CdSe–CdTe Heterojunction Nanorods: Role of CdTe Segment in

Aug 30, 2017 - To achieve the above objective, we designed the CdSe–CdTe heterojunction nanorods with a tunable charge transfer emission and compare...
3 downloads 13 Views 6MB 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

CdSe−CdTe Heterojunction Nanorods: Role of CdTe Segment in Modulating the Charge Transfer Processes Kurukkal Balakrishnan Subila,†,§ Kulangara Sandeep,† Elizabeth Mariam Thomas,† Jay Ghatak,‡ Sonnada Math Shivaprasad,*,‡ and K. George Thomas*,†,‡ †

School of Chemistry, Indian Institute of Science Education and Research Thiruvananthapuram (IISER-TVM), Vithura, Maruthamala (PO), Thiruvananthapuram 695551, India ‡ Chemistry and Physics of Materials Unit, Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Jakkur (PO), Bangalore 560064, India S Supporting Information *

ABSTRACT: Heterojunction nanorods having dissimilar semiconductors possess charge transfer (CT) properties and are proposed as active elements in optoelectronic systems. Herein, we describe the synthetic methodologies for controlling the charge carrier recombination dynamics in CdSe−CdTe heterojunction nanorods through the precise growth of CdTe segment from one of the tips of CdSe nanorods. The location of heterojunction was established through a point-by-point collection of the energy-dispersive X-ray spectra using scanning transmission electron microscopy. The possibilities of the growth of CdTe from both the tips of CdSe nanorods and the overcoating of CdTe over CdSe segment were also ruled out. The CT emission in the heterojunction nanorods originates through an interfacial excitonic recombination and was further tuned to the near-infrared region by varying the two parameters: the aspect ratio of CdSe and the length of CdTe segment. These aspects are evidenced from the emission lifetime and the femtosecond transient absorption studies.

■

INTRODUCTION Heterojunction nanostructures obtained by integrating a semiconductor with another semiconductor display novel functional properties that are fundamentally different from their individual components.1−3 For example, the photoluminescence lifetimes in the heterojunction nanostructures having a staggered band alignment (often called as type II systems) are found to be longer due to the lower degree of overlap between the electron and hole wavefunctions.2,4,5 Photoexcitation in type II semiconductor systems essentially leads to a spatial separation of charge carriers across the interface, making the exciton loosely bound.2,6,7 This unique property qualifies semiconductor heterojunction nanostructures as one of the most promising candidates for photovoltaic applications.8−12 However, effective extraction of electrons and holes is not possible in type II semiconductors having core− shell architectures because one of the charge carriers is trapped in the core.6,13 Indeed, it is possible to overcome such situations by designing one-dimensional heterojunction nanorods having two dissimilar semiconductors (denoted as S1 and S2). The synthesis of such systems having a well-defined interface is challenging: various approaches reported include (i) S2 linked on to the growing tip of S1 (S1−S2),1,2,14 (ii) S2 linked on both the tips of S1 (S2−S1−S2),15−17 and (iii) S2 incorporated in the one-dimensional structures of S1,18−20 forming nanoplatelets21−23 and nanobarbells.24,25 © 2017 American Chemical Society

An ideal type II heterostructure displays unique absorption and emission spectral features originating from the chargeseparated state.2 Theoretical investigations of the CdSe−CdTe heterojunction nanorods have been carried out independently by two groups using the local density approximation: (i) Wang and Wang determined the exciton dissociation dynamics, binding energy, and radiative decay time of charge transfer (CT) excitons in these systems26 and (ii) Ribeiro et al. reported the band structure of nanostructures and their band offset at the interface.27 These theoretical investigations are extremely useful for designing type II heterojunction nanorods with well-defined properties. In a classic work, Scholes and co-workers first demonstrated the design of CdSe−CdTe heterojunction nanorods, which possess two excitonic emission bands, and further investigated their physical properties.1,2,4,28,29 The longwavelength emission band corresponds to an interfacial charge transfer state. The second emission band, blue shifted from the CT emission, stems from the direct excitonic recombination in the CdTe segment.19 The excitons generated through the former process are localized in two segments and hence are loosely bound compared with the latter. The loosely bound excitons in the heterojunction nanostructures are long-lived, Received: July 14, 2017 Accepted: August 8, 2017 Published: August 30, 2017 5150

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 1. HR-TEM image of (A) CdSe(H) and (B) CdSe(H)−CdTe (Te 10) nanorods. Both CdSe(H) and CdSe(H)−CdTe (Te 10) nanorods were allowed to grow for 10 and 5 min, respectively, after the addition of the corresponding precursors (details are provided in Materials and Methods).

Table 1. Stoichiometry of the Precursors Used for Synthesizing Heterojunction Nanorods Cd precursor

a

Se precursor

0.488 mmol

0.22 mmol

1 molar equiv

0.45 molar equiv

Te precursor (mmol)

molar equivalents of Te precursor

0.025 0.05 0.075 0.10 0.15

0.05 0.10 0.15 0.20 0.30

CdSe(X)−CdTe heterojunctiona CdSe(X)−CdTe CdSe(X)−CdTe CdSe(X)−CdTe CdSe(X)−CdTe CdSe(X)−CdTe

(Te (Te (Te (Te (Te

5) 10) 15) 20) 30)

Heterojunction nanorods grown along the c axis of CdSe having a lower aspect ratio (X = L) and a higher aspect ratio (X = H).

and 2.49 (length of ≈10.2 nm and width of ≈4.1 nm), respectively (Figures S1D,E and S2D,E). The first excitonic band of CdSe(L) and CdSe(H) are centered at 559 and 626 nm, respectively, and the corresponding emission maxima were observed at 571 and 635 nm, respectively (Figures S1A and S2A). Highly symmetrical emission spectrum and absence of any band at the long-wavelength region indicate that the emission originates from the direct excitonic recombination rather than from the trap states. The narrow emission, with full width at half-maximum of ≈30 nm, also indicates that CdSe nanorods are highly monodispersed. The variation in the reactivity of different crystal faces forms the basis of the preferential growth of CdTe segment from one of the tips of CdSe. It is reported that the anisotropic growth of the nanocrystal nuclei prefers to be in the wurtzite lattice structure.32,33 The X-ray diffraction patterns of CdSe used for growing CdTe possess a wurtzite structure (Figures S1B and S2B) and details are listed below. Growth of CdSe−CdTe Heterojunction Nanorods. The (001) facets of CdSe nanorods with a wurtzite structure involve alternating Se and Cd layers. As a result, cadmium atoms of the (001̅) face are relatively unsaturated with three dangling bonds per atom, which forms the growing face. In contrast, the opposite (001) face of the c axis exposes relatively saturated Cd atoms having one dangling bond per atom. Another factor responsible for the preferential growth is the lower coordination of ligands along the c axis compared with the stable lateral facets. It is reported that the phosphonic acid ligands bind to the lateral nonpolar facets of the wurtzite structure than the basal polar facets.32 Thus, (i) the higher energy of (001̅) face due to the presence of unsaturated bonds and (ii) the lower coordination ability of the ligands on these polar facets drive the preferential growth of the nanorods along the c axis. The syntheses of CdSe−CdTe nanorods are carried out by the sequential injection of Se and Te precursors into the reaction

which makes the charge carrier extraction more efficient than in the tightly bound excitons.30,31 One of the essential requirements for the efficient design of photovoltaic devices is to identify semiconductor nanostructures that generate long-lived electron−hole pairs. However, the potential of these systems has not been explored may be due to the synthetic difficulties involved in the design of heterojunction nanorods possessing charge transfer states. Also, literature lacks correlative investigations on the dimensions of the semiconductor segments, which can solely produce a charge transfer state. Herein, we report an optimal synthetic procedure for the design of heterojunction nanorods possessing well-defined segments of CdSe and CdTe, which exclusively yield tunable charge transfer emission.

■

RESULTS AND DISCUSSION To achieve the above objective, we designed the CdSe−CdTe heterojunction nanorods with a tunable charge transfer emission and compared their properties with CdSe nanorods. Design of the former was achieved by varying the molar equivalents of the precursors (Cd:Se:Te) and the reaction time. Precise locations of CdSe and CdTe segments as well as the heterojunction were established through a point-by-point elemental mapping and correlation of their photophysical properties with their dimensions. Synthesis of CdSe Nanorods. CdSe nanorods having two different aspect ratios were synthesized and details are provided in Materials and Methods. They are designated as CdSe(L) and CdSe(H), representing the lower and higher aspect ratios, respectively. Details of characterization are presented as the Supporting Information. It is clear from the high-resolution transmission electron microscopy (HR-TEM) images that the nanorods are highly crystalline in nature (Figures 1, S1C, and S2C). The aspect ratios of CdSe(L) and CdSe(H) nanorods are estimated as 1.66 (length of ≈5.0 nm and width of ≈3.0 nm) 5151

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 2. Normalized absorption (A, C) and emission (B, D) spectra of (i) CdSe(L) nanorods (black trace) and (ii, iii) CdSe(L)−CdTe heterojunction nanorods obtained by growing CdTe along the c axis by varying time: (ii) 5 min (red trace) and (iii) 10 min (blue trace). CdSe(L) nanorods are synthesized by injecting Se precursor (0.45 molar equiv) to Cd precursor (1 molar equiv) and arresting the reaction after 5 min. Further CdSe(L)−CdTe heterojunction nanorods are synthesized by adding Te precursor to the above reaction mixture after 5 min: (A, B) 0.05 molar equiv and (C, D) 0.15 molar equiv (designated as CdSe(L)−CdTe (Te 5) and CdSe(L)−CdTe (Te 15), respectively). The absorption spectra are normalized at the corresponding first excitonic band of CdSe and the emission spectra are normalized at the emission maximum.

pot containing Cd2+ ligand complex as mentioned in Table 1. The resultant heterojunction nanorods obtained from CdSe(L) and CdSe(H) are denoted as CdSe(L)−CdTe and CdSe(H)− CdTe, respectively. First, CdSe nanorods are synthesized by injecting Se precursor (0.45 molar equiv) to the reaction pot having excess Cd precursor (1 molar equiv). Further, the growth of CdTe segment is achieved by varying the two factors: (i) concentrations of Te precursor (0.05−0.30 molar equiv; Table 1) and (ii) reaction time. Maintaining the temperature at 280 °C, Te precursor (Te−trioctylphosphine (TOP) solution) is quickly injected to the above solution. The growth of CdTe segment along the c axis of CdSe nanorods is followed as a function of time by withdrawing the aliquots. The concentration of Te precursor is systematically varied (0.05, 0.10, 0.15, 0.20, and 0.30 molar equiv) and the obtained heterojunction nanorods are labeled as indicated in the last column of Table 1. The samples are precipitated from methanol and purified by washing with a mixture (1:1) of methanol and acetone to remove the unreacted precursors and the excess ligands by centrifugation. In the case of CdSe(L)−CdTe (Te 5) nanorods, the excitonic band at 559 nm, corresponding to CdSe(L) segment, is found to be more or less unaffected (Figure 2A) in the absorption spectrum. The growth of CdTe segment results in an increase in the absorption cross section in the near-infrared (NIR) region, suggesting the formation of a charge transfer (CT) state.2,24 This NIR absorption can also have a minor contribution from CdTe segment. Interestingly, the emission spectrum of these systems showed a discrete CT emission in the NIR region with broad spectral features. However, the emission bands corresponding to CdSe as well as CdTe segments are absent in these systems. It is well established in

organic systems that the recombination of electron−hole pair in donor−acceptor systems results in the CT emission.34 With time, the emission maximum of the CT band of CdSe(L)−CdTe (Te 5) shifts to a longer wavelength (traces (ii) and (iii) of Figure 2), indicating the growth of CdTe segment. The CdTe segment is allowed to grow for 5 min and a small aliquot of the product is withdrawn and purified. The emission spectra are recorded by exciting the samples at 450 nm, keeping both the excitation and emission slit width as 2 nm. Similar results are observed when these experiments are repeated by injecting 0.10 molar equiv of Te precursor to CdSe(L) (labeled as CdSe(L)− CdTe (Te 10)). The CT emission shifts to the NIR region with increasing time indicate a progressive growth of the CdTe segment on CdSe(L) nanorods. The CT emission shifts to lower energies as the energy difference between the CdSe conduction band and the CdTe valence band decreases (Figure S3). Further increasing the concentration of Te precursor (0.15− 0.30 molar equiv) yields heterojunction nanorods having two emission bands similar to that reported by Scholes and coworkers: (i) a CT band in the NIR region and (ii) a higher energy band corresponding to the direct recombination of the exciton in CdTe segment. The absorption and emission spectra of the heterojunction nanorods formed on adding 0.15 molar equiv of Te precursor to CdSe(L) nanorods is presented as traces (ii) and (iii) in Figure 2C,D, respectively. The bands corresponding to the emission from CdTe and CT emission from the heterojunction are observed at 640 and 775 nm, respectively, which further are shifted to 663 and 800 nm when allowed to grow for 10 min. Excitation spectrum of CdSe(L)−CdTe (Te 15), followed at the charge transfer emission maximum (775 nm), is compared with the normalized absorption spectrum of CdSe and the corresponding heterojunction nanorod (Figure S4). The 5152

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 3. (A) High-resolution HAADF-STEM images of two CdSe(L)−CdTe (Te 10) heterojunction nanorods and the EDX analysis carried out at their terminal positions, marked as EDX 1−4. The spectral lines corresponding to various elements, measured from the marked positions using EDX analysis, are presented. (B) Elemental analysis performed at closely spaced locations from one end to the other, on a single nanorod, by collecting the EDX spectra. The location 2 is identified as heterojunction wherein the signals from both Se and Te are observed.

nanorods (Figure S6). However, it is difficult to conclusively locate the interface because the lattice mismatch (7.1%) and the difference in the d-spacing values along the c axis plane for CdSe (0.35 nm) and CdTe (0.37 nm) are small.35,36 However, a distinct difference in d-spacing is observed on moving along the longitudinal axis of the heterojunction, which rules out the possibility of overcoating CdTe over CdSe nanorods. Further attempts are being made to identify the heterojunction of CdSe−CdTe nanorods by carrying out the elemental analysis along its c axis using energy-dispersive X-ray (EDX) analysis. Scanning transmission electron microscopy (STEM) mode of TEM (≈0.2 nm beam) is used to carry out these studies using the aberration-corrected Titan (FEI) electron microscope, operated at 300 kV. It is found that severe damage occurred for nanorods (Figure S7) upon prolonged exposure to electron beam during elemental mapping through line scan acquisition method. Hence, the EDX measurements are restricted to a point-by-point analysis. The heterojunction nanorods and their interface are further characterized using scanning transmission electron microscopy

features of the excitation spectrum of the heterojunction nanorods match well with those of the corresponding absorption spectrum. These results together with the lifetime studies (vide infra) indicate that the CT emission originates through the excitonic recombination across the interface. To generalize the design strategy, attempts are made to grow CdTe segments on the reactive tip of CdSe(H). Three independent experiments are being carried out by adding 0.10, 0.20, and 0.30 molar equiv of Te precursors to a solution containing CdSe(H) nanorods. CdTe segments are grown on CdSe nanorods for 5 min and the reaction is arrested by decreasing the temperature. The samples are purified as mentioned earlier. Heterojunctions of CdSe(L)−CdTe and CdSe(H)−CdTe are characterized using various analytical (Supporting Information) and microscopic techniques. Microscopic Characterization of Heterojunctions. It is clear from the HR-TEM images that the length of CdSe nanorods increases on injection of Te precursors (Figures 1, S2D, and S5B). Attempts are made to locate the heterojunction by estimating the d-spacing values along the c axis of the 5153

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 4. HAADF-STEM images of CdSe(H)−CdTe (Te 30) heterojunction nanorods and the EDX analysis carried out at closely spaced locations, from one end to the other, on a single nanorod. The location 3 is identified as a heterojunction wherein the signals from both Se and Te are observed and bending of the nanorods is observed as a result of the strain encountered during the growth of CdTe segment due to lattice mismatch.

high-angle annular dark-field (STEM-HAADF) analysis. In the present case, the contrast between CdSe and CdTe is not high enough to differentiate the materials due to the small difference in atomic number between Se and Te in the HAADF technique. Hence, we performed EDX measurements at specific locations, and these results are presented in Figure 3A. From the STEM-HAADF images, two CdSe(L)−CdTe nanorods are selected and the EDX analysis is performed at two different locations (labeled as 1−4). In all of the locations, we observe a spectral line corresponding to the energy 3.13 keV, which is attributed to the Cd Lα. This result indicates that cadmium is present throughout the nanorod. The EDX spectrum collected from one end (Figure 3A; EDX 1) of the heterojunction nanorod in the STEM image showed a spectral line at 1.37 keV corresponding to Se Lα. The other end of same heterojunction nanorod (Figure 3A; EDX 2) shows a spectral line at 3.76 keV corresponding to Te Lα. The spectral peaks corresponding to Te and Se are not observed in positions 1 and 2, respectively. Similar results are observed for the second nanorod (Figure 3A; EDX 3 and 4). These results conclusively indicate that the nanorods are made up of two dissimilar semiconductors, with one end as CdSe and the other end as CdTe. However, on the basis of the above studies, we could not locate the heterojunction in the nanorods. Hence, similar set of experiments is performed on a single nanorod by collecting the EDX spectra, from one end to the other, in closely spaced locations (Figure 3B). The EDX spectra collected from four locations are presented (Figure 3B, labeled as EDX 1−4) and three important observations are gathered. The spectral lines corresponding to (i) Cd is observed in all of the positions, (ii)

Se is observed at positions 2−4, and (iii) Te is observed at positions 1 and 2. These studies show the presence of Se and Te at position 2, indicating the location of the heterojunction from where CdTe starts to grow on addition of Te precursor. The absence of Te signals at positions 3 and 4 rules out the possibility of the growth of CdTe along the lateral facets. The width remained more or less unchanged, indicating the preferential growth of the nanorods along the c axis. Thus, other possibilities such as (i) the growth of CdTe from both the tips of CdSe nanorods and (ii) overcoating of CdTe over CdSe nanorods are ruled out. Further, we investigated the STEM-HAADF imaging and the EDX analysis of CdSe(H)−CdTe (Te 30) heterojunction nanorods synthesized by using higher molar equivalents of Te precursor. A significant growth of CdTe segment occurred at a higher concentration of Te precursor with a bending of the nanorods. The EDX analysis is performed at six closely spaced locations (Figure 4; positions 1−6) along CdSe(H)−CdTe (Te 30) nanorod. The EDX spectra showed the characteristic lines corresponding to Cd and Se at positions 1 and 2 (EDX 1 and 2). Interestingly, at position 3, Te line was also observed along with Cd and Se lines, indicating the formation of a heterojunction (EDX 3). The EDX analysis performed at positions 4−6 exhibit spectral lines corresponding to Cd and Te (EDX 4−6), indicating the growth of CdTe segment from position 2. On increasing the length of CdTe segment, a curvature is observed at the heterojunction of the nanorods as evidenced in Figure 4. Bending of the nanorods is attributed to the strain encountered during the growth of CdTe segment due to lattice mismatch.37 Strain-induced bending has been earlier reported in CdSe 5154

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 5. (A, C) Absorption and (B, D) emission spectra of CdSe(L) and CdSe(H) nanorods and the corresponding heterojunction nanorods in chloroform. Heterojunction nanorods are obtained by adding varying amounts of Te precursor to (A, B) CdSe(L) and (C, D) CdSe(H) nanorods. The CdTe segments are allowed to grow for 5 min after the addition of Te precursor. The emission spectra are recorded by exciting the solution at 450 nm and by keeping both the excitation and emission slit width as 2 nm.

interface. On further increasing the concentration of Te precursor above 0.15 molar equiv, a new emission band is observed at the high-energy region, originating through the direct recombination of exciton in CdTe segment. The absorption and the emission spectra of CdSe(H) and CdSe(H)−CdTe nanorods are presented in Figure 5C,D (also see Figure S8). Two excitonic emission bands are observed when the growth of CdTe segment exceeds 7 nm, and these aspects are obvious from Figure 5. Following observations are drawn based on the aspect ratio-dependent studies: (i) CT emission originates through the excitonic recombination at the interface at lower concentrations of Te precursor, irrespective of the aspect ratio of CdSe segment; and (ii) a new emission band corresponding to CdTe is observed along with CT emission at higher Te precursor concentrations which is also independent of the aspect ratio of CdSe. Thus, it is concluded that the length of CdTe segment influences the emission properties of these heterojunction nanorods rather than the aspect ratio of CdSe. This is due to the fact that valence and conduction bands of CdTe possess higher energies than those of CdSe and hence the holes will quickly migrate to CdTe segment. As a result, emission corresponding to the CdSe segment is not observed in heterojunction nanorods.

nanorods overcoated with CdTe. In this case, deposition of a few monolayers of CdTe is observed on the lateral faces of CdSe with a faster growth at the tips.38 Further deposition of CdTe results in the bending of the nanorods. In the present case, bending occurs on increasing the length of CdTe and growth occurs only from one of the tips of CdSe nanorod. Photophysical Characterization of Heterojunction Nanorods. After establishing the presence of heterojunction, we further investigated their photophysical properties by varying the length of CdTe segment on CdSe(L) and CdSe(H). The absorption and emission spectra of CdSe(L)− CdTe and CdSe(H)−CdTe synthesized by varying Te precursor concentrations are presented in Figure 5A,B as a stacked plot (Figure S8, for corresponding plots in three-dimensional format). It is evident from the figure that the volume ratio of CdSe and CdTe segments has a major influence on their optical properties. The CdSe(L)−CdTe heterojunction nanorods possess only charge transfer emission when the molar equivalents of Te precursor are maintained at ≤0.10. Based on these results and TEM studies, it is clear that when the length of CdTe segment is maintained below 4 nm in the heterojunction nanorod (CdSe(L) segment: length ≈5.0 nm; width ≈3.0 nm), the emission originates through an excitonic recombination at the 5155

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Figure 6. (A) Scheme illustrating the synthesis of heterojunction nanorods having charge transfer emission and two excitonic emission bands at various Te precursor concentrations. The average dimensions of CdSe and CdTe segments are evaluated from the HR-TEM analysis. (B) Absorption−time profile corresponding to the bleach recovery of (i) CdSe(L) nanorods, (ii) CdSe(L)−CdTe (Te 5), and (iii) CdSe(L)−CdTe (Te 10) heterojunction (Table S2, for kinetic analysis).

The emission lifetime of CdSe(L) and various CdSe(L)−CdTe heterojunction nanorods is estimated using time-correlated single photon counting (TCSPC) measurements. The fitted decay traces along with residual plots are presented in Figure S9. The charge transfer emission observed from the heterojunction nanorods is found to be long-lived. For example, the average lifetime (τ) of CdSe(L) and CdSe(L)−CdTe (Te 5) heterojunction nanorods is estimated as 23 and 136 ns, respectively, which is further enhanced to 178 ns on increasing the length of CdTe segment to 4 nm (denoted as CdSe(L)− CdTe (Te 10)). In the heterojunction nanorods, the electron and hole wavefunctions are located in CdSe and CdTe segments, respectively, and hence possess a lower degree of overlap, making the exciton loosely bound. This effect is more predominant when CdTe segment is longer because the transferred electron becomes more delocalized. A significant growth of CdTe segment is observed when a higher concentration of Te precursor is used (0.15, 0.20, and 0.30 molar equiv). Two excitonic emission bands are observed from these nanorods and emission decay of both the bands are followed. The τavg of the charge transfer band remained more or less unaffected (∼176 ns) in all of the cases. However, the emission emerging as a result of the excitonic recombination in CdTe possesses a much lower lifetime (τavg ∼ 22 ns; details are provided as Table S1). Thus, the longer lifetime for the charge transfer emission is attributed to the reduced overlap between the hole and electron wavefunctions resulting from the slow recombination of excitons across the heterojunction.

Femtosecond transient absorption studies of CdSe(L) and CdSe(L)−CdTe heterojunction nanorods (excited at 400 nm using mode-locked Ti:sapphire laser; Supporting Information) provided information about the dynamics of exciton decay across the interface. Chirp-corrected transient absorption spectra of both the nanorods possess bleach corresponding to the excitonic bands (Figure S10). It is well established that the bleach recovery at the excitonic band position results from the depopulation of the state-filled electrons, which can provide information about various electron relaxation pathways.39 The absorption−time profile of CdSe (trace (i) in Figure 6B and Table S2) in the time window of 500 ps showed a significant recovery, indicating that the state-filled electrons undergo fast relaxation. When investigated under identical experimental conditions, the relaxation of the state-filled electrons is substantially slower on increasing the length of CdTe segment. These results further indicate a lower degree of overlap of the electron and hole wavefunctions, delocalized in the two segments of heterojunction nanorods. Dooley et al. have earlier reported the charge transfer dynamics in CdSe−CdTe heterojunction nanorods by selectively exciting CdTe. The interfacial interconduction band electron transfer from CdTe to CdSe occurs on 500 fs time scale, followed by the formation of a charge transfer band.19 However, in a more recent report, Scholes and co-workers reported the bleaching of CdSe as well as CdTe electronic transitions by pumping of CdSe−CdTe nanoplatelets at CdTe absorption wavelength. This is due to the delocalization of electrons over the entire conduction band of CdSe and CdTe as a result of the quasi-type II band 5156

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

■

alignment.40 Steady-state and time-resolved emission studies presented herein demonstrate that heterojunction nanorods having smaller segments of CdTe possess a well-defined CT emission, originating from the recombination of excitons at the interface. It is easy to extract electrons/holes from such loosely bound excitons in the heterojunction nanorods, thus making the rods useful as active components in photovoltaic devices.

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.7b00995. Experimental techniques; characterization of CdSe nanorods; characterization of CdSe−CdTe heterojunction nanorods; TCSPC analysis; femtosecond transient absorption studies (PDF)

■

■

CONCLUSIONS The synthetic methodology presented here by varying molar equivalents of (Cd:Se:Te) precursors is useful for designing CdSe−CdTe heterojunction nanorods having well-defined segments of CdSe and CdTe. These studies clearly indicate that the length of CdTe segment plays a significant role in controlling the CT band. Design of heterojunction nanorods having an exclusive charge transfer emission has been achieved by maintaining the ratio of the length of the CdTe to CdSe segments below 0.8. The charge transfer emission in these systems originates from the recombination of excitons at the interface. The structures of semiconductor segments and the heterojunction are conclusively established using dark-field scanning electron microscopic studies along with energydispersive X-ray spectroscopy. Absence of Te signal from the segment corresponding to CdSe confirms that the growth of CdTe occurs preferentially along the c axis rather than the lateral facets. The heterojunction nanorods reported herein possess a tunable charge transfer state and are devoid of any direct exciton formation. It is easy to extract electrons/holes from these systems, making them useful as active components in photovoltaic devices.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (S.M.S.). *E-mail: [email protected] (K.G.T.). ORCID

K. George Thomas: 0000-0003-1279-308X Present Address §

Department of Chemistry & Biochemistry, Radiation Laboratory, University of Notre Dame, Notre Dame, Indiana 46556, United States (K.B.S.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Professor C.N.R. Rao for the kind support and encouragement. We thank IISER Thiruvananthapuram, JNCASR Bangalore, and Department of Science and Technology, Government of India (DST-European Commission Project. INT/IRMC/ECSOLAR/LARGECELLS: 1/ 261936/2010) for financial support. K.B.S. and K.S. thank UGC for the fellowship and K.G.T. for the J.C. Bose National Fellowship. We thank Rafeeque P.P. for graphical abstract.

■

■

MATERIALS AND METHODS Synthesis of CdSe(L) and CdSe(H) Nanorods. Cadmium selenide nanorods are synthesized by following high-temperature synthesis (>310 °C) reported by Scholes and co-workers1 with modifications. In brief, CdO (63 mg) dissolved in a mixture of tetradecylphosphonic acid (275 mg) and trioctylphosphine oxide (885 mg) is evacuated in a three-necked round-bottom flask kept at 80 °C for 2 h. The mixture is then heated to 310 °C under nitrogen atmosphere. After obtaining a clear solution, the temperature is reduced to 300 °C and TOP− Se solution (17.5 mg of Se in 0.75 mL TOP) is injected rapidly. The desired aspect ratio of CdSe nanorods is obtained by adjusting the reaction time. CdSe(L) and CdSe(H) nanorods are obtained by increasing the reaction time to 5 and 10 min, respectively. The reaction is arrested by decreasing the temperature. The quantum rods are purified by precipitating with a mixture (1:4) of methanol and acetone, followed by centrifugation. The obtained nanorods are redispersed in CHCl3 (spectroscopic grade) and used for further studies. Synthesis of CdSe−CdTe Heterojunction Nanorods. Heterojunction nanorods are synthesized by injecting TOP−Te to the above reaction mixture at 280 °C, before purification, and allowing the CdTe segment to grow for 5 min. Variation in CdTe length is carried out by increasing the molar equivalents of (Cd:Se:Te) as given in Table 1. The growth of heteronanorods is arrested by decreasing the reaction temperature. The nanorods obtained in all of the cases are purified by repeated precipitation and washing with a mixture (1:4) of methanol and acetone.

REFERENCES

(1) Kumar, S.; Jones, M.; Lo, S. S.; Scholes, G. D. Nanorod Heterostructures Showing Photoinduced Charge Separation. Small 2007, 3, 1633−1639. (2) Lo, S. S.; Mirkovic, T.; Chuang, C.-H.; Burda, C.; Scholes, G. D. Emergent Properties Resulting from Type-II Band Alignment in Semiconductor Nanoheterostructures. Adv. Mater. 2011, 23, 180−197. (3) Selinsky, R. S.; Ding, Q.; Faber, M. S.; Wright, J. C.; Jin, S. Quantum dot nanoscale heterostructures for solar energy conversion. Chem. Soc. Rev. 2013, 42, 2963−2985. (4) He, J.; Lo, S. S.; Kim, J.; Scholes, G. D. Control of Exciton Spin Relaxation by Electron−Hole Decoupling in Type-II Nanocrystal Heterostructures. Nano Lett. 2008, 8, 4007−4013. (5) Ivanov, S. A.; Piryatinski, A.; Nanda, J.; Tretiak, S.; Zavadil, K. R.; Wallace, W. O.; Werder, D.; Klimov, V. I. Type-II Core/Shell CdS/ ZnSe Nanocrystals: Synthesis, Electronic Structures, and Spectroscopic Properties. J. Am. Chem. Soc. 2007, 129, 11708−11719. (6) Zhu, H.; Song, N.; Lian, T. Wave Function Engineering for Ultrafast Charge Separation and Slow Charge Recombination in Type II Core/Shell Quantum Dots. J. Am. Chem. Soc. 2011, 133, 8762− 8771. (7) Tyrakowski, C. M.; Shamirian, A.; Rowland, C. E.; Shen, H.; Das, A.; Schaller, R. D.; Snee, P. T. Bright Type II Quantum Dots. Chem. Mater. 2015, 27, 7276−7281. (8) Kamat, P. V. Quantum Dot Solar Cells. The Next Big Thing in Photovoltaics. J. Phys. Chem. Lett. 2013, 4, 908−918. (9) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Hybrid NanorodPolymer Solar Cells. Science 2002, 295, 2425−2427. (10) Nozik, A. J. Nanoscience and Nanostructures for Photovoltaics and Solar Fuels. Nano Lett. 2010, 10, 2735−2741. (11) Josell, D.; Debnath, R.; Ha, J. Y.; Guyer, J.; Sahiner, M. A.; Reehil, C. J.; Manners, W. A.; Nguyen, N. V. Windowless CdSe/CdTe 5157

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158

ACS Omega

Article

Solar Cells with Differentiated Back Contacts: J−V, EQE, and Photocurrent Mapping. ACS Appl. Mater. Interfaces 2014, 6, 15972− 15979. (12) Kim, N.; Um, H.-D.; Choi, I.; Kim, K.-H.; Seo, K. 18.4%Efficient Heterojunction Si Solar Cells Using Optimized ITO/Top Electrode. ACS Appl. Mater. Interfaces 2016, 8, 11412−11417. (13) Dorfs, D.; Franzl, T.; Osovsky, R.; Brumer, M.; Lifshitz, E.; Klar, T. A.; Eychmüller, A. Type-I and Type-II Nanoscale Heterostructures Based on CdTe Nanocrystals: A Comparative Study. Small 2008, 4, 1148−1152. (14) Bera, A.; Saha, S. K.; Pal, A. J. Heterodimers formed through a partial anionic exchange process: scanning tunneling spectroscopy to monitor bands across the junction vis-a-vis photoinduced charge separation. Nanoscale 2015, 7, 17366−17374. (15) Shim, M.; McDaniel, H.; Oh, N. Prospects for Strained Type-II Nanorod Heterostructures. J. Phys. Chem. Lett. 2011, 2, 2722−2727. (16) Xi, L.; Boothroyd, C.; Lam, Y. M. Controlled Synthesis of CdTe and CdSe Multiblock Heteronanostructures. Chem. Mater. 2009, 21, 1465−1470. (17) Zhao, H.; Koo, B.; Chen, D.; Korgel, B. A.; Kheng, K. Optical characterization of type-II CdTe/CdSe/CdTe heterostructure nanorods. J. Phys.: Conf. Ser. 2010, 210, No. 012029. (18) Sitt, A.; Sala, F. D.; Menagen, G.; Banin, U. Multiexciton Engineering in Seeded Core/Shell Nanorods: Transfer from Type-I to Quasi-type-II Regimes. Nano Lett. 2009, 9, 3470−3476. (19) Dooley, C. J.; Dimitrov, S. D.; Fiebig, T. Ultrafast electron transfer dynamics in CdSe/CdTe donor−acceptor nanorods. J. Phys. Chem. C 2008, 112, 12074−12076. (20) Diroll, B. T.; Gogotsi, N.; Murray, C. B. Statistical Description of CdSe/CdS Dot-in-Rod Heterostructures Using Scanning Transmission Electron Microscopy. Chem. Mater. 2016, 28, 3345−3351. (21) Pedetti, S.; Ithurria, S.; Heuclin, H.; Patriarche, G.; Dubertret, B. Type-II CdSe/CdTe Core/Crown Semiconductor Nanoplatelets. J. Am. Chem. Soc. 2014, 136, 16430−16438. (22) Wu, K.; Li, Q.; Jia, Y.; McBride, J. R.; Xie, Z.-x.; Lian, T. Efficient and Ultrafast Formation of Long-Lived Charge-Transfer Exciton State in Atomically Thin Cadmium Selenide/Cadmium Telluride Type-II Heteronanosheets. ACS Nano 2015, 9, 961−968. (23) Bouet, C.; Tessier, M. D.; Ithurria, S.; Mahler, B.; Nadal, B.; Dubertret, B. Flat Colloidal Semiconductor Nanoplatelets. Chem. Mater. 2013, 25, 1262−1271. (24) Kirsanova, M.; Nemchinov, A.; Hewa-Kasakarage, N. N.; Schmall, N.; Zamkov, M. Synthesis of ZnSe/CdS/ZnSe Nanobarbells Showing Photoinduced Charge Separation. Chem. Mater. 2009, 21, 4305−4309. (25) Halpert, J. E.; Porter, V. J.; Zimmer, J. P.; Bawendi, M. G. Synthesis of CdSe/CdTe nanobarbells. J. Am. Chem. Soc. 2006, 128, 12590−12591. (26) Wang, S.; Wang, L.-W. Exciton Dissociation in CdSe/CdTe Heterostructure Nanorods. J. Phys. Chem. Lett. 2011, 2, 1−6. (27) Ribeiro, M.; Fonseca, L. R. C.; Sadowski, T.; Ramprasad, R. Ab initio calculation of the CdSe/CdTe heterojunction band offset using the local-density approximation-1/2 technique with spin−orbit corrections. J. Appl. Phys. 2012, 111, No. 073708. (28) Scholes, G. D.; Jones, M.; Kumar, S. Energetics of Photoinduced Electron-Transfer Reactions Decided by Quantum Confinement. J. Phys. Chem. C 2007, 111, 13777−13785. (29) Lo, S. S.; Khan, Y.; Jones, M.; Scholes, G. D. Temperature and solvent dependence of CdSe/CdTe heterostructure nanorod spectra. J. Chem. Phys. 2009, 131, No. 084714. (30) Hangarter, C. M.; Debnath, R.; Ha, J. Y.; Sahiner, M. A.; Reehil, C. J.; Manners, W. A.; Josell, D. Photocurrent Mapping of 3D CdSe/ CdTe Windowless Solar Cells. ACS Appl. Mater. Interfaces 2013, 5, 9120−9127. (31) Lek, J. Y.; Xing, G.; Sum, T. C.; Lam, Y. M. Electron Transport Limitation in P3HT:CdSe Nanorods Hybrid Solar Cells. ACS Appl. Mater. Interfaces 2014, 6, 894−902.

(32) Peng, X.; Manna, L.; Yang, W.; Wickham, J.; Scher, E.; Kadavanich, A.; Alivisatos, A. P. Shape control of CdSe nanocrystals. Nature 2000, 404, 59−61. (33) Nair, P. V.; Thomas, K. G. Hydrazine-Induced RoomTemperature Transformation of CdTe Nanoparticles to Nanowires. J. Phys. Chem. Lett. 2010, 1, 2094−2098. (34) Pop, F.; Riobé, F.; Seifert, S.; Cauchy, T.; Ding, J.; Dupont, N.; Hauser, A.; Koch, M.; Avarvari, N. Tetrathiafulvalene-1,3,5-triazines as (Multi)Donor−Acceptor Systems with Tunable Charge Transfer: Structural, Photophysical, and Theoretical Investigations. Inorg. Chem. 2013, 52, 5023−5034. (35) Moon, G. D.; Ko, S.; Xia, Y.; Jeong, U. Chemical Transformations in Ultrathin Chalcogenide Nanowires. ACS Nano 2010, 4, 2307−2319. (36) Smith, A. M.; Mohs, A. M.; Nie, S. Tuning the optical and electronic properties of colloidal nanocrystals by lattice strain. Nat. Nanotechnol. 2009, 4, 56−63. (37) McDaniel, H.; Zuo, J.-M.; Shim, M. Anisotropic Strain-Induced Curvature in Type-II CdSe/CdTe Nanorod Heterostructures. J. Am. Chem. Soc. 2010, 132, 3286−3288. (38) McDaniel, H.; Pelton, M.; Oh, N.; Shim, M. Effects of Lattice Strain and Band Offset on Electron Transfer Rates in Type-II Nanorod Heterostructures. J. Phys. Chem. Lett. 2012, 3, 1094−1098. (39) Burda, C.; Green, T. C.; Link, S.; El-Sayed, M. A. Electron Shuttling Across the Interface of CdSe Nanoparticles Monitored by Femtosecond Laser Spectroscopy. J. Phys. Chem. B 1999, 103, 1783− 1788. (40) Cassette, E.; Pedetti, S.; Mahler, B.; Ithurria, S.; Dubertret, B.; Scholes, G. D. Ultrafast exciton dynamics in 2D in-plane heteronanostructures: delocalization and charge transfer. Phys. Chem. Chem. Phys. 2017, 19, 8373−8379.

5158

DOI: 10.1021/acsomega.7b00995 ACS Omega 2017, 2, 5150−5158