Non-Radiative Energy Transfer from Isolated CdSe/ZnS Quantum Dots ...
Supporting Information for
Non-Radiative Energy Transfer from Isolated CdSe/ZnS Quantum Dots to Single- and Few-Layer Tin Disulfide
Huidong Zang,1 Prahlad K. Routh,1,2 Yuan Huang,1 Jia-Shiang Chen,1,2 Eli Sutter,3 Peter Sutter4* and Mircea Cotlet1,2,* 1
Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton NY 11973,
USA, 2Materials Science Department, Stony Brook University, Stony Brook, NY 11794, USA 3
Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln NE 68588, USA, 4Department of Electrical and Computer Engineering Department, University of Nebraska-Lincoln, Lincoln NE 68588, USA
1. Calculation of energy transfer rate for Qdot-SnS2 hybrids. The calculation of energy transfer rate from quantum dot (Qdot) to layered SnS2 was done according to references 1-5 𝑘"# (𝑧) =
(
123 4
0 5 5 ()*+ℏ- ./ 6 277
*
∆𝐸 *
>∆?@ B
A = ( G ℏC/ 𝑑𝑡𝑒 HIF F
DE (=D >
(S1)
Here, 𝑘"# is the energy transfer rate, ℏ the reduced Planck constant, and 𝜈K is the Fermi velocity of SnS2. In the current work, we use 𝜈K = 1.3256×106 m/s for SnS2.6 𝜇4N is the transition dipole moment of the Qdot under excitation, ∆𝐸 is the energy of the photon emitted by the Qdot (2.36eV), 𝑧H is the distance from the center of Qdot to the ith SnS2 layer, and 𝜀F and 𝜀4PP denote the vacuum permittivity and effective dielectric constant, respectively. The 𝜀4PP is obtained by 𝜀4PP =
5QRS2 T*5 U
(S2)
where 𝜀VWX4 is the dielectric constant of CdSe (6.23)7 and 𝜀 is the dielectric constant of the ligands with a value of 2.67.8 The thickness of a single layer of SnS2 is around 0.6 nm, and the radius of the CdSe/ZnS Qdots is 3.36 nm, as determined by TEM (See section 2, SI). We 1
estimated the center-to-center distance between the Qdot and single layer SnS2 assuming three components: the radius of the Qdot, the length of the ligands (octadecylamine), and the thickness of a single layer SnS2. We did not consider electronic interactions between stacked SnS2 layers such that for few layer SnS2 each layer is considered an additional decay channel and the total rate is a sum of rates of energy transfer of the Qdot with single layers at different distances from the nanocrystal. 2. CdSe/ZnS Qdot size distribution. The size of the CdSe/ZnS QDs was measured by TEM as shown in Figure S1a. The histogram of the radius distribution was analyzed using the ImageJ software package. The average radius is around 3.36 nm, obtained by a Gaussian fit of the histogram in Figure S1b. (a)
(b) 80
Count
60 40 20 0
3
4
5 6 Radius (nm)
7
8
Fig.S1. Transmission electron microscopy (left) and size distribution (right) of core/shell CdSe/ZnS QDs together with a Gauss fit (blue) yielding an average size for the QDs of 3.36nm. 3.
Time-resolved
confocal
photoluminescence
microscopy.
Time-resolved confocal
photoluminescence microscopy measurements were performed on a home-built scanning stage inverted microscope (Olympus IX81, 0.95 NA 100× air objective) coupled to a 440nm pulsed (90 ps FWHM) diode-pumped solid-state laser system (LHD-440 Picoquant) operated at 10 MHz repetition rate. For SnS2-only flake imaging the average power at the sample was kept at about 10µW@10MHz repetition rate. For single nanocrystals studies, this power was reduced by 100x. Photoluminescence was collected in epi-illumination scheme, spectrally separated from the excitation laser light by a dichroic (Semrock, DiO-442) and by a band-pass filter (Semrock FF01-488LP), spatially filtered by a 100 µm pinhole and then imaged onto a single photon counting avalanche photodiode (MPD Picoquant) coupled to a time-analyzer (PicoHarp 300, PicoQuant). Data acquisition and data analysis were performed with the Symphotime 5.32 2
analysis software (Picoquant) and with the BIFL Data Analyzer software (Scientific Software Technologies Center, Belarus). 4. Calculation of P(ton) and P(toff) probability densities . We calculated the probability densities Pon(t) and Poff(t) acording to9-11 where we used a threshold equal to the background signal in single particle PL intensity traces binned at 10 ms dwell time plus three times the standard deviation to separate between on and off events. Probablities were calculated according to 𝑃 𝑡H =
ZA ZA,B\B]^
×
( `D]a3
; 𝑖 = 𝑜𝑛, 𝑜𝑓𝑓
(S3)
with Ni being the number of on- (off-) events with duration time t, Ni,total being the total number of on- (off-) events, and Δtavg being the average duration time between nearest neighbour event. Each probability included blinking data from around 10 isolated QDs, each probed for 60 seconds. 5. Calculation of PL lifetimes and PL lifetime distribution heterogeneity. PL lifetimes constituting the correlograms in Figure 3a-d were estimated by Maximum Likelihood Estimation (MLE) method from PL decays constructed by time sliding (1.5 seconds/decay) over the measured photon stream recorded from a Qdot, with photons spread over 128 channels in a PL decay. PL lifetimes constituting the histrograms in Figure 4a-d were estimated also by the MLE method, this time from PL decays constructed by constant photon number sliding (2000 total photons/decay) over the measured photon stream recorded from a Qdot, with photons spread over 128 channels. Distributions in Figure 4a-d were fitted with Gaussian function using Origin Pro 8.5 to obtain peak position and FWHM. The standard error of means obtained from Gaussian fitting was converted to 95% confidence interval half width and it was used as fitting error for PL lifetime and FWHM. The expected statistical error in determination of PL lifetime using 2000 photons are calculated from the methods described by Maus et al.12 *
𝜎 =
h0 i > Z# >
−
𝑇 𝜏
(1 − 𝑒 )
𝑒n/pq (=𝑒rn/q 𝑒n/pq =(
>
−
i>
=(
(S3)
𝑒n/q =(
3
These expected deviations are much smaller compared to observed spread in PL lifetime values as represented using FWHM of the distribution. The difference in these values are attributed to static heterogeneity and plotted separately in the inset of Figure 4b. Table S1: Gaussian fit parameters of PL lifetime distributions and heterogeneity calculation
PL lifetime
FWHM
Expected FWHM
Heterogeneity
(as calculated) QD only
27 ns ± 0.7
17.5ns ± 1.65
1.77 ns
15.73 ns ± 1.65
1 layer SnS2
17 ns ± 0.39
10.2ns ± 0.92
0.94 ns
9.26 ns ± 0.92
3 layer SnS2
11.6 ns ±0.59
8.7ns ± 1.39
0.61 ns
8.09 ns ± 1.39
6 layer SnS2
9.5 ns ± 0.24
5.9ns ± 0.59
0.49 ns
5.41 ns ± 0.59
6. Calculation of bulk ET Rate between SnS2 layers and QDs Lifetime. A sub-monolayer of CdSe/ZnS Qdots was deposited on multiple flakes using Qdots with concentration of 0.01mg/mL in mixed solvent via spin coating on SnS2/SiO2 /Si flakes at 1500 rpm for 60 seconds. Timeresolved confocal PL microscopy measurements were performed using the same setup as described earlier. FLIM images were recorded with 3ms per pixel (400x400 pixels) and 4x4 pixel binning was carried out during analysis to obtain more than 100 photons per pixel. Point measurements were carried out on selected areas of flakes for at least 30 seconds to record PL decays. These histograms were then fitted with bi-exponential models to obtain amplitude averaged lifetime of bulk Qdots on SnS2. flakes. Multiple points and regions of interest were probed to obtain a distribution of bulk PL lifetime from three different SnS2 flakes. PL lifetimes of Qdots on the SiO2 /Si in the vicinity of each SnS2 flake were used to calculate the energy transfer rate according to eq.(3) main text. These PL lifetimes with corresponding number of SnS2 layers obtained from three different flakes were combined and averaged to obtain bulk energy transfer rates and standard deviations (represented as error bar in the plot of the bulk ET w
rate vs number of SnS2 layers). A stretched exponential function 𝑦 = 𝐴𝑒 =v with z the number of SnS2 layers was used to fit the dependency of bulk ET rate vs number of layers of SnS2. (Fit coefficients: A= 7.56 ± 1.36 and 𝛽 =-0.44)
4
Figure S2: Optical contrast image (left) and confocal FLIM image (right) of flake #1
Figure S3: Optical contrast image (left) and confocal FLIM image (right) of flake #2
Figure S4: Optical contrast image (left) and confocal FLIM image (right) of flake #1 5
KET (107s-1)
6 5 4 3 2
0
1
2
3
4
5
6
7
# of SnS2 Layers
7+
Figure S5: Bulk energy transfer rate vs number of SnS2 layers for Qdot-SnS2 hybrids estimated from flakes shown in Figs.S2-S4 REFERENCES: 1. Chen, Z.; Berciaud, S.; Nuckolls, C.; Heinz, T. F.; Brus, L. E. Energy Transfer from Individual Semiconductor Nanocrystals to Graphene. Acs Nano 2010, 4, 2964-‐2968. 2. Swathi, R. S.; Sebastian, K. L. Long Range Resonance Energy Transfer from a Dye Molecule to Graphene Has (Distance)(-‐4) Dependence. J Chem Phys 2009, 130. 3. Swathi, R. S.; Sebastian, K. L. Resonance Energy Transfer from a Dye Molecule to Graphene. J Chem Phys 2008, 129. 4. Califano, M.; Franceschetti, A.; Zunger, A. Temperature Dependence of Excitonic Radiative Decay in Cdse Quantum Dots: The Role of Surface Hole Traps. Nano Letters 2005, 5, 2360-‐2364. 5. Baer, R.; Rabani, E. Theory of Resonance Energy Transfer Involving Nanocrystals: The Role of High Multipoles. J Chem Phys 2008, 128. 6. Lorenz, T.; Joswig, J.-‐O.; Seifert, G. Combined Sns@ Sns2 Double Layers: Charge Transfer and Electronic Structure. Semiconductor Science and Technology 2014, 29, 064006. 7. Baskoutas, S.; Terzis, A. F. Size-‐Dependent Band Gap of Colloidal Quantum Dots. J Appl Phys 2006, 99, 013708. 8. Baysinger, G. Crc Handbook of Chemistry and Physics. National Institute of Standards and Technology, 2015. 9. Xu, Z., Cotlet, M. Photoluminescence Blinking Dynamics of Colloidal Quantum Dots in the Presence of Controlled External Electron Traps. Small 2012, 8, 253-‐258. 10. Kuno, M.; Fromm, D. P.; Hamann, H. F.; Gallagher, A.; Nesbitt, D. J. "On"/"Off" Fluorescence Intermittency of Single Semiconductor Quantum Dots. J Chem Phys 2001, 115, 1028-‐1040. 11. Song, N. H.; Zhu, H. M.; Jin, S. Y.; Lian, T. Q. Hole Transfer from Single Quantum Dots. Acs Nano 2011, 5, 8750-‐8759. 12. Maus, M.; Cotlet, M.; Hofkens, J.; Gensch, T.; De Schryver, F. C.; Schaffer, J.; Seidel, C. An Experimental Comparison of the Maximum Likelihood Estimation and Nonlinear Least-‐Squares Fluorescence Lifetime Analysis of Single Molecules. Analytical chemistry 2001, 73, 2078-‐2086.
6
Non-Radiative Energy Transfer from Isolated CdSe/ZnS Quantum Dots to Single- and Few-Layer Tin Disulfide
Huidong Zang,1 Prahlad K. Routh,1,2 Yuan Huang,1 Jia-Shiang Chen,1,2 Eli Sutter,3 Peter Sutter4* and Mircea Cotlet1,2,* 1
Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton NY 11973,
USA, 2Materials Science Department, Stony Brook University, Stony Brook, NY 11794, USA 3
Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln NE 68588, USA, 4Department of Electrical and Computer Engineering Department, University of Nebraska-Lincoln, Lincoln NE 68588, USA
1. Calculation of energy transfer rate for Qdot-SnS2 hybrids. The calculation of energy transfer rate from quantum dot (Qdot) to layered SnS2 was done according to references 1-5 𝑘"# (𝑧) =
(
123 4
0 5 5 ()*+ℏ- ./ 6 277
*
∆𝐸 *
>∆?@ B
A = ( G ℏC/ 𝑑𝑡𝑒 HIF F
DE (=D >
(S1)
Here, 𝑘"# is the energy transfer rate, ℏ the reduced Planck constant, and 𝜈K is the Fermi velocity of SnS2. In the current work, we use 𝜈K = 1.3256×106 m/s for SnS2.6 𝜇4N is the transition dipole moment of the Qdot under excitation, ∆𝐸 is the energy of the photon emitted by the Qdot (2.36eV), 𝑧H is the distance from the center of Qdot to the ith SnS2 layer, and 𝜀F and 𝜀4PP denote the vacuum permittivity and effective dielectric constant, respectively. The 𝜀4PP is obtained by 𝜀4PP =
5QRS2 T*5 U
(S2)
where 𝜀VWX4 is the dielectric constant of CdSe (6.23)7 and 𝜀 is the dielectric constant of the ligands with a value of 2.67.8 The thickness of a single layer of SnS2 is around 0.6 nm, and the radius of the CdSe/ZnS Qdots is 3.36 nm, as determined by TEM (See section 2, SI). We 1
estimated the center-to-center distance between the Qdot and single layer SnS2 assuming three components: the radius of the Qdot, the length of the ligands (octadecylamine), and the thickness of a single layer SnS2. We did not consider electronic interactions between stacked SnS2 layers such that for few layer SnS2 each layer is considered an additional decay channel and the total rate is a sum of rates of energy transfer of the Qdot with single layers at different distances from the nanocrystal. 2. CdSe/ZnS Qdot size distribution. The size of the CdSe/ZnS QDs was measured by TEM as shown in Figure S1a. The histogram of the radius distribution was analyzed using the ImageJ software package. The average radius is around 3.36 nm, obtained by a Gaussian fit of the histogram in Figure S1b. (a)
(b) 80
Count
60 40 20 0
3
4
5 6 Radius (nm)
7
8
Fig.S1. Transmission electron microscopy (left) and size distribution (right) of core/shell CdSe/ZnS QDs together with a Gauss fit (blue) yielding an average size for the QDs of 3.36nm. 3.
Time-resolved
confocal
photoluminescence
microscopy.
Time-resolved confocal
photoluminescence microscopy measurements were performed on a home-built scanning stage inverted microscope (Olympus IX81, 0.95 NA 100× air objective) coupled to a 440nm pulsed (90 ps FWHM) diode-pumped solid-state laser system (LHD-440 Picoquant) operated at 10 MHz repetition rate. For SnS2-only flake imaging the average power at the sample was kept at about 10µW@10MHz repetition rate. For single nanocrystals studies, this power was reduced by 100x. Photoluminescence was collected in epi-illumination scheme, spectrally separated from the excitation laser light by a dichroic (Semrock, DiO-442) and by a band-pass filter (Semrock FF01-488LP), spatially filtered by a 100 µm pinhole and then imaged onto a single photon counting avalanche photodiode (MPD Picoquant) coupled to a time-analyzer (PicoHarp 300, PicoQuant). Data acquisition and data analysis were performed with the Symphotime 5.32 2
analysis software (Picoquant) and with the BIFL Data Analyzer software (Scientific Software Technologies Center, Belarus). 4. Calculation of P(ton) and P(toff) probability densities . We calculated the probability densities Pon(t) and Poff(t) acording to9-11 where we used a threshold equal to the background signal in single particle PL intensity traces binned at 10 ms dwell time plus three times the standard deviation to separate between on and off events. Probablities were calculated according to 𝑃 𝑡H =
ZA ZA,B\B]^
×
( `D]a3
; 𝑖 = 𝑜𝑛, 𝑜𝑓𝑓
(S3)
with Ni being the number of on- (off-) events with duration time t, Ni,total being the total number of on- (off-) events, and Δtavg being the average duration time between nearest neighbour event. Each probability included blinking data from around 10 isolated QDs, each probed for 60 seconds. 5. Calculation of PL lifetimes and PL lifetime distribution heterogeneity. PL lifetimes constituting the correlograms in Figure 3a-d were estimated by Maximum Likelihood Estimation (MLE) method from PL decays constructed by time sliding (1.5 seconds/decay) over the measured photon stream recorded from a Qdot, with photons spread over 128 channels in a PL decay. PL lifetimes constituting the histrograms in Figure 4a-d were estimated also by the MLE method, this time from PL decays constructed by constant photon number sliding (2000 total photons/decay) over the measured photon stream recorded from a Qdot, with photons spread over 128 channels. Distributions in Figure 4a-d were fitted with Gaussian function using Origin Pro 8.5 to obtain peak position and FWHM. The standard error of means obtained from Gaussian fitting was converted to 95% confidence interval half width and it was used as fitting error for PL lifetime and FWHM. The expected statistical error in determination of PL lifetime using 2000 photons are calculated from the methods described by Maus et al.12 *
𝜎 =
h0 i > Z# >
−
𝑇 𝜏
(1 − 𝑒 )
𝑒n/pq (=𝑒rn/q 𝑒n/pq =(
>
−
i>
=(
(S3)
𝑒n/q =(
3
These expected deviations are much smaller compared to observed spread in PL lifetime values as represented using FWHM of the distribution. The difference in these values are attributed to static heterogeneity and plotted separately in the inset of Figure 4b. Table S1: Gaussian fit parameters of PL lifetime distributions and heterogeneity calculation
PL lifetime
FWHM
Expected FWHM
Heterogeneity
(as calculated) QD only
27 ns ± 0.7
17.5ns ± 1.65
1.77 ns
15.73 ns ± 1.65
1 layer SnS2
17 ns ± 0.39
10.2ns ± 0.92
0.94 ns
9.26 ns ± 0.92
3 layer SnS2
11.6 ns ±0.59
8.7ns ± 1.39
0.61 ns
8.09 ns ± 1.39
6 layer SnS2
9.5 ns ± 0.24
5.9ns ± 0.59
0.49 ns
5.41 ns ± 0.59
6. Calculation of bulk ET Rate between SnS2 layers and QDs Lifetime. A sub-monolayer of CdSe/ZnS Qdots was deposited on multiple flakes using Qdots with concentration of 0.01mg/mL in mixed solvent via spin coating on SnS2/SiO2 /Si flakes at 1500 rpm for 60 seconds. Timeresolved confocal PL microscopy measurements were performed using the same setup as described earlier. FLIM images were recorded with 3ms per pixel (400x400 pixels) and 4x4 pixel binning was carried out during analysis to obtain more than 100 photons per pixel. Point measurements were carried out on selected areas of flakes for at least 30 seconds to record PL decays. These histograms were then fitted with bi-exponential models to obtain amplitude averaged lifetime of bulk Qdots on SnS2. flakes. Multiple points and regions of interest were probed to obtain a distribution of bulk PL lifetime from three different SnS2 flakes. PL lifetimes of Qdots on the SiO2 /Si in the vicinity of each SnS2 flake were used to calculate the energy transfer rate according to eq.(3) main text. These PL lifetimes with corresponding number of SnS2 layers obtained from three different flakes were combined and averaged to obtain bulk energy transfer rates and standard deviations (represented as error bar in the plot of the bulk ET w
rate vs number of SnS2 layers). A stretched exponential function 𝑦 = 𝐴𝑒 =v with z the number of SnS2 layers was used to fit the dependency of bulk ET rate vs number of layers of SnS2. (Fit coefficients: A= 7.56 ± 1.36 and 𝛽 =-0.44)
4
Figure S2: Optical contrast image (left) and confocal FLIM image (right) of flake #1
Figure S3: Optical contrast image (left) and confocal FLIM image (right) of flake #2
Figure S4: Optical contrast image (left) and confocal FLIM image (right) of flake #1 5
KET (107s-1)
6 5 4 3 2
0
1
2
3
4
5
6
7
# of SnS2 Layers
7+
Figure S5: Bulk energy transfer rate vs number of SnS2 layers for Qdot-SnS2 hybrids estimated from flakes shown in Figs.S2-S4 REFERENCES: 1. Chen, Z.; Berciaud, S.; Nuckolls, C.; Heinz, T. F.; Brus, L. E. Energy Transfer from Individual Semiconductor Nanocrystals to Graphene. Acs Nano 2010, 4, 2964-‐2968. 2. Swathi, R. S.; Sebastian, K. L. Long Range Resonance Energy Transfer from a Dye Molecule to Graphene Has (Distance)(-‐4) Dependence. J Chem Phys 2009, 130. 3. Swathi, R. S.; Sebastian, K. L. Resonance Energy Transfer from a Dye Molecule to Graphene. J Chem Phys 2008, 129. 4. Califano, M.; Franceschetti, A.; Zunger, A. Temperature Dependence of Excitonic Radiative Decay in Cdse Quantum Dots: The Role of Surface Hole Traps. Nano Letters 2005, 5, 2360-‐2364. 5. Baer, R.; Rabani, E. Theory of Resonance Energy Transfer Involving Nanocrystals: The Role of High Multipoles. J Chem Phys 2008, 128. 6. Lorenz, T.; Joswig, J.-‐O.; Seifert, G. Combined Sns@ Sns2 Double Layers: Charge Transfer and Electronic Structure. Semiconductor Science and Technology 2014, 29, 064006. 7. Baskoutas, S.; Terzis, A. F. Size-‐Dependent Band Gap of Colloidal Quantum Dots. J Appl Phys 2006, 99, 013708. 8. Baysinger, G. Crc Handbook of Chemistry and Physics. National Institute of Standards and Technology, 2015. 9. Xu, Z., Cotlet, M. Photoluminescence Blinking Dynamics of Colloidal Quantum Dots in the Presence of Controlled External Electron Traps. Small 2012, 8, 253-‐258. 10. Kuno, M.; Fromm, D. P.; Hamann, H. F.; Gallagher, A.; Nesbitt, D. J. "On"/"Off" Fluorescence Intermittency of Single Semiconductor Quantum Dots. J Chem Phys 2001, 115, 1028-‐1040. 11. Song, N. H.; Zhu, H. M.; Jin, S. Y.; Lian, T. Q. Hole Transfer from Single Quantum Dots. Acs Nano 2011, 5, 8750-‐8759. 12. Maus, M.; Cotlet, M.; Hofkens, J.; Gensch, T.; De Schryver, F. C.; Schaffer, J.; Seidel, C. An Experimental Comparison of the Maximum Likelihood Estimation and Nonlinear Least-‐Squares Fluorescence Lifetime Analysis of Single Molecules. Analytical chemistry 2001, 73, 2078-‐2086.
6
