Exploring Ultrafast Electronic Processes of Quasi-Type II Nanocrystals ...
Exploring Ultrafast Electronic Processes of Quasi-Type II Nanocrystals by Two-Dimensional Electronic Spectroscopy Supporting Information Yoichi Kobayashi†, Chi-Hung Chuang‡, Clemens Burda‡ and Gregory D. Scholes*,† †
Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario, M5S 3H6, Canada
‡
Center for Chemical Dynamics and Nanomaterials Research, Department of Chemistry, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio 44106, United States
*
Corresponding author. E-mail: [email protected]
1
1. Laser pulse spectrum measured by Frequency-resolved optical grating (FROG) All pulse spectra used in the experiments were measured by FROG technique by using methanol and the same setup wit the 2D experiments. The laser pulse spectra for each experiment are shown in Figure S1 a and b. The pulse duration for the experiments with CdTe NCs and CdTe/CdSe NCs were ascertained to be 12.4 and 11.9 fs, respectively.
(a)
(b)
Figure S1. Frequency-resolved optical gating (FROG) surfaces characterizing the pulse of each experiment for a) CdTe NCs and for b) CdTe/CdSe NCs.
2
2. Phasing 2D spectra of CdTe NCs with pump-probe spectra The projections of the 2D absorptive spectra onto detection energy axis along with pump-probe spectra are shown in Figure S2. Figure S2 a and b demonstrates that the projected 2D spectra match well with the pump probe spectra at different t2 time. The phase constant obtained from the fitting shifted very slightly with t2 time due to the experimental condition. We used the average value of the phase constant at different t2 times; Φ = (0.99± 0.02) ×π for phasing the 2D spectra.
Figure S2. The pump-probe spectrum and the projected 2D spectrum onto the detection axis of CdTe NCs at a) 60 fs and at b) 200 fs. c) The phase constant obtained from the phasing as a function of t2 time.
3
3. Phasing 2D spectra of CdTe/CdSe NCs with pump-probe spectra The projections of the 2D absorptive spectra onto detection energy axis along with pump-probe spectra are shown Figure S3a and b. In the case of CdTe/CdSe NCs, we used cresyl violet perchlorate for the phasing because the phasing with the 2D data of CdTe/CdSe NCs is strongly modulated by the stimulated Raman modes of the solvents. The phase constant obtained by the phasing are shown in the figure S3c. We used the average value of the phase constant at different t2 times; Φ = (0.78± 0.02) ×π for phasing the 2D spectra.
Figure S3. The pump-probe spectrum and the projected 2D spectrum onto the detection axis of cresyl violet perchlorate at a) 60 fs and at b) 200 fs. c) The phase constant obtained from the phasing as a function of t2 time.
4
4. The real and imaginary parts of rephasing, nonrephasing and magnitude spectra Figure S4 and S5 show the real and imaginary parts of the rephasing, the nonrephasing, and the total magnitude spectra of CdTe NCs and CdTe/CdSe NCs at 360 fs, respectively. Both spectra are similar irrespective of the shell foromation. The shapes of the rephasing and nonrephasing spectra are similar to the simulated 2D spectra of two states reported earlier,1 which indicates that the phasing of spectra are qualitatively correct.
Figure S4. Real, imaginary and magnitude spectra of rephasing, nonrephasing spectra (a-f) and the absorptive spectrum of CdTe NCs dissolved in hexane (g) at t2 = 360 fs.
Figure S5. Real, imaginary and magnitude spectra of rephasing, nonrephasing spectra (a-f) and the absorptive spectrum of CdTe/CdSe NCs (g) at t2 = 360 fs. 5
5. Steady-state and 2D spectra of CdTe/CdSe NCs2 (larger-core and smaller-shells) Figure S6a shows steady-state absorption (solid line), emission spectra (dashed line) and laser spectra (gray) of large-core/thinner-shell CdTe/CdSe NCs (CdTe/CdSe NCs2). The rephrasing, nonrephasing, magunitude, and absorptive spectra are shown in Figure S6b-h. 2D spectra are very similar to those of CdTe/CdSe NCs irrespective of the shell thickness.
Figure S6. A) Steady-state absorption spectrum and the emission spectrum of CdTe/CdSe NCs2 (orange solid line and dashed line, respectively), and the pulse spectrum (gray). 2D absorptive spectrum (b) and real, imaginary, and the magnitude spectra of the rephrasing and the rephrasing spectra (Figure S4c-h).
6
6. Steady-state and 2D spectra of CdTe/CdSe NCs2 Figure S7 show steady-state absorption and emission spectra (a), 2D absorptive spectrum (b), the real parts and imaginary parts of the rephrasing and the nonrephasing spectra (c-h) of CdTe/ZnS NCs. We observe a large rectangular signal, which did not distinguish the two signals like CdTe NCs. The rectangular spectrum indicates that there are two diagonal peaks and two cross peaks, although they are too broad to resolve. It is due to the larger inhomogeneous broadening of CdTe/ZnS NCs. ESAs are not observed in CdTe/ZnS most probably due to the inhomogeneous broadening and the laser spectrum, which may not fully cover the 2S peak.
Figure S7. A) Steady-state absorption and the emission spectra of CdTe/ZnS NCs (orange solid line and dashed line, respectively), and the pulse spectrum (gray). 2D absorptive spectrum (b) and real, imaginary, and the magnitude spectra of the rephrasing and the rephrasing spectra (Figure S4c-h). 7
7. Solvent response Figure S8 show the rephrasing and the nonrephasing spectra, the t2 dynamics, and the Fourier transformed spectra of toluene. Figure S8a and b show that toluene itself also have a diagonal and a cross-peak 2D signals. The signals clearly oscillates with t2 time and the oscillations are found to be ~800, 1030, and 1200 cm-1. The lowest frequency around 100 cm-1 may not be reliable since our time range is limited to 400 fs. These high frequencies are very similar to those of Raman modes2 and the energy separation between the diagonal peak and the cross peak in the detection axis is similar to these frequency as well. These results indicate that oscillations as a function of t2 time are due to the impulsive Raman modes.
Figure S8. The rephrasing (a) and nonrephasing spectra (d) of toluene at 100 fs. The t2 dynamics of the rephrasing and the nonrephasing spectra probed at the position A and B (b and e), and the Fourier transformed spectra of the t2 dynamics after subtracting the decay components (c and f).
8
8. Analysis of the negative absorption in Figure 2b by using biexciton shift model We analyzed the band-edge negative signal of CdTe/CdSe NCs in Figure 2b assuming that the negative signal only comes from the biexciton shift without any change of the spectral width. Firstly, we integrated 2D spectrum along with the detection energy from 1.86 to 1.98 eV and obtained the 1D spectrum as a function of the excitation energy. Secondly, we fitted the spectrum a t2 = 60 fs with multiple Gaussian functions as below:
2Dspec 60 fs
√2
2
FWHM 2√2&'2 ∙
√2
! 2
where, hνi, σi, and FWHMi are each peak energy, standard deviation, and full width half maximum, respectively. A and ∆ are the amplitude coefficient and spectral shift, respectively. The fitting (Figure S9) shows the spectral shift is ~50 meV. The value is larger as compared to the reported values of similar structures, and therefore other spectral components might be incorporated to the signal.
Figure S9. Fitting result of the integrated 2D spectrum with the biexciton shift model.
9
9. References
(1)
Turner, D. B.; Hassan, Y.; Scholes, G. D. Exciton Superposition States in CdSe Nanocrystals Measured Using Broadband Two-Dimensional Electronic Spectroscopy. Nano Lett. 2012, 12, 880–886.
(2)
Wilmshurst, J. K.; Bernstein, H. J. The Infrared and Raman Spectra of Toluene, Toluene-α-d3, mXylene, and m-Xylene-αα’-d61. Can. J. Chem. 1957, 35, 911–925.
10
Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario, M5S 3H6, Canada
‡
Center for Chemical Dynamics and Nanomaterials Research, Department of Chemistry, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio 44106, United States
*
Corresponding author. E-mail: [email protected]
1
1. Laser pulse spectrum measured by Frequency-resolved optical grating (FROG) All pulse spectra used in the experiments were measured by FROG technique by using methanol and the same setup wit the 2D experiments. The laser pulse spectra for each experiment are shown in Figure S1 a and b. The pulse duration for the experiments with CdTe NCs and CdTe/CdSe NCs were ascertained to be 12.4 and 11.9 fs, respectively.
(a)
(b)
Figure S1. Frequency-resolved optical gating (FROG) surfaces characterizing the pulse of each experiment for a) CdTe NCs and for b) CdTe/CdSe NCs.
2
2. Phasing 2D spectra of CdTe NCs with pump-probe spectra The projections of the 2D absorptive spectra onto detection energy axis along with pump-probe spectra are shown in Figure S2. Figure S2 a and b demonstrates that the projected 2D spectra match well with the pump probe spectra at different t2 time. The phase constant obtained from the fitting shifted very slightly with t2 time due to the experimental condition. We used the average value of the phase constant at different t2 times; Φ = (0.99± 0.02) ×π for phasing the 2D spectra.
Figure S2. The pump-probe spectrum and the projected 2D spectrum onto the detection axis of CdTe NCs at a) 60 fs and at b) 200 fs. c) The phase constant obtained from the phasing as a function of t2 time.
3
3. Phasing 2D spectra of CdTe/CdSe NCs with pump-probe spectra The projections of the 2D absorptive spectra onto detection energy axis along with pump-probe spectra are shown Figure S3a and b. In the case of CdTe/CdSe NCs, we used cresyl violet perchlorate for the phasing because the phasing with the 2D data of CdTe/CdSe NCs is strongly modulated by the stimulated Raman modes of the solvents. The phase constant obtained by the phasing are shown in the figure S3c. We used the average value of the phase constant at different t2 times; Φ = (0.78± 0.02) ×π for phasing the 2D spectra.
Figure S3. The pump-probe spectrum and the projected 2D spectrum onto the detection axis of cresyl violet perchlorate at a) 60 fs and at b) 200 fs. c) The phase constant obtained from the phasing as a function of t2 time.
4
4. The real and imaginary parts of rephasing, nonrephasing and magnitude spectra Figure S4 and S5 show the real and imaginary parts of the rephasing, the nonrephasing, and the total magnitude spectra of CdTe NCs and CdTe/CdSe NCs at 360 fs, respectively. Both spectra are similar irrespective of the shell foromation. The shapes of the rephasing and nonrephasing spectra are similar to the simulated 2D spectra of two states reported earlier,1 which indicates that the phasing of spectra are qualitatively correct.
Figure S4. Real, imaginary and magnitude spectra of rephasing, nonrephasing spectra (a-f) and the absorptive spectrum of CdTe NCs dissolved in hexane (g) at t2 = 360 fs.
Figure S5. Real, imaginary and magnitude spectra of rephasing, nonrephasing spectra (a-f) and the absorptive spectrum of CdTe/CdSe NCs (g) at t2 = 360 fs. 5
5. Steady-state and 2D spectra of CdTe/CdSe NCs2 (larger-core and smaller-shells) Figure S6a shows steady-state absorption (solid line), emission spectra (dashed line) and laser spectra (gray) of large-core/thinner-shell CdTe/CdSe NCs (CdTe/CdSe NCs2). The rephrasing, nonrephasing, magunitude, and absorptive spectra are shown in Figure S6b-h. 2D spectra are very similar to those of CdTe/CdSe NCs irrespective of the shell thickness.
Figure S6. A) Steady-state absorption spectrum and the emission spectrum of CdTe/CdSe NCs2 (orange solid line and dashed line, respectively), and the pulse spectrum (gray). 2D absorptive spectrum (b) and real, imaginary, and the magnitude spectra of the rephrasing and the rephrasing spectra (Figure S4c-h).
6
6. Steady-state and 2D spectra of CdTe/CdSe NCs2 Figure S7 show steady-state absorption and emission spectra (a), 2D absorptive spectrum (b), the real parts and imaginary parts of the rephrasing and the nonrephasing spectra (c-h) of CdTe/ZnS NCs. We observe a large rectangular signal, which did not distinguish the two signals like CdTe NCs. The rectangular spectrum indicates that there are two diagonal peaks and two cross peaks, although they are too broad to resolve. It is due to the larger inhomogeneous broadening of CdTe/ZnS NCs. ESAs are not observed in CdTe/ZnS most probably due to the inhomogeneous broadening and the laser spectrum, which may not fully cover the 2S peak.
Figure S7. A) Steady-state absorption and the emission spectra of CdTe/ZnS NCs (orange solid line and dashed line, respectively), and the pulse spectrum (gray). 2D absorptive spectrum (b) and real, imaginary, and the magnitude spectra of the rephrasing and the rephrasing spectra (Figure S4c-h). 7
7. Solvent response Figure S8 show the rephrasing and the nonrephasing spectra, the t2 dynamics, and the Fourier transformed spectra of toluene. Figure S8a and b show that toluene itself also have a diagonal and a cross-peak 2D signals. The signals clearly oscillates with t2 time and the oscillations are found to be ~800, 1030, and 1200 cm-1. The lowest frequency around 100 cm-1 may not be reliable since our time range is limited to 400 fs. These high frequencies are very similar to those of Raman modes2 and the energy separation between the diagonal peak and the cross peak in the detection axis is similar to these frequency as well. These results indicate that oscillations as a function of t2 time are due to the impulsive Raman modes.
Figure S8. The rephrasing (a) and nonrephasing spectra (d) of toluene at 100 fs. The t2 dynamics of the rephrasing and the nonrephasing spectra probed at the position A and B (b and e), and the Fourier transformed spectra of the t2 dynamics after subtracting the decay components (c and f).
8
8. Analysis of the negative absorption in Figure 2b by using biexciton shift model We analyzed the band-edge negative signal of CdTe/CdSe NCs in Figure 2b assuming that the negative signal only comes from the biexciton shift without any change of the spectral width. Firstly, we integrated 2D spectrum along with the detection energy from 1.86 to 1.98 eV and obtained the 1D spectrum as a function of the excitation energy. Secondly, we fitted the spectrum a t2 = 60 fs with multiple Gaussian functions as below:
2Dspec 60 fs
√2
2
FWHM 2√2&'2 ∙
√2
! 2
where, hνi, σi, and FWHMi are each peak energy, standard deviation, and full width half maximum, respectively. A and ∆ are the amplitude coefficient and spectral shift, respectively. The fitting (Figure S9) shows the spectral shift is ~50 meV. The value is larger as compared to the reported values of similar structures, and therefore other spectral components might be incorporated to the signal.
Figure S9. Fitting result of the integrated 2D spectrum with the biexciton shift model.
9
9. References
(1)
Turner, D. B.; Hassan, Y.; Scholes, G. D. Exciton Superposition States in CdSe Nanocrystals Measured Using Broadband Two-Dimensional Electronic Spectroscopy. Nano Lett. 2012, 12, 880–886.
(2)
Wilmshurst, J. K.; Bernstein, H. J. The Infrared and Raman Spectra of Toluene, Toluene-α-d3, mXylene, and m-Xylene-αα’-d61. Can. J. Chem. 1957, 35, 911–925.
10
