Direct Aqueous Dispersion of Carbon Nanotubes Using Nanoparticle

Apr 25, 2017 - Single-walled carbon nanotubes (SWCNTs) have received much attention because of their potential in optoelectronic applications. Pristin...
0 downloads 0 Views 4MB Size
Download PDF
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article http://pubs.acs.org/journal/acsodf

Direct Aqueous Dispersion of Carbon Nanotubes Using Nanoparticle-Formed Fullerenes and Self-Assembled Formation of p/ n Heterojunctions with Polythiophene Zha Li,†,° Pan He,‡,° Hui Chong,§ Akihiro Furube,∥ Kazuhiko Seki,⊥ Hsiao-hua Yu,# Keisuke Tajima,¶ Yoshihiro Ito,*,†,‡ and Masuki Kawamoto*,†,‡,∇ †

Nano Medical Engineering Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan Emergent Bioengineering Materials Research Team and ¶Emergent Functional Polymers Research Team, RIKEN Center for Emergent Matter Science (CEMS), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan § Chemistry Department, KU Leuven, Celestijnenlaan 200F, P.O. Box 2404, B-3001 Leuven, Belgium ∥ Department of Optical Science, Tokushima University, 2-1 Minami-Josanjima, Tokushima 770-8506, Japan ⊥ Nanofilm Device Group, National Institute of Advanced Industrial Science and Technology, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan # Institute of Chemistry, Academia Sinica, 128 Academia Road Sec. 2, Nankang, Taipei 11529, Taiwan ∇ Photocatalysis International Research Center, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan ‡

S Supporting Information *

ABSTRACT: Single-walled carbon nanotubes (SWCNTs) have received much attention because of their potential in optoelectronic applications. Pristine SWCNTs exhibit substantial van der Waals interactions and hydrophobic characteristics, so precipitation occurs immediately in most organic solvents and water. Highly toxic and hazardous chemicals are often used to obtain well-dispersed SWCNTs. Developing environmentally friendly processing methods for safe and practical applications is a great challenge. Here, we demonstrate direct exfoliation of SWCNTs in pure water only with n-type semiconducting fullerene nanoparticles. The resultant SWCNTs can be well-dispersed in water, where they remain essentially unchanged for several weeks. Adding an aqueous solution of p-type semiconducting water-soluble polythiophene yields self-assembled p/n heterojunctions between polythiophene and the nanoparticles. The aqueous-dispersed SWCNTs yield photocurrent responses in solution-processed thin films as a potential application of water-dispersed carbon nanomaterials.

■

delocalized π stacking interactions, so there is minimal perturbation of the SWCNT electronic structure. Water is a good candidate for green exfoliation methodology7 because it is a biologically safe and abundant solvent. From a greenprocessing point of view, aqueous exfoliation of SWCNTs has been achieved using water-soluble surfactants,8 polymers,9,10 nanocarbon materials,11,12 proteins,13 DNA,14 and peptides.15 In the current study, we develop a clean processing method for “direct” exfoliation of SWCNTs in pure water with n-type semiconducting fullerene nanoparticles composed of [6,6]phenyl-C61-butyric acid methyl ester (PC61BM; Figure 1a) for the first time. Because PC61BM nanoparticles are directly attached to the surface of SWCNTs, a well-dispersed solution can be obtained as a result of electrostatic repulsion of the

INTRODUCTION Single-walled carbon nanotubes (SWCNTs) are cylindrical structures with exceptional electrical, chemical, and mechanical properties1,2 and may contribute to future emerging technologies, including flexible and wearable electronics.3 Solutionprocessed SWCNTs can be easily upscaled to large-area and high-throughput production through well-established spincoating, drop-casting, and inkjet printing techniques.4 Development of exfoliation and dispersion in large quantities at low cost and preferably with clean processing is crucial for the preparation of solution-processed thin-film devices. Unfortunately, SWCNTs form bundles and then immediately precipitate in pure water because of their extended hexagonal lattice of sp2-bonded carbon atoms. Physical modification is a “soft” dispersion approach because dispersion is achieved using noncovalent interactions between SWCNTs and dispersants.5,6 The dispersants are attached to the surface of the SWCNTs through hydrophobic or © 2017 American Chemical Society

Received: February 14, 2017 Accepted: April 10, 2017 Published: April 25, 2017 1625

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

Figure 1. Aqueous dispersion behavior of PC61BM nanoparticles. (a) Molecular structure of PC61BM. (b) TEM image of PC61BM nanoparticles. (c) Size distribution of PC61BM nanoparticles in water. (d) Change in average size (red) and polydispersity index (blue) of PC61BM nanoparticles in water at room temperature. Experimental results in b−d were obtained from the dispersion prepared according to condition 1 in Table S1.

Figure 2. (a) Absorption spectra of aqueous-dispersed PC61BM/SWCNTs composites containing various SWCNT concentrations. (i) 10, (ii) 25, (iii) 50, (iv) 75, (v) 100, and (vi) 200 wt %. (b) Absorbance at 600 nm (blue) and 1178 nm (red) as a function of the SWCNT concentration of PC61BM/SWCNTs. (c) SEM image of a binary composite film (PC61BM/SWCNTs = 100:50 w/w) on an aluminum oxide membrane.

the treatment temperature and the concentration of PC61BM in the mother solution. No significant changes in the sizes of these nanoparticles were observed after 6 months, and their PDI values remained in the range of 0.11−0.21. However, precipitation occurred in water when the concentration of the mother tetrahydrofuran (THF) solution exceeded 1.0 g L−1. To clarify the stability of the PC61BM nanoparticles in water, the zeta potential (ζ) of the aqueous dispersions was measured. A negative ζ value of −41.4 mV was obtained for the aqueous dispersion of PC61BM nanoparticles. This suggests that the PC61BM nanoparticles exhibited a stable dispersion because of electrostatic repulsion. Furthermore, the methyl ester group in PC61BM tends to form hydrogen bonds with water. The interaction between PC61BM and water has been recently investigated using molecular dynamics simulation.17 Radial distribution functions indicated that the distance between the ester group and water (5.4 Å) was smaller than that between the fullerene core and water (8.5 Å) because of hydrogen bonding. The methyl ester group may therefore promote a stable aqueous dispersion by forming a hydration network through hydrogen bonding at the surface of the PC61BM nanoparticles. We then investigated the aqueous exfoliation of SWCNTs using 30 nm diameter PC61BM nanoparticles. Adding raw SWCNTs to an aqueous dispersion of PC61BM nanoparticles and subjecting to ultrasonication formed aqueous-dispersed SWCNTs (Figure S1). TEM images were carefully analyzed to evaluate the dispersion behavior of SWCNTs because ultrasonication for extended durations and/or under harsh conditions can result in very short SWCNTs (5−10 nm).18,19

adsorbed nanoparticles on SWCNTs in water. Further adding an aqueous solution of p-type semiconducting water-soluble polythiophene (WSPT) gives rise to a ternary composite with self-assembled p/n heterojunctions consisting of WSPT and the PC61BM nanoparticles. The resultant composites show photocurrent responses in solution-processed thin films upon illumination. This simple and green exfoliation method is expected to be suitable for printing and coating, offering a clean processing of thin-film devices.

■

RESULTS AND DISCUSSION Nanoparticle-Induced Dispersion of SWCNTs in Pure Water. PC61BM nanoparticles were prepared by modifying a previously reported method.16 Figure 1b shows a transmission electron microscopy (TEM) image of the PC61BM nanoparticles. The sample was prepared by casting the dispersion onto a TEM grid. An average nanoparticle diameter of 24 nm was estimated from 200 nanoparticles in the TEM image. The size distribution of the PC61BM nanoparticles was investigated using dynamic light scattering (DLS), as shown in Figure 1c. DLS indicated an average nanoparticle diameter of 28.8 nm, with a narrow polydispersity index (PDI) of 0.19 in water. To evaluate the stability of PC61BM nanoparticles in water, the time-dependent changes in their average size and PDI were investigated. The PC61BM nanoparticles remained unchanged after 6 months of storage at room temperature (Figure 1d). We also prepared PC61BM nanoparticles with different sizes by varying the experimental conditions (Table S1). TEM images and DLS profiles indicated that nanoparticles with diameters ranging from 28.8 to 106.2 nm could be obtained by changing 1626

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

Figure 3. SWCNTs with WSPT/PC61BM nanoparticle-based p/n heterojunctions. (a) Molecular structure of WSPT. (b) Photographs of aqueous dispersions of raw SWCNTs (i), PC61BM/SWCNTs binary composite (100:4 w/w) (ii), and WSPT/PC61BM/SWCNTs ternary composite (100:100:4 w/w/w) under irradiation using a 633 nm HeNe laser beam (iii). (c) TEM image of the PC61BM/SWCNTs binary composite (100:4 w/ w). (d) Size distribution of the PC61BM nanoparticles in the binary composite. (e) TEM image of the WSPT/PC61BM/SWCNTs ternary composite (100:100:4 w/w/w) on an elastic carbon-coated copper TEM grid. (f) EDS spectra of the ternary composite at various points of (i), (ii), and (iii) in the TEM image. Peaks at 8.040 keV were Cu Kα lines originating from the TEM grid. (g) Magnification of EDS spectra of Figure 3f.

in water. Precipitation was observed at high concentrations of SWCNTs because of their inevitable aggregation.22 As mentioned above, a negative ζ value of −41.4 mV was measured for the PC61BM nanoparticles in water. ζ values for aqueous dispersions of the binary composites (PC61BM/ SWCNTs = 100:X w/w) were slightly lower. The ζ values were −26.5, −32.1, and −30.7 mV for X = 10, 50, and 100, respectively. These results suggest that PC61BM nanoparticles retained stable dispersion properties after being attached to the surface of SWCNTs. Figure 2c shows a scanning electron microscopy (SEM) image of a binary composite film (PC61BM/SWCNTs = 100:50 w/w) on an aluminum oxide membrane. This SEM image reveals that aggregated SWCNTs were dispersed into small bundles by the PC61BM nanoparticles after forming the binary composite film. Unfortunately, we could not find the origin of the dispersion ability of PC61BM nanoparticles for SWCNTs in water. One possibility is that the hydrophobic interaction between fullerene and SWCNT surfaces gives rise to the dispersion of SWCNTs at the molecular level.12 Hilmer et al. reported that fullerene derivatives tend to adsorb onto the SWCNT surface, leading to individually dispersed SWCNTs. By contrast, the large size of our nanoparticle might not penetrate into the individual SWCNTs with diameters of 1−1.3 nm because the diameter of the PC61BM nanoparticle was more than 40 times larger than that of the C60 molecule (0.71 nm). The 30 nm diameter nanoparticle is the smallest as per our present protocol described above. We believe that a smaller diameter of a nanoparticle is expected to be beneficial for the effective dispersion of SWCNTs in water. Self-Assembled Formation of p/n Heterojunctions in Water. Figure 3a shows the chemical structure of WSPT. The synthetic procedure for WSPT is described in Supporting

Judging from TEM images collected from samples with various SWCNT concentrations, small bundles of SWCNTs with diameters of 6−8 nm yielded after ultrasonication. These results suggested that five or six SWCNTs gave rise to entangled structures. Though stable dispersion of SWCNTs was observed in water, we could not find individual SWCNTs (Figure S2). The color of the dispersion changed from yellow to dark gray with increasing SWCNT concentration (Figure S1). This change in absorption spectra was consistent with those expected from the electronic structure of SWCNTs in the visible and near-infrared (NIR) regions (Figure 2a).20 To determine the limit of the dispersion ability of the PC61BM nanoparticles in water, the relationship between absorbance and SWCNT concentration was investigated using the Beer− Lambert law (Figure 2b). The Beer−Lambert law is a linear relationship between the absorbance and concentration of a medium and is defined as A = εcl, where A, ε, c, and l are the absorbance, molar absorption coefficient, concentration, and optical path length, respectively. It was assumed that a good dispersion of SWCNTs occurs if a linear relationship is maintained between A and c. For analysis, we selected two absorption peaks at 600 and 1178 nm, which correspond to the first optical transition of metallic SWCNTs (M11, 400−620 nm) and the first optical transition of semiconducting SWCNTs (S11, 940−1350 nm), respectively.21 Linear relationships between A and c were observed until the binary composite contained 100 wt % SWCNTs, with coefficients of determination (R2) of >0.992 for M11 and S11 transitions. Figure 2b shows that a decrease in slope occurred when the concentration of SWCNTs exceeded 100 wt %. This observation may reflect the saturated dispersion of SWCNTs 1627

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

Figure 4. Photoexcitation dynamics of dispersed SWCNTs with WSPT/PC61BM heterojunctions. (a) Normalized absorption spectra for (i) PC61BM, (ii) WSPT, and (iii) the WSPT/PC61BM/SWCNTs ternary composite film (100:100:4 w/w/w). (b) Transient absorption spectra of the WSPT/PC61BM/SWCNTs film (100:100:4 w/w/w) upon photoexcitation at 400 nm. The absence of data points near 800 nm is an artifact of the equipment using a cutoff filter. (c) Transient decay kinetics for the WSPT/PC61BM/SWCNTs film (100:100:4 w/w/w) (monitoring wavelength: 980 nm) at various light intensities. (d) Transient decay kinetics of composite films at various concentrations of SWCNTs. (e) Analysis of the transient decay kinetics for the WSPT/PC61BM film (100:100 w/w) and the WSPT/PC61BM/SWCNTs film (100:100:4 w/w/w) upon photoexcitation at 400 nm. Black line indicates the smoothed line of the scattering data for the binary composite using a binomial algorithm. Colored lines display analyzed data of the ternary composite at various light intensities using eqs 5 and 6 in Supplementary Note 2. (f) Schematic illustration of carrier separation, electron migration, and carrier recombination of the WSPT/PC61BM/SWCNTs composite.

PC61BM nanoparticles are apparent in the TEM image (Figure S4a). The average size of the nanoparticles in the ternary composite (29.8 nm) was comparable to that in the binary composite. Similar size distributions for the binary and ternary composites revealed no obvious deformation of PC61BM nanoparticles upon coassembly with WSPT (Figures 3d and S4b). To clarify the formation of the ternary composite, we examined energy-dispersive X-ray spectra (EDS) (Figure 3f,g). Sulfur Kα lines (2.307 keV) arising from thiophene rings were observed at points (ii) and (iii) in Figure 3g. The intensity of the signal for the Fe Kα line (6.398 keV) originating from the residual catalyst on the SWCNT surface [point (iii)] was much higher than that observed at point (ii). This indicated that the signals measured at point (iii) resulted from the ternary composite. Characterization. We investigated the carrier generation, separation, and recombination processes of WSPT/PC61BM/ SWCNT composite films using femtosecond-response transient absorption spectroscopy (Figure 4, Supplementary Note 1). The absorption spectrum of the ternary composite film exhibited an absorption maximum at 423 nm. This was attributed to the π−π* transition of the polymer. Further weak absorption at 500−800 nm can be attributed to electronic transitions of both PC61BM23,24 and SWCNTs25 (Figure 4a). After photoexcitation of WSPT at 400 nm, the transient absorption spectra of the ternary composite film (WSPT/

Information. The number-average molecular weight (Mn) and PDI [weight-average molecular weight (Mw)/Mn] were 16 500 and 2.0, respectively, as determined by gel permeation chromatography (GPC) using polystyrene standards and THF as an eluent. The thermal properties of the polymer were characterized using thermogravimetric analysis and differential scanning calorimetry (Figure S3). WSPT is readily soluble in water and common solvents such as THF and chloroform. Figure 3b shows the aqueous dispersion behavior in water. Raw SWCNTs formed a precipitation because of untangled bundles (Figure 3b(i)). Adding raw SWCNTs to an aqueous dispersion of PC61 BM nanoparticles formed dispersed SWCNTs (Figure 3b(ii)). PC61BM nanoparticles were found to be attached to the surface of SWCNTs (PC61BM/SWCNTs = 100:4 w/w) in a TEM image (Figure 3c). From 100 nanoparticles in the TEM image, the average size of the PC61BM nanoparticles was estimated to be 29.9 nm (Figure 3d). This value is consistent with the average size of the PC61BM nanoparticles estimated from the TEM image in Figure 1b (24 nm). Following the addition of WSPT, an aqueous dispersion of the ternary composite (WSPT/PC 61 BM/SWCNTs = 100:100:4 w/w/w) was observed as a homogeneous dispersion, as evidenced by the scattering of a 633 nm HeNe laser beam (Figure 3b(iii)). Coassembled structures of WSPT and 1628

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

Figure 5. Photoresponse behavior of aqueous-dispersed SWCNTs. (a) Schematic illustrations of interdigitated (IDA) gold electrodes and sample configuration. (i) Top view of the electrodes and (ii) side view of the sample configuration. (b) Current−voltage characteristics of the composite film (WSPT/PC61BM/SWCNTs = 100:100:1 w/w/w) upon AM 1.5 photoirradiation (red) and under dark conditions (black). (c) Photocurrent-to-dark current ratio (Iphoto/Idark) of the composite films (WSPT/PC61BM/SWCNTs = 100:100:X w/w/w, X: 0−4) at 1 V. (d) Dark conductivities of composite films containing various SWCNT concentrations. (e) Mechanism of photogenerated carrier-transport and recombination in composite thin films between gold electrodes on IDA electrodes. (i) WSPT/PC61BM binary composite film without SWCNTs. (ii) WSPT/PC61BM/SWCNT ternary composite film containing a SWCNT concentration of up to 1 wt %. (iii) Ternary composite film containing a SWCNT concentration of greater than 1 wt %. Purple and green arrows indicate hole- and electron-transport pathways, respectively.

We theoretically analyzed the transient absorption data of binary and ternary composite films (Figure 4e, Supplementary Note 2). The decay kinetics of the binary composite film (WSPT/PC61BM = 100:100 w/w) was obtained from the smoothed line of the scattered data shown by the black line. The colored lines of the ternary composite film (WSPT/ PC61BM/SWCNTs = 100:100:4 w/w/w) were calculated using eqs 5 and 6 by assuming that the concentration of electrons at time zero is proportional to the light intensity. The theoretical model indicates the effective transport of electrons through SWCNTs. The electrons were able to escape geminate recombination by electron transfer to SWCNTs, but the recombination with the counter-holes surrounding SWCNTs could be induced by the electron transfer. The former effect was dominant if both the concentration of SWCNTs and light intensity were low so that the SWCNTs were likely to be not surrounded by counter-holes. This is consistent with slightly slow transient decay observed for the WSPT/PC61BM/ SWCNTs composite film (100:100:1 w/w/w) compared to the binary component and ternary components with higher concentration of SWCNTs (Figure 4d). Energy levels of WSPT, PC61BM, and SWCNTs were estimated using absorption spectroscopy and photoelectron spectroscopy performed in air (PESA) (Figure S6). WSPT and PC61BM had highest-occupied molecular orbital (HOMO) levels of −5.52 and −5.87 eV under vacuum, respectively (Figure S6a,b). The energy band gaps of WSPT and PC61BM in the films were 2.40 and 2.06 eV, respectively, as determined from Tauc plots.28 The lowest-unoccupied molecular orbital (LUMO) levels of WSPT and PC61BM were calculated to be

PC61BM/SWCNTs = 100:100:4 w/w/w) exhibited a maximum at approximately 980 nm, which was attributed to WSPT polarons generated on timescales shorter than the equipment resolution of 250 fs (Figure 4b). We observed transient absorption peaks at approximately 720, 800, and 950 nm for SWCNT concentrations of 0, 1, and 2 wt %, respectively (SWCNT concentrations in this article refer to the weight percentage of SWCNTs to PC61BM). The changes in the absorption peaks were attributed to different polarons of WSPT in the solid state.26 This indicates that the presence of SWCNTs might have caused the trapping of polarons in the disordered structure. The decay dynamics were then observed for the ternary composite film (WSPT/PC61BM/SWCNTs = 100:100:4 w/w/w) and were found to be affected by the excitation intensity (Figure 4c). Trapped polarons underwent second-order hole−electron recombination because electrons could reach neighboring polarons. The linear relationship between the reciprocal half-life (τ−1) and excitation intensity also suggests second-order recombination on the picosecond timescale (Figure S5a). By contrast, the decay dynamics of the binary composite film (WSPT/PC61BM) were independent of the excitation intensity (Figure S5b). The presence of polarons resulted in geminate recombination because electrons could not escape from their counter-holes in the absence of SWCNTs. The maximum observed at 720 nm for the binary composite film was attributed to more ordered polarons that existed in the more ordered structure of this film (Figure S5c). 27 Furthermore, the hole−electron recombination process depended on the SWCNT concentration (Figure 4d). 1629

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

−3.12 and −3.81 eV under vacuum, respectively. This energy difference between the LUMO of WSPT and the HOMO of PC61BM is the driving force for the dissociation of excitons upon photoirradiation. The work function of SWCNTs was also estimated to be 5.04 eV using PESA, which is comparable with the values reported in the literature (Figure S6c).29 We summarized photoinduced carrier generation, migration, and recombination processes in the WSPT/PC61BM/SWCNT composite film (Figure 4f). Photocurrent Response of the Aqueous-Dispersed SWCNT Solutions. Water-based exfoliation processes will be potentially clean and cost-effective, and aqueous-dispersed SWCNTs with p/n heterojunctions can be exploited to observe photocurrent responses. To explore this hypothesis, thin films (thickness: 630 nm) were prepared by casting SWCNTdispersed solutions as aqueous inks on interdigitated gold electrodes (Figure 5a, Supplementary Note 2). Upon air mass (AM) 1.5 simulated solar irradiation from the backside of the quartz substrate, the ternary composite film exhibited a photocurrent of 8.3 nA at 1 V (Figure 5b). Introducing SWCNTs into the composite films led to an increase in phototo-dark current ratio (I photo /I dark) values for SWCNT concentrations of up to 1 wt % (Figure 5c). Iphoto/Idark then decreased sharply with further increases in the SWCNT concentration. By contrast, the dark conductivity (σ) of the ternary composite gradually increased with increasing SWCNT concentration (Figure 5d). We repeated these experiments using the WSPT/PC61BM/SWCNT films at least twice for each concentration of SWCNTs. The resultant films exhibited good reproducibility and consistent concentration dependence of σs. We assumed that changes in Iphoto/Idark values were primarily caused by changes in the SWCNT network in the active layer (Figure 5e).30 The Iphoto/Idark value for the WSPT/PC61BM binary composite was only 1.2 (Figure 5c). This low value reflected the limited number of electron-transport pathways and suggests that hole−electron recombination was dominant at the interface of WSPT and the PC61BM nanoparticles through the p/n heterojunctions (Figure 5e(i)).31 Increasing the SWCNT concentration to up to 1 wt % formed an SWCNT network in the active layer. This caused an increase in the Iphoto/Idark value because of the formation of electrontransport pathways (Figure 5e(ii)).32 Increasing the SWCNT concentration above 1 wt % resulted in a decrease in the Iphoto/ Idark value because of the electron−hole recombination of the photogenerated carriers (Figure 5e(iii)). This explanation is supported by the transient decay kinetics and second-order electron−hole recombination (Figure 4d and S5a). Fast carrier recombination processes occurred when the SWCNT concentration was 2−4 wt %. To evaluate the effects of the SWCNT network, the photoresponse behavior of a ternary composite containing aggregated SWCNTs was examined as a control experiment (Figure S7). The Iphoto/Idark values of the resultant ternary composites were comparable with those of the WSPT/PC61BM binary composite. Good SWCNT dispersion was necessary for a good photocurrent response with effective carrier-transport pathways in the ternary composite. Low σ values of approximately 1 × 10−9 S cm−1 were obtained, even though SWCNTs were incorporated into the ternary composite. This is because the hole−electron recombination occurred as a result of limited carrier-transport pathways.

Hilmer et al. successfully demonstrated that the amphiphilic fullerenes-SWCNTs heterojunctions exhibited electron transfer in water.12 They found that relative rate constants of electron transfer depended on the fullerene structures: lipid-C61polyethylene glycol (PEG) and lipid-C 71 -PEG showed incomplete fluorescence quenching, indicating that the driving force for electron transfer is small. By contrast, lipid-C85-PEG gave rise to the complete quenching of SWCNT fluorescence. The increase in quenching efficiency was consistent with the deeper LUMO level of lipid-C85-PEG in comparison with lipidC61-PEG and lipid-C71-PEG. In our composites, electron transfer on the picosecond timescale occurred in the presence of SWCNTs (Figure 4). Furthermore, the energy relationships between the LUMO of PC61BM (−3.81 eV) and the work function of SWCNTs (−5.04 eV) are crucial for carrier recombination in the ternary composite films. These results suggest that carrier recombination rates can be tuned using fullerene derivatives showing the different LUMO levels. Researchers have also reported the effects of SWCNTs in organic photovoltaic cells (OPVs) composed of poly(3hexylthiophene) (P3HT) and PC61BM. Specifically, increases in the current densities of OPVs were observed at an SWCNT concentration of 1 wt % compared with OPVs containing no SWCNTs. This increase was attributed to the introduction of electron transport capability. By contrast, the current density decreased at a high SWCNT concentration of 3 wt %, which was due to bimolecular recombination in the presence of metallic SWCNTs.33 Others have reported that P3HT/ PC61BM OPVs exhibit improved device characteristics when they contain 1 wt % SWCNTs because of rapid electron transfer through SWCNTs.34 The current experimental results indicate that the percolation threshold is likely to be in the SWCNT concentration range 1−1.5 wt %. Because the supernatant of the PC61BM/SWCNTs binary composite was used as the mother liquor for the ternary composite, it was difficult to achieve precise control over the SWCNT concentration in our experiments. We assumed that the ternary composite had a percolation threshold at an SWCNT concentration of approximately 1 wt %. The fullerene nanoparticle-assisted dispersion of SWCNTs in water is a good candidate for clean exfoliation methodologies because no complex instrumentation and harmful chemicals are required to form finer SWCNT bundles under ambient conditions. Furthermore, the key function of the semiconducting PC61BM nanoparticles results in the formation of sophisticated p/n heterojunctions by simple step-by-step mixing in water. The modification of the nanostructure is expected to lead to superior photoelectric conversion for future investigation. For example, the exciton diffusion length of fullerenes via the triplet-excited state is estimated to be 8−14 nm,35 so we expect that PC61BM nanoparticles 10−15 nm in diameter will exhibit ideal carrier-transport properties in the coassembled nanostructures. Furthermore, narrow-band gap semiconducting polymers can provide NIR-responsive photodetectors for nondestructive imaging applications.36 Our findings indicate that aqueous-dispersed semiconducting inks are promising for printing and coating materials with simple and clean processing methodologies for future cost-effective industrial and commercial applications. For example, the aqueous processing can potentially allow the fabrication of semiconducting materials for use in high-performance devices. Furthermore, the direct aqueous dispersion of SWCNTs is also applicable in applications such as bioimaging and biosensing, 1630

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

binary composite was obtained from the supernatant. The binary composite film was prepared by filtration of the supernatant onto an aluminum oxide membrane with a pore size of 0.1 μm (Anodisc, SPI Suppliers/Structure Probe Inc., West Chester, PA, USA). WSPT/PC61BM/SWCNTs Ternary Composites. WSPT in water (5 g L−1, 100 μL) was added to the aqueous dispersion of the binary composite (PC61BM/SWCNTs = 100:1 w/w, 5 mL) to yield the WSPT/PC61BM/SWCNT ternary composite (100:100:1 w/w/w) without filtration. Ternary composites were also prepared using raw SWCNTs solutions at concentrations of 0.0001−0.004 g L−1. In a control experiment, samples were obtained by simply mixing WSPT, PC61BM, and SWCNTs in water, without dispersion treatments such as ultrasonication and centrifugation. Precipitation occurred immediately in water because of the large bundles of SWCNTs.

and in medicinal applications that prohibit the use of harmful organic solvents. Water is a chemically stable, biologically safe, and environmentally friendly solvent. This unique processing method contributes to green chemistry and engineering. Evaluation of water-processed flexible thermoelectric devices prepared with aqueous-dispersed SWCNTs using nanoparticleformed fullerenes is ongoing.

■

CONCLUSIONS In conclusion, PC61BM nanoparticles were used as an n-type semiconducting dispersant of SWCNTs in water. SWCNTs could be dispersed at up to 100 wt % in water using this noncovalent bonding methodology, and the dispersion behavior remained unchanged after several weeks. Adding an aqueous solution of the p-type conjugated polymer WSPT formed coassembled p/n heterojunctions consisting of WSPT/ PC61BM nanoparticles on SWCNTs. The p/n heterojunctions in the ternary composite exhibited electron transfer on the picosecond timescale. The all carbon-based composite films exhibited photocurrent responses, upon AM 1.5 irradiation.

■

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.7b00175. Further details of experimental procedures, analysis of the transient decay kinetics, and characterization, including synthetic procedures, DSC and TGA profiles, EDS spectra, plots of reciprocal half-life against excitation intensity, PESA profiles, and ratios of photocurrent to dark current and dark conductivities (PDF)

EXPERIMENTAL SECTION

Materials. Unless otherwise noted, compounds and solvents were purchased from commercial suppliers and used without further purification. Synthesized compounds were purified using flash column chromatography [CombiFlash Companion, Teledyne ISCO, Lincoln, NE, USA; column: Kanto Silica Gel 60 (spherical: 63−210 μm), Kanto Chemical Co., Inc., Tokyo, Japan]. Deuterated chloroform was purchased from Cambridge Isotope Laboratories Inc. (Tewksbury, MA, USA). The highpressure catalytic CO (HiPCO) SWCNTs were purchased from Nano-C Inc. (Westwood, MA, USA; 70 wt % carbon with >85% SWCNTs, 1.0−1.1 μm length, 0.9−1.3 nm diameter) and were used as received. The metallic residue in the raw SWCNTs was estimated to be 18% using inductively coupled plasma (ICP) emission spectrometry. PC61BM was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). General Procedure for Preparing Aqueous-Dispersed PC61BM Nanoparticles. PC61BM (10 mg) was dissolved in THF (10 mL) at room temperature. After filtration using a membrane filter (pore size: 0.45 μm), the resultant solution (1 g L−1, 10 mL) was added dropwise to water (100 mL). To remove the THF, the solution was purged with nitrogen for 3 h, which yielded the aqueous dispersion of PC61BM nanoparticles. The concentration of PC61BM nanoparticles in this aqueous dispersion was 0.1 g L−1. Detailed protocols of various experimental conditions are described in Table S1. PC61BM/SWCNTs Binary Composites. A suspension of raw SWCNTs in water (0.1 g L−1, 0.1 mL) was added to the aqueous dispersion of PC61BM nanoparticles (0.1 g L−1, 10 mL). The mixture was sonicated for 15 min using a tip-type ultrasonic homogenizer (Branson Sonifier 250, Branson Ultrasonics, Danbury, CT, USA; power output: 40 W) in a water bath at 25 °C. The resultant solution was allowed to stand overnight, and then the binary composite composed of PC61BM/SWCNTs (100:1 w/w, 10 mL) was collected as the supernatant without filtration. Binary composites with different SWCNT concentrations were also prepared, at concentrations of up to PC61BM/SWCNTs = 100:200 w/w. For high SWCNT concentrations above PC61BM/SWCNTs = 100:10 w/w, the mixtures after ultrasonication were centrifuged at 5000g for 5 min to remove large bundles of SWCNTs. A homogeneous

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (Y.I.). *E-mail: [email protected] (M.K.). ORCID

Kazuhiko Seki: 0000-0001-9858-2552 Keisuke Tajima: 0000-0003-1590-2640 Masuki Kawamoto: 0000-0003-3101-4416 Author Contributions °

Z.L. and P.H. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Daisuke Hashizume, Daishi Inoue, and Tomoka Kikitsu of the Materials Characterization Support Unit, RIKEN CEMS, for SEM, TEM, and TEM EDS measurements and for their valuable comments and discussion. We also thank Prof. Yu Nagase and Tomoki Mimura of Tokai University for GPC measurements, and their comments. We acknowledge Dr. Zhaomin Hou and Dr. Masayoshi Nishiura of the Organometallic Chemistry Laboratory, RIKEN, for TGA and DSC measurements. We also thank the Chemical Analysis Group, Materials Characterization Support Unit, RIKEN CEMS, for ICP emission spectrometry. This work was partially supported by a Grant-in-Aid for Scientific Research (C) (no. 15K05639) for M.K. from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

■

REFERENCES

(1) De Volder, M. F. L.; Tawfick, S. H.; Baughman, R. H.; Hart, A. J. Carbon Nanotubes: Present and Future Commercial Applications. Science 2013, 339, 535−539.

1631

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632

ACS Omega

Article

(2) Jariwala, D.; Sangwan, V. K.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C. Carbon Nanomaterials for Electronics, Optoelectronics, Photovoltaics, and Sensing. Chem. Soc. Rev. 2013, 42, 2824−2860. (3) Park, S.; Vosguerichian, M.; Bao, Z. A Review of Fabrication and Applications of Carbon Nanotube Film-Based Flexible Electronics. Nanoscale 2013, 5, 1727−1752. (4) Chen, K.; Gao, W.; Emaminejad, S.; Kiriya, D.; Ota, H.; Nyein, H. Y. Y.; Takei, K.; Javey, A. Printed Carbon Nanotube Electronics and Sensor Systems. Adv. Mater. 2016, 28, 4397−4414. (5) Vázquez, E.; Giacalone, F.; Prato, M. Non-Conventional Methods and Media for the Activation and Manipulation of Carbon Nanoforms. Chem. Soc. Rev. 2014, 43, 58−69. (6) Samanta, S. K.; Fritsch, M.; Scherf, U.; Gomulya, W.; Bisri, S. Z.; Loi, M. A. Conjugated Polymer-Assisted Dispersion of Single-Wall Carbon Nanotubes: The Power of Polymer Wrapping. Acc. Chem. Res. 2014, 47, 2446−2456. (7) Kawamoto, M.; He, P.; Ito, Y. Green Processing of Carbon Nanomaterials. Adv. Mater. 2017, DOI: 10.1002/adma.201602423. (8) Richard, C.; Balavoine, F.; Schultz, P.; Ebbesen, T. W.; Mioskowski, C. Supramolecular Self-Assembly of Lipid Derivatives on Carbon Nanotubes. Science 2003, 300, 775−778. (9) Kang, Y. K.; Lee, O.-S.; Deria, P.; Kim, S. H.; Park, T.-H.; Bonnell, D. A.; Saven, J. G.; Therien, M. J. Helical Wrapping of SingleWalled Carbon Nanotubes by Water Soluble Poly(p-phenyleneethynylene). Nano Lett. 2009, 9, 1414−1418. (10) Numata, M.; Asai, M.; Kaneko, K.; Bae, A.-H.; Hasegawa, T.; Sakurai, K.; Shinkai, S. Inclusion of Cut and As-Grown Single-Walled Carbon Nanotubes in the Helical Superstructure of Schizophyllan and Curdlan (β-1,3-Glucans). J. Am. Chem. Soc. 2005, 127, 5875−5884. (11) Qiu, L.; Yang, X.; Gou, X.; Yang, W.; Ma, Z.-F.; Wallace, G. G.; Li, D. Dispersing Carbon Nanotubes with Graphene Oxide in Water and Synergistic Effects between Graphene Derivatives. Chem.Eur. J. 2010, 16, 10653−10658. (12) Hilmer, A. J.; Tvrdy, K.; Zhang, J.; Strano, M. S. Charge Transfer Structure−Reactivity Dependence of Fullerene−SingleWalled Carbon Nanotube Heterojunctions. J. Am. Chem. Soc. 2013, 135, 11901−11910. (13) Grigoryan, G.; Kim, Y. H.; Acharya, R.; Axelrod, K.; Jain, R. M.; Willis, L.; Drndic, M.; Kikkawa, J. M.; DeGrado, W. F. Computational Design of Virus-Like Protein Assemblies on Carbon Nanotube Surfaces. Science 2011, 332, 1071−1076. (14) Tu, X.; Manohar, S.; Jagota, A.; Zheng, M. DNA Sequence Motifs for Structure-Specific Recognition and Separation of Carbon Nanotubes. Nature 2009, 460, 250−253. (15) Wang, S.; Humphreys, E. S.; Chung, S.-Y.; Delduco, D. F.; Lustig, S. R.; Wang, H.; Parker, K. N.; Rizzo, N. W.; Subramoney, S.; Chiang, Y.-M.; Jagota, A. Peptides with Selective Affinity for Carbon Nanotubes. Nat. Mater. 2003, 2, 196−200. (16) Deguchi, S.; Alargova, R. G.; Tsujii, K. Stable Dispersions of Fullerenes, C-60 and C-70, in Water. Preparation and Characterization. Langmuir 2001, 17, 6013−6017. (17) Varanasi, S. R.; Guskova, O. A.; John, A.; Sommer, J.-U. Water around Fullerene Shape Amphiphiles: A Molecular Dynamics Simulation Study of Hydrophobic Hydration. J. Chem. Phys. 2015, 142, 224308. (18) Sun, X.; Zaric, S.; Daranciang, D.; Welsher, K.; Lu, Y.; Li, X.; Dai, H. Optical Properties of Ultrashort Semiconducting Single-Walled Carbon Nanotube Capsules Down to Sub-10 nm. J. Am. Chem. Soc. 2008, 130, 6551−6555. (19) Liu, L.; Yang, C.; Zhao, K.; Li, J.; Wu, H.-C. Ultrashort SingleWalled Carbon Nanotubes in a Lipid Bilayer as a New Nanopore Sensor. Nat. Commun. 2013, 4, 2989. (20) Liu, H.; Nishide, D.; Tanaka, T.; Kataura, H. Large-Scale SingleChirality Separation of Single-Wall Carbon Nanotubes by Simple Gel Chromatography. Nat. Commun. 2011, 2, 309. (21) Hirano, A.; Tanaka, T.; Urabe, Y.; Kataura, H. Purification of Single-Wall Carbon Nanotubes by Controlling the Adsorbability onto Agarose Gels Using Deoxycholate. J. Phys. Chem. C 2012, 116, 9816− 9823.

(22) Attal, S.; Thiruvengadathan, R.; Regev, O. Determination of the Concentration of Single-Walled Carbon Nanotubes in Aqueous Dispersions using UV−Visible Absorption Spectroscopy. Anal. Chem. 2006, 78, 8098−8104. (23) Cook, S.; Ohkita, H.; Kim, Y.; Benson-Smith, J. J.; Bradley, D. D. C.; Durrant, J. R. A Photophysical Study of PCBM Thin Films. Chem. Phys. Lett. 2007, 445, 276−280. (24) Wang, J.; Larsen, C.; Wågberg, T.; Edman, L. Direct UV Patterning of Electronically Active Fullerene Films. Adv. Funct. Mater. 2011, 21, 3723−3728. (25) Blanch, A. J.; Lenehan, C. E.; Quinton, J. S. Optimizing Surfactant Concentrations for Dispersion of Single-Walled Carbon Nanotubes in Aqueous Solution. J. Phys. Chem. B 2010, 114, 9805− 9811. (26) Ohkita, H.; Ito, S. Transient Absorption Spectroscopy of Polymer-Based Thin-Film Solar Cells. Polymer 2011, 52, 4397−4417. (27) Guo, J.; Ohkita, H.; Yokoya, S.; Benten, H.; Ito, S. Bimodal Polarons and Hole Transport in Poly(3-hexylthiophene):Fullerene Blend Films. J. Am. Chem. Soc. 2010, 132, 9631−9637. (28) Tauc, J. Optical Properties and Electronic Structure of Amorphous Ge and Si. Mater. Res. Bull. 1968, 3, 37−46. (29) Liu, P.; Sun, Q.; Zhu, F.; Liu, K.; Jiang, K.; Liu, L.; Li, Q.; Fan, S. Measuring the Work Function of Carbon Nanotubes with Thermionic Method. Nano Lett. 2008, 8, 647−651. (30) Li, J.; Ma, P. C.; Chow, W. S.; To, C. K.; Tang, B. Z.; Kim, J.-K. Correlations between Percolation Threshold, Dispersion State, and Aspect Ratio of Carbon Nanotubes. Adv. Funct. Mater. 2007, 17, 3207−3215. (31) Dittmer, J. J.; Marseglia, E. A.; Friend, R. H. Electron Trapping in Dye/Polymer Blend Photovoltaic Cells. Adv. Mater. 2000, 12, 1270−1274. (32) Bryning, M. B.; Islam, M. F.; Kikkawa, J. M.; Yodh, A. G. Very Low Conductivity Threshold in Bulk Isotropic Single-Walled Carbon Nanotube-Epoxy Composites. Adv. Mater. 2005, 17, 1186−1191. (33) Liu, L. M.; Stanchina, W. E.; Li, G. Y. Effects of Semiconducting and Metallic Single-Walled Carbon Nanotubes on Performance of Bulk Heterojunction Organic Solar Cells. Appl. Phys. Lett. 2009, 94, 233309. (34) Kymakis, E.; Kornilios, N.; Koudoumas, E. Carbon Nanotube Doping of P3HT:PCBM Photovoltaic Devices. J. Phys. D: Appl. Phys. 2008, 41, 165110. (35) Peumans, P.; Yakimov, A.; Forrest, S. R. Small Molecular Weight Organic Thin-Film Photodetectors and Solar Cells. J. Appl. Phys. 2003, 93, 3693−3723. (36) Rauch, T.; Böberl, M.; Tedde, S. F.; Fürst, J.; Kovalenko, M. V.; Hesser, G.; Lemmer, U.; Heiss, W.; Hayden, O. Near-Infrared Imaging with Quantum-Dot-Sensitized Organic Photodiodes. Nat. Photonics 2009, 3, 332−336.

1632

DOI: 10.1021/acsomega.7b00175 ACS Omega 2017, 2, 1625−1632