Supporting Information

Transcription

1 Supporting Information Highly luminescent water-dispersible NIR-emitting wurtzite CuInS 2 /ZnS core/shell colloidal quantum dots Chenghui Xia, a,b Johannes D. Meeldijk, c Hans C. Gerritsen, b and Celso de Mello Donegá a * a. Condensed Matter and Interfaces, Debye Institute for Nanomaterials Science, Utrecht University, P.O. Box 80000, 3508 TA Utrecht, The Netherlands b. Molecular Biophysics, Debye Institute for Nanomaterials Science, Utrecht University, 3508 TA Utrecht, Netherlands. c. Electron Microscopy Utrecht, Debye Institute for Nanomaterials Science, Utrecht University, 3584 CH Utrecht, Netherlands. *Corresponding author: c.demello-donega@uu.nl S1

2 Table S1. Composition of the reaction mixture used to synthesize CIS QDs by partial Cu + for In 3+ cation exchange in template Cu 2-x S NCs at 20 C (slow CE). Cu 2-x S NCs Toluene In(NO 3 ) 3 H 2 O Methanol TOP [TOP] TOP/In 1 1 ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 0 μl ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 40 μl 15 mm ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 80 μl 30 mm ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 140 μl 52 mm ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 200 μl 73 mm ml, 0.1 mmol Cu + 3 ml 0.1 mmol 2 ml 300 μl 108 mm 6.8 Supporting Method. Measurement of the PLQYs. The PLQYs of the CIS and CIS/ZnS core/shell QDs were estimated by using indocyanine green (ICG, PLQY= 12%) as a standard, according to the following equation: Gradx x = ST n x 2 Grad ST where the subscripts ST and X denote standard and sample respectively, Φ is the fluorescence quantum yield, Grad the gradient from the plot of integrated fluorescence intensity versus absorbance, and n the refractive index of the solvent (DMSO for ICG, and toluene or water for the QDs). To reduce random error, five different concentrations of ICG and QDs were measured. 2 n ST Figure S1. Absorption (black line) and PL (red line, excitation wavelength= 678 nm) spectra of ICG in anhydrous DMSO. The absorption and PL spectra of CIS or CIS/ZnS QDs dissolved in toluene (or Milli-Q water) were measured under the same conditions. S2

3 Figure S2. TEM images and corresponding size histograms of Cu 2-x S NCs obtained by controlling the reaction time from 20 to 180 min at 210 C. The upper panel shows a photograph of a series of vials containing solutions of the imaged samples. The size distribution histograms were constructed by measuring over 100 NCs per image and are fitted to Gaussian functions. The NCs are nearly spherical and grow from 3 nm to 8 nm as the reaction time increases from 20 to 160 min. Longer reaction times result in hexagonal platelets. S3

4 Figure S3. XRD pattern of 6 nm diameter Cu 2-x S NCs obtained after 90 min at 210 C. The red lines indicate the high chalcocite Cu 2 S diffraction pattern (JCPDS Card ). The XRD pattern was obtained on Bruker D2 Phaser, equipped with a Co Kα X-ray source ( Å). Figure S4. (a) absorption and (b) PL spectra of CIS QDs prepared by Cu + for In 3+ cation exchange in 6 nm diameter template Cu 2-x S NCs at 20 C using different amounts of TOP (see Table S1 above). The features in the PL spectra at 1150 nm and 1200 nm are due to absorption by vibrational overtones of the solvent toluene. (c-f) Corresponding TEM images of the product CIS QDs. The insert in panel (a) shows vials containing crude reaction mixtures with increasing TOP concentrations from left to right. S4

5 Figure S5. TEM images and size histograms of (a) template Cu 2-x S NCs and (b) product CIS QDs synthesized by partial Cu + for In 3+ CE at 20 C in the NCs shown in (a). Figure S6. XRD pattern of 6 nm diameter product CIS QDs synthesized by partial Cu + for In 3+ CE at 20 C in 6 nm template Cu 2-x S NCs. The black line indicates the wurtzite CIS diffraction pattern (JCPDS Card ), while the red line indicates the chalcopyrite CIS diffraction pattern (JCPDS Card ). The XRD pattern was obtained on Bruker D2 Phaser, equipped with a Co Kα X-ray source ( Å). S5

6 Figure S7. EDS spectra of (a) 6 nm template Cu 2-x S NCs (Cu/S ratio= 2±0.2); (b) 6 nm product CIS QDs synthesized by partial Cu + for In 3+ CE at 20 C in the NCs shown in (a). The elemental ratios are Cu:In:S= 1.4±0.2: 1±0.14: 2.6±0.5. EDS analyses were performed on wide areas (~ nm 2 ), encompassing several hundreds of NCs, and were repeated for three different areas of the TEM grid. The elemental ratios were obtained by averaging over the different measurement spots. S6

7 Figure S8. TEM images and corresponding size histograms of product CIS QDs prepared by partial Cu + for In 3+ exchange in template Cu 2-x S NCs using different indium-sources at 125 C for 1 h. The reaction conditions were the same in all cases: 4 ml ODE, 40 μl (0.9 mmol) TOP, 0.1 mmol In-salt, and 0.1 mmol Cu + (upper limit estimate) in Cu 2-x S NCs (6.5 nm diameter). Figure S9. PL spectra of product CIS QDs prepared by partial Cu + for In 3+ exchange in template Cu 2-x S NCs using different indium-sources at 125 C for 1 h. The samples are the same as shown in Figure S8 above. The features in the PL spectra at 1150 nm and 1200 nm are due to absorption by vibrational overtones of the solvent toluene. S7

8 Figure S10. XRD pattern of CIS QDs prepared by partial Cu + for In 3+ exchange in 6 nm template Cu 2-x S NCs at 125 C for 1 h using TOP-In(Ac) 3. The black line indicates the wurtzite CIS diffraction pattern (JCPDS Card ), while the red line indicates the chalcopyrite CIS diffraction pattern (JCPDS Card ). The XRD pattern was obtained on Bruker D2 Phaser, equipped with a Co Kα X-ray source ( Å). The extra peaks at 36.9 and 53.3 indicated by the green dashed line are possibly due to excess indium acetate, which was not washed out completely. Figure S11. TEM images and corresponding size histograms of product CIS QDs prepared by partial Cu + for In 3+ exchange in 6 nm Cu 2-x S NCs using TOP-In(Ac) 3 at 125 C for 1 h. (a) non-washed Cu 2-x S NCs templates; (b) washed Cu 2-x S NCs templates. S8

9 Figure S12. (a) XRD patterns of slow CE CIS QDs prior to and after ZnS shell overgrowth. (b). XRD patterns of fast CE CIS QDs prior to and after ZnS shell overgrowth. The black lines in both panels indicate the wurtzite CIS diffraction pattern (JCPDS Card ), while the green lines indicate the chalcopyrite CIS diffraction pattern (JCPDS Card ). The XRD pattern was obtained on Bruker D2 Phaser, equipped with a Co Kα X-ray source ( Å). The extra peaks are due to the crystalline impurities in the final products, probably excess Zn stearate. Extensive washing removes this impurity (see Figure S13 below), but also decreases the colloidal stability of the QDs and leads to the formation of aggregates. S9

10 Figure S13. XRD patterns of fast CE CIS QDs prior to and after ZnS shell overgrowth. The blue line indicates the wurtzite CIS diffraction pattern (JCPDS Card ). The XRD pattern was obtained on Bruker D2 Phaser, equipped with a Co Kα X-ray source ( Å). The samples are the same as shown in Figure S12(b) above, but were washed more extensively (6 times) to remove excess precursors. Although extensive washing successfully removes the crystalline impurity observed in Figure S12, it also decreases the colloidal stability of the QDs and leads to the formation of aggregates. S10

11 Figure S14. (a) EDS spectrum of 6 nm CIS QDs prepared by slow CE (sample was drop-cast on a carboncoated 300 mesh Ni grid). The elemental ratios are Cu:In:S= 1.4±0.2: 1±0.14: 2.6±0.5. (b) EDS spectrum of CIS/ZnS QDs obtained by ZnS overgrowth on the CIS NCs shown in (a) (sample was drop-cast on a carbon-coated 300 mesh Al grid). The elemental ratios are Cu:In:Zn:S= 1.3±0.3: 1±0.15: 0.6±0.2: 5.6±0.8. TEM images of the same samples are shown in Figure 3 of the main text. EDS analyses were performed on wide areas (~ nm 2 ), encompassing several hundreds of NCs, and were repeated for three different areas of the TEM grid. The elemental ratios were obtained by averaging over the different measurement spots. S11

12 Figure S15. (a) EDS spectrum of CIS QDs prepared by fast CE. The elemental ratios are Cu:In:S= 1.8±0.4: 1±0.2: 2.7±0.5. (b) EDS spectrum of CIS/ZnS QDs obtained by ZnS overgrowth on the CIS NCs shown in (a). Samples were drop-cast on carbon-coated 300 mesh Al grids. The elemental ratios are Cu:In:Zn:S= 1.3±0.3: 1±0.2: 0.4±0.2: 3.5±0.8. TEM images of the same samples are shown in Figure 3 of the main text. EDS analyses were performed on wide areas (~ nm 2 ), encompassing several hundreds of NCs, and were repeated for three different areas of the TEM grid. The elemental ratios were obtained by averaging over the different measurement spots. S12

13 Figure S16. Experimental band gap estimate of (a) 6.3 nm slow CE CIS QDs and (b) 6.4 nm fast CE CIS QDs before (black lines) and after (red lines) ZnS shell overgrowth using a Tauc plot. The absorption coefficient of wurtzite CuInS 2 is not well known, so an approximation of the Tauc relationship for direct band-gap semiconductors was used: 1 ( h ) 2 = [hν*(1-abs) 2 /2Abs] 2. [1] X. Lu, Z. Zhuang, Q. Peng, Y. Li, Chem. Commun. 2011, 47, Figure S17. Absorption and PL spectra of CIS QDs obtained by slow CE before and after ZnS shell growth. S13

14 Figure S18. (a) TEM image and corresponding size distribution histogram of 6.5 nm CIS QDs synthesized by slow CE. (b) Absorption and PL spectra of 6.5 nm CIS QDs obtained by slow CE before (black line) and after (red line) ZnS shell growth. (c) TEM image and corresponding size distribution histogram of 7.0 nm CIS QDs synthesized by slow CE. (d) Absorption and PL spectra of 7.0 nm CIS QDs obtained by slow CE before (black line) and after (red line) ZnS shell growth. The size distribution histograms were constructed by measuring over 100 NCs per image and are fitted to Gaussian functions. Table S2. Shift observed in the experimental band gaps and PL maxima of WZ CIS/ZnS QDs with respect to those of the WZ CIS QDs used as cores. The bandgaps were estimated from the absorption spectra by using a Tauc plot (see Figure S16 above for examples). CIS Core size (nm) Synthesis Method Cores Band Gap Shift (mev) PL Shift (mev) Abs. Spectra PL Spectra PLQYs of CIS cores (%) PLQYs of CIS/ZnS (%) 5.4 Fast CE Fig. 4b Fig. 4b 22 ± 1 75 ± Fast CE Not shown Not shown 10.6 ± ± Fast CE Not shown Not shown 3.4 ± ± Fast CE Not shown Not shown 0.31 ± ± Slow CE Fig. S17 Fig. S ± ± Slow CE Fig. S18b Fig. S18b 0.24 ± ± Slow CE Fig. S18d Fig. S18d 0.22 ± ± 2 S14

15 Figure S19. (a) PL decay curves of different sizes (5.4 nm and 6.1 nm) of WZ CIS QDs synthesized via fast CE. (b) PL Decay curves of 5.4 nm WZ CIS QDs before and after ZnS shell overgrowth. The PL decay constants are not significantly different from those previously reported in the literature for both CP and WZ CIS QDs (reference numbers refer to the main text): [23] 3 nm diameter CP CIS QDs: 6.5 and 190 ns. After ZnS shelling: 500 ns (PL at 650 nm) [31] 2.5 nm diameter CP CIS QDs: 25 and 207 ns. After ZnS shelling (1 monolayer): 201 and 360 ns (PL at 660 nm) [30] ~10 nm diameter WZ CIS platelets: 30 and 272 ns. After ZnS shelling (partial alloying): 42 and 529 ns at 950 nm. [28] 2.5 nm diameter WZ CIS QDs: 26 ns and 205 ns at 665 nm and 42 ns and 500 ns at 900 nm. [28] 4 nm diameter WZ CIS QDs: 68 ns and 408 ns at 870 nm. [42] 2.7 nm CP CIS-ZnS gradient alloy QDs: 200 ns S15

16 Figure S20. (a) FTIR spectra of CIS/ZnS core/shell QDs before (black line) and after (red line) exchange of the native ligands by deprotonated MUA. (b) Hydrodynamic size (14.2 ± 2.7 nm) and Zeta potential (-60 mv) of water-soluble CIS/ZnS core/shell QDs. The inserted digital image shows hydrophilic CIS/ZnS core/shell QDs in 1 ml deionized H 2 O. Figure S21. TEM images of CIS/ZnS core/shell QDs before (a) and after (b) exchange of the native ligands by deprotonated MUA. S16

17 Figure S22. (a) TEM images of CIS/ZnS core/shell QDs (9.2 nm) before (left) and after (right) exchange of the native ligands by deprotonated MUA. The MUA capped CIS/ZnS QDs were not washed prior to the TEM imaging. (b) Absorption and PL spectra of 9.2 nm CIS/ZnS core/shell QDs (core diameter: 8.1 nm) before and after phase transfer into water. S17