Impact of Alkyl Chain Length on Small Molecule Crystallization and

In this work, we have investigated a series of aniline-based squaraines, with varying solubilizing alkyl chains, as donor materials in bulk heterojunc...
0 downloads 0 Views 9MB Size
Subscriber access provided by University of Newcastle, Australia

Article

Impact of Alkyl Chain Length on Small Molecule Crystallization and Nanomorphology in Squaraine-Based Solution Processed Solar Cells Chenyu Zheng, Ishita Jalan, Patrick Cost, Kyle Oliver, Anju Gupta, Scott T. Misture, Jeremy Alan Cody, and Christopher J. Collison J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 21 Mar 2017 Downloaded from http://pubs.acs.org on March 22, 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.

The Journal of Physical Chemistry C 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 45

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

The Journal of Physical Chemistry

Impact of Alkyl Chain Length on Small Molecule Crystallization and Nanomorphology in Squaraine-Based Solution Processed Solar Cells Chenyu Zheng1,2, Ishita Jalan3, Patrick Cost3, Kyle Oliver3, Anju Gupta4, Scott Misture5, Jeremy A. Cody3, and Christopher J. Collison1,2,3* 1

Microsystems Engineering, 2NanoPower Research Laboratory (NPRL), 3School of Chemistry

and Materials Science, 4Chemical Engineering, Rochester Institute of Technology, Rochester, NY 14623; 5Inamori School of Engineering, Alfred University, Alfred, NY 14802.

Abstract In this work, we have investigated a series of aniline based squaraines, with varying solubilizing alkyl chains, as donor materials in bulk heterojunction (BHJ) solar cells. Although these squaraine molecules exhibit similar absorbance spectra and crystal structure, the difference in properties that drive the OPV performance becomes apparent when blending each squaraine with PCBM. Thin film X-ray diffraction results demonstrate a disruption of squaraine crystallization in the presence of PCBM, more so for shorter side chain squaraines. As a result, the holemobilities of BHJ films of shorter side chain squaraines show the largest drop when compared to their neat films, whereas the mobility decrease for the longer side chain counterparts is small. However, morphological studies have shown that the phase separation rapidly happens during the spin casting process for longer side chain squaraines. Ultimately it is the extent of phase separation that dominates the final device efficiency. Therefore, rational design can greatly be influenced as a result of our systematic materials properties overview for anilinic squaraines targeted for OPV.

ACS Paragon Plus Environment

The Journal of Physical 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

Introduction Solution processed organic photovoltaic (OPV) devices are attractive because they promise low cost, flexible and aesthetically pleasing solar energy harvesting devices.1,2 The power conversion efficiency of OPV has been improved significantly to an unprecedented 10-12% owing to the development of novel donor and acceptor materials.3–5 Although the recent progress in efficiency has been made through using non-fullerene acceptor materials,6–8 fullerene derivatives, such as PC61BM or PC71BM, remain the benchmark acceptors for developing efficient donor polymers and small molecules (SM). Small donor molecules have the advantages of well-defined structure, and thus a better batch-to-batch reproducibility over their polymer counterparts. Up to now, power conversion efficiencies of 8-10% can be achieved with a few SM donors, including those based on benzodithiophene,9–11 indacenodithiophene12,13 dithienosilole14–16 and porphyrin units.17,18 Squaraine (SQ) dyes are highly investigated small molecule donor materials for solution processed OPV due to their ease of high purity synthesis and high extinction coefficients in the near-infrared region.19–24 With a simple molecular structure, DiBSQ(OH)2 (or DIBSQ) based OPV cells have achieved a power conversion efficiency of 5-7% in solution processed and vacuum deposited photovoltaic cells.25,26 It has also been reported that the high stability of this SQ allows the active layer to be solution processed in ambient conditions.27 Recently, research efforts on molecular design,28–30 device architecture31 and use in ternary blends32 has led to rapid and significant advances in SQ-based OPV research. Power conversion efficiency of over 7% has now been achieved for solution processed solar cells33 and over 10% for quaternary OPV34. The simple and high-yield synthesis, high efficiency and strong stability of SQ materials mark the potential of this material for commercialization of OPV.

ACS Paragon Plus Environment

Page 2 of 45

Page 3 of 45

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

The Journal of Physical Chemistry

Successful OPV application of SQ molecules is at least partially related to their ability to aggregate or crystallize in thin films either upon spin casting or after subsequent annealing treatment. Highly crystalline SQ films lead to a significant increase in exciton diffusion length and charge mobility, and thus improved power conversion efficiency (PCE).25,35 For example, Wei at el has reported an increase in exciton diffusion length of SQ by a factor of 3 after thermally annealing the DiBSQ(OH)2 thin films before C60 and aluminum cathode deposition.35 For solution processed bulk heterojunction (BHJ) cells using DiBSQ(OH)2, the cell PCE was significantly increased by over 100% after solvent vapor annealing (SVA) in dichloromethane (DCM) vapor.25 In our previous study, however, a decrease in device PCE was found after thermal annealing for some other SQ:PCBM systems, in which the only molecular structure difference of our SQs as compared to DiBSQ(OH)2 is the choice of side chains.20,36 Our recent study has shed light on discrepancies regarding the influence of aggregation (or crystallization) on solar cell performance where we pointed out that the efficiency roll-off of our SQ:PCBM BHJ solar cells is due to over-developed phase separation.37 Therefore, controlling donor acceptor phase separation is critical for highly efficient SQ:PCBM based solar cells. Modification of alkyl side groups for molecular or polymeric donor materials, has been shown to be an effective way to control the phase separation between donor and acceptor in BHJ.38–41 For example, Gadisa et al. studied the effect of alkyl side chain length of poly(3-alkyl thiophene) (P3AT), specifically with butyl (i.e. P3BT), pentyl (i.e. P3PT) and hexyl (i.e. P3HT) side chains, on morphology and charge transport in P3AT:PCBM BHJ.38 The results showed that the P3HT:PCBM BHJ, with a higher degree of phase separation, has a more balanced bipolar charge transport in BHJ and thus a better device performance. In a previous work by Nguyen et al, a further increase in alkyl side groups, for example to octyl (i.e. P3OT) and decyl (i.e. P3DT), was

ACS Paragon Plus Environment

The Journal of Physical 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

found to significantly aggravate the phase separation and led to unsatisfactory morphology and device efficiency.39 In the present paper, we study the influence of solubilizing alkyl side groups on hole mobility, film crystallinity, phase separation and, subsequently, solar cell performance of SQ:PCBM BHJ. A series of aniline based SQ molecules with n-propyl (i.e. DPrSQ(OH)2), n-butyl (i.e. DBSQ(OH)2), n-pentyl (i.e. DPSQ(OH)2) and n-hexyl (i.e. DHSQ(OH)2) carbon chains were synthesized according to the procedure described previously.42 These SQ molecules show very similar properties in solution and as neat films, as they share the same conjugated backbones and the same crystal packing motif. However, we observe different degrees of aggregate disruption in as-cast SQ:PCBM blend films; for SQs with longer alkyl groups, the aggregation is less disrupted in the presence of fullerenes, leading to a higher degree of film crystallinity. Subsequently, the phase separation is more profound for SQs with longer alkyl chains. As a result, OPV performance is in the order of DBSQ(OH)2 > DPSQ(OH)2 > DHSQ(OH)2. The low efficiency of devices employing DHSQ(OH)2 is related to its non-optimal BHJ nanomorphology with extensive phase separation, while the best performing BHJ of DBSQ(OH)2:PCBM is essentially a uniform mixing of two components (X-ray diffraction study has demonstrated that the optimal BHJ films are essentially amorphous). Although hierarchical phase separation and polymer crystallization has often been considered as critical to achieve high efficiency in polymer/fullerene solar cells,43,44 our observations suggest a different case for SQ based OPV. This is likely due to the smaller (when compared to polymers) exciton diffusion length, Ld = 1~5 nm,35,45 measured in typical SQ films; efficient exciton dissociation cannot be achieved once the domain size is above this value. Our work has shown that longer side chain length (from butyl to hexyl) will exacerbate the phase separation between squaraine donor and fullerene acceptor in

ACS Paragon Plus Environment

Page 4 of 45

Page 5 of 45

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

The Journal of Physical Chemistry

spin-cast BHJ films, leading to a large decrease in solar cell performance. However, shortening the side chain will significantly reduce the solubility of SQ in conventional organic solvents (as proven by the poor solubility of DPrSQ(OH)2). Thus, we highlight that to shorten the alkyl chains while maintaining enough solubility is more strategic for future designing efficient squaraine donor molecule.

Experimental section Three SQ molecules: DBSQ(OH)2, (2,4-bis[4-(N,N-dibutylamino)-2,6-dihydroxyphenyl]squaraine), DPSQ(OH)2, (2,4-bis[4-(N,N-dipentylamino)-2,6-dihydroxyphenyl]squaraine) and DHSQ(OH)2 (2,4bis[4-(N,N-dihexylamino)-2,6-dihydroxyphenyl]squaraine) were synthesized according to a one-pot twostep procedure.46 The corresponding amine was purchased and condensed at reflux with 1,3,5trihydroxylbenzene in a toluene / n-butanol (3:1, v/v) mixed solution. The yielded aniline intermediates were directly mixed with half equivalent squaric acid for the second condensation. All final products are green solids. PC61BM ([6,6]-phenyl C61 butyric acid methyl ester, >99.5%) and PC71BM ([6,6]-phenyl C71 butyric acid methyl ester, >99%) were purchased from Solenne BV; MoO3 (molybdenum trioxide, >99.5%) was purchased from Sigma Aldrich. All the materials are stored in a N2-filled glovebox and used as received. In the following content, the PC61BM is referred to as PCBM. For absorbance spectra, thin films were spin cast from 16 mg/mL chloroform solution at a spin speed of 1500 RPM. Thermal treatment, if indicated in the text, was performed by placing the film on a hot plate under a nitrogen environment with temperatures accurate to ± 5 ºC. Melting characteristics of the dyes were studied using a TA Instruments Q2000 Differential Scanning Calorimeter (DSC). Decomposition of the dyes was studied using a TA Instruments Q500 Thermogravimetric analyzer (TGA). Data was analyzed using the Universal Analysis 2000 software (TA

ACS Paragon Plus Environment

The Journal of Physical 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

Instruments). Each sample was heated at 10 °C/min from 25°C to 270°C for DSC and at 5°C/min from 25°C to 400°C for TGA under nitrogen atmosphere (40 mL/min) on a hermetic pan. For organic Photovoltaic devices, patterned ITO substrates were consecutively cleaned in an acetone and an isopropanol ultrasonic bath. A thin layer (8 nm) of MoO3 was evaporated onto the pre-cleaned ITO substrates at a rate of 0.5 Å/s. The ITO/MoO3 substrates were then transferred directly into a N2-filled glovebox for active layer spin casting. Solutions of SQ and PCBM (varying concentration corresponding to blend ratios) were prepared in chloroform as described above. The active layer solutions were spin coated at 1500 RPM in the glovebox, and were placed in a dark vacuum chamber to allow further evaporation of residual solvent. Finally, a shadow mask was applied and a 100 nm aluminum layer was vacuum evaporated through the shadow mask under low pressure (< 10-6 torr) to form the cathode. J-V characterization was performed on a Newport 91192 solar simulator at a power density of 100 mW/cm-2 (calibrated against standard InGaP solar cells fabricated in NASA Glenn Research Center, Photovoltaics Branch 5410) and by using a Keithley 2400 sourcemeter. For X-ray diffraction study, SQ or SQ:PCBM films were prepared on ITO/MoO3 (8 nm) substrates with the same procedure as used in preparing OPV devices. The X-ray diffraction was performed using a Bruker D8 Advanced system with the Bragg-Brentano geometry set up. For room temperature measurements, the X-ray diffraction signal is measured using a LYNXEYE XE position sensitive detector while the films were slowly spinning at a rate of 30 RPM. In-situ high temperature X-ray diffraction measurements (in-situ HTXRD) were performed with an Anton PAAR high temperature control system. The X-ray source is Cu Kα1 and Kα2 lines for both room temperature and high temperature. X-ray diffraction peaks are shown specifically for squaraine crystallinity. This was determined by scans on substrate controls with only ITO substrate and ITO/MoO3 substrate. Transmission electron microscopy images were obtained using a JEM-2010 instrument performed at 200 kV. Samples were prepared by removing the spin-cast active layer analogue films from their glass substrates by dipping them into water, and then placing them onto copper TEM grids covered with an amorphous carbon film. The water was

ACS Paragon Plus Environment

Page 6 of 45

Page 7 of 45

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

The Journal of Physical Chemistry

allowed to evaporate under ambient conditions. Average domain size was quantified using ImageJ software.

Results and discussion

Scheme 1. Molecular structure of aniline based SQ molecule. R = n-propyl, n-butyl, n-pentyl and n-hexyl for DPrSQ(OH)2, DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2 respectively.

All SQs were synthesized according to a one-pot two-step procedure.46 The generalized molecular structure is shown in Scheme 1. DBSQ(OH)2 and DPSQ(OH)2 are small needle-like crystallites, and DHSQ(OH)2 is a dull fiber-like solid. Except for DPrSQ(OH)2, the solubilities of all SQs studied here, in chloroform, are > 24 mg/mL. Generally, SQs with longer alkyl substituents have higher solubility in the organic solvents. The present study will focus on DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2. The thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) results for these SQs are shown in Figure 1. The onset degradation temperature is measured to be 260-265 0C for all three. It is noteworthy that DPrSQ(OH)2 exhibits a slightly higher onset degradation temperature of 283 0C. A higher onset degradation temperature would be more advantageous for device stability but its dramatically decreased solubility (3-4 mg/mL) in common organic solvents limits the practical benefits of DPrSQ(OH)2 in solution processes.

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 1. a) Thermogravimetric analysis and b) differential scanning calorimetry results of SQ materials. The asterisks highlight a unique endothermic peak at 100 0C of DBSQ(OH)2 powder solids.

Differential scanning calorimetry (DSC) studies demonstrate that the melting and crystallization temperatures of SQ materials decrease as the side chain increases (Figure 1b). The onset melting temperatures are 235 0C, 225 0C and 205 0C respectively, and the onset crystallization temperatures are 198 0C, 181 0C and 178 0C respectively for dibutyl, dipentyl and dihexyl SQs. As an aside, there is a small endothermic peak at 95-110 0C for DBSQ(OH)2 and we investigated this further using in-situ XRD. We found evidence for a phase change at 90 0C (Figure S1), with a corresponding reverse to the original phase upon cooling back down to room temperature. These data demonstrate the existence of two polymorphs for DBSQ(OH)2 powders. We therefore assign the endothermic peak found in the DSC to this phase change with the assumption that there are no surface bound solvents in the powder sample. Most importantly, the clear trend in our data is for both melting point and crystallization temperature to decrease as the squaraine side chain length is increased. SQs are known to have a high propensity to aggregate upon spin casting from solutions.20,47–49 One immediate phenomenon accompanying this aggregation is a change in absorbance spectrum.

ACS Paragon Plus Environment

Page 8 of 45

Page 9 of 45

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

The Journal of Physical Chemistry

In a previous study, we have investigated SQ aggregation in solvent mixtures (i.e. water and dimethyl sulfoxide), as a model for the thin film condition.42 We found that the typical absorbance spectrum associated with these SQ aggregates can be described as a broad “doublehump”; one of the peaks is blue shifted and has a higher oscillator strength, while the other peak is slightly red-shifted with respect to the monomer peak in solution. Importantly, the red-shifted peak is not associated with SQ monomer or J-aggregate. In this case, the aggregates of these aniline-based SQs cannot be simply explained by H- or J-aggregates. Rather, the unusual absorbance spectra result from significant intermolecular charge transfer (ICT) contributions when the molecules are tightly packed.42 Therefore, we assign this double hump absorbance feature to SQ aggregates, in general.

Figure 2. Normalized absorbance spectra of three SQs in chloroform solution (dashed line) and as neat films (solid line). The absorbance spectra for all three solutions overlap each other with the same peak position. The SQ neat films are annealed at 90 0C for 5 min to ensure a complete aggregation of SQ molecules.

All three SQs share typical absorbance spectra in solution and films, as shown in Figure 2. The solution absorbance is sharp and intense, with a strong 0-0 transition peak at 650 nm (ε = 3 ~ 4 ×

ACS Paragon Plus Environment

The Journal of Physical 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

105 M-1 cm-1) and a weak 0-1 transition peak at 600 nm. In the solid state, the neat films all show the broad “double hump” absorption feature, each with a blue-shifted peak at 550 nm and another peak at 660 nm. After annealing the neat films at 90 0C, the absorbance spectra show little change for DPSQ(OH)2 and DHSQ(OH)2; but for DBSQ(OH)2, the absorption changes a little with further splitting between the peaks (Figure S2), which is consistent with previous studies.20 This change in absorbance indicates that DBSQ(OH)2 may have re-arranged itself in a tighter crystal packing structure under annealing. The lack of significant spectral change upon annealing indicates that the crystallization is immediately complete after spin casting of DPSQ(OH)2 and DHSQ(OH)2 solutions. Overall, the neat film absorbance spectra of all three SQs are very similar, which undoubtedly results from the shared molecular backbone, as well as the similar crystal packing motif.

Figure 3. a): Molecular packing of DBSQ(OH)2 single crystal, featuring a π-π stacking with slippages in both long and short molecular axes; other SQs adopt a similar slip stacking motif in the single crystal. b): a simple cartoon illustrates the slip stack of the SQ molecules with an artificial Cartesian coordinate; x and y axes are along the short and long molecular backbones respectively, so the z axis is perpendicular to the parallell π-stacked molecular planes. Red dots: oxygen atoms on the squaric acid moiety; blue dots: nitrogen atoms to which the alkyl groups attached (omitted).

ACS Paragon Plus Environment

Page 10 of 45

Page 11 of 45

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

The Journal of Physical Chemistry

The single crystal structure of DBSQ(OH)242 is shown in Figure 3a. Single crystal structures of DPSQ(OH)2 and DHSQ(OH)2 exhibit the same stacking motif and can be found in the supporting information and ref [50] respectively. For each SQ molecule, the backbone is planarized by intramolecular hydrogen bonds between the hydroxyl group on the aniline ring and oxygen on the squarylium; the alkyl chains are pointing up and down with respect to the backbone plane for each side, respectively. The overall molecular packing can be described as a slip stack. The cartoon in Figure 3b illustrates one way to quantify this slip stack structure using artificial Cartesian coordinates. The y-axis is defined by the molecular long axis on which are located the two nitrogen atoms (blue dots), the x-axis is defined by the short axis on which are located the two squaric oxygen atoms (red dots). The z-axis is perpendicular to the xy plane. In this way, the small differences in crystal structure of three SQs can be quantified and the results are summarized in Table 1. Overall, the difference in π-π stacking distance, ∆z, is less than 0.1 Å and the differences in slips in (x, y) axes, ∆x and ∆y, are less than 0.2 Å. The shared molecular backbone and similar crystal structures can explain the overlapping absorbance spectra of these SQs in solution and film as shown in Figure 2. Table 1. Unit cell dimensions measured for DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2 single crystals, along with the calculated slippages in the short and long molecular axes (i.e. ∆x and ∆y respectively) and π-π interplanar distances (i.e. ∆z). Compound

a/Å

b/Å

c/Å

∆x / Å

∆y / Å

∆z / Å

DBSQ(OH)2[42]

5.169(4)

10.846(9)

13.538(11)

1.879

3.470

3.353

DPSQ(OH)2

5.227(2)

10.694(4)

15.412(5)

1.688

3.621

3.371

DHSQ(OH)2[50]

5.097

10.746

16.604

1.734

3.498

3.276

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 4. (a) Absorbance spectra of as-cast SQ:PCBM blend films; (b) SQ monomer absorbance peak in 98 wt% PCBM. The inset corresponds to raw absorbance spectra. The dashed line represents the absorbance spectra of SQ in chloroform solution. (c) Absorbance spectra of SQ:PCBM blend films after annealing at 90 0C for 5 min.

In contrast to the similar absorbance spectra of solution and neat films in Figure 2, the absorbance spectra of as-cast SQ:PCBM blend films are distinct from each other, as shown in Figure 4a. Here, PCBM is used instead of PC71BM to open a wide spectral window (500-750 nm) through which the absorbance of SQ aggregates can be clearly seen. For future brevity, the absorbance of SQ aggregates refers to the “double hump” absorbance from 500-800 nm, while the absorbance of the SQ monomer in spin cast SQ:PCBM films refers to the absorbance peaks at 678 nm (for all SQs) with 2 wt% of SQ and 98 wt% of PCBM, as shown in Figure 4b. It should be noted that the squaraine monomer absorbance peaks in the SQ:PCBM blends are redshifted by 28 nm from the solution absorbance peaks. This is attributed to the different dielectric constants of the environments that the SQ molecules are subjected to (i.e. PCBM vs. chloroform). With absorbance peaks properly assigned, we can see that DBSQ(OH)2 shows a disruption of aggregation in the presence of 50 wt% PCBM, as evidenced by a loss of the double hump and a peak emerging at 677 nm, similar to the peak at 678 nm of the (monomer) blend films with 2

ACS Paragon Plus Environment

Page 12 of 45

Page 13 of 45

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

The Journal of Physical Chemistry

wt% of SQ. But if one compares these two spectra, the SQ absorbance peak in the 50 wt% PCBM films is broader than that in 98 wt% PCBM films, with a distinguishable “shoulder” at 600 nm. This shoulder decreases as the PCBM weight ratio is increased and, therefore, it is ascribed to the weak interaction between SQ molecules in the SQ:PCBM blend films.20 On the other hand, the DHSQ(OH)2:PCBM blend film exhibits an absorbance spectrum similar to that of the neat film, indicating the aggregates are formed completely. Interestingly, for the DPSQ(OH)2:PCBM blend film, an intermediate spectrum is exhibited, with a distinguishable aggregate double hump as well as a monomer absorption feature. Similar absorbance spectra were observed for other blend ratios, from 5:5 to 2:8, as shown in Figure S3. Specifically, DBSQ(OH)2 aggregates are completely disrupted throughout different blend ratios (with the “shoulder” decreasing as the PCBM ratio increases), DPSQ(OH)2 shows absorbance spectra indicating mixtures of aggregate and monomer and the aggregates of DHSQ(OH)2 are gradually disrupted by increasing the weight ratio of PCBM, but a significant amount of aggregates are still formed in a 2:8 SQ:PCBM blend ratio. Even in 2:98 blend ratios, a small DHSQ(OH)2 aggregate absorbance peak at 550 nm can clearly be observed in the spectrum (Figure 4b). Thus, SQ aggregate formation, when blended with PCBM, is favored by longer alkyl side groups. Upon annealing the SQ:PCBM films at 90 0C for 5 min, the absorbance spectra of all three blend films start to resemble those of the neat films, as seen in Figure 4c, with the biggest spectral change being observed for the DBSQ(OH)2:PCBM blend. This spectral change suggests i) initial disorder of DBSQ(OH)2 and DPSQ(OH)2 molecules in the as-cast blend films and ii) these SQ molecules can easily self-assemble or aggregate in films upon thermal annealing.

ACS Paragon Plus Environment

The Journal of Physical 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

The results of Figure 4 are explained as follows. The disruption of aggregation in as-cast DBSQ(OH)2:PCBM blends is due to the rapid loss of the solvent environment during spin casting with the materials being frozen into a kinetically stable mixing state, not too dissimilar to the homogeneous solution phase. DPSQ(OH)2 and DHSQ(OH)2, however, manage to form ordered structures (i.e. aggregates) in the blend films even during spin casting. Due to such differences in aggregation for these three SQ molecules when blended with PCBM, the heterojunction blends are expected to have varying film crystallinity, hole mobility properties, film morphologies and subsequently different organic photovoltaic performances.

Figure 5. Hole mobilities of SQ neat films and SQ:PCBM blend films as a function of the number of side chain carbons of the SQ molecule. The measurements were done on unannealed films.

Charge mobility is a critical factor that influences the free charge extraction versus recombination in the bulk heterojunction layer, and thus can significantly affect the solar cell short-circuit current and fill factor.51,52 Previous work by Proctor et al53 has shown that in solution processed small molecule (SM) solar cells with fill factor > 0.65, the hole mobilities of the donor acceptor blends are almost exclusively on the order of 10-4 cm2/V·s. Indeed, the hole mobility is often measured to be the lower limit of the bipolar charge transport in SM:PCBM

ACS Paragon Plus Environment

Page 14 of 45

Page 15 of 45

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

The Journal of Physical Chemistry

bulk heterojunctions.53–56 Therefore, it is considered a critical property to evaluate small donor molecules for OPV application. Here, the hole mobilities of the pristine squaraine films as well as of the SQ:PCBM blend films are extracted by fitting the current-voltage curves of single carrier devices to the Mott-Gurney law for the voltage region where the current is space-charge limited.57 The hole only devices are fabricated in the structure of ITO/MoO3/SQ or SQ:PCBM /MoO3/Al.28 The experimental data and fitting curves can be found in Table S1 and Figures S4S6. The average mobility data vs. number of side chain carbons is shown in Figure 5. The hole mobilities of SQ neat films are measured to be 3.1×10-4 cm2/V·s, 3.0 ×10-4 cm2/V·s, 1.4×10-4 cm2/V·s for DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2 respectively. In neat films we recall that the absorption spectra are similar for each SQ, representing a similar extent of crystallinity. Therefore we can attribute the decrease in hole mobility, as the side chain length increases, to the detrimental impact of non-conductive alkyl groups. Overall, the hole mobilities of SQ neat films are all above 10-4 cm2/V·s. Yet, after blending the SQ with PCBM, the mobility immediately drops by up to an order of magnitude, to ߤ௛ = 4.2×10-5 cm2/V·s, 5.5 ×10-5 cm2/V·s, 6.7×10-5 cm2/V·s for DBSQ(OH)2:PCBM, DPSQ(OH)2:PCBM and DHSQ(OH)2:PCBM blends respectively at the 5:5 ratio. Interestingly, if one compares the mobility and absorbance spectra of neat films to SQ:PCBM blend films, a trend is readily observed; for SQs with larger spectral change when blended with fullerene (as compared to SQ neat films), the drop in hole mobility is also more significant. Specifically, DHSQ(OH)2 is able to retain 50% of its neat film mobility in SQ:PCBM 5:5 (w/w) blend films, whereas the mobility of DBSQ(OH)2 drops to only 10% of its neat film mobility for films with the same weight to weight blends. As a result, the hole mobilities of the blended films are similar to each other, with DHSQ(OH)2 blends taking the highest value. Therefore, we attribute this decrease in mobility to the disruption of the packing

ACS Paragon Plus Environment

The Journal of Physical 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

order in the SQ donor phases. We are reminded that in Figure 2, the aggregation of DBSQ(OH)2 is completely disrupted by PCBM at a 5:5 weight ratio, whereas DHSQ(OH)2 molecules remain completely aggregated in the same blend ratio. To further validate this interpretation, we investigated the crystallinity of these blend films by using X-ray diffraction (XRD). In the XRD study, the films were prepared by spin casting neat SQ solution or SQ:PCBM solution onto MoO3 treated ITO substrates, to resemble the condition of the films prepared for hole-only and solar cell devices. The XRD system is configured in a Bragg-Brentano geometry. The XRD patterns of blend films before (i.e. as-cast) and after thermal annealing are shown in Figure 6. The XRD patterns of ITO/MoO3 substrates were recorded and weak diffraction peaks of the ITO layer were identified at 2θ = 21.50 (corresponding to (2 1 1) plane), 2θ = 30.60 (corresponding to (2 2 2) plane) and 2θ = 35.50 (corresponding to (4 0 0) plane) 58; no diffraction peaks were identified for the thin MoO3 layer (thickness = 8 nm). Besides these peaks, the crystalline SQ diffraction peaks of neat and blend films are found at 2θ = 50 ~ 60 for all three SQs.

ACS Paragon Plus Environment

Page 16 of 45

Page 17 of 45

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

The Journal of Physical Chemistry

Figure 6. X-ray diffraction (XRD) patterns of a) SQ neat films and b) SQ:PCBM blend films (5:5 w/w) before (black) and after (red) thermal annealing. The ITO substrate diffraction pattern is also provided in a) (green). Thermal annealing was done in-situ with temperature ramped up every 30 0C from 30 0C to 150 0C, then cooled down to 30 0C (data can be found in Figure S7). The results for annealed films in (a) and (b) refer to the films after they have been cooled back down to 30 0C. In order to clearly resolve all the diffraction peaks, the peak intensities were plotted logarithmically in both a) and b).

In thin film XRD patterns of neat SQ films (Figure 6a), the SQ crystalline structures can be identified with a sharp peak at 2θ = 6.010, 5.680 and 5.010 for thermally annealed DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2 neat films, respectively. The DHSQ(OH)2 neat film pattern also features a weak peak at 2θ = 9.910, suggesting a higher degree of crystallinity of this SQ in spincast films. The diffraction peaks at 2θ = 50 ~ 60 are close in position with the (0 0 1) plane powder diffraction peak, calculated based on SQ single crystal structures. Unfortunately, there is no further strong evidence (such as other diffraction peaks) to prove that the SQ crystalline

ACS Paragon Plus Environment

The Journal of Physical 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

structure in neat films resembles the single crystal structure. Here, we only use this diffraction peak to mark the crystallinity of the films; the more intense and narrow this peak, the greater the extent of crystallinity in the films. When comparing the peak intensity and full width at half maximum (FWHM), we conclude that DHSQ(OH)2 neat films exhibit the highest level of crystallinity. The as-cast DPSQ(OH)2 neat films are also highly crystalline. The thermal annealing does not significantly change the intensity and shape of the diffraction peaks of these two SQ neat films, consistent with the observation in absorbance spectra after annealing. The XRD patterns of as-cast DBSQ(OH)2 neat films only show a very weak XRD peak at 2θ = 6.010 and multiple shoulders at 2θ = 70 ~ 80, consistent with the as-cast DBSQ(OH)2 neat films being more amorphous. A more crystalline film is obtained after annealing, indicated by a significant increase in the diffraction peak at 2θ = 6.010 and disappearance of the shoulder peaks. This data complements the endothermic peak found in DSC data (Figure 1) and the spectral change observed after annealing in absorbance (Figure 4). In general, SQs with longer side chains, i.e. DHSQ(OH)2 and DPSQ(OH)2, have a higher degree of crystallinity in the as-cast films. It is, interestingly, the opposite trend to that from the reported polymer side chain study by Nguyen et al39 where, as the side chain length increases, the XRD peak heights of the as-cast neat P3AT polymer films decrease. For XRD patterns of blend films (with SQ:PCBM 5:5 w/w) shown in Figure 6b, the results correspond very well with the absorbance spectral data in Figure 4. First, for DBSQ(OH)2:PCBM blends, the absorbance spectra indicate a complete disruption of SQ aggregation. Correspondingly, there are no observable diffraction peaks in the XRD patterns of the DBSQ(OH)2:PCBM films. Second, DHSQ(OH)2 is fully aggregated in blend films as indicated by absorbance. Consistently, a sharp peak at 2θ = 5.140 is observed in the XRD

ACS Paragon Plus Environment

Page 18 of 45

Page 19 of 45

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

The Journal of Physical Chemistry

patterns together with a weak peak at 2θ = 9.990. These two peaks for the blended films are consistent with the diffraction patterns of neat DHSQ(OH)2 films (2θ = 5.010 and 9.910), indicating that this SQ can retain highly crystalline structure in the films even in the presence of PCBM. Third, DPSQ(OH)2:PCBM blend films act in an intermediate way, having a certain degree of crystallinity as predicted by absorbance spectra. Fourth, after thermal annealing, XRD patterns of all three SQ:PCBM blend films exhibit a high film crystallinity with their peaks resembling those of the annealed neat SQ films. These trends are wholly consistent with our observations in absorbance spectra. With the XRD data, the hole mobility results can be explained. In neat SQ films, where aggregation is nearly complete, increasing non-conductive alkyl chain length leads intuitively to a decrease in the hole mobility. Therefore, DBSQ(OH)2 has the highest hole mobility in neat film SCLC measurements. This is consistent with the large body of research on small molecules15,40,59 and polymers38,39. In SQ:PCBM blend films, mobility is also related to the degree of crystallinity of the donor phases. The mobility for the DHSQ(OH)2:PCBM blend is comparable to the mobility in its neat films, associated with an ability to retain the SQ crystalline structures in the as-cast blend films. For DBSQ(OH)2:PCBM blend films, the SQ crystallization is disrupted by PCBM and the films are amorphous with no distinguishable diffraction peaks in X-ray diffractograms, leading to a large drop in hole mobility from neat SQ films to blend films. One other important observation is that the two diffraction peaks of DHSQ(OH)2:PCBM films correspond exactly to the neat DHSQ(OH)2 films, suggesting that the crystallite dimension formed in both films is essentially the same. It indicates that the exclusion of PCBM from the SQ phase can effectively take place during the spin casting process.

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 7. Transmission electron microscope images of DBSQ(OH)2:PCBM (left), DPSQ(OH)2:PCBM (middle) and DHSQ(OH)2:PCBM (right) blend films. The SQ:PCBM weight ratio is 5:5. Scale bars: 100 nm.

We further investigate the phase separation using bright field transmission electron microscopy (BF-TEM) and the images of SQ:PCBM blend films are shown in Figure 7. The origin of the contrast in the TEM images is ascribed to the density difference between SQ (1.2-1.3 g cm-3 as measured for SQ single crystals and considered the upper limit for the density of SQ films) and PCBM (ߩ ൎ 1.5 g cm-3)60–62. Thus, the bright phases are assigned to the SQ-rich domains and the dark phases are PCBM-rich domains, due to the relatively higher electron scattering density of PCBM as compared to SQs. Consistent with the XRD results, a very clear phase separation between DHSQ(OH)2 and PCBM is seen in the TEM image (Figure 7c). The diameters of the dark phases are estimated to be 60-80 nm. The domain sizes of DBSQ(OH)2:PCBM and DPSQ(OH)2:PCBM films cannot be determined but, nevertheless, one can clearly see a much more uniform morphology in these two blends. The SQ:PCBM bulk morphology is also evaluated by atomic force microscopy (AFM) operated in tapping mode, with the height images being shown in Figure S8. For AFM height images, the roughness of the surface of SQ:PCBM films is measured and decreases in the order of DHSQ(OH)2 > DPSQ(OH)2 > DBSQ(OH)2. DHSQ(OH)2:PCBM has the coarsest surface with

ACS Paragon Plus Environment

Page 20 of 45

Page 21 of 45

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

The Journal of Physical Chemistry

Rq = 3.07 nm. Amorphous films are expected to have finer granularity and to be smoother, whereas crystalline films are expected to show a courser topography associated with the larger domains. Therefore, these data are consistent with the aforementioned interpretations from both TEM and XRD results. Therefore, we conclude that SQs with longer side groups have a higher degree of crystallinity and subsequently more extensive phase separation in the as-cast films. Similar observations have been reported by Nguyen et al13 and Gadisa et al38 for polymer-fullerene systems and by Min et al40 for oligomer-fullerene systems. In particular, Nguyen et al reported a higher degree of polymer crystallization and phase separation in as-cast poly(3-alkylthiophene):PCBM films when the length of alkyl chain is increased. One explanation is that the longer side groups create more space between the rigid conjugated backbones, thus allowing improved diffusion of fullerenes towards increasing phase purity. Overall, we interpret our data in a similar way. The XRD results in Figure 6 show that peak position, 2θ, decreases (or d-spacing increases) as the side chain length is increased for SQ materials. If we assume the sharp XRD peak represents the (0 0 1) plane of SQ crystallites, then the c-axis configuration is calculated to be 13.52 Å, 14.31 Å and 16.22 Å for DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2, respectively. We also consider that as the side chains increase in size, the overall rigidity of the molecule is reduced, with a lower relative contribution of the intramolecular hydrogen bonding. This is consistent with the decreased melting and crystallization temperatures as shown in Figure 1. Thus, the flexibility towards diffusion of molecules (in particular, the fullerenes) is increased. In other words, the materials will be more “liquid-like” as the side chains increase, especially at the early stage of the solution drying process during spin casting. As fullerene self-assembly takes place, this leads to the enrichment of the pure phases and therefore the phase separation is accelerated, evidenced

ACS Paragon Plus Environment

The Journal of Physical 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 45

by the TEM images in Figure 7. After the fullerene diffuses out from the SQ:PCBM mixed phase, the remaining SQ molecules can aggregate or crystalize into pure phases with crystallite dimensions similar to that of the neat SQ films as shown in the absorbance spectra in Figure 4. Organic photovoltaic devices are constructed with the same conditions for all three squaraines, with a simple device structure of ITO/MoO3/SQ:PCBM or SQ:PC71BM/Al. The main results are listed in Table 2 and detailed blend ratio results are summarized in Table S2. Table 2. OPV parameters of best performing devices of DBSQ(OH)2, DPSQ(OH)2 and DHSQ(OH)2 when blended with PCBM and PC71BM, along with the averaged power conversion efficiencies. Donor

Acceptor

DBSQ(OH)2

PCBM PC71BM

DPSQ(OH)2

Jsc, mA/cm2

Voc, V

FF

PCE, % (ave.)a

5:5

7.6

0.85

0.50

3.1 (3.0)

1:2 (best)

8.5

0.84

0.52

3.7 (3.6)

1:2

10.2

0.85

0.55

4.8 (4.5)

5:5

6.2

0.86

0.52

2.8 (2.6)

3:7 (best)

7.6

0.85

0.51

3.3 (3.2)

3:7

7.6

0.85

0.50

3.3 (3.1)

5:5

6.7

0.86

0.44

2.6 (2.4)

3:7 (best)

6.8

0.84

0.50

2.9 (2.7)

3:7

7.2

0.82

0.42

2.5 (2.3)

PCBM PC71BM

DHSQ(OH)2

Blend ratios

PCBM PC71BM

a

The averaged data point is obtained by averaging over 12 devices on different films, as well as in different batches of device fabrication. The devices (without encapsulation) are tested in a N2 filled glove box under a solar simulator at AM 1.5G, 100 mW/cm2.

All three SQs are able to output moderate efficiencies (PCE ~ 2.6 - 3.7%) when blended with PCBM. Similar values for VOC are achieved for all three SQs (~ 0.85 V). This is consistent with the assumption that the alkyl side chains do not significantly perturb the electronic energy levels of these molecules. In general, higher device performance is achieved in the order of DBSQ(OH)2 > DPSQ(OH)2 > DHSQ(OH)2. The differences in fill factor and short circuit

ACS Paragon Plus Environment

Page 23 of 45

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

The Journal of Physical Chemistry

current in these solar cells should be highlighted. We recall that even though DBSQ(OH)2 and DPSQ(OH)2 each have a higher hole mobility when compared to DHSQ(OH)2, the hole mobilities of SQ:PCBM BHJ films are measured to be similar for all three SQs. Therefore, we cannot use mobility alone to explain the difference in solar cell performances. In fact, the difference in solar cell performance can only be well explained by considering BHJ morphology. For 5:5 ratio, the short circuit current decreases from 8.5 mA/cm2 to 7.6 mA/cm2, and then to 6.8 mA/cm2 as side chain length decreases, and the fill factors of DBSQ(OH)2 and DPSQ(OH)2 based devices are above 0.5 while the DHSQ(OH)2 based devices only exhibited FF = 0.44. This is related to the dissatisfactory nanomorphology of DHSQ(OH)2:PCBM at the 5:5 ratio, as shown in Figure 7. Specifically, the DHSQ(OH)2:PCBM blend phase separates extensively while the other two blends are well mixed. Noticeably, the DHSQ(OH)2:PCBM 3:7 blends yield a competitive FF of 0.5, leading to a slight increase in the efficiency to 2.9%. This is because the DHSQ(OH)2 aggregation is, to a certain degree, disrupted in the 3:7 blend, as compared to the 5:5 blend. We have previously shown that the SQ phase domain size is greatly reduced as the PCBM weight ratio increases.37 As a result, the DHSQ(OH)2:PCBM 3:7 w/w BHJ morphology has been slightly improved towards a more well mixed state, similar to DBSQ(OH)2:PCBM and DPSQ(OH)2:PCBM blends at the 5:5 weight ratio. Therefore, SQs with longer side chains need more PCBM to disrupt the aggregation and thus to suppress the phase separation. We then switched the fullerene acceptor from PCBM to PC71BM. PC71BM has a higher absorptivity in the visible region of the spectrum, and thus is expected to enhance the contribution of fullerene absorption to photo-generated charges. The efficiency is further increased by using PC71BM for DBSQ(OH)2, mainly due to the improved short-circuit current from 8.5 mA/cm2 to 10.2 mA/cm2, resulting in an increased power conversion efficiency to

ACS Paragon Plus Environment

The Journal of Physical 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

4.8%. Interestingly, the solar cell characteristics remain the same for DPSQ(OH)2 based devices independent of fullerene choice. For DHSQ(OH)2, the solar cell efficiency even decreases from 2.9% to 2.5%, after switching the acceptor from PCBM to PC71BM. This could be related to the enhanced phase separation when using PC71BM as the acceptor.59 Although the DBSQ(OH)2:PC71BM

blend has achieved the best solar cell efficiency, the XRD data suggest that the blend is

essentially amorphous, resulting from highly mixed SQ and fullerene. This amorphous state leads to a large drop in hole mobility in the blend films (3.1×10-4 cm2/V·s) as compared to the neat films (4.2×10-5 cm2/V·s), which is a significant drawback for charge transport and collection. We noticed that there are a few small molecules, such as the dithienosilole (DTS) based donors, which are able to retain high hole mobilities when blended with PCBM.53,63,64 As a result, the OPV devices produced by such materials are more efficient compared to our SQ-based OPV cells. We thus highlight future study to improve the hole mobility of SQ materials when blended with fullerenes.

Conclusion In this work, we have provided a comprehensive description of the properties of a series of squaraines with varying side-chain length, pertinent to their use in organic photovoltaics. Despite the molecular structure differences, the molecules pack with the same slip-stack motif, and absorbance spectra of neat films are very similar for each material. Nevertheless, when these squaraines are blended with PCBM, the differences in properties that drive OPV efficiency become apparent. Absorbance spectra indicate well a qualitative disruption of crystallinity, more so for short chain squaraines. For longer side-chain squaraines, phase separation is more

ACS Paragon Plus Environment

Page 24 of 45

Page 25 of 45

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

The Journal of Physical Chemistry

significant and some evidence of crystal structure is retained by DHSQ(OH)2:PCBM blends. Melting point and crystallization temperatures decrease as the side chains become larger, which leads to a greater ease of squaraine and PCBM self-assembly. Mobility is increased for neat films made with squaraines of short side chain length but when crystal packing is disrupted in blends, this mobility drops. For squaraines of longer side chain length, the blended films retain a higher mobility than their short chain counterparts, as a result of phase separation even during the spin casting process. However, ultimately it is the extent of phase separation itself that dominates the power conversion efficiency. When phase separation is too rapid, the interfacial heterojunction area will decrease and the films will be less bicontinuous with substantially more islands being distributed throughout. These results and interpretations culminate in an ongoing strategy to maximize ordered molecular packing while maintaining smaller domain sizes that nevertheless connect in a bicontinuous network. Shorter side-chain squaraines are therefore the best selection in our series. When devices are optimized with PC71BM the PCE is seen to rise to 4.8%. Solubility for squaraines decreases quickly as chain length is reduced further. However, by taking this into account rational design can greatly be influenced as a result of our systematic materials properties overview for aniline based squaraines targeted for OPV.

ASSOCIATED CONTENT Acknowledgement This work was carried out with the financial support from National Science Foundation (CBET1603372, DMR-1461063 and CBET-1236372). The XRD work was performed at Alfred University with the help from S.M. We thank Prof. Richard Hailstone for TEM measurements. C.Z thanks Dr. Anju Gupta from Rochester Institute of Technology (RIT) for TGA & DSC

ACS Paragon Plus Environment

The Journal of Physical 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

measurements and discussion, and Dr. Surendra Gupta from RIT for AFM and film thickness measurements. C.Z gratefully thanks Dr. Daobin Yang and Dr. Taojun Zhuang from Yamagata University for constructive discussion.

Supporting Information. Additional absorbance spectra, in-situ XRD patterns, J-V measurements and SCLC fitting curve for hole only devices and AFM images are provided in supporting information.

AUTHOR INFORMATION Corresponding Author* Correspondence should be addressed to Christopher J. Collison *Email: [email protected] Author Contributions P.C., K.O. and J.A.C. designed and synthesized the SQ molecules. C.Z. designed the experiments. C.Z. and I.J. performed the OPV device fabrication and testing, absorbance experiments, etc. S.M. and C.Z. performed XRD experiments. C.Z analyzed the results. C.Z. and C.J.C wrote the manuscript. All authors have reviewed and given approval to the final version of the manuscript.

References (1) (2) (3)

(4)

Carlé, J. E.; Krebs, F. C. Technological Status of Organic Photovoltaics (OPV). Sol. Energy Mater. Sol. Cells 2013, 119, 309–310. Darling, S. B.; You, F. The Case for Organic Photovoltaics. RSC Adv. 2013, 3, 17633–17648. Wang, Q.; Zhang, S.; Xu, B.; Li, S.; Yang, B.; Yuan, W.; Hou, J. Efficient Fullerene-Free Polymer Solar Cells Based on Alkylthio Substituted Conjugated Polymers. J. Phys. Chem. C 2017, 121, 4825–4833. Liu, Y.; Zhao, J.; Li, Z.; Mu, C.; Ma, W.; Hu, H.; Jiang, K.; Lin, H.; Ade, H.; Yan, H. Aggregation and Morphology Control Enables Multiple Cases of High-Efficiency Polymer Solar Cells. Nat. Commun. 2014, 5, 5293.

ACS Paragon Plus Environment

Page 26 of 45

Page 27 of 45

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

The Journal of Physical Chemistry

(5)

(6)

(7) (8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

Baran, D.; Kirchartz, T.; Wheeler, S.; Dimitrov, S.; Abdelsamie, M.; Gorman, J.; Ashraf, R. S.; Holliday, S.; Wadsworth, A.; Gasparini, N.; et al. Reduced Voltage Losses Yield 10% Efficient Fullerene Free Organic Solar Cells with >1 V Open Circuit Voltages. Energy Environ. Sci. 2016, 9, 3783–3793. Yang, L.; Zhang, S.; He, C.; Zhang, J.; Yao, H.; Yang, Y.; Zhang, Y.; Zhao, W.; Hou, J. New Wide Band Gap Donor for Efficient Fullerene-Free All-Small-Molecule Organic Solar Cells. J. Am. Chem. Soc. 2017, 139, 1958–1966. Zhao, W.; Li, S.; Zhang, S.; Liu, X.; Hou, J. Ternary Polymer Solar Cells Based on Two Acceptors and One Donor for Achieving 12.2% Efficiency. Adv. Mater. 2017, 29, 1604059. Ye, L.; Zhao, W.; Li, S.; Mukherjee, S.; Carpenter, J. H.; Awartani, O.; Jiao, X.; Hou, J.; Ade, H. High-Efficiency Nonfullerene Organic Solar Cells: Critical Factors That Affect Complex MultiLength Scale Morphology and Device Performance. Adv. Energy Mater. 2016, 1602000. Kan, B.; Zhang, Q.; Li, M.; Wan, X.; Ni, W.; Long, G.; Wang, Y.; Yang, X.; Feng, H.; Chen, Y. Solution-Processed Organic Solar Cells Based on Dialkylthiol-Substituted Benzodithiophene Unit with Efficiency near 10%. J. Am. Chem. Soc. 2014, 136, 15529–15532. Cui, C.; Guo, X.; Min, J.; Guo, B.; Cheng, X.; Zhang, M.; Brabec, C. J.; Li, Y. High-Performance Organic Solar Cells Based on a Small Molecule with Alkylthio-Thienyl-Conjugated Side Chains without Extra Treatments. Adv. Mater. 2015, 27, 7469–7475. Sun, K.; Xiao, Z.; Lu, S.; Zajaczkowski, W.; Pisula, W.; Hanssen, E.; White, J. M.; Williamson, R. M.; Subbiah, J.; Ouyang, J.; et al. A Molecular Nematic Liquid Crystalline Material for HighPerformance Organic Photovoltaics. Nat. Commun. 2015, 6, 6013. Wang, J.-L.; Liu, K.-K.; Yan, J.; Wu, Z.; Liu, F.; Xiao, F.; Chang, Z.-F.; Wu, H.-B.; Cao, Y.; Russell, T. P. Series of Multifluorine Substituted Oligomers for Organic Solar Cells with Efficiency over 9% and Fill Factor of 0.77 by Combination Thermal and Solvent Vapor Annealing. J. Am. Chem. Soc. 2016, 138, 7687–7697. Wang, J.-L.; Xiao, F.; Yan, J.; Wu, Z.; Liu, K.-K.; Chang, Z.-F.; Zhang, R.-B.; Chen, H.; Wu, H.B.; Cao, Y. Difluorobenzothiadiazole-Based Small-Molecule Organic Solar Cells with 8.7% Efficiency by Tuning of π-Conjugated Spacers and Solvent Vapor Annealing. Adv. Funct. Mater. 2016, 26, 1803–1812. Ni, W.; Li, M.; Liu, F.; Wan, X.; Feng, H.; Kan, B.; Zhang, Q.; Zhang, H.; Chen, Y. Dithienosilole-Based Small-Molecule Organic Solar Cells with an Efficiency over 8%: Investigation of the Relationship between the Molecular Structure and Photovoltaic Performance. Chem. Mater. 2015, 27, 6077–6084. Liu, X.; Sun, Y.; Hsu, B. B. Y.; Lorbach, A.; Qi, L.; Heeger, A. J.; Bazan, G. C. Design and Properties of Intermediate-Sized Narrow Band-Gap Conjugated Molecules Relevant to SolutionProcessed Organic Solar Cells. J. Am. Chem. Soc. 2014, 136, 5697–5708. Huang, Y.; Wen, W.; Mukherjee, S.; Ade, H.; Kramer, E. J.; Bazan, G. C. High-Molecular-Weight Insulating Polymers Can Improve the Performance of Molecular Solar Cells. Adv. Mater. 2014, 26, 4168–4172. Gao, K.; Li, L.; Lai, T.; Xiao, L.; Huang, Y.; Huang, F.; Peng, J.; Cao, Y.; Liu, F.; Russell, T. P.; et al. Deep Absorbing Porphyrin Small Molecule for High-Performance Organic Solar Cells with Very Low Energy Losses. J. Am. Chem. Soc. 2015, 137, 7282–7285. Gao, K.; Miao, J.; Xiao, L.; Deng, W.; Kan, Y.; Liang, T.; Wang, C.; Huang, F.; Peng, J.; Cao, Y.; et al. Multi-Length-Scale Morphologies Driven by Mixed Additives in Porphyrin-Based Organic Photovoltaics. Adv. Mater. 2016, 28, 4727–4733. Yang, D.; Yang, Q.; Yang, L.; Luo, Q.; Huang, Y.; Lu, Z.; Zhao, S. Novel High Performance Asymmetrical Squaraines for Small Molecule Organic Solar Cells with a High Open Circuit Voltage of 1.12 V. Chem. Commun. 2013, 49, 10465–10467. Zheng, C.; Penmetcha, A. R.; Cona, B.; Spencer, S. D.; Zhu, B.; Heaphy, P.; Cody, J. A.; Collison, C. J. Contribution of Aggregate States and Energetic Disorder to a Squaraine System Targeted for Organic Photovoltaic Devices. Langmuir 2015, 31, 7717–7726.

ACS Paragon Plus Environment

The Journal of Physical 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

(21)

(22)

(23)

(24)

(25) (26) (27) (28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

Spencer, S.; Hu, H.; Li, Q.; Ahn, H.-Y.; Qaddoura, M.; Yao, S.; Ioannidis, A.; Belfield, K.; Collison, C. J. Controlling J-Aggregate Formation for Increased Short-Circuit Current and Power Conversion Efficiency with a Squaraine Donor. Prog. Photovolt. Res. Appl. 2014, 22, 488–493. Wei, G.; Xiao, X.; Wang, S.; Zimmerman, J. D.; Sun, K.; Diev, V. V.; Thompson, M. E.; Forrest, S. R. Arylamine-Based Squaraine Donors for Use in Organic Solar Cells. Nano Lett. 2011, 11, 4261–4264. Chen, G.; Sasabe, H.; Wang, Z.; Wang, X.; Hong, Z.; Kido, J.; Yang, Y. Solution-Processed Organic Photovoltaic Cells Based on a Squaraine Dye. Phys. Chem. Chem. Phys. 2012, 14, 14661– 14666. Mayerhöffer, U.; Deing, K.; Gruß, K.; Braunschweig, H.; Meerholz, K.; Würthner, F. Outstanding Short-Circuit Currents in BHJ Solar Cells Based on NIR-Absorbing Acceptor-Substituted Squaraines. Angew. Chem. Int. Ed. 2009, 48, 8776–8779. Wei, G.; Wang, S.; Sun, K.; Thompson, M. E.; Forrest, S. R. Solvent-Annealed Crystalline Squaraine: PC70BM (1:6) Solar Cells. Adv. Energy Mater. 2011, 1, 184–187. Chen, G.; Sasabe, H.; Wang, Z.; Wang, X.-F.; Hong, Z.; Yang, Y.; Kido, J. Co-Evaporated Bulk Heterojunction Solar Cells with >6.0% Efficiency. Adv. Mater. 2012, 24, 2768–2773. Varma, P. C. R.; Namboothiry, M. A. G. Squaraine Based Solution Processed Inverted Bulk Heterojunction Solar Cells Processed in Air. Phys. Chem. Chem. Phys. 2016, 18, 3438–3443. Yang, D.; Jiao, Y.; Yang, L.; Chen, Y.; Mizoi, S.; Huang, Y.; Pu, X.; Lu, Z.; Sasabe, H.; Kido, J. Cyano-Substitution on the End-Capping Group: Facile Access toward Asymmetrical Squaraine Showing Strong Dipole–dipole Interactions as a High Performance Small Molecular Organic Solar Cells Material. J. Mater. Chem. A 2015, 3, 17704–17712. Yang, D.; Yang, Q.; Yang, L.; Luo, Q.; Chen, Y.; Zhu, Y.; Huang, Y.; Lu, Z.; Zhao, S. A Low Bandgap Asymmetrical Squaraine for High-Performance Solution-Processed Small Molecule Organic Solar Cells. Chem. Commun. 2014, 50, 9346–9348. Xiao, X.; Wei, G.; Wang, S.; Zimmerman, J. D.; Renshaw, C. K.; Thompson, M. E.; Forrest, S. R. Small-Molecule Photovoltaics Based on Functionalized Squaraine Donor Blends. Adv. Mater. 2012, 24, 1956–1960. Zimmerman, J. D.; Lassiter, B. E.; Xiao, X.; Sun, K.; Dolocan, A.; Gearba, R.; Vanden Bout, D. A.; Stevenson, K. J.; Wickramasinghe, P.; Thompson, M. E.; et al. Control of Interface Order by Inverse Quasi-Epitaxial Growth of Squaraine/Fullerene Thin Film Photovoltaics. ACS Nano 2013, 7, 9268–9275. Huang, J.-S.; Goh, T.; Li, X.; Sfeir, M. Y.; Bielinski, E. A.; Tomasulo, S.; Lee, M. L.; Hazari, N.; Taylor, A. D. Polymer Bulk Heterojunction Solar Cells Employing Forster Resonance Energy Transfer. Nat Photon 2013, 7, 479–485. Yang, D.; Sasabe, H.; Jiao, Y.; Zhuang, T.; Huang, Y.; Pu, X.; Sano, T.; Lu, Z.; Kido, J. An Effective π-Extended Squaraine for Solution-Processed Organic Solar Cells with High Efficiency. J. Mater. Chem. A 2016, 4, 18931–18941. Goh, T.; Huang, J.-S.; Yager, K. G.; Sfeir, M. Y.; Nam, C.-Y.; Tong, X.; Guard, L. M.; Melvin, P. R.; Antonio, F.; Bartolome, B. G.; et al. Quaternary Organic Solar Cells Enhanced by Cocrystalline Squaraines with Power Conversion Efficiencies >10%. Adv. Energy Mater. 2016, 6, 1600660. Wei, G.; Lunt, R. R.; Sun, K.; Wang,hompson, M. E.; Forrest, S. R. Efficient, Ordered Bulk Heterojunction Nanocrystalline Solar Cells by Annealing of Ultrathin Squaraine Thin Films. Nano Lett. 2010, 10, 3555–3559. Spencer, S.; Cody, J.; Misture, S.; Cona, B.; Heaphy, P.; Rumbles, G.; Andersen, J.; Collison, C. Critical Electron Transfer Rates for Exciton Dissociation Governed by Extent of Crystallinity in Small Molecule Organic Photovoltaics. J. Phys. Chem. C 2014, 118, 14840–14847. Zheng, C.; Bleier, D.; Jalan, I.; Pristash, S.; Penmetcha, A. R.; Hestand, N. J.; Spano, F. C.; Pierce, M. S.; Cody, J. A.; Collison, C. J. Phase Separation, Crystallinity and Monomer-Aggregate

ACS Paragon Plus Environment

Page 28 of 45

Page 29 of 45

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

The Journal of Physical Chemistry

(38)

(39)

(40)

(41)

(42)

(43) (44) (45)

(46)

(47)

(48)

(49) (50)

(51) (52)

(53) (54)

Population Control in Solution Processed Small Molecule Solar Cells. Sol. Energy Mater. Sol. Cells 2016, 157, 366–376. Gadisa, A.; Oosterbaan, W. D.; Vandewal, K.; Bolsée, J.-C.; Bertho, S.; D’Haen, J.; Lutsen, L.; Vanderzande, D.; Manca, J. V. Effect of Alkyl Side-Chain Length on Photovoltaic Properties of Poly(3-Alkylthiophene)/PCBM Bulk Heterojunctions. Adv. Funct. Mater. 2009, 19, 3300–3306. Nguyen, L. H.; Hoppe, H.; Erb, T.; Günes, S.; Gobsch, G.; Sariciftci, N. S. Effects of Annealing on the Nanomorphology and Performance of Poly(alkylthiophene):Fullerene Bulk-Heterojunction Solar Cells. Adv. Funct. Mater. 2007, 17, 1071–1078. Min, J.; Luponosov, Y. N.; Gasparini, N.; Richter, M.; Bakirov, A. V.; Shcherbina, M. A.; Chvalun, S. N.; Grodd, L.; Grigorian, S.; Ameri, T.; et al. Effects of Alkyl Terminal Chains on Morphology, Charge Generation, Transport, and Recombination Mechanisms in SolutionProcessed Small Molecule Bulk Heterojunction Solar Cells. Adv. Energy Mater. 2015, 5, 1500386. McCulloch, B.; Ho, V.; Hoarfrost, M.; Stanley, C.; Do, C.; Heller, W. T.; Segalman, R. A. Polymer Chain Shape of Poly(3-Alkylthiophenes) in Solution Using Small-Angle Neutron Scattering. Macromolecules 2013, 46, 1899–1907. Hestand, N. J.; Zheng, C.; Penmetcha, A. R.; Cona, B.; Cody, J. A.; Spano, F. C.; Collison, C. J. Confirmation of the Origins of Panchromatic Spectra in Squaraine Thin Films Targeted for Organic Photovoltaic Devices. J. Phys. Chem. C 2015, 119, 18964–18974. Brabec, C. J.; Heeney, M.; McCulloch, I.; Nelson, J. Influence of Blend Microstructure on Bulk Heterojunction Organic Photovoltaic Performance. Chem. Soc. Rev. 2011, 40, 1185–1199. Vandewal, K.; Himmelberger, S.; Salleo, A. Structural Factors That Affect the Performance of Organic Bulk Heterojunction Solar Cells. Macromolecules 2013, 46, 6379–6387. Wei, G.; Xiao, X.; Wang, S.; Sun, K.; Bergemann, K. J.; Thompson, M. E.; Forrest, S. R. Functionalized Squaraine Donors for Nanocrystalline Organic Photovoltaics. ACS Nano 2012, 6, 972–978. Dirk, C. W.; Herndon, W. C.; Cervantes-Lee, F.; Selnau, H.; Martinez, S.; Kalamegham, P.; Tan, A.; Campos, G.; Velez, M. Squarylium Dyes: Structural Factors Pertaining to the Negative ThirdOrder Nonlinear Optical Response. J. Am. Chem. Soc. 1995, 117, 2214–2225. Chen, G.; Sasabe, H.; Sasaki, Y.; Katagiri, H.; Wang, X.-F.; Sano, T.; Hong, Z.; Yang, Y.; Kido, J. A Series of Squaraine Dyes: Effects of Side Chain and the Number of Hydroxyl Groups on Material Properties and Photovoltaic Performance. Chem. Mater. 2014, 26, 1356–1364. Spencer, S. D.; Bougher, C.; Heaphy, P. J.; Murcia, V. M.; Gallivan, C. P.; Monfette, A.; Andersen, J. D.; Cody, J. A.; Conrad, B. R.; Collison, C. J. The Effect of Controllable Thin Film Crystal Growth on the Aggregation of a Novel High Panchromaticity Squaraine Viable for Organic Solar Cells. Sol. Energy Mater. Sol. Cells 2013, 112, 202–208. Zheng, C.; Jalan, I.; Cody, J. A.; Collison, C. J. Small Molecule with Extended Alkyl Side Substituents for Organic Solar Cells. MRS Adv. 2016, 1–7. Bruck, S.; Krause, C.; Turrisi, R.; Beverina, L.; Wilken, S.; Saak, W.; Lutzen, A.; Borchert, H.; Schiek, M.; Parisi, J. Structure-Property Relationship of Anilino-Squaraines in Organic Solar Cells. Phys. Chem. Chem. Phys. 2014, 16, 1067–1077. Proctor, C. M.; Kuik, M.; Nguyen, T.-Q. Charge Carrier Recombination in Organic Solar Cells. Prog. Polym. Sci. 2013, 38, 1941–1960. Proctor, C. M.; Albrecht, S.; Kuik, M.; Neher, D.; Nguyen, T.-Q. Overcoming Geminate Recombination and Enhancing Extraction in Solution-Processed Small Molecule Solar Cells. Adv. Energy Mater. 2014, 4, 1400230. Proctor, C. M.; Love, J. A.; Nguyen, T.-Q. Mobility Guidelines for High Fill Factor SolutionProcessed Small Molecule Solar Cells. Adv. Mater. 2014, 26, 5957–5961. Walker, B.; Tamayo, A. B.; Dang, X.-D.; Zalar, P.; Seo, J. H.; Garcia, A.; Tantiwiwat, M.; Nguyen, T.-Q. Nanoscale Phase Separation and High Photovoltaic Efficiency in SolutionProcessed, Small-Molecule Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2009, 19, 3063– 3069.

ACS Paragon Plus Environment

The Journal of Physical 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

(55)

(56)

(57) (58) (59)

(60)

(61)

(62)

(63)

(64)

Liu, Y.; Chen, C.-C.; Hong, Z.; Gao, J.; (Michael) Yang, Y.; Zhou, H.; Dou, L.; Li, G.; Yang, Y. Solution-Processed Small-Molecule Solar Cells: Breaking the 10% Power Conversion Efficiency. Sci. Rep. 2013, 3, 3356. Proctor, C. M.; Kim, C.; Neher, D.; Nguyen, T.-Q. Nongeminate Recombination and Charge Transport Limitations in Diketopyrrolopyrrole-Based Solution-Processed Small Molecule Solar Cells. Adv. Funct. Mater. 2013, 23, 3584–3594. Mott, N. F.; Gurney, R. W. Electronic Processes in Ionic Crystals; 1940. Leem, J. W.; Yu, J. S. Glancing Angle Deposited ITO Films for Efficiency Enhancement of aSi:H/Μc-Si:H Tandem Thin Film Solar Cells. Opt. Express 2011, 19, A258–A269. Zhou, J.; Wan, X.; Liu, Y.; Zuo, Y.; Li, Z.; He, G.; Long, G.; Ni, W.; Li, C.; Su, X.; et al. Small Molecules Based on Benzo[1,2-b:4,5-B′]dithiophene Unit for High-Performance SolutionProcessed Organic Solar Cells. J. Am. Chem. Soc. 2012, 134, 16345–16351. Bulle-Lieuwma, C. W. T.; van Gennip, W. J. H.; van Duren, J. K. J.; Jonkheijm, P.; Janssen, R. A. J.; Niemantsverdriet, J. W. Characterization of Polymer Solar Cells by TOF-SIMS Depth Profiling. Appl. Surf. Sci. 203–204, 2003, 547–550. Geens, W.; Martens, T.; Poortmans, J.; Aernouts, T.; Manca, J.; Lutsen, L.; Heremans, P.; Borghs, S.; Mertens, R.; Vanderzande, D. Modelling the Short-Circuit Current of Polymer Bulk Heterojunction Solar Cells. Thin Solid Films 451–452, 2004, 498–502. Yang, X.; Loos, J.; Veenstra, S. C.; Verhees, W. J. H.; Wienk, M. M.; Kroon, J. M.; Michels, M. A. J.; Janssen, R. A. J. Nanoscale Morphology of High-Performance Polymer Solar Cells. Nano Lett. 2005, 5, 579–583. Love, J. A.; Nagao, I.; Huang, Y.; Kuik, M.; Gupta, V.; Takacs, C. J.; Coughlin, J. E.; Qi, L.; van der Poll, T. S.; Kramer, E. J.; et al. Silaindacenodithiophene-Based Molecular Donor: Morphological Features and Use in the Fabrication of Compositionally Tolerant, High-Efficiency Bulk Heterojunction Solar Cells. J. Am. Chem. Soc. 2014, 136, 3597–3606. Takacs, C. J.; Sun, Y.; Welch, G. C.; Perez, L. A.; Liu, X.; Wen, W.; Bazan, G. C.; Heeger, A. J. Solar Cell Efficiency, Self-Assembly, and Dipole–Dipole Interactions of Isomorphic NarrowBand-Gap Molecules. J. Am. Chem. Soc. 2012, 134, 16597–16606.

ACS Paragon Plus Environment

Page 30 of 45

Page 31 of 45

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

The Journal of Physical Chemistry

Table of Contents Graphic

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 1. a) Thermogravimetric analysis results of SQ materials. 65x55mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 32 of 45

Page 33 of 45

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

The Journal of Physical Chemistry

Figure 1. b) differential scanning calorimetry results of SQ materials. The Asterisks highlight an unique endothermic peak at 100 0C of DBSQ(OH)2 powder solids 65x56mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 2. Normalized absorbance spectra of three SQs in chloroform solution (dashed line) and as neat films (solid line). The solution absorbance spectra are overlapping with the same peak position. 61x49mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 34 of 45

Page 35 of 45

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

The Journal of Physical Chemistry

Figure 3. a): Molecular packing of DBSQ(OH)2 single crystal, featuring a π-π stacking with slippages in both long and short molecular axes; other SQs adopt a similar slip stacking motif in the single crystal. 232x179mm (96 x 96 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 3. b): a simple cartoon illustrates the slip stack of the SQ molecules with an artificial Cartesian coordinate. 177x127mm (136 x 136 DPI)

ACS Paragon Plus Environment

Page 36 of 45

Page 37 of 45

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

The Journal of Physical Chemistry

84x70mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

84x70mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 38 of 45

Page 39 of 45

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

The Journal of Physical Chemistry

84x70mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 5. Hole mobilities of SQ neat films and SQ:PCBM blend films as a function of the number of side chain carbons of the SQ molecule. The measurements were done on unannealed films. 75x64mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 40 of 45

Page 41 of 45

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

The Journal of Physical Chemistry

Figure 6. X-ray diffraction (XRD) patterns of a) SQ neat films before (black) and after (red) thermal annealing. 91x110mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 6. X-ray diffraction (XRD) patterns of b) SQ:PCBM blend films (5:5 w/w) before (black) and after (red) thermal annealing. 91x110mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 42 of 45

Page 43 of 45

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

The Journal of Physical Chemistry

Figure 7. Transmission electron microscope images of DBSQ(OH)2:PCBM blend films. The SQ:PCBM weight ratio is 5:5. Scale bars: 100 nm. 16743x12512mm (5 x 5 DPI)

ACS Paragon Plus Environment

The Journal of Physical 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

Figure 7. Transmission electron microscope images of DPSQ(OH)2:PCBM blend films. The SQ:PCBM weight ratio is 5:5. Scale bars: 100 nm. 16743x12537mm (5 x 5 DPI)

ACS Paragon Plus Environment

Page 44 of 45

Page 45 of 45

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

The Journal of Physical Chemistry

Figure 7. Transmission electron microscope images of DHSQ(OH)2:PCBM blend films. The SQ:PCBM weight ratio is 5:5. Scale bars: 100 nm. 16743x12517mm (5 x 5 DPI)

ACS Paragon Plus Environment