between the directions of donor transition moment in emission and acceptor transition moment in absorption, uD - the angle between donor transition moment in emission and the axis connecting the interacting molecules, uA the angle between acceptor transition moment in absorption and the axis connecting the interacting molecules. If the index DA in formula (1) is replaced by DD we obtain respective formula for critical distance and spectral overlap for energy migration between donors. It can be seen from Table 1 that the critical distances for energy migration and transfer are different in disordered and ordered systems for all the systems studied. These differences result mostly from a significant change in the averaged orientation factors. This is in agreement with results obtained from Monte Carlo simulations for disordered and uniaxially stretched films. It occurs that for the system DOCI – DTCI (well orientating donor and acceptor transition moments towards the axis of stretching) all values of hj2 i hj2DA i hj2DD i hj2AA i are comparable and in a given uniaxially stretched film exceed significantly that in disordered system. For all disordered systems (RS = 1) hj2 i ¼ 0:476 irrespectively of the system, which is in perfect agreement with the well known result obtained analytically for statistical distribution of immobile dipoles [10,24]. Higher values of hj2 i at higher RS result from enhanced orientation of transition moments towards the axis of polymer stretching. This situation
is different for the system acridone – DOCI. In the case of energy migration in the donor ensemble (acridone) the averaged orientation factor hj2DD i and the respective critical distance R0DD gradually decrease with increasing matrix stretching RS, which means that energy migration is somewhat weakened in that case compared to the random system. This is because the transition moment of acridone is localized perpendicular to long molecular axis and upon matrix stretching it also tends towards the perpendicular direction to the axis of matrix stretching. In a theoretically interesting case of extremely high (experimentally unavailable) matrix stretching both energy migration among acridone molecules and energy transfer from acridone to DOCI molecules gets very weak as respective transition moments of donors and acceptors form statistically large angles. This is also reflected by extremely long calculation time of Monte Carlo simulations of energy transfer due to very low energy transfer efficiency. In turn, for the DOCI – R101 system it can be seen that the critical distance R0AA (for energy migration in the acceptor ensemble) remains almost unaffected upon matrix stretching as the orientations of rhodamine 101 transition moments remain almost random despite matrix stretching. The critical distance R0DA (donor – acceptor energy transfer) increases with RS, which is mostly due to the strong order effect of donor (DOCI) transition moments upon film stretching. Figs. 2 and 3 show in turn selected results of donor and acceptor emission anisotropy decays for DOCI – DTCI and acridone – DOCI system, respectively. It can be seen from Fig. 2 that the matrix stretching changes completely the relation between dynamical behaviour in donor and acceptor ensemble. As expected, for disordered system (RS = 1) donor emission anisotropy decays much slower than sensibilized acceptor fluorescence anisotropy decay. In partly ordered polymer film acceptor emission anisotropy observed at 600 nm does not decay over the whole time window and even small repolarization of acceptor fluorescence can be observed, whereas the corresponding donor emission anisotropy observed at 510 nm (no acceptor emission signal) tends to decay slowly with time. The lack of decrease of acceptor emission anisotropy with time may be qualitatively understood as a result of very good orientation of acceptor molecules towards the direction of stretching and very weak influence of depolarization factors in this case. This very effective orientation (even significantly better than that of the donor molecules) was evidenced by the increase in the acceptor emission anisotropy band (Fig. 1a) and linear dichroism data. Very good mean orientation of acceptor molecules means also that they practically no not undergo even hindered rotational movements in the stretched matrix, which usually leads to some depolarization.
Table 1 Critical distances and averaged orientation factors for the systems: DOCI – DTCI, DOCI – R101, acridone – DOCI obtained for unstretched and uniaxially stretched PVA matrices. System
RS
R0DD (Å)
R0AA (Å)
R0DA (Å)
hj2DD i
hj2AA i
hj2DA i
DOCI-DTCI
1 3 5
42 45.6 49.3
48.8 53.3 57.4
47.8 52.1 51.6
0.476 0.799 1.19
0.476 0.805 1.26
0.476 0.801 1.25
DOCI-R101
1 3 5
42 45.6 49.3
56 56.2 56.3
40.1 41.8 42.2
0.476 0.799 1.19
0.476 0.476 0.476
0.476 0.61 0.68
Acridone-DOCI
1 3 5
27.4 26.5 26.0
42 45.6 49.3
46.1 47.8 45.4
0.476 0.388 0.350
0.476 0.799 1.19
0.476 0.493 0.432
Fig. 2. Emission anisotropy decays for DOCI – DTCI system at RS = 1 and RS = 5 measured within donor (kobs = 510 nm) and acceptor (kobs = 600 nm) fluorescence bands for kexc = 445 nm. CDOCI = 0.001 mol/l, CDTCI = 0.0005 mol/l.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
5
In particular, strong repolarization effect was found in partly ordered systems DOCI – DTCI and acridone – DOCI after energy transfer (acceptor fluorescence band) both in steady-state and time-resolved experiments. Such a behaviour of emission anisotropy is completely different compared to disordered systems, where strong depolarization is always observed as a result of energy transfer to randomly distributed acceptors. Monte Carlo simulations results of emission anisotropy of molecules excited via energy transfer and averaged orientation factor support our analysis. We believe that the analysis performed in stretched polymer films will help to design specific optical materials with controllable spectrally tuned polarization properties. Fig. 3. Emission anisotropy decays of acridone – DOCI system for uniaxially stretched polymer film (RS = 5) taken in the donor (kobs = 440 nm) and acceptor (kobs = 540 nm) fluorescence band for kexc = 375 nm. Cacridone = 0.00055 mol/l, CDOCI = 0.001 mol/l.
Also small acceptor concentration (0.0005 M) ensures the lack of fluorescence depolarization due to energy migration. At such small concentration (about one order of magnitude smaller than the critical concentration for this process) energy transfer acts in the ensemble of acceptors are very seldom. Therefore, there is no energy migration process and the influence of the concentration depolarization of fluorescence in this case can be neglected. However, small but clearly observable repolarization effect in the acceptor band over time may be explained by the energy transfer effect from initially excited donors to acceptors more effectively oriented than the donors relatively to the axis of stretching. Acceptors are ‘‘pumped” through energy transfer from donors at different time moments after initial donor excitation. The repolarization effect in the time – domain can be also illustrated by a somewhat different presentation visible in Fig. 3 for acridone – DOCI system, where relative emission anisotropy decays in the donor and acceptor band are shown. Sensibilized acceptor emission anisotropy (kobs = 540 nm) decays significantly slower that of the donor (kobs = 440 nm) due to the preferential orientation of acceptor transition moments towards the direction of stretching. Mention should be made that in this case acceptor emission anisotropy does not stay constant over the time window presented as effectiveness of ordering of DOCI transition moments is lower than those of DTCI (as revealed by linear dichroism data) [22]. 4. Conclusions Excitation energy transfer was studied in selected donoracceptor systems of controllable order degree and different orientation of donor-acceptor transition moments. Emission anisotropy spectra occurred drastically different between stretched and unstretched matrices as a result of induced reorientation of transition moments of molecules participating in energy transfer. Each of three ordered systems behaved differently as a result of specific orientation of donor and/or acceptor transition moments.
Acknowledgment This research has been supported by the grant 2015/17/B/ ST5/03143 financed by National Science Centre (P. Bojarski, A. Synak, B. Grobelna). I. Gryczynski acknowledges support from Fulbright Foundation; S. Rangełowa - Jankowska acknowledges support from European Social Fund and the Foundation for the Development of Gdansk University. References [1] S. Ramakrishna, Z. Ma, T. Matsurra, Polymer Membranes in Biotechnology, Imperial College Press, 2011. [2] V. Mittal (Ed.), Advanced Polymer Nanoparticles, CRC Press, 2011. [3] J. Michl, E.W. Thulstrup, Spectroscopy With Polarized Light, VCH Publishers Inc., 1986. [4] I. Gryczynski, Z. Gryczynski, W. Wiczk, J. Kusba, J.R. Lakowicz, SPIE 1640 (1992) 622. [5] Stephen Z.D. Cheng, Fuming Li, E.P. Savitski, F.W. Harris, Trends Polym. Sci. 5 (1997) 51. [6] C. Bojarski, K. Sienicki, Energy transfer and migration in fluorescent solution, in: J.A. Rabek (Ed.), Photophysics and Photochemistry, CRC Press, Boca Raton, 1990, p. 1. [7] P. Bojarski, L. Kułak, Chem. Phys. 220 (1997) 323. [8] A. Jabłon´ski, Acta Phys. Pol., A 38 (1970) 453; A. Jabłon´ski, Acta Phys. Pol., A 39 (1971) 87. [9] P. Bojarski, A. Kamin´ska, L. Kułak, M. Sadownik, Chem. Phys. Lett. 375 (2003) 547. [10] M. Sadownik, P. Bojarski, Chem. Phys. Lett. 396 (2004) 293. [11] P. Bojarski, I. Gryczyn´ski, L. Kułak, A. Synak, A. Barnett, Chem. Phys. Lett. 431 (2006) 94. [12] L. Kułak, Chem. Phys. Lett. 457 (2008) 259. [13] A. Synak, P. Bojarski, B. Grobelna, L. Kułak, A. Lewkowicz, J. Phys. Chem. C 117 (2013) 11385. [14] M. Maeda, Lasers Dyes, Academic Press, Tokyo, 1984. [15] M. Krieg, R.W. Redmond, Photobiology 57 (57) (1993) 472. [16] I. Ketskemety, J. Dombi, R. Horvai, J. Hevesi, L. Kozma, Acta Phys. Chem. (Szeged) 7 (1961) 17. [17] A. Kawski, G. Piszczek, B. Kuklin´ski, T. Nowosielski, Z. Naturforsch. 49a (1994) 824. [18] P. Bojarski, A. Kawski, J. Fluorescence 2 (1992) 133. [19] P. Bojarski, L. Kułak, C. Bojarski, A. Kawski, J. Fluorescence 5 (1995) 307. [20] J. Michl, E.W. Thulstrup, Spectroscopy with Polarized Light, V.C.H. Publ. Inc., 1986. [21] A. Kawski, Z. Gryczyn´ski, Z. Naturforsch. 41a (1986) 1195; A. Kawski, Z. Gryczyn´ski, Z. Naturforsch. 42a (1987) 808. [22] A. Synak, P. Bojarski, Chem. Phys. Lett. 416 (2005) 300. [23] Th. Förster, Ann. Physik 2 (1948) 55. [24] I.Z. Steinberg, Chem. Phys. 48 (1968) 2411.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Excitation energy transfer in partly ordered polymer films differing in donor and acceptor transition moments orientation A. Synak a,⇑, P. Bojarski a, M. Sadownik a, L. Kułak b, I. Gryczynski a, B. Grobelna c, S. Rangełowa-Jankowska a, D. Jankowski a, A. Kubicki a a b c
University of Gdansk, Faculty of Mathematics, Physics and Informatics, Institute of Experimental Physics, Wita Stwosza 57, 80-952 Gdansk, Poland Gdansk University of Technology, Faculty of Technical Physics and Applied Mathematics, Narutowicza 11/12, 80-952 Gdansk, Poland University of Gdansk, Faculty of Chemistry, Wita Stwosza 63, 80-308 Gdansk, Poland
a r t i c l e
i n f o
Article history: Received 27 October 2015 Received in revised form 19 January 2016 Accepted 1 March 2016 Available online xxxx Keywords: Energy transfer Emission anisotropy PVA films Partly ordered systems Monte Carlo simulations
a b s t r a c t Based on spectroscopic measurements selected properties of nonradiative Förster energy transport are studied in uniaxially stretched polyvinyl alcohol thin films for three systems differing in donor and acceptor transition moments orientation relative to the axis of stretching. In particular, donor – acceptor emission anisotropy spectra yield completely different regularities for these systems in uniaxially stretched films, whereas they are similar in unstretched films. In particular it is shown that acceptor fluorescence can be either strongly polarized after nonradiative energy transfer in stretched films or depolarized depending on the angular distribution of acceptor transition moments in the matrix. Donor and acceptor emission anisotropy decays exhibit similar regularities to those of steady-state measurements. The obtained results are analyzed with the help of Monte Carlo simulations. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction Polymer films have numerous applications in various industrial and biomedical sectors such as functional and protective coatings, biocompatible medical implants, advanced membranes, sensors [1,2]. Ordered polymer films doped with fluorophores are well known materials used in polarization spectroscopy, to determine transition moments directions of molecules, their conformations or to study properties of nonradiative energy transport process [3–5]. Excitation energy transport has been widely studied for donor and acceptor transition moments randomly distributed in space [6,7]. It is well known that for such disordered systems sensibilized fluorescence emitted by acceptors excited exclusively through energy transfer from donors is almost totally depolarized [8]. This is because only primarily excited molecules (by linearly polarized light) contribute to emission anisotropy. However, as shown recently, quite different observations have been reported for uniaxially oriented polymer films, where emission anisotropy of molecules excited via energy migration or transfer remain high [9–11]. As a result of mechanical uniaxial stretching of the polymer film along fixed axis, the directions of transition dipole moments of ⇑ Corresponding author.
elongated fluorophores loose their random orientation and exhibit certain nonrandom distribution relative to the axis of stretching. This means, in turn, that energy transfer takes place from an initially excited donor to an acceptor, transition moment direction of which is correlated with that of donor. On a macroscale, we can consider the value of the averaged orientation factor as a certain measure of these correlations. The averaging over all molecular configurations can lead to a much more pronounced change of the orientation factor in such a macroscopically nonrandom system compared to the random systems or systems which are locally nonrandom. The angular correlations between the directions of transition dipole moments are mostly responsible for the effect of emission anisotropy preservation in uniaxially stretched one component polymer films, where the excitation energy visits many chemically identical sites an arbitrary number of times [9–11]. Unlike one – component systems, one can imagine different physical situations in uniaxially oriented polymer matrices doped with donors and acceptors of excitation energy. Chemically nonidentical donors and acceptors can be selected in such a way that their transition moments are located differently towards respective long molecular axes. For example, if transition moments of both donor and acceptor are parallel to their long molecular axes, then they orientate quite efficiently upon matrix stretching towards this axis of stretching. On the other hand, if transition moment of the donor is perpendicular to long molecular axis, but
E-mail address: [email protected] (A. Synak). http://dx.doi.org/10.1016/j.optmat.2016.03.005 0925-3467/Ó 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
2
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
acceptor transition moment is parallel to its long molecular axis, then upon matrix stretching both orientations of donor and acceptor transition moments are totally different. As a result, quite different emission anisotropy courses and the values of orientation factors should be obtained in both discussed cases. Such manipulations on donor – acceptor orientations may be interesting as they can deliver valuable lacking information on the mechanism of energy transfer processes in partly ordered systems. The aim of this work is to partly fill this gap and investigate energy transfer mechanism in two-component donor – acceptor systems in uniaxially oriented polymer films. To perform this both steady state and time resolved emission anisotropy measurements were carried out for disordered and uniaxially stretched polymer matrices. The analysis of the experimental results is supported by Monte Carlo simulations of emission anisotropy upon energy transfer and the orientation factor between the interacting species. The Monte Carlo method applied for the analysis of energy transfer mechanism has been in detail described previously [12,13]. For investigations three differently orientating donor – acceptor systems were selected: system I: Diethyloxycyanine Iodide (donor, DOCI) – Diethylthiacyanine Iodide (acceptor, DTCI) with parallel orientation of both donor and acceptor transition moments towards respective long molecular axis; system II: Diethyloxycyanine Iodide (donor, DOCI) – Rhodamine 101 (acceptor, R101) with parallel orientation of donor transition moments and no specific orientation of acceptor transition moments (plane, not elongated R101 molecule); and system III: Acridone (donor) – DOCI (acceptor) with the donor transition moment perpendicular towards the long molecular axis and acceptor transition moment parallel to long molecular axis. Molecular objects used in this study as donors and acceptors of excitation energy are highly fluorescent, stable and important probes in many applications. They are for example known as membrane potential probes, play a role of sensitizers in photodynamic therapy and are used in optical materials and laser technology to name just a few [14,15].
2. Experimental Analytically pure acridone, DOCI, DTCI, R101 and poly (vinyl alcohol) (PVA) were obtained from Fluka. The dyes were dissolved in 5% water – ethanol solution of PVA at temperature T = 323 K to obtain homogeneous solution. Then samples were left in a clean place to allow water evaporation and seasoned for about a week before stretching and measurements. Uniaxial stretching of the films was performed at a temperature T = 313 K using a simple device designed in our laboratory. The fluorescence signal was always recorded from the central small area of the sample. The optical density of the film was low enough to neglect the inner filter effects [16]. Absorption spectra were measured using Shimadzu 1650 spectrophotometer. Fluorescence spectra were measured upon so called front face excitation and observation of sample fluorescence using spectrofluorometer constructed in our laboratory and described previously [17]. Steady-state emission anisotropy was measured with two-channel single photon counting apparatus [18]. Time resolved emission anisotropy measurements were performed in the usual way with our pulsed spectrofluorometer especially dedicated to measurements from thin layers. Similarly to other measurements, the emission anisotropy decays were performed in the front face mode upon the excitation kexc = 375 nm and at kexc = 445 nm with the use of laser head LDH-D-C-375 with controller PDL 800-D, PCI-board for TCSPC TimeHarp 200 (PicoQuant, Germany). Fluorescence light was recorded by the H10721P-01 photomultiplier (Hamamatsu Photonics K.K., Japan)
combined with the slit of Czerny-Tuner spectrograph Shamrock 303i-B (Andor Technology, UK). Analysis was obtained with FluoFit Pro version (PicoQuant, Germany). The measurements were also independently performed with Fluotime 200 spectrofluorometer (PicoQuant).
3. Results and discussion Fig. 1abc shows absorption, fluorescence spectra recorded for (1a) DOCI and DTCI, (1b) DOCI and R101, (1c) acridone and DOCI. The fluorescence spectra were obtained for the following excitation wavelengths: DOCI (kexc = 460 nm), DTCI (kexc = 520 nm), R101 (kexc = 520 nm), acridone (kexc = 380 nm). The spectra presented in Fig. 1abc were taken in disordered polymer matrix. The process of stretching did not induce significant change in the steady-state absorption and fluorescence spectra at low concentration. Fig. 1abc presents also emission anisotropy spectra recorded for disordered (RS = 1–wheels) and uniaxially stretched PVA films (RS = 3 – triangles and RS = 5 – squares,) for (1a) DOCI – DTCI (1b) DOCI – R101 and (1c) acridone – DOCI systems. RS is the stretching factor of the film along the selected direction of stretching. The emission anisotropy spectra were obtained for the following excitation wavelengths: DOCI–DTCI (kexc = 450 nm), DOCI–R101 (kexc = 450 nm), acridone – DOCI (kexc = 380 nm). Mention should be made that the overlaps between the fluorescence spectrum of acceptor and absorption spectrum of the donor for DOCI–R101 and acridone –DOCI systems are negligible in each case considered, therefore the effect of reverse energy transfer from acceptor to donor does not occur [7,18]. Moreover, in the case of DOCI (donor) –DTCI (acceptor) system the overlap between the emission spectrum of DTCI and absorption spectrum of DOCI is also insignificant. The critical concentration (i.e. the concentration at which the probability of transfer is 1/2) for reverse energy transfer resulting from this overlap exceeds 0.1 M which means that at donor dye concentration in this work (CD = 0.001 M) the reverse energy transfer is unlikely to occur effectively. It can be seen from Fig. 1abc that for all disordered films emission anisotropy decreases rapidly when passing from the donor to the acceptor fluorescence band and at long wavelengths, where mostly acceptors emit fluorescence, it attains very low values (r ffi 0.05). This fact results from random distribution of fluorophores to which energy transfer from initially excited donors takes place in unstretched polymer matrix. This observation is in satisfactory agreement with former theoretical predictions [8,19]. It has been calculated that acceptors excited via single step energy transfer emit practically depolarized light r1 0.01. Our results obtained by high precision Monte Carlo technique are similar. For acceptors excited after single step energy transfer we obtained r1 = 0.0104 and after two steps further depolarization takes place r2 = 0.0011. Somewhat higher (r 0.05) experimental emission anisotropies recorded in the acceptor fluorescence band result from residual contribution of highly polarized donor fluorescence. However, emission anisotropy spectra differ completely for each uniaxially stretched system. Instead of fluorescence depolarization one can see in Fig. 1a that the emission anisotropy is preserved in the acceptor fluorescence band both for RS = 3 (moderately ordered system) and RS = 5 (strongly ordered system). The effect of emission anisotropy preservation after energy transfer results from the preferential orientation of transition dipole moments of elongated donors and acceptors taking part in energy transfer. It is well known that the transition moments of most of the elongated molecules are parallel to the long molecular axis and tend to orientate towards the direction of matrix stretching [20,21]. Slight repolarization effect observed in the acceptor
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
Fig. 1. Absorption, fluorescence spectra recorded for (1a) DOCI and DTCI (1b) DOCI and R101, (1c) acridone and DOCI. Figure 1a-c presents also emission anisotropy spectra recorded for disordered and uniaxially stretched PVA films for (1a) DOCI – DTCI (1b) DOCI – R101 and (1c) acridone – DOCI systems. The arrows illustrate respectively increasing or decreasing of the value of the emission anisotropy with increasing Rs. Measurements have been done for the following concentrations: (1a) CDOCI = 0.001 mol/l, CDTCI = 0.00025 mol/l, (1b) CDOCI = 0.0016 mol/l, CR101 = 0.0005 mol/l, (1c) Cacridone = 0.00055 mol/l, CDOCI = 0.001 mol/l.
fluorescence band results in this case from somewhat better ability of acceptor transition moments to orientate in the matrix as revealed by linear dichroism data (linear dichroism ratio RD = 1.8 for DOCI and RD = 3.2 for DTCI at Rs = 5) [22]. Simulated donor emission anisotropy for Rs = 5 at C = 0.001 M yield rD = 0.54,
3
whereas after single step energy transfer to acceptor rA = 0.65, which qualitatively reflects the tendency observed for experimental data (Fig. 1a). Fig. 1b shows in turn the results obtained for DOCI – R101 system. Transition moment of DOCI (the same donor as previously) exhibits quite effective orientation towards the direction of stretching. However, R101 is a nonlinear molecule which does not orientate effectively in the polymer matrix upon its stretching (RD = 1.2 at RS = 5). It means that despite matrix stretching the angular distribution of transition moments of all acceptors is only slightly nonrandom. Therefore, energy transfer from highly ordered set of transition moments of DOCI molecules takes place to almost randomly distributed transition moments of R101 resulting in pronounced fluorescence depolarization in the acceptor fluorescence band. In the case of rhodamine 101 the absorption and fluorescence transition moments are also practically parallel. This is clear as the measurement of limiting emission anisotropy of R101 in anhydrous glycerol gives r = 0.39 (C = 0.00001 M, T = 283 K), which is close to the fundamental value r = 0.4. The arrows in Fig. 1b describe the changes of emission anisotropy in the donor and acceptor emission bands. Due to much better ability of DOCI transition moments to orientate with the increasing matrix stretching, donor emission anisotropy increases much more significantly (longer left hand side arrow). However, it is worth of notice that the fluorescence observed in the acceptor emission band is partly polarized despite weak acceptor ability to orientate in the matrix. This fact is mostly due to the contribution of highly polarized residual donor fluorescence. It results also from high sensitivity of emission anisotropy to even relatively small order effects of transition moments due to the matrix stretching. Our results of Monte Carlo simulations yield that even for RD = 1.2 emission anisotropy increases in the case studied about 30% compared to disordered system (RD = 1). Fig. 1c shows in turn distinct fluorescence repolarization in the band of acceptor emission both for RS = 3 and RS = 5 in the system acridone (donor) – DOCI (acceptor). In this case energy transfer takes place from donor molecules with transition moments localized perpendicular to long molecular axis (parallel to short molecular axis) to acceptors, transition moments of which are localized parallel to long molecular axis. Again, long molecular axes of acridone and DOCI tend to orientate towards the axis of matrix stretching, however, transition moments of both species orientate quite differently upon matrix stretching. It can be seen from Fig. 1c that emission anisotropy in the donor emission band decreases with the increase of matrix stretching in view of more and more effective orientation of acridone transition moment in the direction perpendicular to the direction of stretching. This regularity is symbolically indicated by down pointing arrow in Fig. 1c and it is quite opposite compared to those presented in Fig 1a and b. The long up pointing arrow (on the right hand side of the figure) shows how distinct is the effect of fluorescence repolarization in the acceptor emission band when passing from disordered to uniaxially stretched polymer matrix. As a result, information on the orientation of the electric vector of the exciting light is partly regained when the observation is made in the acceptor fluorescence band. Again, simulated emission anisotropy of acceptor molecules receiving excitation energy after energy transfer from acridone are 0.37 and 0.39 for RS = 3 and RS = 5, respectively, which exceed strongly respective values of the donor 0.22 and 0.17. The comparison performed for the three systems shows the major role of donor and acceptor transition moments orientation in the emission anisotropy spectra course of two – component partly ordered systems in which energy transfer takes place. From the discussion made it is also clear that contrary to most of disordered systems the measurements of acceptor emission anisotropy in macroscopically ordered systems can deliver valuable
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
4
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
information on the mechanism of energy transport and fluorophores distribution. Table 1 presents the values of energy transfer parameters for the three systems of interest. In order to conduct systematically energy transfer studies it is necessary to determine so called critical distance R0 . The physical meaning of R0 is that this is the distance between molecules at which the transfer probability is equal to 1/2 [23]. The values of critical distances R0 can be easily obtained from the measurements of donor and acceptor emission and absorption spectra and determined from the following formula:
R60 ¼
9hj2 iðln 10Þg0D Iv ðD; AÞ 128p5 n4 N0
where
Z
Iv ðD; AÞ ¼
1
0
f D ðv ÞeA ðv Þ
ð1Þ
dv
ð2Þ
v4
is the overlap integral and goD is the absolute donor fluorescence quantum yield in the absence of acceptors, N0 is the Avogadro number per one mM (6.02 1020); hj2 i is the orientation factor averaged over all molecular configurations (averaged orientation factor), where j2 is a real orientation factor defined for a given donor– acceptor pair as:
j2 ¼ ðcos uDA 3 cos uD cos uA Þ2
ð3Þ
where uDA denotes the angle between the directions of donor transition moment in emission and acceptor transition moment in absorption, uD - the angle between donor transition moment in emission and the axis connecting the interacting molecules, uA the angle between acceptor transition moment in absorption and the axis connecting the interacting molecules. If the index DA in formula (1) is replaced by DD we obtain respective formula for critical distance and spectral overlap for energy migration between donors. It can be seen from Table 1 that the critical distances for energy migration and transfer are different in disordered and ordered systems for all the systems studied. These differences result mostly from a significant change in the averaged orientation factors. This is in agreement with results obtained from Monte Carlo simulations for disordered and uniaxially stretched films. It occurs that for the system DOCI – DTCI (well orientating donor and acceptor transition moments towards the axis of stretching) all values of hj2 i hj2DA i hj2DD i hj2AA i are comparable and in a given uniaxially stretched film exceed significantly that in disordered system. For all disordered systems (RS = 1) hj2 i ¼ 0:476 irrespectively of the system, which is in perfect agreement with the well known result obtained analytically for statistical distribution of immobile dipoles [10,24]. Higher values of hj2 i at higher RS result from enhanced orientation of transition moments towards the axis of polymer stretching. This situation
is different for the system acridone – DOCI. In the case of energy migration in the donor ensemble (acridone) the averaged orientation factor hj2DD i and the respective critical distance R0DD gradually decrease with increasing matrix stretching RS, which means that energy migration is somewhat weakened in that case compared to the random system. This is because the transition moment of acridone is localized perpendicular to long molecular axis and upon matrix stretching it also tends towards the perpendicular direction to the axis of matrix stretching. In a theoretically interesting case of extremely high (experimentally unavailable) matrix stretching both energy migration among acridone molecules and energy transfer from acridone to DOCI molecules gets very weak as respective transition moments of donors and acceptors form statistically large angles. This is also reflected by extremely long calculation time of Monte Carlo simulations of energy transfer due to very low energy transfer efficiency. In turn, for the DOCI – R101 system it can be seen that the critical distance R0AA (for energy migration in the acceptor ensemble) remains almost unaffected upon matrix stretching as the orientations of rhodamine 101 transition moments remain almost random despite matrix stretching. The critical distance R0DA (donor – acceptor energy transfer) increases with RS, which is mostly due to the strong order effect of donor (DOCI) transition moments upon film stretching. Figs. 2 and 3 show in turn selected results of donor and acceptor emission anisotropy decays for DOCI – DTCI and acridone – DOCI system, respectively. It can be seen from Fig. 2 that the matrix stretching changes completely the relation between dynamical behaviour in donor and acceptor ensemble. As expected, for disordered system (RS = 1) donor emission anisotropy decays much slower than sensibilized acceptor fluorescence anisotropy decay. In partly ordered polymer film acceptor emission anisotropy observed at 600 nm does not decay over the whole time window and even small repolarization of acceptor fluorescence can be observed, whereas the corresponding donor emission anisotropy observed at 510 nm (no acceptor emission signal) tends to decay slowly with time. The lack of decrease of acceptor emission anisotropy with time may be qualitatively understood as a result of very good orientation of acceptor molecules towards the direction of stretching and very weak influence of depolarization factors in this case. This very effective orientation (even significantly better than that of the donor molecules) was evidenced by the increase in the acceptor emission anisotropy band (Fig. 1a) and linear dichroism data. Very good mean orientation of acceptor molecules means also that they practically no not undergo even hindered rotational movements in the stretched matrix, which usually leads to some depolarization.
Table 1 Critical distances and averaged orientation factors for the systems: DOCI – DTCI, DOCI – R101, acridone – DOCI obtained for unstretched and uniaxially stretched PVA matrices. System
RS
R0DD (Å)
R0AA (Å)
R0DA (Å)
hj2DD i
hj2AA i
hj2DA i
DOCI-DTCI
1 3 5
42 45.6 49.3
48.8 53.3 57.4
47.8 52.1 51.6
0.476 0.799 1.19
0.476 0.805 1.26
0.476 0.801 1.25
DOCI-R101
1 3 5
42 45.6 49.3
56 56.2 56.3
40.1 41.8 42.2
0.476 0.799 1.19
0.476 0.476 0.476
0.476 0.61 0.68
Acridone-DOCI
1 3 5
27.4 26.5 26.0
42 45.6 49.3
46.1 47.8 45.4
0.476 0.388 0.350
0.476 0.799 1.19
0.476 0.493 0.432
Fig. 2. Emission anisotropy decays for DOCI – DTCI system at RS = 1 and RS = 5 measured within donor (kobs = 510 nm) and acceptor (kobs = 600 nm) fluorescence bands for kexc = 445 nm. CDOCI = 0.001 mol/l, CDTCI = 0.0005 mol/l.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005
A. Synak et al. / Optical Materials xxx (2016) xxx–xxx
5
In particular, strong repolarization effect was found in partly ordered systems DOCI – DTCI and acridone – DOCI after energy transfer (acceptor fluorescence band) both in steady-state and time-resolved experiments. Such a behaviour of emission anisotropy is completely different compared to disordered systems, where strong depolarization is always observed as a result of energy transfer to randomly distributed acceptors. Monte Carlo simulations results of emission anisotropy of molecules excited via energy transfer and averaged orientation factor support our analysis. We believe that the analysis performed in stretched polymer films will help to design specific optical materials with controllable spectrally tuned polarization properties. Fig. 3. Emission anisotropy decays of acridone – DOCI system for uniaxially stretched polymer film (RS = 5) taken in the donor (kobs = 440 nm) and acceptor (kobs = 540 nm) fluorescence band for kexc = 375 nm. Cacridone = 0.00055 mol/l, CDOCI = 0.001 mol/l.
Also small acceptor concentration (0.0005 M) ensures the lack of fluorescence depolarization due to energy migration. At such small concentration (about one order of magnitude smaller than the critical concentration for this process) energy transfer acts in the ensemble of acceptors are very seldom. Therefore, there is no energy migration process and the influence of the concentration depolarization of fluorescence in this case can be neglected. However, small but clearly observable repolarization effect in the acceptor band over time may be explained by the energy transfer effect from initially excited donors to acceptors more effectively oriented than the donors relatively to the axis of stretching. Acceptors are ‘‘pumped” through energy transfer from donors at different time moments after initial donor excitation. The repolarization effect in the time – domain can be also illustrated by a somewhat different presentation visible in Fig. 3 for acridone – DOCI system, where relative emission anisotropy decays in the donor and acceptor band are shown. Sensibilized acceptor emission anisotropy (kobs = 540 nm) decays significantly slower that of the donor (kobs = 440 nm) due to the preferential orientation of acceptor transition moments towards the direction of stretching. Mention should be made that in this case acceptor emission anisotropy does not stay constant over the time window presented as effectiveness of ordering of DOCI transition moments is lower than those of DTCI (as revealed by linear dichroism data) [22]. 4. Conclusions Excitation energy transfer was studied in selected donoracceptor systems of controllable order degree and different orientation of donor-acceptor transition moments. Emission anisotropy spectra occurred drastically different between stretched and unstretched matrices as a result of induced reorientation of transition moments of molecules participating in energy transfer. Each of three ordered systems behaved differently as a result of specific orientation of donor and/or acceptor transition moments.
Acknowledgment This research has been supported by the grant 2015/17/B/ ST5/03143 financed by National Science Centre (P. Bojarski, A. Synak, B. Grobelna). I. Gryczynski acknowledges support from Fulbright Foundation; S. Rangełowa - Jankowska acknowledges support from European Social Fund and the Foundation for the Development of Gdansk University. References [1] S. Ramakrishna, Z. Ma, T. Matsurra, Polymer Membranes in Biotechnology, Imperial College Press, 2011. [2] V. Mittal (Ed.), Advanced Polymer Nanoparticles, CRC Press, 2011. [3] J. Michl, E.W. Thulstrup, Spectroscopy With Polarized Light, VCH Publishers Inc., 1986. [4] I. Gryczynski, Z. Gryczynski, W. Wiczk, J. Kusba, J.R. Lakowicz, SPIE 1640 (1992) 622. [5] Stephen Z.D. Cheng, Fuming Li, E.P. Savitski, F.W. Harris, Trends Polym. Sci. 5 (1997) 51. [6] C. Bojarski, K. Sienicki, Energy transfer and migration in fluorescent solution, in: J.A. Rabek (Ed.), Photophysics and Photochemistry, CRC Press, Boca Raton, 1990, p. 1. [7] P. Bojarski, L. Kułak, Chem. Phys. 220 (1997) 323. [8] A. Jabłon´ski, Acta Phys. Pol., A 38 (1970) 453; A. Jabłon´ski, Acta Phys. Pol., A 39 (1971) 87. [9] P. Bojarski, A. Kamin´ska, L. Kułak, M. Sadownik, Chem. Phys. Lett. 375 (2003) 547. [10] M. Sadownik, P. Bojarski, Chem. Phys. Lett. 396 (2004) 293. [11] P. Bojarski, I. Gryczyn´ski, L. Kułak, A. Synak, A. Barnett, Chem. Phys. Lett. 431 (2006) 94. [12] L. Kułak, Chem. Phys. Lett. 457 (2008) 259. [13] A. Synak, P. Bojarski, B. Grobelna, L. Kułak, A. Lewkowicz, J. Phys. Chem. C 117 (2013) 11385. [14] M. Maeda, Lasers Dyes, Academic Press, Tokyo, 1984. [15] M. Krieg, R.W. Redmond, Photobiology 57 (57) (1993) 472. [16] I. Ketskemety, J. Dombi, R. Horvai, J. Hevesi, L. Kozma, Acta Phys. Chem. (Szeged) 7 (1961) 17. [17] A. Kawski, G. Piszczek, B. Kuklin´ski, T. Nowosielski, Z. Naturforsch. 49a (1994) 824. [18] P. Bojarski, A. Kawski, J. Fluorescence 2 (1992) 133. [19] P. Bojarski, L. Kułak, C. Bojarski, A. Kawski, J. Fluorescence 5 (1995) 307. [20] J. Michl, E.W. Thulstrup, Spectroscopy with Polarized Light, V.C.H. Publ. Inc., 1986. [21] A. Kawski, Z. Gryczyn´ski, Z. Naturforsch. 41a (1986) 1195; A. Kawski, Z. Gryczyn´ski, Z. Naturforsch. 42a (1987) 808. [22] A. Synak, P. Bojarski, Chem. Phys. Lett. 416 (2005) 300. [23] Th. Förster, Ann. Physik 2 (1948) 55. [24] I.Z. Steinberg, Chem. Phys. 48 (1968) 2411.
Please cite this article in press as: A. Synak et al., Opt. Mater. (2016), http://dx.doi.org/10.1016/j.optmat.2016.03.005