European Polymer Journal
-
Upload
rafael-cossiello -
Category
Technology
-
view
983 -
download
1
description
Transcript of European Polymer Journal
1
2
3
4
5
6
789
10
11
1 3
141516171819
2021222324252627
2 8
53
54
55
56
57
European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
Contents lists available at ScienceDirect
European Polymer Journal
journal homepage: www.elsevier .com/locate /europol j
PR
OO
FTemperature dependence of molecular dynamics and supramolecularaggregation in MEH-PPV films: A solid-state NMR, X-ray and fluorescencespectroscopy study
A.A. Souza a, R.F. Cossiello b, T.S. Plivelic c, G.L. Mantovani a, G.C. Faria a, T.D.Z. Atvars b,I.L. Torriani c,d, T.J. Bonagamba a, E.R. deAzevedo a,*
a Instituto de Física de São Carlos, Universidade de São Paulo, Caixa Postal 369, 13560-970 São Carlos, SP, Brazilb Instituto de Química, Universidade Estadual de Campinas, Caixa Postal 6154, 13084-971 Campinas, SP, Brazilc Laboratório Nacional de Luz Síncrotron, Caixa Postal 6192, 13083-970 Campinas, SP, Brazild Instituto de Física, Universidade Estadual de Campinas, Caixa Postal 6165, 13084-971 Campinas, SP, Brazil
2930313233343536373839404142
a r t i c l e i n f o
Article history:Received 29 June 2008Received in revised form 23 August 2008Accepted 16 September 2008Available online xxxx
Keywords:MEH-PPV filmsNMRDIPSHIFTWAXSMolecular aggregationFluorescence
U
434445464748495051
0014-3057/$ - see front matter � 2008 Elsevier Ltddoi:10.1016/j.eurpolymj.2008.09.030
* Corresponding author. Tel.: +55 16 33738086; fE-mail address: [email protected] (E.R. deAzev
Please cite this article in press as: SouzaEur Poly J (2008), doi:10.1016/j.eurpolym
OR
REC
TED
a b s t r a c t
This article presents an investigation of the temperature induced modification in themicrostructure and dynamics of poly[2-methoxy-5-(20-ethylhexyloxy)-1,4-phenylenevin-ylene] (MEH-PPV) cast films using Wide-Angle X-ray Scattering (WAXS), solid-stateNuclear Magnetic Resonance (NMR), and Fluorescence Spectroscopy (PL). MEH-PPV chainmotions were characterized as a function of temperature by NMR. The results indicatedthat the solvent used to cast the films influences the activation energy of the side-chainmotions. This was concluded from the comparison of the activation energy of the toluenecast film, Ea = (54 ± 8) kJ/mol, and chloroform cast film, Ea = (69 ± 5) kJ/mol, and could beattributed to the higher side-chain packing provided by chloroform that preferentially sol-vates the side chain in contrast to toluene that solvates mainly the backbone. Concerningthe backbone mobility, it was observed that the torsional motions in the MEH-PPV haveaverage amplitude of �10� at 300 K, which was found to be independent of the solventused to cast the films. In order to correlate the molecular dynamics processes with thechanges in the microstructure of the polymer, in situ WAXS experiments as a function oftemperature were performed and revealed that the interchain spacing in the MEH-PPVmolecular aggregates increases as a function of temperature, particularly at temperatureswhere molecular relaxations occur. It was also observed that the WAXS peak associatedwith the bilayer spacing, narrows and their by increases intensity whereas the peak asso-ciated with the interbackbone planes reduces its intensity for higher temperatures. Thislast result could be interpreted as a decrease in the number of aggregates and the reductionof the interchain species during the MEH-PPV relaxation processes. These WAXS resultswere correlated with PL spectra modifications observed upon temperature treatments.
� 2008 Elsevier Ltd. All rights reserved.
52
C58
59
60
61
62
N1. Introduction
Conjugated polymers have potential to be employed inseveral applications such as materials for light-emittingdiodes, lasers and thin-film transistors due to their elec-
63
64
65
. All rights reserved.
ax: +55 1633739876.edo).
AA et al., Temperature dj.2008.09.030
tron and hole transport properties [1–4]. The large-scaleelectroluminescent devices can be feasible due to theirsimple production. However, the efficiency of the electro-luminescence (EL) and photoluminescence (PL) is reducedwhen the conjugated polymers form interchain speciesthat arise as a result of molecular aggregation [5,6]. Molec-ular aggregation and, consequently, the electro- andphoto-luminescence emissions and charge transport
ependence of molecular dynamics and supramolecular ...,
E
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
2 A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
UN
CO
RR
properties may be partially controlled by the processingconditions which change the polymer microstructure andmorphology [7–9].
Photoluminescence and electroluminescence are corre-lated phenomena not only strongly dependent on themicrostructure of the material but also on the dynamicsof the polymer chains [7–9]. For example, the temperaturedependence of the electroluminescent devices with MEH-PPV showed changes [10–12] associated with specificmovements of the polymer chain such as motions of thelateral groups at 220 K and related with the glass transition(Tg) at 330 K [13,14]. In our previous reports [13,14] it waspostulated that the blue shift of the PL spectra at temper-atures above the glass transition could be explained bythe interchain dissociation induced by thermal motions,although no further experimental evidence have beengiven.
Wide-Angle X-ray Scattering (WAXS) and Transmis-sion Electron Microscopy (TEM) showed that MEH-PPVfilms form molecular aggregates that present local nano-scopic order with structural anisotropy [15,16]. The dif-fraction patterns obtained in stretched films wereindexed proposing an orthorhombic unit cell withparameters a = 7.12 Å, b = 16.05 Å and c = 6.47 Å [16]. Afurther description of the molecular aggregates was pre-sented recently [2], where it was shown that MEH-PPVchains are locally ordered with phenyl rings partially or-ganized parallel to each other and also parallel to thefilm plane. This molecular packing was described as asuperstructure unit cell consisting of a bilayer arrange-ment with repeating distance of 24.2 Å along the b-axis,a characteristic distance of 6.3 Å assigned to repeatedunit along the backbone (c-axis), and the regular spacingbetween the backbones of the coplanar phenylene rings(a = 4.3 Å). Moreover, the size of the chain-packedMEH-PPV nanostructured domain was estimated fromthe diffraction line profile and assumed to consist of nomore than 4d-spacings in each crystallographic direction.
Although the structure of MEH-PPV films and the con-formation of the polymer chains have been well de-scribed in the literature and their influence on the PLand EL were also well understood, the relationship be-tween structure and dynamics of the polymer chainsare not well known [16]. Therefore, the aim of this re-port is the study of the temperature dependence of thedynamics, supramolecular organization, and short-rangechain ordering of MEH-PPV films, when the film is castfrom two solvents, chloroform and toluene, with distinctsolvation abilities. The polymer chains dynamics wasstudied by a set of very convenient solid-state NMR tech-niques to detect possible differences in the moleculardynamics due to molecular aggregation: DIPSHIFT(DIPolar-chemical SHIFT correlation) [17–20] and CODEX(Centerband-Only Detection of EXchange) [21–24]. Tem-perature evolution of the supramolecular and theshort-range structures was studied by WAXS as well asSteady-state Fluorescence Spectroscopy allows to checkwhether structural changes had happen (due to its sensi-tivity to short range interaction among the polymerchains).
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
CTED
PR
OO
F
2. Experimental
MEH-PPV with average molar weight Mn = 86 kg/moland polydispersity Mn/Mw = 4.9 was obtained fromSigma–Aldrich Co. Toluene and chloroform solvents withspectrophotometric grade were purchased from Acros. Allmaterials were used as received.
MEH-PPV solutions were prepared by dissolving thepolymer samples in each solvent. After that, the solutionswere maintained in the dark in a sealed flask. Films wereprepared by casting the solutions in a Petri dish, with slowevaporation under a saturated solvent atmosphere, atroom temperature, for 30 h. Later, the films were dried inan oven under dynamic vacuum at a temperature of ca.323 K for 24 h. Film thicknesses were approximately30–40 lm. For WAXS experiments, multilayer samples ofup to 400 lm thicknesses were prepared by stackingpieces cut from a single film.
Steady-state fluorescence spectra of MEH-PPV filmswere recorded using a PC1TM Photon Counting Spectroflu-orimeter from ISS Inc. The spectral range was from 600 to800 nm for the emission spectra. Slits were selected for aspectral resolution of ±0.5 nm. Excitation wavelength waskex = 490 nm.
WAXS experiments were performed at the D11A-SAXSbeamline of the LNLS (Brazilian Synchrotron Light Labora-tory). The wavelength used was 1.608 Å and the sampledetector distance was approximately 182 mm in all cases.The films were set-up in two configurations: with the inci-dent X-ray beam perpendicular (\) and near-parallel (||) tothe film plane.
The samples were first examined at room temperaturefor two-dimensional (2D) patterns. Data were recorded inFuji film image plates and 30 min exposures were takenin all cases. Average radial intensity profiles were obtainedintegrating an arbitrary 30� angular sector in the case ofthe isotropic scattering pattern (\ incidence) and a similarsector centered around the maximum in the oriented scat-tering ring (|| incidence). Intensities were normalized bythe integrated intensity incident on the sample duringthe exposure and by sample absorption. Parasitic scatter-ing was subtracted from each pattern.
Afterwards, one-dimensional (1D) patterns for thein situ thermal treatment were recorded using a linear po-sition sensitive detector (PSD). The samples were placed ina hot stage cell specially designed for X-ray scattering mea-surements (THM 600, Linkam Ltda [25]). For each of theexperimental geometries (\ and || incidences) the filmswere placed in a sample support adapted to the hot stage.The difference between the values of the temperature onthe sample and the values set on the controller was lessthan 5 K for all scans in the range 123–423 K, allowing afairly precise determination of the thermal state of thesample. The in situ measurements were performed allow-ing 5 min of stabilization and 15-min data acquisition foreach desired temperature. From room temperature thesamples were rapidly quenched (60 K/min) down to123 K. Next, after each step of a heating ramp (10 K/min),WAXS patterns were obtained for the temperatures 198,273, 313, 348 and 423 K. The samples were kept at the
ependence of molecular dynamics and supramolecular ...,
PR
OO
F
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx 3
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
highest temperature during 1 h. After this isothermal treat-ment, additional data were taken. Finally, the sampleswere cooled down to 348 and 293 K at the rate of10 K/min, with additional exposures being obtained atthese temperatures. The 1 h annealing at 423 K, waschosen well above the MEH-PPV glass transition tempera-ture (Tg = 330 K) [13].
NMR experiments were performed using a VARIANINOVA spectrometer at 13C and 1H frequencies of 100.5and 400.0 MHz, respectively. A VARIAN 7-mm MAS dou-ble-resonance probe head with variable temperature (VT)was used. The spinning speeds, varying between 4 and6 kHz, were controlled by a VARIAN pneumatic system thatensures a rotation stability of ±2 Hz. Typical p/2 pulseslengths of 3.5 and 4.5 ls were applied for 13C and 1H,respectively. Time Proportional Phase Modulated (TPPM)proton decoupling with field strength of 70 kHz, cross-polarization time of 1 ms and recycle delays varying be-tween 3 and 5 s were used. Amplitude of slow molecularmotions were investigated using CODEX technique[21,22] with mixing time tm of 200 ms and evolution times(Ntr) ranging from 333 to 2500 ls. The temperature depen-dence of 13C–1H dipolar coupling were measured usingDIPSHIFT technique [18], where 1H–1H homonucleardecoupling was achieved by the Phase-Modulated-Lee-Goldburg (PMLG) sequence [26,27], using field strengthsof approximately 80 kHz.
EC
TED
Fig. 1. Steady-state fluorescence spectra of MEH-PPV films. (a) Roomtemperature spectra of chloroform (square symbols) and toluene (circlesymbols) cast films before and after annealing by 12 h at 363 K (full orempty symbols, respectively). The arrow stands for the blue shiftobserved as a function of temperature. (b) Steady-state fluorescencespectra of MEH-PPV films cast from chloroform as a function oftemperature from 53 to 393 K. The inset show the integrated intensityof the corresponding spectra of MEH-PPV films cast from chloroform as afunction of temperature.
R3. Results and discussions
This article is organized as follow. The room tempera-ture photoluminescence spectra of MEH-PPV films castform toluene and chloroform is presented and comparedtogether with a short discussion about the temperaturedependence of the PL. In sequence, WAXS are analyzed asa function of temperature in order to give some insightabout the temperature dependence of the microstructureof the MEH-PPV aggregates. Then, to elucidate the charac-teristics of the molecular dynamics processes in the poly-mer chains, solid-state NMR measurements arepresented. Finally, a correlation between the strucuturaland dynamics results are presented and discussed in thecontext of the change in the PL spectra.
R 241242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
UN
CO3.1. Photoluminescence spectroscopy
Steady-state fluorescence spectra (PL) of MEH-PPV filmscast from toluene and chloroform solutions before andafter annealing at 363 K for 12 h are shown in Fig. 1a. Ingeneral, these spectra are composed by a higher intensityband around 650 nm and a second overlapped bandaround 700 nm (Fig. 1). This spectrum, usually observedfor the MEH-PPV films, is assigned to the interchain speciesseparated by a backbone interplanar distance of �4.05 Å[28–31]. The longer red-edge tail is also observed whichresults from the overlap of two contributions, the vibronicprogression of the aggregate emission and from the emis-sion of excimeric/interchain species [32]. Spectral broad-ening was estimated and the full-width at the half-maximum (FWHM) showed that those values are in the
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
range of 2000 cm�1 which are orders of magnitude largerthan expected for homogeneous broadening [33]. As itcan be observed in Fig. 1a, there is a slightly difference inthe intensity of the band assigned to the excimeric/inter-chain species upon the different solvents, which are mostlyerased with thermal annealing. We also noticed that thereis an increase of the FWHM with the annealing, which canbe attributed to the increase of the relative amount ofaggregates compared with the non-annealed sample.
In order to provide information about the tempera-ture dependence of the photoluminescence, the PL spec-tra was also recorded from 53 to 393 K for MEH-PPVfilms cast from chloroform, Fig. 1b. The behavior of thespectra as a function of temperature is similar to theone reported in Ref. [13], i.e., at temperatures below150 K little variation is observed either in the integratedintensity (see figure inset) or in the position of the PL
ependence of molecular dynamics and supramolecular ...,
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
4 A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
bands. Between 150 and 300 K there is an increase in theintegrated intensity, which is accompanied by a blueshift. Above 300 K a stronger intensification and also ablue shift is observed. Toluene cast films presents a sim-ilar behavior [13].
3.2. Wide-Angle X-Ray Scattering (WAXS)
The microstructure of the MEH-PPV films as well asthe modifications induced by temperature was studiedby WAXS. Fig. 2 shows integrated intensity scans (left)from the 2D-WAXS images (right) for as cast MEH-PPVfree-standing films obtained from chloroform solutionusing perpendicular and parallel incidences, at 293 K.The anisotropy features of these films are already con-firmed comparing the 2D patterns. These patterns aresimilar in shape, showing differences in relative peakintensities and scattering contributions according to thegeometry of the experiment. Similar to Ref. [2], preferen-tial orientation of the ordered domains can be clearlynoted in the 2D images of the parallel (||) incidence pat-terns, in which the reflections corresponding to q values�0.27 �1 and �1.5 �1 present arcs of strongerintensity.
UN
CO
RR
E
Fig. 2. MEH-PPV films cast from chloroform: WAXS intensity profiles obtaiPerpendicular and (c and d) parallel incidence to the film plane. Red line: fitlogarithmic baseline for the background.
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
RO
OF
For the interpretation of the structural changes, thecurves were fitted using Gaussian functions for the peaksand a logarithmic baseline for the background. The individ-ually fitted contribution of each peak is plotted on the bot-tom of the Fig. 2a and b whereas the total fit is representedas a continuous red line in the same plots. The most impor-tant wide angle peaks were referred to the molecular pack-ing parameters proposed by Jeng et al. [2]. The firstreflection observed for q = 0.27 Å�1 (d1 = 2p/q = 23.2 Å) isassumed to represent the bilayer chain packing distance.The next important characteristic length observed forq = 1.03 Å�1 (d2 = 6.1 Å) corresponds to the monomeric re-peat unit. These two parameters define the chain packedplanar layers of the MEH-PPV films. Finally, the peak inour patterns located at q = 1.49 Å�1 (d3 = 4.2 Å) would cor-respond to the inter-backbone distance, in the directionnormal to the coplanar phenylene rings, as reported byJeng et al. [2]. It is worth mentioning that, due to the exist-ing disorder, the scattering peaks are quite broad, but theirpositions could be well determined from the fittings. Fromthe aforementioned results, is possible to conclude that themajor difference with the data reported in the literature[2,16] is that corresponding to the d1 parameter. Nonethe-less, discrepancies in the bilayer packing parameter are not
CTED
P
ned from the corresponding 2D patterns (shown at right). (a and b)of the experimental data using Gaussian functions for the peaks and a
ependence of molecular dynamics and supramolecular ...,
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx 5
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
surprising, since more than one type of chain packing maycoexist for the as-cast films.
Fig. 3 shows the WAXS intensity profiles at some se-lected temperatures, below and above Tg, obtained forMEH-PPV films cast from chloroform at parallel and per-pendicular incidences. All the intensity profiles were nor-malized by an arbitrary scale factor to observe moreclearly the changes in intensity and peak positions. Themain modifications of the first diffraction maximum arebetter seen in the parallel incidence geometry (Fig. 3a).A shift in the peak position of this reflection, associatedwith the bilayer spacing d1, and a narrowing of the linewidth as a function of increasing temperature is ob-served. The d1 values varied from 19.3 to 21.3 Å. A shoul-der in this peak is hardly noticed at the lowertemperatures, but becomes more evident for 423 K. Themaximum intensity values and narrowest line profileswere obtained for 423 K and after the samples were keptat that temperature for 1 h (see Fig. 3c). The peak areaincreases as a function of thermal treatment (tempera-ture and annealing time). This result indicates an in-crease in the number of molecular aggregates in thebilayer normal direction.
Furthermore, in Fig. 3a, the peak associated with thed3 parameter seems to shift to lower q values, whichwould indicate an increase in the inter-backbone spacingperpendicular to the plane of the phenylene rings. As areduction in the peak intensity with the temperature is
UN
CO
RR
E
Fig. 3. WAXS intensity curves of MEH-PPV film cast from chloroform for in situcurves for temperatures from 123 to 423 K before annealing. (c) Parallel and (dannealing and for 293 K at the end of the temperature cycle.
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
PR
OO
F
also observed (see also Fig. 3b), indicating a decrease inthe number of the scattering objects or chain dissocia-tion in the perpendicular direction to the ring planes.
Finally, the behavior of the d2 parameter can onlybe clearly followed in the perpendicular incidence pat-terns (see Fig. 3b and d). Its value of (approximately6.0 Å) for all temperatures ranging from 123 to 423 K,confirms the stability of this molecular packing param-eter. Note that this parameter represents the distancebetween two benzene units and is only related to bondlengths.
The explicit dependence of d1, d2 and d3 parameterswith the temperature and the annealing time is shown inFig. 4a for the chloroform cast film. The most noticeable ef-fect is observed for d1 parameter. It increases continuouslyas the temperature is raised from 123 to 318 K. Suchbehavior is in good agreement with the increase in side-chain mobility, as we are going to show in sequence. Thesubsequent increase in d1 values from �20.5 to �21.5 Åwhen the temperature reaches 348 K, can be correlatedwith the onset of the polymer glass transition at 323 K,that produces higher free volume for the relaxation ofthe aggregate supramolecular structures. From 348 to423 K and after 1 h annealing at 423 K, the d1 parameterdecreases slightly. After cooling to room temperature, a va-lue of �20.9 Å is obtained. Thus, the overall increase of thisparameter is �0.4 Å. A similar behavior in the WAXSprofiles and d1, d2 and d3 parameters was observed for
CTED
thermal treatment. (a) Parallel and (b) perpendicular incidence scattering) perpendicular incidence scattering curves at 423 K before and after 1 h
ependence of molecular dynamics and supramolecular ...,
OR
REC
TED
PR
OO
F
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
Fig. 4. Temperature dependence of the structural parameters d1, d2 and d3 for MEH-PPV: (a) cast from chloroform, (b) cast from toluene. (c) Average size hDiof the ordered domains in the bilayer normal direction. Dotted lines: visual guide of the behavior.
6 A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
UN
Cthe toluene cast films except for the higher d1 variation athigher temperatures (see Fig. 4b).
Using Scherrer equation [34], the average size of the or-dered domains <D> in the bilayer normal direction couldbe roughly estimated from the line width (FWHM) of thed1 diffraction profile. As shown in Fig. 4c for chloroformcast samples, almost constant values of <D> (around60 Å) are found for temperatures up to 348 K. An increaseof 20% is obtained for 423 K and a maximum value isreached after 1 h of annealing (90 Å). For the films obtainedfrom toluene solution, initially lower <D> values areachieved compared with the films obtained from chloro-
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
form casting, but bigger values (around 100 Å) are attainedafter the temperature of 423 K is reached.
3.3. Solid-state NMR
Fig. 5 shows the repeat unit, the orientation of theprincipal values of the CSA tensor, and typical 13C CP/MAS spectra of MEH-PPV at 303 K from chloroform andtoluene cast films. The line assignments are also shown.All the spectra are basically identical (including those atlow temperatures, 233 K), showing that there are nodrastic different in the chemical conformational state of
ependence of molecular dynamics and supramolecular ...,
E
PR
OO
F
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
Fig. 5. (a) MEH-PPV chemical structure; (b) chemical shift tensor principal axis orientation and typical 13C CP/MAS spectra of MEH-PPV at 303 K for filmscast from (c) chloroform, and (d) toluene. *denotes the spinning sidebands.
A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx 7
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
UN
CO
RR
the polymer chains the films prepared with the differentsolvents.
NMR can provide specific information about the side-chain molecular dynamics in MEH-PPV [14]. The degreeof molecular dynamics of a particular molecular group inthe side-chain can be characterized by measuring thestrength of the effective 13C–1H dipolar coupling for thatsegment, which can be provided by DIPSHIFT technique[17,20]. Such experiments, performed under Magic-An-gle-Spinning (MAS), provide a measurement of the13C–1H magnetic dipolar coupling for each chemical group.This is done by measuring the dependence of the signalamplitude with the evolution period (t1), used for codifyingthe 13C–1H dipolar coupling, which produces a typicalcurve that depends on the strength of the averaged dipolarcoupling, hmdipi. Since motions with correlation timesshorter than �100 ls average the dipolar coupling be-tween 1H and 13C, from the measurement of this parameterit is possible to distinguish rigid from mobile segments andestimate the amplitude of the molecular rotation. Molecu-lar order parameters S for each chemical group can also beobtained as the ratio between this averaged dipolar cou-pling and its respective rigid-lattice value S ¼ hmdipi=mrigid
dip .Besides, measuring the ratio hmdipðTÞi=mrigid
dip versus temper-ature allows qualitatively monitoring the increase of themolecular dynamic rate as a function of T. Typical DIPSHIFTcurves for the CH (labeled 11 in Fig. 5a) and CH2 (labeled14 and 15 in Fig. 5a) side-chain groups of MEH-PPV filmsat 293 K cast from chloroform and toluene are shown inFig. 6. These chemical groups were chosen as probes tothe molecular dynamics because they are placed in thehead (close to the backbone) and tail (at the end) of theside-chain, respectively, allowing detecting possible differ-ences in the dynamics at different side-chain positions. Thecorresponding hmdipðTÞi=mrigid
dip parameters are also shown in
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
CTED
Fig. 7. It is possible to observe that the effectivehmdipðTÞi=mrigid
dip is smaller for films where toluene was usedas solvent, confirming the higher degree of side-chainmobility in films cast from toluene. Comparing the CHand CH2 DIPSHIFT curves (Fig. 6a–b), it is also seen thatthere is a clear difference for chloroform than for toluenecast films. This also indicates a more loosely packed side-chain in the toluene cast films. This feature can be betterobserved in Fig. 7 that shows the temperature behaviorof the CH and CH2 hmdipðTÞi=mrigid
dip parameters for both films.For films cast from chloroform the CH group S parameter ishigher than for CH2 in the whole temperature range, indi-cating a significant difference in the degree of mobility be-tween the head and the tail of the side-chain. In contrast,for the film cast from toluene the S parameters for theCH and CH2 are mostly identical in the temperature range,suggesting that the mobility in the head and tail of theside-chain are much similar. This corroborates the abovefindings that point to a looser packing in the side-chainin films cast from toluene. Note that the packing of theside-chain would avoid motion of the whole side-chain,but not anisotropic motion of specific segments in theside-chain. Moreover, the fact that the effective S parame-ters tend to a plateau different from zero at higher temper-atures is due to the presence of a residual dipolar coupling,indicating that the side-chains in MEH-PPV do not rotateisotropically, but execute rotations around a local axis. Thisresidual coupling is observed as a plateau even at temper-atures well above Tg, which indicates that the motions ob-served are not really associated with free side-chains in theamorphous region of the polymer, but with ‘‘trapped” side-chains in the aggregated regions of the polymer. To con-firm that this effect is really solvent related we performedthermal annealing of the samples at 363 K during 12 h un-der dynamic vacuum. As it can be observed in Fig. 7b and d,
ependence of molecular dynamics and supramolecular ...,
UN
CO
RR
E
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
Fig. 6. (a) and (b) Typical 13C DIPSHIFT curves for the CH and CH2 side-chain groups of MEH-PPV films cast from chloroform and toluene at293 K, respectively. (c) Arrhenius plot of the correlation time (sc)extracted from the CH DIPSHIFT curves for the samples cast fromchloroform and toluene.
8 A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
the behavior of both films becomes much similar, confirm-ing that the differences observed are associated with mem-ory effect due to the solvent.
The above statements reveal that the side-chaindynamics of MEH-PPV films can be affected by the solvent
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
CTED
PR
OO
F
characteristics, which can be easily erased by thermalannealing. However, this was only a qualitative discussionand it would be worth to quantify the change in the kineticparameter that characterizes the molecular dynamics.Concerning the geometry of the motions, the fact that theS parameter does not go to zero at high temperatures (fastmotion limit) indicates that the motion is anisotropic, i.e.,occurs about a specific axis or only part of the segmentstake part of the motion in the DIPSHIFt time scale. How-ever, the DIPSHIFT data of the CH2 group attached to thebackbone (labeled 10 in Fig. 5a) shows that this group israther rigid (no temperature dependence were detectedin the DIPSHIFT curves of this group from 213 to 353 K).This is largely consistent with a motional model wherethe CH group executes n-site jumps around theH2C10AC11H bond, i.e., n-site jumps on a cone with anopening defined by the relative orientation of the CH bondwith respect to the CAC bond (ideally 109�). As describedin Ref. [35], using this previous knowledge of the motionalgeometry one can use a spin dynamics simulation program[36] to simulate the experimental DIPSHIFT curves for car-bon 11 and extract the correlation times as a function oftemperature. Besides, the shape of the DIPSHIFT curvecan also be related with the non-exponentiality of the mo-tion correlation function, which in some cases can betranslated as a distribution of correlation times. In Fig. 6awe show the temperature dependence of DIPSHIFT curvesfor the MEH-PPV CH group in films cast from toluene andchloroform with the corresponding simulations. The corre-sponding correlation times as a function of temperatureare shown in the Arrhenius plot of Fig. 6c. The activationenergy was evaluated as (69 ± 5) kJ/mol for films cast fromchloroform and (54 ± 8) kJ/mol for films cast from toluene.Thus, the results show that the energy barrier for the onsetof side-chain motion is higher for chloroform than for tol-uene cast films. This seems the only effect of the solventmemory interferes only on the local side-chain motion,since the results do not point to a modification on thegeometry of the motion (same S parameter at high temper-ature and rigid CH2 attached to the backbone).
Despite DIPSHIFT experiments are sensitive to motionsin the kHz frequency range, the presence of molecular mo-tions that occur with rates beyond the detection limit ofthese experiments cannot be ruled out. Thus, to bettercharacterize the dynamics of these systems it is attractiveto perform experiments capable of providing informationabout motions in other frequency ranges. One of suchexperiments is the CODEX [21,22] technique that makespossible to characterize the slow motion (with rates inthe Hz scale) of different chemical groups with a consider-able degree of details. Essentially, the experiment detectsthe signal reduction resulting from changes in the orienta-tion-dependent chemical-shift frequencies due to segmen-tal reorientations during a waiting time also denoted asmixing time tm. Information about the amplitude (meanreorientation angles) of the motion is obtained by thedependence of E(tm, Ntr), as a function of Ntr.
In previous work, it was observed the presence of slowmolecular motions in the backbone of MEH-PPV at 293 K[13,14]. These motions were assigned as small angle ringrotations around the 1–4 axis. Fig. 8 shows E(tm, Ntr) as a
ependence of molecular dynamics and supramolecular ...,
EC
TED
PR
OO
F518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
Fig. 7. Thermal behavior of the order parameter S for CH and CH2 side-chain groups in MEH-PPV cast from (a) chloroform before annealing, (b) chloroformafter annealing at 363 K by 12 h, (c) toluene before annealing, and (d) toluene after annealing at 363 K by 12 h.
A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx 9
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
UN
CO
RR
function of Ntr for the para-carbons (labeled 1 and 4 inFig. 5a) of MEH-PPV films cast from chloroform and tolu-ene at 293 K. The average rotation angles were obtainedby simulating the experimental curves based on principalvalues and orientation of the chemical shift tensor. Forthat, the chemical shift anisotropy principal values weremeasured using a standard Herszfeld and Berger analysis.[37] The orientation of the chemical shift principal axissystems of phenyl rings para-carbons is similar in differentsystems. Thus, we assumed the same orientation of p-xy-lene, i.e., the principal value r33 (z-axis) is 1� tilted fromthe normal to the phenylene ring plane and the x-axis, cor-responding to r11 is 1� away from the 1–4 axis, as shown inFig. 5b. Because the molecular motions in polymers do notinvolve single but distributed rotation angles, the simula-tions were performed considering Gaussian shaped distri-butions of reorientation angles centered at 0� and withfull-width at half-maximum (FWHM), hwi, that representsthe average reorientation angle.
It can be observed in Fig. 8 that, within the experimen-tal uncertainty, all curves can be fitted using the same dis-tribution of reorientation angles with hwi �10�. This meansthat no significant differences among the overall motionalamplitudes of the backbone phenylene rings could be ob-
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
served. The fact that no solvent effects were observed inthe CODEX measurements suggests that the overall tor-sional motions are similar in all cases. If we associate theamplitude of these torsional motions with the conforma-tional disorder along the polymer backbone, we might con-clude that the motional induced conformational disorder issimilar for both studied films. However, it should bepointed out that the average rotation angle obtained byCODEX cannot be associated with a particular segment inthe polymer backbone, but reflects the average behaviorof all backbone segments, including those poorly conju-gated that do not contribute to the photoluminescence. Be-sides that, because the motional amplitude is rather small,proton-driven spin-diffusion [38] may have a non-negligi-ble contribution to the CODEX exchange intensity, whichmasks the possible differences that may exist among thetorsional motion of the samples.
4. Conclusions
This work provided new results of distinct aspects ofthe dynamics and supramolecular organization in MEH-PPV films cast from two different solvents. The NMR exper-iments were particularly useful for describing general and
ependence of molecular dynamics and supramolecular ...,
E
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655
Fig. 8. E(tm,Ntr) intensities versus Ntr for para-carbons (labeled 1 and 4 inFig. 5a) at 293 K in MEH-PPV films cast from (a) chloroform, and (b)toluene.
10 A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
UN
CO
RR
specific aspects of the chain dynamics. It has been shownthat the dynamics of side-chains (b-relaxation) in MEH-PPV films tend to be less hindered (smaller activation en-ergy) in films as cast from toluene than in chloroform. Thiswas attributed to the higher side-chain packing providedby chloroform that preferentially solvates the side chainin contrast to toluene that solvates mainly the backbone.It was not possible to observe solvent induced changes inthe backbone conformational disorder using CODEX NMRtechnique. In fact, the results have shown that the torsionthermal motions at room temperature have average ampli-tude of approximately 10�, but no significant variation wasobserved between the different samples.
Temperature dependent WAXS experiments providedinformation about the modifications in chain packing in-side the aggregated domains. As a general trend, d1 andd3 values increase with temperature but a significantgrowth is specially observed for temperatures higher thanthe polymer glass transition temperature. The increase inthe d1 peak intensity and the reduction of his FWHM wasinterpreted as due to larger number of scattering objectsand bigger average sizes of the ordered domains <D> inthe direction normal to the bilayer normal plane for highertemperatures. On the other hand, the decrease in the peakintensity associated with the d3 parameter can be taken as
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
CTED
PR
OO
F
a direct evidence of the presence of interchains dissocia-tion and increased disorder of the phenyl rings. Consider-ing that the increase of conformational disorder mayquench some interchain processes that contributes to theluminescence, the results confirms our previously pro-posed model (dissociation of the interchain species duringthe MEH-PPV relaxation processes) based on PL and NMRresults [13]. More specifically, the intensity increase ob-served at 200 K is associated with in appearing of motionalring torsions due to the onset of the side-chain motionsand the stronger intensification at 300 K to the increasingof these motions due to the onset of the glass transition.In both relaxation processes the blue shift is attributed tothe increase of the motional induced conformationaldisorder.
Acknowledgements
Authors thank FAPESP, CNPq, CAPES and MCT/PADCT/IMMP for the financial support and fellowships. TDZAand RFC thank FAEPEX/Unicamp for financial support andfellowships. ERdA thanks Prof. Kay Saalwachter for provid-ing the spin dynamics simulation program and for helpfuldiscussions. WAXS data were collected under proposalsD11A-SAXS1 #4247 and #4293 of the Brazilian Synchro-tron Light Laboratory (LNLS).
References
[1] Inigo AR, Chiu H-C, Fann W, Huang Y-S, Jeng U, Linn TL, et al. PhysRev B 2004;69.
[2] Jeng U, Hsu CH, Sheu HS, Lee HY, Inigo AR, Chiu HC, et al.Macromolecules 2005;38:6566–74.
[3] Tan CH, Inigo AR, Fann W, Wei PK, Perng GY, Chen SA. Org Electronics2002;3:81–8.
[4] Schwartz BJ, Nguyen TQ, Wu JJ, Tolbert SH. Synth Met2001;116:35–40.
[5] Jakubiak R, Collison CJ, Wan WC, Rothberg LJ, Hsieh BR. Braz J PhysChem A 1999;103:2394–8.
[6] Jakubiak R, Rothberg LJ, Wan W, Hsieh BR. Syn Metals1999;101:230–3.
[7] Nguyen TQ, Kwong RC, Thompson ME, Schwartz BJ. Appl Phys Lett2000;76:2454–6.
[8] Lee TW, Park OO. Adv Mater 2000;12:801–4.[9] Chen SH, Su AC, Han SR, Chen SA, Lee YZ. Macromolecules
2004;37:181–6.[10] Bozano L, Tutle SE, Carter SA, Brock PJ. Appl Phys Lett 1998;73:3911.[11] Bozano L, Carter SA, Scott JC, Malliaras GC, Brock PJ. Appl Phys Lett
1999;74:1132.[12] Lupton JM, Samuel IDW. Syn Met 2000;111:381.[13] Cossiello RF, Kowalski E, Rodrigues PC, Akcelrud L, Bloise AC,
deAzevedo ER, et al. Macromolecules 2004;38:925–32.[14] Bloise AC, deAzevedo ER, Cossiello RF, Bianchi RF, Balogh D, Faria
RM, et al. Phys Rev B 2004;71:17120021–024.[15] Chen S, Su A, Chou H, Pengand K, Chen S. Macromolecules
2004;37:167–73.[16] Yang CY, Hide F, Diaz-Garcia MA, Heeger AJ, Cao Y. Polymer
1998;39:2209–304.[17] Munowitz MG, Griffin RG, Bodenhausen G, Huang TH. J Am Chem
Soc 1981;103:2529–33.[18] Munowitz M, Aue WP, Griffin RG. J Chem Phys 1982;77:1686–9.[19] Hong M, Gross JD, Griffin RG. J Phys Chem B 1997;101:5869–74.[20] Huster D, Xiao LS, Hong M. Biochemistry 2001;40:7662–74.[21] deAzevedo ER, Hu WG, Bonagamba TJ, Schmidt-Rohr K. J Am Chem
Soc 1999;121:8411–2.[22] deAzevedo ER, Hu WG, Bonagamba TJ, Schmidt-Rohr K. J Chem Phys
2000;112:8988–9001.[23] Schmidt-Rohr K, deAzevedo ER, Bonagamba TJ. Encyclopedia of NMR
2002;9:633–42.[24] deAzevedo ER, Bonagamba TJ, Reichert D. Prog Nucl Magn Reson
Spectrosc 2005;47:137–64.
ependence of molecular dynamics and supramolecular ...,
656657658659660661662663664665
666667668669670671672673674675
A.A. Souza et al. / European Polymer Journal xxx (2008) xxx–xxx 11
EPJ 4755 No. of Pages 11, Model 3G
14 October 2008 Disk UsedARTICLE IN PRESS
[25] Bras W, Derbyshire GE, Devine A, Clark SM, Komanschek BE, Ran AJ. JAppl Crystallogr 1995;28:26–32.
[26] Bielecki A, Kolbert AC, Levitt MH. Chem Phys Lett 1989;155:341–6.[27] Vinogradov E, Madhu PK, Vega S. Chem Phys Lett 1999;314:443–50.[28] Conwell EM, Perlstein J, Shaik S. Phys Rev B 1996;54:R2308–10.[29] Schwartz BJ. Ann Rev Phys Chem 2003;54:141–72.[30] Huser T, Yan MJ. Photochem Photobiol A-Chem 2001;144:43–51.[31] Lemmer U, Heun S, Mahrt RF, Scherf U, Hopmeier M, Siegner U, et al.
Chem Phys Lett 1995;240:373–8.[32] Cossiello RF, Akcelrud L, Atvars DZ. J Braz Chem Soc 2005;16:74–86.
UN
CO
RR
E
676
Please cite this article in press as: Souza AA et al., Temperature dEur Poly J (2008), doi:10.1016/j.eurpolymj.2008.09.030
[33] Birks B. Photophysics of aromatic molecules. New York: Wiley;1970.
[34] Alexander LE. X-ray diffraction methods in polymer science. NewYork: Wiley-Interscience; 1969.
[35] deAzevedo ER, Saalwachter K, Pascui O, Souza AA, Bonagamba TJ,Reichert D. J Chem Phys 2008;28:104505.
[36] Saalwächter K, Fischbach I. J Magn Reson 2002;157:17–30.[37] Herzfeld J, Berger AE. J Chem Phys 1980;73:6021–30.[38] Reichert D, Bonagamba TJ, Schmidt-Rohr K. J Magn Reson
2001;151:129–35.
CTED
PR
OO
F
ependence of molecular dynamics and supramolecular ...,