Plasmonic Nanospectroscopy for Thermal Analysis of Organic

Jan 23, 2017 - Thermal analysis of organic (semiconductor) thin films is limited to a small number of highly specialized and sophisticated techniques...
0 downloads 11 Views 992KB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Plasmonic Nanospectroscopy for Thermal Analysis of Organic Semiconductor Thin Films Ferry Anggoro Ardy Nugroho, Amaia Diaz de Zerio Mendaza, Camilla Lindqvist, Tomasz J. Antosiewicz, Christian Müller, and Christoph Langhammer Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04807 • Publication Date (Web): 23 Jan 2017 Downloaded from http://pubs.acs.org on January 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Plasmonic Nanospectroscopy for Thermal Analysis of Organic Semiconductor Thin Films

Ferry A. A. Nugroho1, Amaia Diaz de Zerio Mendaza2, Camilla Lindqvist2, Tomasz J. Antosiewicz1,3, Christian Müller2,*, and Christoph Langhammer1,*

1

Department of Physics and 2Department of Chemistry and Chemical Engineering, Chalmers University of Technology, 412 96 Göteborg, Sweden 3

Centre of New Technologies, University of Warsaw, Banacha 2c, 02-097 Warsaw, Poland

E-mail: *(CM) [email protected]; *(CL) [email protected]

ACS Paragon Plus Environment

1

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

ABSTRACT

Organic semiconductors are key materials for the next generation thin film electronic devices like field-effect transistors, light-emitting diodes and solar cells. Accurate thermal analysis is essential for the fundamental understanding of these materials, for device design, stability studies and quality control because desired nanostructures are often far from thermodynamic equilibrium and therefore tend to evolve with time and temperature. However, classical experimental techniques are insufficient because the active layer of most organoelectronic device architectures is typically only on the order of hundred nanometers or less. Scrutinizing the thermal properties in this size range is, however, critical because strong deviations of the thermal properties from bulk values due to confinement effects and pronounced influence of the substrate become significant. Here, we introduce plasmonic nanospectroscopy as an experimental approach to scrutinize the thickness dependence of the thermal stability of semi-crystalline, liquid-crystalline and glassy organic semiconductor thin films down to the sub-100 nm film thickness regime. As the main result we find a pronounced thickness dependence of the glass transition temperature of ternary polymer:fullerene blend thin films, and their constituents, which can be resolved with exceptional precision by the plasmonic nanospectroscopy method, that relies on remarkably simple instrumentation.

ACS Paragon Plus Environment

2

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

INTRODUCTION Organic semiconductors receive tremendous interest as active materials in a wide range of thin film electronic devices, ranging from field-effect transistors and sensors to organic light-emitting diodes and solar cells. Precise control over the active layer nanostructure is critical for optimal device performance. However, the desired nanostructure is often in a state far away from thermodynamic equilibrium and therefore tends to evolve with time and temperature. Hence, accurate thermal analysis is essential for the full development cycle of organic electronic components from fundamental studies to device design, stability studies and quality control. Classical experimental techniques are often insufficient because the active layer of most device architectures is typically only (several) hundred nanometers thin, which leads to strong deviations of transition temperatures from bulk values due to confinement effects, and a pronounced influence of the substrate.1–3

Thermal analysis of organic (semiconductor) thin films is limited to a small number of highly specialized and sophisticated techniques.4 Chip calorimetry, a variant of differential scanning calorimetry (DSC), is attractive because it allows rapid heating/cooling rates.5 In-situ grazingincidence wide-angle X-ray scattering (GIWAXS) requires access to synchrotron beamlines and is particularly sensitive for the case of strongly scattering samples.6 Variable-temperature fieldeffect transistor measurements can be employed to monitor the thermal behavior in complete devices.7,8 The by far most widely employed technique, however, is variable-temperature ellipsometry.4,9 Although thermal transitions can be detected by simply plotting the ellipsometric angles, i.e. the raw data, as a function of temperature,9,10 more detailed analysis can require

ACS Paragon Plus Environment

3

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

intricate modeling. A number of ellipsometry studies have confirmed that the thermal behavior of conjugated polymer thin films can strongly deviate from that of the bulk.10–13

One technique that, so far, has not been considered for thermal analysis of organic semiconductor thin films is plasmonic nanospectroscopy.14–17 Conversely, for the vibrant fields of bio-18 and chemosensing19,20 the same technique is widely used because of its exceptional sensitivity. Here we demonstrate that a versatile benchtop-type setup that exploits localized surface plasmon resonance (LSPR)18,21 is readily suitable for the characterization of semicrystalline, liquid-crystalline or glassy organic semiconductor thin films down to the sub-100 nm film thickness regime. Our approach relies on sensors that comprise metal nanoparticles (usually Au or Ag), which sustain LSPR that gives rise to a locally enhanced electric field surrounding the particles. This field typically extends a few tens of nanometers from the plasmonic nanoparticle surface. Through coupling of this enhanced near field with the local environment, the LSPR frequency is very sensitive towards the refractive index (RI) of the surrounding medium.22 In the far field, a minute change of the local RI is thus detected by a spectral shift of the characteristic “peak” associated with the LSPR in an optical extinction or scattering spectrum, ∆ߣ௣௘௔௞ . To utilize this signal for thermal analysis we exploit the fact that thermal expansion of a thin film causes a change in density and hence RI. Since phase changes are accompanied by a change in thermal expansion coefficient ߙ = ‫ି ݐ‬ଵ ∙ ∆‫ݐ‬/∆ܶ, where t is the film thickness, we can detect thermal transitions (e.g. glass transition and melting) as a change in the rate by which ∆ߣ௣௘௔௞ varies with temperature, i.e. ݀(∆ߣ௣௘௔௞ )/݀ܶ.15 Here, we apply this concept to organic semiconductor thin films to scrutinize their thermal stability down to the sub-100 nm thickness range. Furthermore, we carefully examine the details of LSPR-based thermal analysis with

ACS Paragon Plus Environment

4

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Finite-Difference Time-Domain (FDTD) simulations to develop a thorough understanding of the sensing mechanism and the role of the intrinsic surface corrugation of nanoplasmonic sensors in the present application.

EXPERIMENTAL SECTION Plasmonic Sensor Chip Fabrication: All sensors were fabricated directly on a 1x1 cm glass substrate (Borofloat, Schott Scandinavia AB) or on a 1x1 cm silicon wafer for SEM imaging by hole-mask colloidal lithography.23 The fabrication details are presented in the Supporting Information SI. Materials and Sample Preparation: PMMA (Microchem Corporation, 4 wt.% diluted in anisole, Mw ~950 kg/mol), P3HT (Ossila, regio-regularity ~96%, Mw ~64 kg/mol, PDI ~2.2), APFO3 (Mw ~12 kg/mol, PDI ~2.4), F8BT (American Dye Source, Mw ~217 kg/mol, PDI ~6.6), TQ1 (Mw ~263 kg/mol, PDI ~3.7 as measured with size exclusion chromatography (SEC) in 1,2,4-trichlorobenzene (1,2,4-TCB) at 150ºC using an Agilent PL-GPC 220 system and relative calibration against polystyrene standards), and PCBM (Solenne BV, purity > 99%) were used. Thin films were prepared by spin-coating ortho-dichlorobenzene (o-DCB, Sigma Aldrich, purity 99%) solutions. The film thickness was adjusted by varying the spin-coating speed (from 300– 800 rpm, followed by 3000 rpm) and/or solution concentration (12.5–50 g/L solid content), followed by drying at ambient temperature. The film thickness and topography were measured with a surface profiler (VeecoDektak 150) and an AFM (Bruker Dimension 3100 SPM) in tapping mode, respectively. Plasmonic ܶ௚ Measurements: The samples were mounted in an insulated quartz tube gas flow reactor system with optical access (see Figure S1 for details and schematic depiction of the

ACS Paragon Plus Environment

5

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

setup). The sequence of ܶ௚ measurements was as follows: i) The sample was heated to 80oC and dwelled for 1.5 h. ii) Temperature was reduced to 40oC. iii) Three cycles of heating/cooling were run in sequence: heating to 200oC (or 300oC for P3HT, P3HT:PC61BM and F8BT samples) with 5 oC/min rate and cooling back to 40oC. All of the processes described were run under 50 mL/min constant flow of Ar to ensure dry and inert conditions. The results reported in the main text were obtained from the second heating scan. However, results from the subsequent third heating scan reveal very similar ܶ௚ values (Figure S10), demonstrating stable and reversible sample conditions in our experiments. Differential Scanning Calorimetry (DSC): Thermograms were recorded under nitrogen using a heat-flux DSC from Mettler Toledo (DSC2) equipped with a high-sensitivity sensor (HSS9+) and a TC-125MT intracooler. Mettler 70 µl Al crucible light sample pans were used with sample weight of ~6 mg. The heating rate was 20 °C/min. Ternary

Blend

Optical

Properties:

The

RI

and

extinction

coefficient

of

2:1:1

TQ1:PC61BM:PC71BM were calculated by using the Bruggeman effective medium approximation, based on data for neat TQ1, PC61BM and PC71BM reported elsewhere.24 To convert between weight and volume fraction, densities of 1 g/cm3 and 1.3 g/cm3 were assumed for TQ1 and PCBM, respectively. Numerical Simulations: FDTD method was used for the electrodynamics simulations of the experimentally investigated systems. Due to the inherent long-range randomness of amorphous arrays of the nanodisks on the plasmonic sensor chips used, even a small subset of such a structure is impractical to calculate using numerical tools. Thus, we simulated the optical response of just one disk. However, electromagnetic interactions within amorphous arrays shift

ACS Paragon Plus Environment

6

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the resonance with respect to that of the base element.25 To match the spectral position of the simulated plasmonic resonance to that of the measured one to within 10 nm we slightly varied the geometrical parameters of the silver disk (r = 80 nm, h = 25 nm) and RI of the substrate and support layer. As we were unable to account for interparticle interaction in an amorphous array, we used periodic boundary conditions to mimic the effective result of this interaction. Matching of the interaction strength is assumed when the calculated plasmon peak matches the experimental resonance. This matching is important as it assures that the resonance is close to the spectral location of the experimentally measured one, and the enhanced electric fields feel the “correct” RI of the sensed polymer, i.e. the sensing volume of the simulated particle is as close as possible to that of the real one. The disks were placed on top of a glass substrate (RI of 1.465) and covered by a conformal support layer of Si3N4 (RI of 1.7, appropriate at wavelength of ca. 1 µm). On top of the support layer the polymer was placed semi-conformally for thin layers, following the profile obtained by AFM measurements (see Figure 4), or in a flat manner for thick layers. The permittivity of silver is taken from Palik26 and the optical properties of the polymers from ellipsometric measurements described above. RESULTS AND DISCUSSION As the first step, to benchmark the method, we applied plasmonic nanospectroscopy on six different polymer thin films (thickness ~150–250 nm) ranging from completely amorphous (PMMA), to liquid-crystalline (APFO3, TQ1 and F8BT), to semi-crystalline (P3HT and P3HT:PC61BM), and exhibiting diverse thermal properties. The plasmonic sensor used consists of an array of Ag nanodisks on a transparent glass substrate uniformly covered with a 10 nm thin Si3N4 support layer (Figure 1). This layer provides a chemically uniform surface and separates the plasmonically active Ag nanodisks from the studied films, as well as efficiently protects them

ACS Paragon Plus Environment

7

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

from oxidation and corrosion. We also note that the support layer can be tailored to achieve a desired condition and/or functionality, and that alternative plasmonic sensor surfaces that feature narrow plasmonic peaks due to interference effects and Fano resonances27 can be used equally well to potentially further increase the sensitivity.

For the thermal analysis experiments, the polymer films were spin-coated onto the sensor surface. First we compare their extinction spectra, which, except for PMMA, comprise two regions, i.e. absorption bands below and above 750 nm (Figure 2a). The first band at high photon energies is associated with the polymer itself, whose band gap varies between 450 to 700 nm depending on the polymer, whereas the second band, at low photon energies, is associated with the LSPR of the sensor. The spectral separation between the two absorption bands is critical for the plasmonic measurements because it guarantees that the LSPR readout can be achieved without convolution from the absorption of the polymer. It can be conveniently achieved by tailoring the dimensions and/or the material of the plasmonic particles used (here Ag nanodisks of 170 nm and 20 nm in diameter and height, respectively).

After spin coating, the sample is heated in a gas flow chamber with optical access that enables continuous recording of the spectral position of the LSPR peak, ߣ௣௘௔௞ (see Experimental Sections). The plasmonic particles respond to the temperature-induced (even very minute) changes in the RI of the polymer via shifts of their LSPR peak, Δߣ௣௘௔௞ (Figure S3, and the intrinsic temperature response of a blank sensor shown in Figure S4). The measured Δߣ௣௘௔௞ thus directly reflects the thermal evolution of the polymer film since thermal effects induced by light absorption in the plasmonic nanoparticle are negligible at the low light intensity used.28

ACS Paragon Plus Environment

8

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

For all six systems we observe that the Δߣ௣௘௔௞ initially varies linearly upon heating up to a material-dependent temperature where the rate of change, ݀(Δߣ௣௘௔௞ )/݀ܶ, quite abruptly adopts a different value (Figure 2b-g). We also note that the magnitude of ∆ߣ௣௘௔௞ is different for the studied systems as it depends on, e.g., the RI of the film. Nonetheless, at least two linear regions with different ݀(Δߣ௣௘௔௞ )/݀ܶ can be identified for all samples. Thus, for PMMA, APFO3, TQ1, P3HT:PC61BM and F8BT we follow the procedure detailed in Ref. 15 to obtain the glass transition temperature ܶ௚ by finding the intersection point of two straight lines fitted to the two linear regions (solid black lines in Figure 2b-d and f-g). This approach is also further corroborated by FDTD simulations (Figure S5 and corresponding text). To put the accuracy of this analysis into perspective, we relate our results to the 95% confidence boundaries for each of the fitted data sets (dashed lines in Figure 2b-d and f-g), which define the possible range of error in determining the intersection, and thus ܶ௚ (Figure S6). For P3HT, P3HT:PC61BM and F8BT we also define the onset and endset of the melting temperature ܶ௠ as the points where ∆ߣ௣௘௔௞ deviates from the straight lines fitted to linear regions (green areas in Figure 2e-g). Lastly, for F8BT, we also derive a nematic-isotropic transition ܶ௟௖ by following the same procedure as for obtaining the ܶ௚ (Figure 2g). The characteristic temperatures derived in this way from the ∆ߣ௣௘௔௞ scans of these six different polymer systems (Table 1) correspond to their known ܶ௚ s (PMMA,29

APFO3,30

TQ1,31

P3HT:PC61BM32,33

and

F8BT34),

ܶ௠ ‫ݏ‬

(P3HT,32–34

P3HT:PC61BM32,33 and F8BT34), and ܶ௟௖ (F8BT).34 Moreover, we find that the thermal behavior of the thin films (thickness ~150–250 nm), monitored here with plasmonic nanospectrosocopy, is in good agreement with those of bulk samples (cf. Table 1 and Figure S7).

ACS Paragon Plus Environment

9

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

With the method established for the assessment of single-constituent polymer thin film thermal properties, we turn to a more complex system of a polymer blend of high relevance for organic photovoltaic devices and study the film thickness dependence of its thermal properties, as well as its constituents, down to the sub-100 nm film thickness regime. We chose to focus on the thiophene-quinoxaline copolymer TQ135 which permits solar cell efficiencies of up to 6–7%.36,37. We blend TQ1 with an alloy of the two most widely used fullerene acceptors, phenyl-Cx-butyric acid methyl ester with x = 61 and 71, respectively (PC61BM and PC71BM). Fullerene alloys are an attractive acceptor material because they improve the processability,38–40 costeffectiveness,38,41,42 thermal stability42–45 and mechanical properties of bulk-heterojunction blends,41 and can give rise to solar cell efficiencies of 10%.46

In a first set of experiments for this system, we compared the glass transition temperature ܶ௚ of ~150 nm thick films of neat TQ1 with a 4:1 PC61BM:PC71BM fullerene mixture and a 5:4:1 TQ1:PC61BM:PC71BM ternary blend. As shown again in Figure 3a we found ܶ௚்ொଵ ~117°C for the TQ1 film. This is somewhat higher than a value of ~100°C previously measured with variable-temperature ellipsometry for a TQ1 batch with a similar Mn ~46 kg/mol, but with a different thermal history (i.e. first heating scan of an as-cast film vs. second heating scan of a film initially annealed at 80ºC in this study).31 Both the fullerene mixture and ternary blend display a single ܶ௚ , which can be anticipated for finely intermixed blends.47 For the fullerene ௉஼లభ ஻ெ:௉஼ళభ ஻ெ

mixture we deduce a ܶ௚

~128°C (Figure 3b), which is somewhat lower than the ௉஼లభ ஻ெ

bulk values previously reported for the two neat fullerenes, i.e. ܶ௚ ௉஼ళభ ஻ெ

ܶ௚

~110–150°C and

~163°C.4,48 Lastly, the ternary blend features a single ܶ௚ with a value of ~123°C

ACS Paragon Plus Environment

10

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(Figure 3c), which lies in between the transitions measured for the individual blend components, i.e. neat TQ1 and the fullerene mixture.

As the second set of experiments we move on to examining the influence of the film thickness on the ܶ௚ of the ternary blend and its constituents in the thickness range of ~20–350 nm. However, prior to discussing the corresponding experimental data, we will analyze in detail the possible consequences of the fact that, for the thinnest studied films, their thickness becomes comparable to the height of the plasmonic nanodisks, by combining AFM analysis of the thin films formed on our sensors with FDTD simulations. Figure 4 shows the topographical profile of the polymer films for four different cases: i) a blank sensor and sensors coated with ii) 21 nm TQ1:PC61BM:PC71BM, iii) 58 nm PC61BM:PC71BM and iv) 65 nm TQ1. Here we focus only on the thinnest films studied for the respective ternary blend and its constituents, to represent the “extreme” conditions where (significant) deviations from an ideally flat surface are expected. The characterization was done on polymer films prior to the ܶ௚ measurements, and we also note that, while not shown here, the profile of the polymers after measurements did not show any significant change. AFM images reveal that for a 21 nm thin TQ1:PC61BM:PC71BM ternary blend, a semi-conformal coating is obtained, with a film thickness of ca. 10 nm on the top of the nanodisk and the anticipated ca. 21 nm in between. For thicker films of TQ1 and PC61BM:PC71BM the surface corrugation caused by the plasmonic sensor nanodisks is then already not significant anymore.

With this information at hand, we performed FDTD simulations of these systems to analyze the scattered field intensity distributions in order to derive where locally the thermal behavior of the

ACS Paragon Plus Environment

11

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

thin polymer films is probed by the plasmonic sensorFigure 5 shows the field scattered/induced by the silver nanodisk at the resonance wavelength of four relevant cases derived from the AFM analysis, i.e. a Ag plasmonic sensor nanodisk with: a) Si3N4 support layer but without polymer; b) with a 23 nm semi-conformal polymer coating (real case, Figure 4); c) a fully conformal 23 nm polymer coating with constant thickness; and d) an infinite polymer film reminiscent of thick coatings in which the sensor height does not affect the surface of the polymer. As can be seen, the enhanced field penetrates the immediate surrounding for the bulk polymer case (Figure 5d) in a rather uniform manner around the edges of the disk since the RIs of the substrate and the polymer do not differ significantly. However, in the case of a blank sensor, the field is significantly stronger in air at the sides of the Si3N4 support layer than at the top, even close to the disk’s edges (Figure 5a). This is caused by the difference in the RIs of the substrate and Si3N4 support layer vs. the air above the sensor. Thus, one can expect that for polymer thicknesses ranging from zero to infinity, the field will smoothly change from the blank sensor to the bulk polymer case. Nevertheless, the dependence may not be monotonic, as indeed is shown for the real case of semi-conformal coverage where the field at the top of the Si3N4 support layer above the disk is considerably weaker than at the side (Figure 5b). As the key conclusion, this means that in the experiment, the measured signal is to the largest extent stemming from the polymer located to the side of the disk and that the contribution from the thinner layer on top of the disk is negligible. In other words, we indeed measure the thermal behavior of the anticipated nominal film thickness. We further support these qualitative conclusions with calculations of the split of electromagnetic energy, which is the factor determining the local LSPR sensitivity around the nanodisk,49 between the bottom 25 nm of the polymer (up to the top of the Si3N4 support layer), as well as everything else above that line (Figure S8). The obtained split of

ACS Paragon Plus Environment

12

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

electromagnetic energy is summarized in Table S1. For the experimentally relevant case of a 23 nm semi-conformal film (10 nm on top of the disk), it turns out that only 21% of the energy is contained in the polymer on top of the disk. The rest is localized in the nominal 23 nm thick film on the side of the disk. For a hypothetical, perfectly conformal film (Figure 5c), the split is more in favor of the film on top of the disk although it still only contains 33% of the energy. The second limiting case (i.e. bulk polymer) shows that the split is approximately 1:1. These calculations thus further and quantitatively corroborate that the local thinning of the polymer “cap” on top of the Ag nanodisk for the case of very thin polymer layers only has a small influence on the overall measured response, and thus what is measured are indeed the properties of the deposited film of nominal thickness located in between the nanodisks. For further experimental confirmation see our recent work50 and also Figure S11.

Having now discussed in detail the sensing mechanism and local sensitivity of our sensors, we measured three thickness series for TQ1, the 4:1 PC61BM:PC71BM fullerene mixture and the 5:4:1 TQ1:PC61BM:PC71BM ternary blend (Figure S9). The obtained ܶ௚ as function of film thickness is plotted in Figure 6 and shows a similar trend for all three systems, i.e. an increase in ܶ௚ with decreasing film thickness. To further analyze these data, we use a model proposed by Kim et al.51 and fit our data with the following equation to describe the thickness dependence of ܶ௚ : ܶ௚ (‫ܶ = )ݐ‬௚௕௨௟௞

௧(ଶ௞ା௧)

(1)

(ఌା௧)మ

where t is the film thickness, and k, ߝ and ܶ௚௕௨௟௞ are fitting parameters. Specifically, ܶ௚௕௨௟௞ represents the bulk glass transition temperature and ߝ is a parameter correlating with the statistical segment length of the polymer. We find that for all three investigated systems

ACS Paragon Plus Environment

13

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

Equation (1) describes the observed trend in ܶ௚ very well. Furthermore, the extrapolated bulk ܶ௚ from the fit nicely approaches the experimentally deduced ܶ௚௕௨௟௞ values for t ≈ 150 nm (cf. Figure 3). Overall, we find a ܶ௚௕௨௟௞ ~120oC for the ternary blend, which lies in between those of its constituents, i.e. ~113oC for TQ1 and ~125oC for the fullerene mixture (Table 2). Interestingly, the best fits to our data for TQ1 and the fullerene mixture reveal a ߝ parameter of 4.4 nm and 1.2 nm, respectively. For TQ1 the deduced value implies that the polymer is semiflexible, which is in agreement with the liquid-crystalline behavior of TQ1,35 whereas for the fullerene mixture we note an excellent agreement with the van der Waals diameter of C60 of 1.0 nm. Moreover, we find that the two fitting parameters are very similar for TQ1 and TQ1:PC61BM:PC71BM, which implies a major influence of the TQ1 constituent on the characteristics of the ternary blend.

CONCLUSIONS In summary, we have shown that plasmonic nanospectroscopy is an efficient and accurate experimental technique for thermal analysis of semi-crystalline, liquid-crystalline or glassy organic semiconductor thin films, including conjugated polymers and their blends with fullerenes. Specifically, using this experimental methodology in the present study we were able to reveal how the film thickness of much less than 100 nm impacts the glass transition of a polymer:fullerene ternary blend, as well as its components, the thiophene-quinoxaline copolymer TQ1 and a PC61BM:PC71BM alloy. We anticipate that plasmonic nanospectroscopy will open up the possibility to study the evolution of organic semiconductor thin films with exceptional precision, and thus will aid the development of thermally robust organic electronic devices.

ACS Paragon Plus Environment

14

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

ASSOCIATED CONTENT Supporting Information Sensor fabrication and the gas flow reactor details, extinction spectra of sensors coated with PC61BM:PC71BM and TQ1:PC61BM:PC71BM, LSPR peak evolution upon heating, blank sensor reference, determination of ܶ௚ error range, DSC measurements, FDTD simulations of the polymer thermal response, electromagnetic energy distribution in the polymer films near plasmonic antenna, raw dataset for TQ1, TQ1:PC61BM:PC71BM and PC61BM:PC71BM, ܶ௚ from third heating scan, ܶ௚ measurement on a flat sensor.

AUTHOR INFORMATION Corresponding Author *(CM): [email protected] (CL): [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENTS We acknowledge financial support from the Swedish Research Council, the Swedish Energy Agency, the Swedish Foundation for Strategic Research Framework Programs RMA11-0037 and RMA15-0052 and the Chalmers Area of Advance for Nanoscience and Nanotechnology. We

ACS Paragon Plus Environment

15

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 28

thank the Knut and Alice Wallenberg Foundation for their support of the infrastructure in the MC2 nanofabrication laboratory at Chalmers, and the Swedish Research Council for their support of the µ-fab cleanroom infrastructure in Sweden. TJA thanks the Polish National Science Center for support via the project 2012/07/D/ST3/02152. Moreover, we thank Dr. Ergang Wang (Chalmers University of Technology) and Prof. Mats. Andersson (University of South Australia) for providing the TQ1 polymer used in this study, and Dr. Mariano Campoy-Quiles (ICMABCSIC) for modeling of optical constants.

REFERENCES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)

(15) (16)

Forrest, J.; Dalnoki-Veress, K.; Stevens, J.; Dutcher, J. Phys. Rev. Lett. 1996, 77, 2002– 2005. Ellison, C. J.; Torkelson, J. M. Nat. Mater. 2003, 2, 695–700. Priestley, R. D.; Ellison, C. J.; Broadbelt, L. J.; Torkelson, J. M. Science 2005, 309, 456– 459. Müller, C. Chem. Mater. 2015, 27, 2740–2754. Van den Brande, N.; Van Assche, G.; Van Mele, B. Polymer (Guildf). 2015, 57, 39–44. Lu, X.; Hlaing, H.; Germack, D. S.; Peet, J.; Jo, W. H.; Andrienko, D.; Kremer, K.; Ocko, B. M. Nat. Commun. 2012, 3, 795. Labram, J. G.; Domingo, E. B.; Stingelin, N.; Bradley, D. D. C.; Anthopoulos, T. D. Adv. Funct. Mater. 2011, 21, 356–363. Mattias Andersson, L.; Hedstrm, S.; Persson, P. Appl. Phys. Lett. 2013, 103, 213303. Campoy-Quiles, M.; Alonso, M. I.; Bradley, D. D. C.; Richter, L. J. Adv. Funct. Mater. 2014, 24, 2116–2134. Campoy-Quiles, M.; Sims, M.; Etchegoin, P. G.; Bradley, D. D. C. Macromolecules 2006, 39, 7673–7680. Wang, T.; Pearson, A. J.; Dunbar, A. D. F.; Staniec, P. A.; Watters, D. C.; Coles, D.; Yi, H.; Iraqi, A.; Lidzey, D. G.; Jones, R. A. L. Eur. Phys. J. E. Soft Matter 2012, 35, 129. Müller, C.; Andersson, L. M.; Peña-Rodríguez, O.; Garriga, M.; Inganäs, O.; CampoyQuiles, M. Macromolecules 2013, 46, 7325–7331. Liu, D.; Osuna Orozco, R.; Wang, T. Phys. Rev. E 2013, 88, 22601. Syrenova, S.; Wadell, C.; Nugroho, F. A. A.; Gschneidtner, T. A.; Diaz Fernandez, Y. A.; Nalin, G.; Świtlik, D.; Westerlund, F.; Antosiewicz, T. J.; Zhdanov, V. P.; Moth-Poulsen, K.; Langhammer, C. Nat. Mater. 2015, 14, 1236–1244. Langhammer, C.; Larsson, E. M.; Kasemo, B.; Zorić, I. Nano Lett. 2010, 10, 3529–3538. Larsson, E. M.; Langhammer, C.; Zorić, I.; Kasemo, B. Science 2009, 326, 1091–1094.

ACS Paragon Plus Environment

16

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31)

(32) (33) (34) (35)

(36) (37) (38) (39) (40)

(41)

Nugroho, F. A. A.; Xu, C.; Hedin, N.; Langhammer, C. Anal. Chem. 2015, 87, 10161– 10165. Mayer, K. M.; Hafner, J. H. Chem. Rev. 2011, 111, 3828–3857. Larsson, E. M.; Syrenova, S.; Langhammer, C. Nanophotonics 2012, 1, 249–266. Tittl, A.; Giessen, H.; Liu, N. Nanophotonics 2014, 3, 157–180. Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light by Small Particles; Wiley-VCH: New York, 2007. Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442–453. Fredriksson, H.; Alaverdyan, Y.; Dmitriev, A.; Langhammer, C.; Sutherland, D. S.; Zäch, M.; Kasemo, B. Adv. Mater. 2007, 19, 4297–4302. Campoy-Quiles, M.; Müller, C.; Garriga, M.; Wang, E.; Inganäs, O.; Alonso, M. I. Thin Solid Films 2014, 571, 371–376. Antosiewicz, T. J.; Apell, S. P.; Zäch, M.; Zorić, I.; Langhammer, C. Phys. Rev. Lett. 2012, 109, 247401. Handbook of Optical Constants of Solids; Palik, E. D., Ed.; Elsevier, 1997. Wu, C.; Khanikaev, A. B.; Adato, R.; Arju, N.; Yanik, A. A.; Altug, H.; Shvets, G. Nat. Mater. 2011, 11, 69–75. Zhdanov, V. P.; Zorić, I.; Kasemo, B. Phys. E Low-dimensional Syst. Nanostructures 2012, 46, 113–118. Porter, C. E.; Blum, F. D. Macromolecules 2000, 33, 7016–7020. Müller, C.; Wang, E.; Andersson, L. M.; Tvingstedt, K.; Zhou, Y.; Andersson, M. R.; Inganäs, O. Adv. Funct. Mater. 2010, 20, 2124–2131. Kroon, R.; Gehlhaar, R.; Steckler, T. T.; Henriksson, P.; Müller, C.; Bergqvist, J.; Hadipour, A.; Heremans, P.; Andersson, M. R. Sol. Energy Mater. Sol. Cells 2012, 105, 280–286. Müller, C.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Frost, J. M.; Bradley, D. D. C.; Smith, P.; Stingelin-Stutzmann, N.; Nelson, J. Adv. Mater. 2008, 20, 3510–3515. Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande, D.; Van Mele, B. J. Phys. Chem. B 2009, 113, 1587–1591. Wu, L.; Casado, S.; Romero, B.; Otón, J. M.; Morgado, J.; Müller, C.; Xia, R.; CabanillasGonzalez, J. Macromolecules 2015, 48, 8765–8772. Wang, E.; Bergqvist, J.; Vandewal, K.; Ma, Z.; Hou, L.; Lundin, A.; Himmelberger, S.; Salleo, A.; Müller, C.; Inganäs, O.; Zhang, F.; Andersson, M. R. Adv. Energy Mater. 2013, 3, 806–814. Kim, Y.; Yeom, H. R.; Kim, J. Y.; Yang, C. Energy Environ. Sci. 2013, 6, 1909. Wang, E.; Hou, L.; Wang, Z.; Hellström, S.; Zhang, F.; Inganäs, O.; Andersson, M. R. Adv. Mater. 2010, 22, 5240–5244. Diaz de Zerio Mendaza, A.; Bergqvist, J.; Bäcke, O.; Lindqvist, C.; Kroon, R.; Gao, F.; Andersson, M. R.; Olsson, E.; Inganäs, O.; Müller, C. J. Mater. Chem. A 2014, 2, 14354. Angmo, D.; Bjerring, M.; Nielsen, N. C.; Thompson, B. C.; Krebs, F. C. J. Mater. Chem. C 2015, 3, 5541–5548. de Zerio Mendaza, A. D.; Melianas, A.; Rossbauer, S.; Bäcke, O.; Nordstierna, L.; Erhart, P.; Olsson, E.; Anthopoulos, T. D.; Inganäs, O.; Müller, C. Adv. Mater. 2015, 27, 7325– 7331. Savagatrup, S.; Rodriquez, D.; Printz, A. D.; Sieval, A. B.; Hummelen, J. C.; Lipomi, D.

ACS Paragon Plus Environment

17

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(42) (43)

(44)

(45) (46) (47) (48) (49) (50) (51)

Page 18 of 28

J. Chem. Mater. 2015, 27, 3902−3911. Lindqvist, C.; Bergqvist, J.; Bäcke, O.; Gustafsson, S.; Wang, E.; Olsson, E.; Inganäs, O.; Andersson, M. R.; Müller, C. Appl. Phys. Lett. 2014, 104, 153301. Schroeder, B. C.; Li, Z.; Brady, M. A.; Faria, G. C.; Ashraf, R. S.; Takacs, C. J.; Cowart, J. S.; Duong, D. T.; Chiu, K. H.; Tan, C.-H.; Cabral, J. T.; Salleo, A.; Chabinyc, M. L.; Durrant, J. R.; McCulloch, I. Angew. Chem. Int. Ed. Engl. 2014, 53, 12870–12875. Lindqvist, C.; Bergqvist, J.; Feng, C.-C.; Gustafsson, S.; Bäcke, O.; Treat, N. D.; Bounioux, C.; Henriksson, P.; Kroon, R.; Wang, E.; Sanz-Velasco, A.; Kristiansen, P. M.; Stingelin, N.; Olsson, E.; Inganäs, O.; Andersson, M. R.; Müller, C. Adv. Energy Mater. 2014, 4, 1301437. Richards, J. J.; Rice, A. H.; Nelson, R. D.; Kim, F. S.; Jenekhe, S. A.; Luscombe, C. K.; Pozzo, D. C. Adv. Funct. Mater. 2013, 23, 514–522. Cheng, P.; Yan, C.; Wu, Y.; Wang, J.; Qin, M.; An, Q.; Cao, J.; Huo, L.; Zhang, F.; Ding, L.; Sun, Y.; Ma, W.; Zhan, X. Adv. Mater. 2016, 28, 8021–8028. Kalogeras, I. M.; Brostow, W. J. Polym. Sci. Part B Polym. Phys. 2009, 47, 80–95. Leman, D.; Kelly, M. A.; Ness, S.; Engmann, S.; Herzing, A.; Snyder, C.; Ro, H. W.; Kline, R. J.; DeLongchamp, D. M.; Richter, L. J. Macromolecules 2015, 48, 383–392. Antosiewicz, T. J.; Apell, S. P.; Claudio, V.; Käll, M. Opt. Express 2012, 20, 524–533. Nugroho, F. A. A.; Frost, R.; Antosiewicz, T. J.; Fritzsche, J.; Larsson Langhammer, E. M.; Langhammer, C. ACS Sensors 2016. doi: 10.1021/acssensors.6b00612 Kim, J. H.; Jang, J.; Zin, W.-C. Langmuir 2000, 16, 4064–4067.

ACS Paragon Plus Environment

18

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

FIGURES

Figure 1. a) Configuration schematic and working principle of plasmonic nanospectroscopy. Upon temperature change the phase (or RI) evolution of the studied polymer thin film induces a spectral shift of the LSPR extinction peak position of the plasmonic sensor. b) SEM images of a typical sensor surface taken at 70o (top) and 90o (bottom) tilt angle. The dashed yellow lines outline the 10 nm thick Si3N4 support layer that encapsulates the plasmonic silver nanoparticle sensors. The scale bar is 200 nm.

ACS Paragon Plus Environment

19

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 28

Figure 2. a) Extinction spectra of a plasmonic sensor coated with >150 nm PMMA (purple), APFO3 (brown), TQ1 (red), P3HT (light green), P3HT:PC61BM (magenta) and F8BT (orange). The extinction peaks in the white- and blue-shaded areas originate from absorption bands associated with the polymer and the sensor, respectively. Determination of thermal properties (glass transition ܶ௚ , melting transition ܶ௠ , and nematic-isotropic transition ܶ௟௖ ) from plasmonic nanospectroscopy measurements of b) PMMA, c) APFO3, d) TQ1, e) P3HT, f) P3HT:PC61BM and g) F8BT. The gray- and green-shaded areas denote the uncertainty in the derived ܶ௚ and ܶ௟௖ based on the 95% confidence interval of the linear fits (Figure S6) and the onset and endset of ܶ௠ , respectively.

ACS Paragon Plus Environment

20

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 3. Determination of ܶ௚ from plasmonic nanospectroscopy measurements of ~150 nm thick a) TQ1, b) TQ1:PC61BM:PC71BM and c) PC61BM:PC71BM films. The gray-shaded areas denote the uncertainty in the derived ܶ௚ based on the 95% confidence interval of the linear fits. Inset: Chemical structure of PC61BM and PC71BM.

ACS Paragon Plus Environment

21

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 28

Figure 4. AFM images of a) a blank Ag nanodisk sensor, and sensors coated with b) 21 nm TQ1:PC61BM:PC71BM, c) 58 nm PC61BM:PC71BM, and d) 65 nm TQ1. The thicknesses of the polymer and fullerene alloy represent the thinnest investigated films in this study. b) Topographical profiles obtained from the images for the blank sensor (black) and TQ1 (red), PC61BM:PC71BM (blue) and TQ1:PC61BM:PC71BM (green). Note that the polymer height profiles are shifted upward according to their nominal thicknesses to portray the actual configuration. In the case of the 21 nm TQ1:PC61BM:PC71BM ternary blend, a semi-conformal coating is observed with only ~10 nm film on top of the nanodisk. For the thicker films of TQ1 and PC61BM:PC71BM surface corrugation induced by the disk is much less pronounced and thus negligible.

ACS Paragon Plus Environment

22

Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 5. FDTD-calculated scattered electromagnetic field distribution at resonance in the polarization plane of incident light around a silver nanodisk sensor coated with a 10 nm Si3N4 support layer for four different scenarios: a) blank sensor (ߣ௣௘௔௞ = 810 nm), b) semi-conformal 23 nm thick TQ1:PC61BM:PC71BM blend film with a 10 nm cap on top of the disk (ߣ௣௘௔௞ = 890 nm). This corresponds to the experimental situation derived from AFM imaging (cf. Figure 4) for the thinnest measured system. c) fully conformal 23 nm thick TQ1:PC61BM:PC71BM blend film (ߣ௣௘௔௞ = 910 nm), and d) bulk TQ1:PC61BM:PC71BM blend film (ߣ௣௘௔௞ = 980 nm). The white lines mark interfaces between media.

ACS Paragon Plus Environment

23

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 28

Figure 6. ܶ௚ as function of film thickness for TQ1 (◊), TQ1:PC61BM:PC71BM (□) and PC61BM:PC71BM (○). The error bars correspond to the uncertainty in the procedure used to determine the ܶ௚ , as discussed in the main text. The solid and dashed lines indicate the best fit based on the model51 to the measured data points and the derived ܶ௚௕௨௟௞ , respectively.

ACS Paragon Plus Environment

24

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Analytical Chemistry

Table 1. Thermal Transition Temperatures Measured with Plasmonic Nanospectroscopy and DSC.a polymer structure

ܶ௚ (°C)

plasmonic

ܶ௠ (°C)

DSC

nanospectroscopy

plasmonic

ܶ௟௖ (°C)

DSC

nanospectroscopy

plasmonic

DSC

nanospectroscopy

PMMA n

108

124

-

-

-

-

124

109

-

-

n.m.

197

113c

n.m.

-

-

n.m.

300d

n.m.

12-14

190-235

235

-

-

74

60

166-205

198e

-

-

APFO3b

n

TQ1

n

P3HT n

1:1 P3HT: PC61BM

ACS Paragon Plus Environment

25

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 26 of 28

F8BTf n

109

120

199-226

223

259

270

a

glass transition temperature ܶ௚ (inflection point for DSC), melting temperature ܶ௠ (onset/endset for plasmonic nanospectroscopy and peak value for DSC), and nematic to isotropic transition temperature ܶ௟௖ (peak value for DSC). n.m. = not measured. bsame APFO3 batch and DSC measurement as in Ref. 4; cextrapolated bulk values (cf. Table 2); dvalue from Ref. 35; eܶ௠ (cf. SI and Ref. 32); fsame F8BT batch and DSC measurement as in Ref. 34.

ACS Paragon Plus Environment

26

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Table 2. Fitting Parameters for the Model by Kim et al.51 TQ1

TQ1:PC61BM:PC71BM

PC61BM:PC71BM

113

120

125

k (nm)

6.9

6.7

2.8

ߝ (nm)

3.9

4.4

1.2

ܶ௚௕௨௟௞ (℃)

ACS Paragon Plus Environment

27

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 28

TOC

ACS Paragon Plus Environment

28